ISPS2009 Proceedings PDF

International Symposium on Performance Science The Author 2009, Published by the AEC
ISBN 978-94-90306-01-4 All rights reserved
Proceedings of
ISPS 2009
002 WWW.PERFORMANCESCIENCE.ORG
INTERNATIONAL SYMPOSIUM ON PERFORMANCE SCIENCE 003
Proceedings of the
International Symposium on
Performance Science 2009
Edited by
AARON WILLIAMON
Royal College of Music, London
SHARMAN PRETTY
The University of Melbourne
and
RALPH BUCK
The University of Auckland
AEC, PO Box 805, NL-3500 AV Utrecht, The Netherlands
Published worldwide in The Netherlands by the

Association Europenne des Conservatoires, Acadmies de Musique et Musikhochschulen (AEC)
The AEC is a European network of institutions in higher music education

www.aecinfo.org
Copyright AEC 2009

First published 2009
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or
transmitted, in any form or by any means, without the prior permission in writing of the publisher, or
as expressly permitted by law, or under terms agreed with the appropriate reprographics rights
organization. Enquiries concerning reproduction outside the scope of the above should be sent to
Aaron Williamon, Centre for Performance Science, Royal College of Music, Prince Consort Road,
London SW7 2BS, United Kingdom.
Disclaimer. Statements of fact and opinion in the articles in the Proceedings of ISPS 2009 are those
of the respective authors and contributors and not of the editors or the AEC. Neither the AEC nor the
editors make any representation, express or implied, in respect of the accuracy of the material in this
volume and cannot accept any legal responsibility or liability for any errors or omissions that may be
made. The reader should make her or his own evaluation as to the appropriateness or otherwise of
any experimental technique described.
ISBN 978-94-90306-01-4
Cover design by Christopher Tomlin

Printed by GEON (www.geongroup.com)
in New Zealand
Preface and acknowledgments
The second International Symposium on Performance Science, ISPS 2009,

explored the theme Performing Excellence from interdisciplinary perspec-
tives across the arts, as well as the natural, social, and applied sciences.
Hosted by the National Institute of Creative Arts and Industries at The Uni-
versity of Auckland, the conference brought together delegates from over 30
countries to debate and to experience performing excellence through a wide
array of theoretical, practical, and methodological perspectives.
This proceedings volume represents the lively discussion that ensued.
Covering such topics as motivation and the development of expertise, the
psychology and physicality of performance, performers health, and the per-
ception, analysis, and evaluation of performance, the 100 articles contained
herein offer a glimpse into a growing and vibrant field.
We are grateful to several individuals and organizations whose generous
support and contributions have made ISPS 2009 possible. Firstly, we wish to
thank the many artists and scientists who have contributed their valuable
work to both the conference and to this volume. We are also deeply indebted
to The University of Auckland, Royal College of Music, and members of the
Scientific and Organizing Committees for providing unwavering support for
this event, from the early planning stages onwards. We are delighted to ac-
knowledge the following partners of ISPS 2009: the Asia-Pacific Society for
the Cognitive Sciences of Music (APSCOM), British Council, European Asso-
ciation of Conservatoires (AEC), International Association for Dance Medi-
cine and Science (IADMS), International Society for Music Education
(ISME), Performing Arts Medicine Association (PAMA), and Society for the
Psychology of Aesthetics, Creativity, and the Arts. Finally, we would like to
acknowledge the following individuals, who have given unreservedly of their
time and ingenuity in putting this event together: Claire Speedy and Steve
Burns at The University of Auckland for their tireless work in bringing this
event together, and Rosie Burt-Perkins and Terry Clark at the Royal College
of Music for invaluable help in editing and revising the proceedings manu-
script.
Aaron Williamon
Sharman Pretty
Ralph Buck
Contents
Scientific committee page ix
TUESDAY, 15 DECEMBER 2009
Keynote paper
Performance, science, and society page 5
Robert Winston
WEDNESDAY, 16 DECEMBER 2009
Keynote paper
Discovering deliberate practice activities that overcome
plateaus and limits on improvement of performance page 11
K. Anders Ericsson
Symposium
Physical and psychological vulnerabilities
in music and dance students page 23
Thematic sessions
The vocabulary of performance page 43
Performance analysis page 63
Symposium
Performance science: Implications for
educational and professional practice page 83
Thematic sessions
Expression and interpretation I page 105
Expertise development page 125
Expression and interpretation II page 145
Memory and performance page 165
Graduate award paper
Fast feedforward error-detection mechanisms in
highly skilled performance page 187
Mara Herrojo Ruiz
THURSDAY, 17 DECEMBER 2009
Keynote paper
A balanced approach to excellence: Life skill
intervention and elite performance page 203
Deidre Anderson
Poster session page 215
Symposium
Musicians dystonia: New aspects in
pathophysiology and treatment page 373
Thematic sessions
Understanding performance page 401
The perception of technique page 421
Emotion in performance page 437
Workshops page 457
FRIDAY, 18 DECEMBER 2009
Symposium
Student musicians motivation, learning, and performance page 467
Thematic sessions
Performers health page 493
Performance and life factors page 513
Physicality of performance page 529
Defining performance page 545
Psychology of performance page 565
Perceiving performance page 581
Keynote paper
The dominant artistic discourse as a health determinant page 599
Sylvie Fortin
Author index page 611

Scientific committee
Aaron Williamon, co-chair

Royal College of Music, London (UK)
Sharman Pretty, co-chair
The University of Melbourne (Australia)
Ralph Buck, co-chair
The University of Auckland (New Zealand)
Mayumi Adachi
Hokkaido University (Japan)
Eckart Altenmller
Hanover University of Music and Drama (Germany)
Emmanuel Bigand
University of Burgundy (France)
Jane Boston
Royal Academy of Dramatic Arts (UK)
Daniela Coimbra
ESMAE, Porto Polytechnic Institute (Portugal)
Hubert Eiholzer
Conservatory of Italian Switzerland (Switzerland)
Dianna Kenny
The University of Sydney (Australia)
Alison McGregor
Imperial College London (UK)
Caroline Palmer
McGill University (Canada)
Miikka Peltomaa
Finnish Musicians Medicine Association (Finland)
Emma Redding
Trinity Laban Conservatoire of Music and Dance (UK)
Dale Speedy
The University of Auckland (New Zealand)
Tuesday
15 December 2009
Keynote paper
Performance, science, and society
Robert Winston
Imperial College London, UK
The evaluation of music and musicality is of growing interest to scien-

tists. Various technologies, particularly functional brain imaging, have
increased our understanding of the perception of music, pitch, and
rhythm, and research into the physiology of performance and the physics
of acoustics are now important academic disciplines. There is now
awareness that inspiration, intuition, and emotion are as important to
the exploration of neuroscience as they are to composing and perform-
ing. Music is an art that gets closest to the basis of our humanity and
opens windows of perception in unique and varied ways. This is why mu-
sic is so important in human experiencefrom Shostakovich reflecting
on death in intensive care at the end of his Fifteenth Symphony, to the
Prince ludicrously elevated from his melancholy in Prokovievs March of
the Love for Three Oranges. And this aspect of music, its ability to ma-
nipulate our mind, has been used in various waysfrom musak in su-
permarkets, to the pipes at Culloden or the drums and trumpets of
soldiers marching into battle, from Furtwangler conducting Wagner in
front of the Nazi faithful, to Puccini at a football match. This paper ex-
amines some of the relationships between music, science, and society.
Keywords: brain imaging; physiology; physics; musical perception; in-

spiration
Address for correspondence
Robert Winston, Imperial College London, Faculty Building, South Kensington Campus,
London SW7 2AZ, UK; Email: r.winston@imperial.ac.uk
Wednesday
16 December 2009
Keynote paper
Discovering deliberate practice activities that

overcome plateaus and limits on improvement
of performance
K. Anders Ericsson
Department of Psychology, Florida State University, USA
Since Sir Francis Galtons book on Hereditary Genius, many scientists

have argued that heritable factors set limits of performance and only al-
low a select few individuals to attain exceptional levels. However, recent
research rejects the associated learning theory and its implied perform-
ance plateaus and shows that expert performance is mediated by ac-
quired complex cognitive mechanisms. It describes different types of
deliberate practice activities that develop and refine mental representa-
tions, which in turn permit attained performance to exceed performance
resulting from extensive experience only. Empirical investigations are re-
viewed to show that expert performance and outstanding achievements
will be primarily constrained by individuals engagement in deliberate
practice and the quality of the available training resources.
Keywords: deliberate practice; expert performance; innate talent; cogni-

tive mechanisms; skill acquisition
Everyone has experienced the excitement associated with being introduced to

a new activity and rapidly reaching an acceptable level of performance. In the
beginning there are often dramatic improvements within the first few hours
of introductory engagement in popular games, such as darts, volleyball, and
shuffle board. Sir Francis Galton (1869/1979) summarized the popular view
that performance improvements are rapid only in the start of initial training
and that subsequent increases become increasingly smaller, until the per-
formance reaches a plateau and [M]maximal performance becomes a rigidly
determinate quantity (p. 15). According to this view, heritable capacities and
innate talents set the upper bounds for an individuals physical and mental
achievements. Once the performance has become immediate (automatic),
then it does not seem possible that any additional amount of practice can
increase the performance further. This assumed inability to influence per-

formance through any type of practice or training provided the foundation for
Galtons compelling arguments that individual differences in performance
must be determined by stable unmodifiable factors, such as individual differ-
ences in innate endowment.
In the nineteenth and early-twentieth centuries, most scientists assumed
an important boundary between mind and body. Consequently, when a per-
sons performance was reported to be automatic, it no longer would reflect
conscious thinking, and therefore, the speed of the execution had to be lim-
ited by physical and biological characteristics of ones body and nervous sys-
tem. In his pioneering book, Hereditary Genius (1869/1979), Galton
presented evidence that height and body size were genetically determined
and, thus, could not be altered by practice. In direct analogy he proposed that
similar genetic mechanisms must determine all other aspects of ones physi-
ology, such as size of brain and speed of neurons and, therefore by inference,
all mental capacities.
The most important contemporary theories of skill acquisition (Anderson
1982, Fitts and Posner 1967) are consistent with Galtons general assump-
tions and fit with casual observations on the development of everyday activi-
ties. After being introduced to activities such as driving a car, typing, or
playing golf, an individuals primary goal is to reach an acceptable level of
performance. During the first phase of learning (Fitts and Posner 1967), nov-
ices try to understand the activity and concentrate on completing their at-
tempts successfully, as is illustrated by the cognitive phase (black segment) of
the lower arm of Figure 1.
Figure 1. An illustration of the qualitative difference between the course of improve-

ment of expert performance and of everyday activities. (Adapted from Ericsson 1998, p.
90. European Council for High Ability, used by permission.)
With more experience in the associative phase (grey segment in Figure 1),
large mistakes become increasingly rare, performance appears smoother, and
learners no longer need to concentrate as hard to perform the task. After a
limited period of training and experiencefrequently less than 50 hours for
most recreational activitiesan acceptable standard of performance is at-
tained. Once performance is autonomous (white segment in Figure 1), indi-
viduals no longer attempt further modifications and improvements, and this
typically leads to a stable plateau of performance, which is consistent with
Galtons (1869/1979) description. Galton, however, would argue that these
stable plateaus exist not only for the acquisition of acceptable performance in
casual, recreational activities, but also for the acquisition of exceptional abil-
ity in most domains.
This view of acquisition of skill is consistent with findings for amateurs
performance in tennis and golf, where they improve initially but then stay at a
stable of performance level for decades of regular engagement. Performance
in everyday skills, such as driving and typing (Keith and Ericsson 2007), also
are remarkably stable across decades of regular activity. Finally, professional
performance does not improve with experience beyond the first few years of
initial experience (Ericsson 2004, Ericsson and Lehmann 1996, Ericsson et
al. 2007). In fact, sometimes job performance decreases as function of job
experience since graduation (Choudhry et al. 2005).
MAIN CONTRIBUTION
In most domains of professional and leisure activity, the majority of people

reach stable levels of performance (plateaus) that are maintained for years
and decades. These attained levels of performance are not rigidly limited by
some upper bound. In their review, Ericsson et al. (1993) found many docu-
mented examples of individuals who, for requirements for promotion in the
job, dramatically increased their performance by trainingsuch as typing
speed. In leisure activities, it is not uncommon that an adult starts taking
lessons with a golf or tennis coach and, after a period of training, improves
their performance. In the same review, Ericsson et al. (1993) proposed that
for many different activities there existed deliberate practice activities that
could change aspects of performance and over time increase the overall per-
formance on tasks representative of the activity, such as time to type an un-
familiar paragraph, run 100 meters, or win chess games.
Deliberate practice differs from the mere experience of doing the task in
many different ways. Perhaps the most striking way concerns the mental at-
titude of the individual. During deliberate practice the individual has the goal
of improving some measurable aspect of their performance. For example, a

recreational golfer aims their putt toward the hole on the green and either
misses or drops the ball in the hole. Every time the golfer putts the ball, it is a
different situation and the golfer would not know whether the mistake was
caused by their putting technique, the slope of the green, the resistance of the
grass, or whatever. During deliberate practice on a practice green, the golfer
has the opportunity to make the same putt many times (Ericsson 2001). If the
golfer closes her eyes and strokes 10 putts without seeing the results, the balls
will not have the exact same trajectory and will not stop at the exact same
spot on the green. In fact, the balls will form a cluster and the diameter of this
cluster will be a very good predictor of golf putting performance. Elite golf
players will have a tighter cluster than less skilled recreational players, but
every player will show considerable variability. Hence, even world-class play-
ers can never be certain that they would sink putts over 10 feet. In fact, all
that they can control is that the putt will stop within a circle near the hole.
Several of the best golf players in twentieth century reported that realizing
that they could not control whether a given putt would drop in the hole and
therefore should not be upset with themselves as long as the putt stopped
near the hole in the circle (defined by their reproducible accuracy). Based on
this analysis, it is clear that training ones putting stroke, so it can be con-
trolled to give the same/similar result for putts of different lengths is an as-
pect that can be much better developed on the practice green by repeated
shots and systematically varied putts. Similarly, learning to read the varied
shape of the green with hills and valleys so one can image the path of
ones putt is also a skill that can be improved more effectively by immediate
feedback and opportunities for repeated shots to explore the consequences of
differences in speed over inclined planes and slopes. In a similar manner,
golfers learn to better plan the ball trajectories of drives and other longer
shots by getting repeated opportunities with immediate feedback.
Deliberate practice can be focused on those aspects of the game that are
weakest and have the most room for improvement. During typical everyday
experience in the domain, the probability that a golfer would encounter a
sequence of sand trap shots in a given round of golf is small, but during prac-
tice the coach could work on sand trap shots for a full hour. Likewise, a delib-
erate practice session can be designed to focus on any particular aspect of
someones performance.
Identifying the aspects that should be the focus of deliberate practice re-
quires some assessment, typically by coach or teacher. It is possible to design
tasks that the individual performs several times that allow the teacher to bet-
ter assess current problems or anticipated future issues with some parts of
performance. In everyday life, beginners are searching for quick fixes that
allow them to rapidly reach an acceptable performance. In contrast, teachers
are focused on future performance and on the student acquiring the appro-
priate fundamentals, so that the student can master more complex techniques
and reach higher levels of performance. In many domains, such as music,
ballet, and gymnastics, teachers over time have developed a curriculum where
there is consensus on the best ordering of techniques to be mastered in order
to reach the highest levels of performance (Bloom 1985). For many domains
of expertise, such as scientific research, writing, and art, there is currently
less consensus on the preferred curriculum and even if a curriculum would
benefit the development of creative performance.
Deliberate practice should be scheduled when the performers are rested
and maximally alert, and the duration of training must be adapted so they
avoid fatigue and thus are able to maintain their highest level of concentra-
tion and performance during the entire training session.
Deliberate practice requires a close connection between the actual per-
formance and the training tasks. It is essential that some aspect of current
performance is taken as the focus and that training tasks are intermittently
exchanged for the real-world context so that transfer of improvements during
training is successfully incorporated in the corresponding real-world per-
formance. It is possible that the training tasks lead to performance improve-
ments that depend on the crutches and the scaffolding during training and
thus cannot be connected to the aspects of the real-world performance that
were the original stimuli for the design of the practice activities.
In the rest of the paper I will discuss how these ideas have allowed re-
searchers to identify the deliberate practice activities that have a high corre-
lation with attained performance in numerous domains (Ericsson 2006).
Overcoming plateaus in chess
If you were a chess player in a chess club in the 1950s and were able to beat
all the other players, could you, and if so how would you be able to improve
your performance? At that time there were no chess computers that played
chess at a world class level, and it was not that common that people travelled
to tournaments to play the best players in the state, country, or around the
world. Based on some informal interviews and analyses, Ericsson et al. (1993)
proposed that it was essential for effective learning that one encountered
challenging situations, where the probability of making mistakes and failure
was relatively high. We proposed that one method for doing this in chess
would be to buy books and chess magazines with published games between
international level and world-class chess players. The aspiring chess expert
would then simulate playing against these players by analyzing each position
in the chess game and by trying to find the best move for every position in the
game. If the chess player selected a move that matched the one picked by
chess master then the chess player did as well as the master. More impor-
tantly, if the chess players picked a different move from the master then there
is an opportunity for learning and improvement. Through further analysis,
the chess player could attempt to figure out why the chess masters move was
superior. The next step would be to think through how the chess players
processes involving search and move selection should be changed to be able
to find this move as well as similar moves in other chess positions. A few as-
piring chess experts reported spending 3-5 hours every day engaging in this
type of solitary analysis along with studies of variants of chess openings. Sub-
sequent studies have found that serious chess players spend as much as four
hours every day engaged in this type of solitary study (Charness et al. 2005).
Most important, these studies show that the accumulated hours of solitary
analysis of chess playing is the best predictor of someones chess skill. Some-
what surprisingly the amount chess playing with friends is not associated
with increased chess-playing skill. Furthermore, these studies show that the
size of someones chess library, that is the number of chess books and chess
magazines, is correlated with better chess skillperhaps because they were
necessary for solitary chess study before the emergence of chess playing com-
puter programs and internet data bases. By spending additional time analyz-
ing the consequences of moves for a chess position, players can increase the
quality of their selections of moves. With more study, individuals refine their
representations and can access or generate the same information faster. As a
result, chess masters can typically recognize a superior move virtually imme-
diately, whereas a competent club player requires much longer to find the
same move by successive planning and evaluation rather than recognition (de
Groot 1966). With additional time the master can often generate even better
moves.
In the classical model of skill acquisition (Fitts and Posner 1967), more
experience allows the person to generate the same move faster through auto-
mation. In contrast, the nature of the improvement in chess concerns the
ability to generate different and better moves based on refined acquired rep-
resentations and associated analysis and search (de Groot 1946/1978). The
same type of improvement, based on deliberate practice and the acquisition
of complex representations supporting planning, evaluation, and online
monitoring of performance (cf. Ericsson and Kintsch 1995) can explain grad-
ual and extended increases in performance in a wide range of domains, such

as billiards, golf, music, Scrabble, darts, and surgery (Ericsson 2006).
Deliberate practice in typing
Most adults are able to type, yet there are often large individual differences in
their style and efficiency. Adults typically do not think about their typing and
simply do it, thus typing would seem to meet the criterion of low effort or
even autonomous performance. A recent review (Keith and Ericsson 2007)
has found that estimates of how much someone has been typing appear to be
a poor predictor of measurable typing speed under standardized conditions.
How could someone improve their typing performance after having typed at a
similar speed for decades? The answer is clearly supported by a number of
training studies. The key to improved speed in typing is to find some time
during the day (Dvorak et al. 1936) when one is able to fully concentrate for
15-25 minutes. During that time, one finds something to copy and increases
ones comfortable typing speed to a speed 15-30% faster than normal. During
this faster speed it becomes apparent that some key strokes or key-stroke
combinations are slower and associated with hesitations. One then works on
mastering these weaknesses by focused practice and then by interweaving
them into speeded typing. In an interactive fashion, we speed up ones typing
to find the problems causing slower typing and then practice until the speed
of typing has increased and then repeat the process. The general approach of
this type of deliberate practice is to find methods to push performance be-
yond its normal speedeven if that performance can be maintained only for a
short time. This methodology offers the potential for identifying and correct-
ing weaker components that will improve performance.
How is it possible to improve the speed of habitual and effortless behav-
ior? Researchers have studied the individual differences in typing speed of
skilled typists and unskilled participants by having them type passages from a
collection of unfamiliar texts as fast as they can without making errors. High-
speed films of finger movements show that the faster typists start moving
their fingers toward their desired locations on the keyboard well before the
keys are struck. The superior typists speed advantage is linked to their per-
ceptual processing of the text beyond the word that they are currently typing
(Salthouse 1984). By looking ahead in the text to identify letters to be typed,
they can prepare future keystrokes in advance. This evidence for anticipation
has been confirmed by experimental studies where expert typists have been
restricted from looking ahead. Under such conditions their typing speed is
dramatically reduced and approaches the speed of less skilled typists.
In sum, the superior speed of reactions by expert performers, such as

typists and athletes, appears to depend primarily on cognitive representations
mediating skilled anticipation, rather than faster basic speed of their nervous
system (Abernethy 1991). For instance, expert tennis players are able to an-
ticipate where a tennis players shots will land even before the players rac-
quet has contacted the ball (Williams et al. 2002). Eye-movements of expert
tennis players show that they are able to pick up predictive information from
subtle, yet informative, motion cues, such as hip and shoulder rotation, com-
pared to their novice counterparts. They can also use later occurring and
more deterministic cues, such as racket swing, to confirm or reject their ear-
lier anticipations.
Deliberate practice in other domains
In many domains performers and teachers have accumulated knowledge

about effective methods of training and deliberate practice. They have devel-
oped curricula in a wide range of domains of expertise, such as music, ballet,
gymnastics, martial arts, and so on. This implies that at some point in history
one or more individuals must have discovered one of these methods for the
first time. There are many documented instances of such discoveries in
sports. For example, the famous long distance runners during the middle of
the twentieth century, such as Emil Zatopek, developed different variants of
interval training, where the runners alternate periods of fast running speed
much faster than their regular speed for endurance racesand slower speeds
or even just rest (Billat 2001). These runners were vastly superior to their
contemporaries, and it is likely that their superiority could be attributed to
their superior training, as more recent runners are able to clearly surpass
these classic runners based on the current running techniques that rely on
their training innovations. Similar training innovations are found in virtually
every domain of expertise.
In my early work with Bill Chase (Ericsson et al. 1980) on the acquisition
of exceptional memory, I was fortunate to observe how our first participant,
SF, encountered a plateau of performance. We were convinced that he had
reached his highest level of performance at digit span after not making any
improvements beyond lists with around 50 digits for over 10 hours of contin-
ued training. To test if he had reached an unmodifiable limit we presented
him much longer digits (around 60 digits), and he found that he could almost
recall these much longer lists. We also presented the digits at a somewhat
slower rate (at 1.5-2 s per digit instead of the standard 1 s per digit) to help
him test out better methods for encoding the digits. These tests convinced Bill
and me, and most importantly SF, that he could still improve his perform-
ance. SF later reached a digit span of 82 digits.
For the last few decades, I have been searching for documented instances
where motivated healthy individuals have reached unmodifiable plateaus that
constrain their performance in a given domain of expertise. So far I have
mostly encountered people who gave up their efforts because they did not
think that they could reach the performance of other people, rather than that
they had reached their own limits.
IMPLICATIONS
For a long time, the belief that individuals innate capacity limits their attain-
able level of performance has been accepted based on indirect and weak evi-
dence. In contrast, the theoretical framework of deliberate practice asserts
that improvement in performance of aspiring experts does not result from
automation due to further experience. By increasing the challenge of training,
individuals can remain in the cognitive phase (see the upper curve in Figure
1) and keep engaging in deliberate practice to acquire and refine complex
cognitive mechanisms that mediate how the brain and nervous system control
performance.
The time has come to seek out detailed information on performance pla-
teaus encountered by individuals motivated to improve. Scientists should
document their existence and examine their structure with experiments and
analyze the past training and current performance. This evidence on per-
formance limits will allow us to evaluate different theories of the determi-
nants of expert level performance, as well as motivate the development of new
training curricula and associated deliberate practice activities in order to as-
sess whether these plateaus reflect unmodifiable limits to the development of
performance.
Acknowledgments
I want to thank Len Hill for comments on an earlier draft of this paper.
K. Anders Ericsson, Department of Psychology, Florida State University, Campus Box

3064701, Tallahassee, Florida 32306, USA; Email: ericsson@psy.fsu.edu
References
Abernethy B. (1991). Visual search strategies and decision-making in sport. Interna-

tional Journal of Sport Psychology, 22, pp. 189-210.
Anderson J. R. (1982). Acquisition of cognitive skill. Psychological Review, 89, pp. 369-
406.
Billat L. V. (2001). Interval training for performance: A scientific and empirical practice.
Sports Medicine, 31, pp. 13-31.
Bloom B. S. (1985). Generalizations about talent development. In B. S. Bloom (ed.),
Developing Talent in Young People (pp. 507-549). New York: Ballantine Books.
Charness N., Tuffiash M. I., Krampe R., et al. (2005). The role of deliberate practice in
chess expertise. Applied Cognitive Psychology, 19, pp. 151-165.
Choudhrey N. K., Fletcher R. H., and Soumerai S. B. (2005). Systematic review: The
relationship between clinical experience and quality of health care. Annals of Inter-
nal Medicine, 142, pp. 260-273.
de Groot A. D. (1966). Perception and memory versus thought: Some old ideas and
recent findings. In B. Kleinmuntz (ed.), Problem Solving (pp. 19-50). New York;
John Wiley and Sons.
de Groot A. (1946/1978). Thought and Choice in Chess. The Hague, The Netherlands:
Mouton. (Original work published in 1946.)
Dvorak A., Merrick N. L., Dealey W. L., and Ford G. C. (1936). Typewriting Behavior.
New York: American Book Company.
Ericsson K. A. (1998). The scientific study of expert levels of performance: General im-
plications for optimal learning and creativity. High Ability Studies, 9, pp. 75-100.
Ericsson K. A. (2001). The path to expert golf performance: Insights from the masters
on how to improve performance by deliberate practice. In P. R. Thomas (ed.), Op-
timising Performance in Golf (pp. 1-57). Brisbane, Australia: Australian Academic
Press.
Ericsson K. A. (2004). Deliberate practice and the acquisition and maintenance of ex-
pert performance in medicine and related domains. Academic Medicine, 10, pp.
S70-S81.
Ericsson K. A. (2006). The influence of experience and deliberate practice on the devel-
opment of superior expert performance. In K. A. Ericsson, N. Charness, P. Fel-
tovich, and R. R. Hoffman (eds.), Cambridge Handbook of Expertise and Expert
Performance (pp. 685-706). Cambridge: Cambridge University Press.
Ericsson K. A., Chase W. G., and Faloon S. (1980). Acquisition of a memory skill. Sci-
ence, 208, pp. 1181-1182.
Ericsson K. A. and Kintsch W. (1995). Long-term working memory. Psychological Re-
view, 102, pp. 211-245.
Ericsson K. A., Krampe R. T., and Tesch-Rmer C. (1993). The role of deliberate practice
in the acquisition of expert performance. Psychological Review, 100, pp. 363-406.
Ericsson K. A. and Lehmann A. C. (1996). Expert and exceptional performance: Evi-
dence on maximal adaptations on task constraints. Annual Review of Psychology,
47, pp. 273-305.
Ericsson K. A., Whyte J., and Ward P. (2007). Expert performance in nursing: Review-
ing research on expertise in nursing within the framework of the expert-perform-
ance approach. Advances in Nursing Science, 30, pp. E58-E71.
Fitts P. and Posner M. I. (1967). Human Performance. Belmont, California, USA:
Brooks/Cole.
Galton F. Sir (1869/1979). Hereditary Genius. London: Julian Friedman Publishers.
(Original work published in 1869.)
Keith N. and Ericsson K. A. (2007). A deliberate practice account of typing proficiency
in everyday typists. Journal of Experimental Psychology: Applied, 13, pp. 135-145.
Salthouse T. A. (1984). Effects of age and skill in typing. Journal of Experimental Psy-
chology: General, 113, pp. 345-371.
Williams A. M., Ward P., Knowles J. M., and Smeeton N. J. (2002). Anticipation skill in
a real-world task: Measurement, training, and transfer in tennis. Journal of Ex-
perimental Psychology: Applied, 8, pp. 259-270.
Symposium:
Physical and psychological vulnerabilities
in music and dance students
Suffering for ones art: Performance related

musculoskeletal disorders in tertiary
performing arts students in music and dance
Dianna T. Kenny1,2, Justine Cormack2, and Rosemary Martin2
1 Australian Centre for Applied Research in Music Performance,

Faculty of Arts, University of Sydney, Australia
2 National Institute of Creative Arts and Industries,
University of Auckland, New Zealand
The study reports the results of an intake questionnaire assessing point

prevalence of performance related musculoskeletal disorders (PRMD)
and pain in 151 tertiary level performing arts students in music and
dance. Forty percent reported having received a medical diagnosis for a
condition related to playing their instrument or performing their dance
style. The most common diagnoses for music students were repetitive
sprains and strains such as tennis elbow, carpel tunnel syndrome, ten-
donitis, tenosynovitis, and muscle tightness, including temporo-man-
dibular joint syndrome, tight embouchure, muscle spasm, numbness,
and cramping. Sixty-four percent of the dance students had serious inju-
ries such as broken bones, joint dislocations, cruciate ligament and me-
niscus tears and strains, other muscle tears, and sprains/strains in
biceps, hamstrings, and ankles, tightening in hamstrings and hips, and
bunions. Only 27% of the total sample reported no current pain from a
PRMD; 11.3% reported suffering daily or almost daily from a PRMD.
Given the high injury rates among performing arts students, injury sur-
veillance systems may be a cost effective way to identify high risk injuries
in specific cohorts, and to apply appropriately specific management and
prevention strategies for musicians and dancers during their training in
order to assist in the development of sustainable careers.
Keywords: performance related musculoskeletal injury; musicians;

dancers; tertiary students; injury surveillance
Musicians and dancers (secondary, tertiary, and professional) are at high risk
of strain and injury in the execution of their art (Barton et al. 2008). Ap-
proximately half of professional musicians and music students experience
significant symptoms (Norris 1993, Zaza 1998) with a point prevalence of
performance related musculoskeletal disorders (PRMD) ranging from 39% to
87% in adult musicians and from 34% and 47% in secondary students (Zaza
1998). Musculoskeletal injury is also the most frequent medical problem
among classical and modern dancers (McBryde et al. 2007), with 97% of all
dancers surveyed sustaining injuries over an eight month period (Ostwald et
al. 1994). Overuse injuries account for 60-76% of all dance injuries (Bronner
et al. 2003). Potential risk factors associated with developing PRMD in musi-
cians include gender (females at greater risk; Barton et al. 2008), years of
playing, instrument played (string and keyboard players at greater risk;
Bruno et al. 2008), playing-related physical (long hours, over-practicing) and
psychological stressors (self-pressure/academic), lack of preventative well-
ness behaviors (taking breaks), and previous trauma (Wu 2007). This study
reports the results of a PRMD questionnaire to provide information regarding
the point prevalence of PRMD and pain experienced by tertiary level per-
forming arts students in music and dance as a basis for subsequent targeted
preventive action.
METHOD
Participants
The sample comprised 151 students, aged from 17-50 years, with a mean age
of 21.44 years (SD=5.9); 109 (72%) music and 42 (28%) dance students54
(36%) males and 97 (64%) femalesattending the National Institute of Crea-
tive Arts and Industries, University of Auckland, in March 2009. These num-
bers represented 48% (music) and 79.2% (dance) of the total cohort. Students
completed a comprehensive questionnaire at the commencement of the aca-
demic year.
Materials
Performance related musculoskeletal variables: Performance related mus-

culoskeletal disorders (PRMDs) were defined as any pain, weakness, numb-
ness, tingling or any other symptoms that interfere with your ability to play
your instrument/perform your dance routines at the level to which you are
accustomed. This definition does not include mild short-lived aches or pains.
The following variables were selected as the dependent measures of PRMDs:
PRMD frequency: assessed on a 10-point Likert scale (0=never, 10=daily)

PRMD severity (worst ever): assessed on a 10-point Likert scale (0=no
pain, and 10=worst imaginable)
Current pain or injury from any cause
PRMD duration of symptoms (in days): students indicated the duration of
a current PRMD if they had one
Pain severity for pain right now
Pain severity for pain at its most severe
Body Mass Index (BMI): BMI is a measure of body fat based on height and
weight that applies to adult men and women. People may be categorized into
four weight groups based on their BMI as follows: underweight: <18.5, nor-
mal weight: 18.5-24.9, overweight: 25.0-29.9, obese: 30.0. Students were
assessed for BMI to ascertain whether there were any differences in PRMD
and pain according to weight.
Procedure
Ethics approval for the study was granted by the University of Auckland Hu-
man Participants Ethics Committee. The study was introduced to students in
the first lectures of the semester through brief presentations to class groups
by the lecturers and researchers. Participation information sheets were dis-
tributed to all eligible students, and those who were interested were invited to
attend classes set aside for the completion of the surveys.
RESULTS
Demographics: There were 151 students [109 (72%) music and 42 (28%)
dance students; 54 (36%) males and 97 (64%) females; mean age=21.44 years
(SD=5.9), range 17-50 years]; 52 students (34%) were in their first year, 54
(36%) in second year, 27 (18%) in third year, 1 in fourth year (0.7%), and 17
(11%) were postgraduate students; piano students (24%), guitar (13%), violin
(9%), saxophone (9%), cello (6%), drums (6%), trumpet (6%), viola (6%),
flute (5%), and clarinet (5%). The remainder was studying double bass,
French horn, trombone, tuba, and bassoon. The majority of dance students
were studying contemporary (38%) or hip hop (29%), then jazz (14%), classi-
cal (7%), cultural (7%), tap (2%), and ballroom (2%). Of the total group, 63
(40%) reported having received a medical diagnosis for a condition related to
playing their instrument or performing their dance style. Of the 109 music
students, 37 (33%) reported one or more diagnoses from medical practitio-
ners that comprised mainly repetitive sprains and strains (tennis elbow, car-
Table 1. PRMD and pain factors and descriptive and F-statistics by group.
N Mean SD Max Min F Sig

PRMD frequency Music 109 2.79 2.84 0 10 2.997 0.086
Dance 42 3.71 3.21 0 10
Total 151 3.05 2.97 0 10
PRMD severity Music 109 3.00 2.60 0 8 1.583 0.210
Dance 42 3.62 2.97 0 10
Total 151 3.17 2.71 0 10
Pain severity now Music 48 2.79 2.09 0 8 0.374 0.543
Dance 27 3.11 2.31 0 8
Total 75 2.91 2.16 0 8
Pain most severe Music 48 6.25 2.10 2 9 5.852 0.018
Dance 27 7.33 1.33 4 10
Total 75 6.64 1.92 2 10
Pain duration of Music 48 388.92 472.77 1 1680 0.003 0.960
symptoms Dance 27 395.33 613.33 2 2184
Total 75 391.23 523.58 1 2184
pel tunnel syndrome, tendonitis, tenosynovitis, and muscle tightness,

including temporo-mandibular joint syndrome, tight embouchure, muscle
spasm, numbness, and cramping). Of the 42 dance students, 27 (64%) re-
ported medical diagnoses that included serious injuries such as broken bones,
joint dislocations, cruciate ligament and meniscus tears and strains, other
muscle tears, and sprains/strains in biceps, hamstrings, and ankles, tighten-
ing in hamstrings and hips, and bunions. Thirteen students (8.6%) reported
that they had undergone surgery that they believed had an impact on their
ability to play music or dance. Three nominated abdominal surgeries (appen-
dectomy, bowel, and caesarean section), one had surgery for the removal of a
cancerous tumor and the remainder nominated limb surgeries on broken
bones in arms and fingers, and knee reconstructions. On the 10 point scale for
PRMD frequency, where 10 represents daily occurrence, 27% (n=40) of stu-
dents reported no current PRMD; 23.8% rated their PRMD frequency as 5 or
higher, of whom 17 (11.3%) reported daily or almost daily occurrence of a
PRMD. Forty-four students (22.5%) rated their PRMD severity above 5, and
10 students (6.6%) rated their PRMD severity to be as close to or actually the
worst severity that they could imagine. Seventy-five students (50%; 48 (44%)
music students and 27 (64%) dance students) reported experiencing a current
Table 2. PRMD and pain factors and descriptive and F-statistics by gender.
N Mean SD Max Min F Sig

PRMD frequency Male 54 2.04 2.53 0 10 10.331 0.002
Female 97 3.61 3.05 0 10
Total 151 3.05 2.97 0 10
PRMD severity Male 54 2.20 2.31 0 10 11.443 0.001
Female 97 3.71 2.78 0 9
Total 151 3.17 2.71 0 10
Pain severity now Male 24 2.33 1.97 0 7 2.530 0.116
Female 51 3.18 2.22 0 8
Total 75 2.91 2.16 0 8
Pain most severe Male 24 5.67 2.08 2 9 10.177 0.002
Female 51 7.10 1.68 2 10
Total 75 6.64 1.88 2 10
Pain duration of Male 24 187.75 302.98 1 1008 5.667 0.020
symptoms Female 51 486.98 578.16 1 2184
Total 75 391.23 523.58 1 2184
pain, 11.3% of whom reported suffering daily or almost daily pain. Fifty-three
percent (n=40) said that their current pain was caused by playing their in-
strument/dancing. Seventy-two percent (72%; n=54) indicated that their
current pain had a negative impact on their ability to play their instru-
ment/dance.
Subgroup analyses: Subgroup analyses were conducted to ascertain
whether PRMD and pain patterns were related to group (music or dance),
gender, or body mass index (BMI<18.50 or >18.50). Dance students recorded
higher mean ratings than music students for the most severe pain ever ex-
perienced (see Table 1). With the exception of current pain severity, females
reported higher mean ratings on all the other PRMD and pain factors (see
Table 2). Fifteen percent (n=22) of students were underweight, 70% (n=104)
were of normal weight, 12% (n=18) were overweight, and 3% (n=5) were
obese. Chi square analysis indicated that there were no differences in the
proportions of underweight, overweight, or obese students by group or sex,
and no relationship between BMI and any PRMD or pain measure.
DISCUSSION
Findings confirmed high injury rates and PRMD prevalence of performing

arts students; dancers (64%) had double the rates of musicians (33%); 53%
reported that their current pain was caused by their instrument/dancing. No
significant differences were found between instrument groups or dance styles
in terms of PRMD and pain prevalence. Females reported higher rates of
PRMD and pain. BMI was not associated with either PRMD or pain. These
results reinforce the need for injury management programs and injury pre-
vention education to support sustainable careers in the performing arts.
Translational research, development, and implementation are needed in both
educational and professional contexts to address the unacceptably high
prevalence of PRMD in performing arts students (Wynn Parry 2004).
Dianna Kenny, University of Sydney, PO Box 170, Lidcombe, New South Wales 1825,
Australia; Email: d.kenny@usyd.edu.au
References
Barton R., Killian C., Bushee M. et al. (2008). Occupational performance issues and
predictors of dysfunction in college instrumentalists. Medical Problems of Per-
forming Artists, 23, pp. 72-78.
Bronner S., Ojofeitimi, S., and Rose D. (2003). Injuries in a modern dance company.
American Journal of Sports Medicine, 31, pp. 365-373.
Bruno S., Lorusso A., and LAbbate N. (2008). Playing-related disabling musculoskele-
tal disorders in young and adult classical piano students. International Archives of
Occupational and Environmental Health, 81, pp. 855-860.
McBryde A. M., Rodriguez R. F., and Dugas J. R. (2007). Dance injuries. Sports Medi-
cine Update, Sept/Oct, pp. 2-6.
Norris R. (1993). The Musicians Survival Manual. St. Louis, USA: MMB Music.
Wynn Parry C. B. (2004). Managing the physical demands of musical performance. In
A. Williamon (ed.), Musical Excellence (pp. 41-60). Oxford: Oxford Univ. Press.
Ostwald P. F., Baron B. C., Byi N. M., and Wilson F. R. (1994). Performing arts medi-
cine. Western Journal of Medicine, 160, pp. 48-52.
Wu S. J. (2007). Occupational risk factors for musculoskeletal disorders in musicians: A
systematic review. Medical Problems of Performing Artists, 22, pp. 43-51.
Zaza C. (1998) Playing-related musculoskeletal disorders in musicians: A systematic
review of incidence and prevalence. Canadian Medical Association Journal, 158,
pp. 1019-1025.
Practicing perfection: The physical costs of

practice in tertiary music and dance students
Dianna T. Kenny1,2, Rosemary Martin2, and Justine Cormack2

Despite the many physical demands involved in practice in music and

dance, little attention has been paid to the impact of practice on the mus-
culoskeletal system of young performers. We, therefore, assessed
whether the amount of daily practice and the practice and rehearsal rou-
tines of tertiary music and dance students were related to the frequency
and severity of reported performance related musculoskeletal disorders
(PRMD) in 109 music and 42 dance (36% males) from The National In-
stitute of Creative Arts and Industries, University of Auckland. Music
students practiced, on average, 156 minutes per day, compared with
107.5 minutes for dance students. Music students spent an average of 401
minutes and dance students spent an average of 369 minutes per week
rehearsing with others. Contrary to prediction, linear regression analyses
showed no relationship between PRMD frequency, severity or duration,
and any of the practice factors assessed. Although excessive practice is
frequently cited by performing artists as one of the contributors to
PRMD, this study did not confirm a relationship between practice and
PRMD in this sample. It is possible that the amount of practice (2.6
hours per day in music students and 1.8 hours for dancers) fell below the
threshold for a pain inducing practice period, particularly as these stu-
dents reported taking rest breaks after 48 (dancers) and 60 (music) min-
utes practice. Further research is needed to assess the relationship
between PRMD and practice.
Keywords: practice; practice routines; musicians; dancers; performance

related musculoskeletal disorders
Tertiary students in music and dance practice for many hours per day over
many years, often commencing in early childhood, to perfect their art, with
the hope that the years of practice will gain them a place in a leading orches-
tra or dance company. To achieve this goal, dancers and musicians place their
bodies under tremendous stress on a regular basis, with pain and injury often
being considered an occupational hazard (Brandfonbrener 2003, Bruno et al.
2008, Garrick and Lewis 2001, Hoppmann 2001, Hoppmann and Reid 1995,
Kelman 2000). However, despite the many physical demands involved in
practicing music and dance, little attention has been paid to the impact of
practice on the musculoskeletal system. Systematic reviews of literature
(Bragge et al. 2006, Wu 2007) have identified practice as a risk factor, citing
long hours and over-practicing as contributors to injury, in particular the
development of overuse syndromes. Hoppman and Patrone (1989) showed
that longer daily practice time was significantly related to development of
musculoskeletal problems, and Zetterberg et al. (1998) found that practice
hours were related to problems in the neck and upper extremity of musicians.
Bejjani et al. (1996) observed that overuse syndrome occurs in up to 50%
professional symphony orchestra musicians and 21% music students. Morse
et al. (2000) found that the relationship of hours played to pain is complex,
and may well be affected by risks from non-music jobs, which are very com-
mon among amateur musicians (p. 85). Musculoskeletal injury is the most
frequent medical problem among classical and modern dancers (Milan 1994,
Ostwald et al. 1994, McBryde et al. 2007). Ostwald et al. (1994) revealed that
over an eight month period, 97% of all dancers surveyed had sustained inju-
ries. Overuse injuries account for the majority (60-76%) of all dance injuries
and these are most likely to occur when fatigued or overworked (Bronner et
al. 2003). However, there are no studies assessing the relationship between
musculoskeletal disorders and practice in dance. This study assessed whether
amount of daily practice and practice and rehearsal routines of tertiary music
and dance students were related to frequency, severity, and duration of re-
ported PRMDs.
METHOD
Participants
The sample comprised music and dance students in all years who were at-
tending the National Institute of Creative Arts and Industries, University of
Auckland, in March 2009. The sample comprised 151 students, 109 (72%)
music and 42 (28%) dance students. These numbers represented 48% (mu-
sic) and 79.2% (dance) of the total cohort. There were 54 (36%) males and 97
(64%) females. They ranged in age from 17-50 years, with a mean age of 21.44
years (SD=5.9). Students completed a comprehensive questionnaire at the
commencement of the academic year.
Materials
A comprehensive questionnaire was developed to explore the relationship

between practice and performance related musculoskeletal disorders in this
population. The variables of interest in this paper are described below.
Practice variables: Six variables were selected as the dependent measures
of practice as follows: (1) days per week of practice, (2) hours/minutes prac-
tice in one day, (3) length of practice before taking a rest, (4) time in minutes,
weekly, rehearsing with others, (5) length of rest break before resuming prac-
tice, and (6) total practice, calculated using the algorithm: (days per week of
practice x minutes practice in one day x rehearsal time).
Performance related musculoskeletal variables: Performance related
musculoskeletal disorders (PRMDs) were defined as any pain, weakness,
numbness, tingling or any other symptoms that interfere with your ability to
play your instrument or dance at the level to which you are accustomed. This
definition does not include mild short-lived aches or pains. Three variables
were selected as the dependent measures of performance related muscu-
loskeletal disorders as follows: (1) PRMD frequency (n=151): rated on a 10-
point Likert scale (never=0, daily=10), (2) PRMD severity (most ever)
(n=151): rated on a 10-point Likert scale (no pain=0, worst imaginable
pain=10), and (3) PRMD duration of symptoms (in days) (n=75): students
rated the duration of a current PRMD if they had one.
Procedure
by lecturers and researchers. Participation information sheets were distrib-
uted to all eligible students and those who were interested were invited to
RESULTS
Descriptives, practice variables: Music students practiced, on average, 5.83

days per week (SD=1.23, range=2-7 days). Dance students practiced on aver-
age 3.64 days per week (SD=1.87, range 1-7 days). Music students practiced,
on average, 156 minutes (SD=73.67, range=45-360 mins) per day, compared

with 107.5 minutes (SD=71.1, range=30-300 mins) for dance students. This
difference was statistically significant (F1,149=13.42, p=0.001). Music students
reported practicing on average for 61.1 mins (SD=40.4, range=0-240 mins);
dance students practiced on average 47.9 minutes (SD=32.4, range=0-120
mins) before taking a break. This difference fell just short of significance
(F1,149=3.6, p=0.06). Music students reported that they took, on average, a 19
min rest (SD=29.8, range=0-240 mins) compared with dance students, who
broke for an average of five minutes (SD=4, range=0-15 mins) before resum-
ing practice. This difference was statistically significant (F1,149=9.1, p=0.003).
Music students spent an average of 401.2 mins per week (SD=308.2,
range=0-1800 mins) rehearsing with others; dance students spent an average
of 368.9 mins (SD=380.4, range=0-1800 mins) rehearsing with others. This
difference was not statistically significantly different (F1,149=0.287, p=0.59).
Data on total practice were available for 147 students. The mean total in-
volvement in some form of weekly practice was 1180 mins per week (19.67
hours) [SD=673.2 mins (11.2 hours), range=60-2880 mins (1- 48 hours)].
Descriptives, performance related musculoskeletal variables: Music stu-
dents average rating for PRMD frequency was 2.79 (SD=2.84, range=0-10),
PRMD severity was 3.0 (SD=2.60, range=0-8), and duration of symptoms for
the 48 reporting a current PRMD was 388.9 days (SD=472.8 days, range=0-
1680). For dance students, PRMD frequency was 3.8 (SD=3.2, range=0-10),
PRMD severity was 3.7 (SD=2.9; range=0-10), and duration of symptoms for
the 27 reporting a current PRMD was 401.6 days (SD=632.96 days, range=2-
2354).
Analyses: A series of linear regression analyses were conducted to assess
the association between practice and rehearsal routines and frequency, se-
verity, and duration of PRMDs. Because they were assessed on different
scales, the dependent measures were converted to Z scores prior to analysis.
The dependent variables were PRMD frequency, severity, and duration, and
the independent variables were minutes practiced per day, length of rest
breaks between practice sessions, amount of rehearsal time per week, and
total practice. As predicted, there was a significant relationship between the
amount of practice and length of rest breaks before resuming practice
(R2=0.56), with longer practice sessions tending to be followed by longer rest
breaks. The relationship between PRMD frequency and PRMD severity was
also significant and in the expected direction (R2=0.48, F1,149=136.7,
p=0.001). However, contrary to prediction, there was no relationship be-
tween PRMD frequency, severity, or duration, and any of the practice factors
assessed. For the dependent variable total practice, there was no relationship
between practice and any of the PRMD variables (R2=0.83, F1,149=1.26,

p=0.29). Inspection of the standardized Beta coefficients indicated that the
strongest relationship was between PRMD severity and total practice (=.26,
t=1.58, p=0.12). Because of the very great differences in amount of weekly
practice in this sample, two groups were created by median split into prac-
ticed less than 1080 minutes per week (52.4% of sample) and practiced
more than 1080 minutes per week (47.6% of sample). One way analysis of
variance (ANOVA) was conducted on these groups with PRMD duration, fre-
quency, and severity as the dependent variables. There were no significant
differences on any of these three variables between low and high practice
groups: PRMD duration (F1,73=0.33, p=0.07), frequency (F1,145=0.12, p=0.73),
and severity (F1,145=0.97, p=0.33). This analysis was run separately for musi-
cians and dancers, and again there were no significant differences between
high and low music or dance practicers and PRMD outcomes.
DISCUSSION
Contrary to prediction, there was no significant relationship between practice

and PRMD frequency, severity, or duration in this sample. It is likely that the
amount of practice fell below the threshold for a pain-inducing practice pe-
riod. For those students suffering PRMDs, practice may act as a risk factor or
contributor in exacerbating their injuries, although may not necessarily be the
root cause of their injury. A number of factors not assessed in this study may
contribute to the relationship between practice and PRMDs. These include
correct performance technique, technically correct practice, physical charac-
teristics, and level of fitness of participants and work history. Further re-
search is needed to assess the relationship between PRMD and practice.
References
Bejjani F. J., Kaye G. M., and Benham M. (1996). Musculoskeletal and neuromuscular
conditions of instrumental musicians. Archives of Physical Medicine and Reha-
bilitation, 77, pp. 406-413.
Bragge P., Bialocerkowski A., and McMeeken J. (2006). A systematic review of preva-
lence and risk factors associated with playing-related musculoskeletal disorders in
pianists. Occupational Medicine, 56, pp. 28-38.
Brandfonbrener A. G. (2003). Musculoskeletal problems of instrumental musicians.

Hand Clinics, 19, pp. 231-239.
Bronner S., Ojofeitimi, S., and Rose D. (2003). Injuries in a modern dance company.
Bruno S., Lorusso A., and LAbbate N. (2008). Playing-related disabling musculoskele-
tal disorders in young and adult classical piano students. International Archives of
Occupational and Environmental Health, 81, pp. 855-860.
Garrick J. G. and Lewis S. L. (2001). Career hazards for the dancer. Occupational Medi-
cine, 16, pp. 609-618.
Hoppmann R. A. (2001). Instrumental musicians hazards. Occupational Medicine, 16,
pp. 619-631.
Hoppmann R. A. and Patrone N. A. (1989). A review of musculoskeletal problems in
instrumental musicians. Seminars in Arthritis and Rheumatism Journal, 10, pp.
117-126.
Hoppmann R. A. and Reid R. R. (1995). Musculoskeletal problems of performing artists.
Current Opinion in Rheumatology, 7, pp. 147-150.
Kelman B. B. (2000). Occupational hazards in female ballet dancers. American Asso-
ciation of Occupational Health Nurses Journal, 48, pp. 430-434.
McBryde A. M., Rodriguez R. F., and Dugas J. R. (2007). Dance injuries. Sports Medi-
cine Update, Sept/Oct, pp. 2-6.
Milan K. (1994). Injury in ballet: A review of relevant topics for the physical therapist.
Journal of Orthopaedic Sports Physical Therapy, 19, pp. 121-129.
Morse T., Ro J., Cherniack M., and Pelletier S. R. (2000). A pilot population study of
musculoskeletal disorders in musicians. Medical Problems of Performing Artists,
15, pp. 81-85.
Ostwald P., Baron B., Bly N., and Wilson F. (1994). Performing arts medicine. Western
Journal of Medicine, 160, pp. 48-52.
Wu S. J. (2007). Occupational risk factors for musculoskeletal disorders in musicians: A
systematic review. Medical Problems of Performing Artists, 22, pp. 43-51.
Zetterberg C., Backlund H., Karlsson J., et al. (1998). Musculoskeletal problems among
male and female music students. Medical Problems of Performing Artists, 13, pp.
160-166.
The factor structure of the revised Kenny

Music Performance Anxiety Inventory
Dianna T. Kenny1,2

This study assessed the factor structure of the revised Kenny Music Per-
formance Anxiety Inventory (KMPAI) using a sample (n=151) of tertiary
music and dance students attending the National Institute of Creative
Arts and Industries, University of Auckland. The scale consists of 40
items that assess the factor structure of the revised, expanded KMPAI to
ascertain whether it captured the latent etiological factors identified by
emotion theory underlying performance anxiety. Students completed the
(revised) Kenny Music Performance Anxiety Inventory at the com-
mencement of the 2009 academic year. Principal component analysis
(with varimax rotation) of the KMPAI revealed three latent factors and 12
underlying factors, as follows: early relationship context comprising gen-
erational transmission of anxiety and parental empathy; psychological
vulnerability comprising controllability, depression, hopelessness, and
trust; and proximal performance concerns comprising somatic anxiety,
pre- and post-performance rumination, self/other scrutiny, performance
outcome concerns, memory reliability, and commitment to performance.
These results provide initial evidence of the complex structure of music
performance anxiety, particularly in its severe form, and indicate that
management and treatment of the condition will need to be tailored to
the individuals pattern of contributing causal features.
Keywords: music performance anxiety; Kenny Music Performance Anxi-

ety Inventory; factor structure; musicians; dancers
Current conceptualizations of music performance anxiety tend to focus on

state cognitive and somatic anxiety prior to or during performance and fail to
take account of the etiological complexity of the condition (Kenny and Os-
borne 2006). Kenny (2004, 2009) developed the Kenny Music Performance
Anxiety Inventory (KMPAI) to include the assessment of underlying psycho-
logical vulnerabilities according to Barlows emotion based theory of the eti-
ology of anxiety disorders, as well as pre-performance experiences, as a step
in aiding the development of a more comprehensive conceptualization of the
condition and to provide a more appropriate focus for comprehensive treat-
ments for performing artists suffering from performance anxiety. The aim of
this study was to assess the factor structure of the revised, expanded KMPAI
to ascertain whether it captured the latent etiological factors identified by
emotion theory (Barlow 2000).
METHOD
Participants
The sample comprised music and dance students in all years who were at-
tending the National Institute of Creative Arts and Industries, University of
Auckland, in March 2009. The sample comprised 151 students: 109 (72%)
music and 42 (28%) dance students. These numbers represented 48% (mu-
sic) and 79.2% (dance) of the total cohort. There were 54 (36%) males and 97
(64%) females. They ranged in age from 17-50 years, with a mean age of 21.44
years (SD=5.9).
Materials
Students completed the revised Kenny Music Performance Anxiety Inventory

(KMPAI, Kenny 2009) as part of a comprehensive survey of at the com-
mencement of the academic year. This is a 40 item version of the earlier in-
ventory (see Table 1).
Procedure
by lecturers and researchers. Participation information sheets were distrib-
uted to all eligible students and those who were interested were invited to
Table 1. Rotated factor structure of the revised 40 item Kenny Music Performance Anxi-
ety Inventory.
Factor Factor loading

1. Depression/hopelessness (psychological vulnerability)
I often feel that I am not worth much as a person 0.665
Sometimes I feel depressed without knowing why 0.646
I often feel that I have nothing to look forward to 0.602
I often feel that life has not much to offer me 0.542
I often find it difficult to work up the energy to do things 0.474
Sometimes I feel anxious for no particular reason 0.460
I worry that one bad performance may ruin my career 0.426
I am often concerned about a negative reaction from the audience 0.335
I find it easy to trust others (-)* 0.332
2. Worry/dread (negative cognitions)
Thinking about the evaluation I may get interferes with my 0.630
performance
During a performance I find myself thinking about whether Ill get 0.613
through it
I often prepare for a concert with a sense of dread and impending 0.600
disaster
Even in the most stressful performance situations, I am confident 0.586
that I will perform well (-)*
My worry and nervousness about my performance interferes with 0.540
my focus and concentration
Even if I work hard in preparation for a performance, I am likely to 0.417
make mistakes
3. Proximal somatic anxiety
Prior to, or during a performance, I experience increased heart rate 0.761
like pounding in my chest.
Prior to, or during a performance, I experience shaking or trembling 0.609
or tremor
Prior to, or during a performance, I feel sick or faint or have a 0.582
churning in my stomach
Prior to, or during a performance, I get feelings akin to panic 0.573
Prior to, or during a performance, I have increased muscle tension 0.425
Prior to, or during a performance, I experience dry mouth 0.411
I remain committed to performing even though it terrifies me 0.399
Note. (-)* indicates reverse scored items.
Table 1 (cont.)
Factor Factor loading

4. Parental empathy
My parents were mostly responsive to my needs (-)* 0.836
My parents always listened to me (-)* 0.704
My parents encouraged me to try new things (-)* 0.660
5. Memory
When performing without music, my memory is reliable (-)* 0.901
I am confident playing from memory (-)* 0.802
6. Pre- and post-performance rumination
After the performance, I replay it in my mind over and over 0.619
I worry so much before a performance, I cannot sleep 0.524
7. Generational transmission of anxiety
One or both of my parents were overly anxious 0.638
Excessive worrying is a characteristic of my family 0.624
As a child, I often felt sad 0.431
8. Self/other scrutiny
I am concerned about my own judgment of how well I will perform 0.638
After the performance, I worry about whether I played well enough 0.413
I am concerned about being scrutinized by others 0.323
9. Controllability
I generally feel in control of my life (-)* 0.609
I never know before a concert whether I will perform well 0.478
10. Opportunity cost
I give up worthwhile performance opportunities due to anxiety 0.534
11. Trust
I find it difficult to depend on others 0.641
12. Pervasive anxiety
From early in my music studies, I remember being anxious about 0.515
performing
RESULTS
Principal axis factoring (with varimax rotation) of the KMPAI revealed 12

underlying factors, which can be subsumed under the following categories
(number in parentheses indicates subscale number as indicated in Table 1):
Early relationship context: (7) Generational transmission of anxiety, (4)

Parental empathy
Psychological vulnerability: (1) Depression/hopelessness, (9) Controlla-
bility, (11) Trust, (12) Pervasive performance anxiety
Proximal performance concerns: (3) Proximal somatic anxiety, (2)
Worry/dread (negative cognitions), (6) Pre- and post-performance rumi-
nation, (8) Self/other scrutiny, (10) Opportunity cost, (5) Memory reli-
ability
DISCUSSION
These results provide initial evidence of a complex structure for music per-
formance anxiety, particularly in its severe form, that is consistent with the
emotion based theory of the anxiety disorders. These data indicate that man-
agement and treatment of music performance anxiety will need to take ac-
count of multiple factors in its etiology and maintenance, with a broader
focus than proximal performance concerns. Results of this study will need to
be replicated on other samples and tested on professional musicians to cross-
validate the factor structure of the revised KMPAI beyond student popula-
tions.
References
Barlow H. (2000). Unraveling the mysteries of anxiety and its disorders from the per-
spective of emotion theory. American Psychologist, 55, pp. 1245-1263.
Kenny D. T. and Osborne M. S. (2006). Music performance anxiety: New insights from
young musicians. Advances in Cognitive Psychology, 2, pp. 103-112.
Kenny D. T. (2009). Kenny Music Performance Anxiety Inventory-Revised. Unpub-
lished manuscript.
Kenny D. T., Davis P. and Oates J. (2004). Music performance anxiety and occupational
stress amongst opera chorus artists and their relationship with state and trait anxi-
ety and perfectionism. Journal of Anxiety Disorders, 18, pp. 757-777.
Thematic session:
The vocabulary of performance
Linguistic limitations of describing sound: Is

talking about music like dancing about
architecture?
Helen F. Mitchell1 and Raymond A. R. MacDonald2
1 Sydney Conservatorium of Music, University of Sydney, Australia

2 Department of Psychology, Glasgow Caledonian University, UK
Verbalizing sound quality presents a challenge to musicians and peda-

gogues in describing a complex sensory phenomenon. Verbal descrip-
tions may only be effective when a performers sound translates easily
and completely into words. A verbal overshadowing (VO) effect may oc-
cur when a verbal description distorts the recall of the original aural
memory. The aim of this study is to examine the impact of verbal over-
shadowing in a music performance context. This exploratory pilot project
builds on VO research in other fields to assess the value and limitations
of language in describing performers sound quality. Outcomes will have
implications for musicians and pedagogues in the use of language in mu-
sic teaching and learning.
Keywords: music perception; verbalization; auditory recognition; music

performance; singing voice
Verbal overshadowing (VO) can occur when we use words to describe sensory
experiences (such as sight, taste, or hearing). Numerous studies have exam-
ined this phenomenon in face, taste, and sound recognition and report that
verbal description, or verbal encoding, impairs later recall of the respective
experience. Individuals were less likely to identify the target stimuli from a
line-up (Parr et al. 2002, Perfect et al. 2002, Schooler and Engstler-Schooler
1990). The perception and description of music performance presents a
similar sensory challenge. Can words adequately capture sound quality in a
music performance? And if so, which terms or group of terms are most effec-
tive in expressing sound quality?
The VO effect has been most apparent in face recognition, where wit-
nesses to crimes have to describe the perpetrator of a robbery in detail and
later select their face from a photographic line-up. The act of verbal descrip-
tion has negatively influenced individuals ability to recognize the face they
described in words (Schooler and Engstler-Schooler 1990). Empirical studies
have confirmed this VO effect in facial recognition by comparing verbal de-
scription groups with memory-only control groups. More recent studies have
demonstrated that there is a similar VO effect on listeners ability to recognize
spoken voices following verbal description while no-description control
groups were able to isolate the voice from a line-up (Perfect et al. 2002, Va-
nags et al. 2005). Interestingly, expert wine-tasters were more likely than
novices to recognize wine-relevant odors by their verbal labels (Parr et al.
2002) despite the fact all tasters had knowledge of language used to describe
wine. Novices may have lacked the ability to separate perceptual and linguis-
tic skills while experts superior perceptual skills protected them from lin-
guistic interference to taste recognition.
In music, the modality of the stimuli (aural) does not match the task (ver-
bal description), yet we rely on verbal descriptors to explain and communi-
cate our perception of sound quality in training and assessment. There is
growing interest in the way listeners perceive, rate, and describe music per-
formances. Like the witnesses above, listeners form a global impression of the
performance or performer and use a limited selection of verbal descriptors to
explain the reasons for their judgments. They focus on the more easily ver-
balized technical and performance components rather than describing overall
quality (Davidson and Coimbra 2001, Stanley et al. 2002). In fact, examiners
have even noted candidates dress and stage manner to facilitate later recall of
individual singers performances for discussion and assessment (Davidson
and Coimbra 2001). It is more natural to conceptualize sound as belonging to
a sound-producing object, rather than analyze the components that make up
the sound (Ekholm et al. 1998). The use of language may effectively hinder
our recall of performers and performances.
While there is a growing body of knowledge focused on categorizing ter-
minology used in describing vocal quality, there is still much to be learned
about verbal descriptors in listeners perception of music performance. Com-
munication may focus on the most easily verbalized characteristics in the
sound rather than individual composition of these characteristics that makes
a sound unique (Kenny and Mitchell 2006, Mitchell and Kenny 2006). Ver-
balization, rather than enhancing interpretation of sound qualities, may have
a disruptive effect on listeners recall and communication of music perform-
ance. The main objective of this project is to examine impact of verbal de-
scription on the recall of music performances and, specifically, singers. It
investigates whether verbal overshadowing (VO), or putting a sound into
words, distorts listeners memory and subsequent recall of the original per-
formance or performer. This paper presents work in progress, the full details
of which will be reported in subsequent publications.
METHOD
Participants
Singers and listeners were recruited from the staff and students at the Sydney
Conservatorium of Music (SCM).
Materials
Singers (n=6) performed two short song excerpts in a sound-treated studio at

SCM and recorded using a matched pair of stereo microphones (Neumann
KU100, ORTF configuration). Samples of their recordings were played to
listeners in the perceptual test.
Procedure
Listeners (n=24) were asked to attend a single listening session at SCM and
were informed that the purpose of the study was to investigate the effect of
visual and verbal tasks on the ability to recognize aural stimuli. Audio sam-
ples were played to listeners from a CD player via circum-aural closed-back
stereo monitoring headphones (Sennheiser HD 270) to ensure that each lis-
tener heard the same quality of sample. Listener participants were assigned to
either a verbal description group or a non-description control group.
Listeners were presented with a short sample of a single singer perform-
ing song 1 (for encoding) on CD. Singer presentation was randomized from
the six singers. Listeners were then asked to perform a visual maze task on
paper (a filler-task) for 10 mins. In the following 5 mins, the experimental
group was asked to write a detailed description of the voice they heard at the
start of the test, while the control group completed an anagram puzzle. Fi-
nally, both groups were presented a line-up of six voices singing a short ex-
cerpt from song 2 and were asked to identify the original voice they heard and
to rate their confidence in their response on a scale of 1-10 (1=not confident,
10=very confident).
RESULTS
Listeners responses were examined for correct voice identification by each

group (verbal description and non-description control). Ten of 24 listeners
Table 1. Correct and incorrect voice line-up identifications, and mean confidence scores
for the verbal description and non-description groups.
Verbal description Non-description

group control group
Correct identification 4 6
Incorrect identification 8 6
Mean score (SD) for confidence in choice 5.7 (2.1) 6.3 (2.2)
correctly identified their target voice from the line-up. Table 1 shows the dis-
tribution of responses by experimental group. Fifty percent of the control
group and 33% of the verbal description group correctly identified their target
voice.
Verbal descriptions
For the written descriptions of target voices, listeners were instructed to focus
on the characteristics of the voice singing the target song rather than the song
the singer sang. Most listeners identified that the singer was a soprano, al-
though some listeners debated whether their target singer might be a mezzo
soprano, with comments such as: slightly darker sound than for soprano
and voice in a lower timbre to that of a soprano. Eight listeners mentioned
their target singers use (or lack) of vibratofor instance, singer does not use
an abundance of vibrato, smooth even vibrato on long notes but none on
shorter notes, and vibrato enlarged at the end of phrase; vibrato change
may be for emotional purpose?.
Seven listeners mentioned the singers diction or articulation of language:
good diction, the articulation of the words was extremely clear, and when
singing, the words are clearly understandable. Listeners made general judg-
ments about the quality of their target voices without expressing the reasons
that motivated their statements: not very expressive at all, after thinking
about it; quite boring, not the style of music I listen to but singer is good,
quality of the instrument has potential, needs refinement of technique to
bring out the individual voice, but she is still young, and sings in an un-
emotional way (possibly she doesnt know/understand what shes saying).
Some listeners mentioned words usually associated with classical singing
voice, such as rich and smooth, bell-like, round, and covered while
others used technical descriptors like support and projection. For two
listeners, the sound they heard evoked a mental picture of the target voice
singing: English accent/Anglo-Saxon, blonde, quite tall and image evoked

based on sound: larger brunette femalesolid, fairly tall.
DISCUSSION
This is the first study to investigate the effect of VO in the perception of music
performers. Verbal description decreased listeners likelihood of identifying
the target voice but only showed a slight reduction in listeners confidence in
their identification choice. Listeners were less likely to pick the correct voice
but were not aware that verbalizing their perception of the target voice re-
duced their ability to achieve correct identification.
Listeners responses in this study follow a similar test of spoken voice rec-
ognition (Perfect et al. 2002), where only 50% of listeners in the non-verbal
group and 21.4% of the verbal description group correctly identified the target
voice from a line-up. This project will confirm if music listeners are similarly
susceptible to the effects of VO when they describe singing voices.
Listeners descriptions of the target voices illustrated the limits of lan-
guage to communicate music perception. Examination of the comments fur-
ther confirmed that vocal quality is difficult to articulate or itemize (Davidson
and Coimbra 2001, Ekholm et al. 1998). While verbalizing sound quality in
this context proved an unusual challenge, listeners still resorted to descrip-
tions of the song, or indeed of the person singing, rather than attempt to de-
scribe the idiosyncrasies of each voice.
These preliminary results form part of a more extensive study investigat-
ing the nature of VO effects in music listeners and the extent to which the
results from VO studies in other domains generalize to the musical context.
Moreover, this ongoing work is also highlighting the complexities and ambi-
guities in describing singing voices linguistically.
Acknowledgments
The first author was funded by an early career researcher grant from the Sydney Con-
servatorium of Music, University of Sydney. We wish to thank the singers and listeners
who took part in this study.
Helen Mitchell, Sydney Conservatorium of Music, Building C41, University of Sydney,

Sydney, New South Wales 2006, Australia; Email: h.mitchell@usyd.edu.au
References
Davidson J. W. and Coimbra D. (2001). Investigating performance evaluation by asses-

sors of singers in a music college setting. Musicae Scientiae, 5, pp. 33-53.
Ekholm E., Papagiannis G. C., and Chagnon F. P. (1998). Relating objective measure-
ments to expert evaluation of voice quality in Western classical singing: Critical per-
ceptual parameters. Journal of Voice, 12, pp. 182-196.
Kenny D. T. and Mitchell H. F. (2006). Acoustic and perceptual appraisal of vocal ges-
tures in the female classical voice. Journal of Voice, 20, pp. 55-70.
Mitchell H. F. and Kenny D. T. (2006). Can experts identify open throat technique as a
perceptual phenomenon? Musicae Scientiae, 10, pp. 33-58.
Parr W., Heatherbell D., and White K. G. (2002). Demystifying wine expertise: Olfac-
tory threshold, perceptual skill and semantic memory in expert and novice wine
judges. Chemical Senses, 27, pp. 747-755.
Perfect T. J., Hunt L. J., and Harris C. M. (2002). Verbal overshadowing in voice recog-
nition. Applied Cognitive Psychology, 16, pp. 973-980.
Schooler J. W. and Engstler-Schooler T. Y. (1990). Verbal overshadowing of visual
memories: Some things are better left unsaid. Cognitive Psychology, 22, pp. 36-71.
Stanley M., Brooker R., and Gilbert, R. (2002). Examiner perceptions of using criteria in
music performance assessment. Research Studies in Music Education, 18, pp. 43-
52.
Vanags T., Carroll M., and Perfect T. J. (2005). Verbal overshadowing: A sound theory
in voice recognition? Applied Cognitive Psychology, 19, pp. 1127-1144.
The influence of listeners singing experience

and the number of singers on the
understanding of sung text
Philip Fine1, Jane Ginsborg2, and Chris Barlow3
1 Department of Psychology, University of Buckingham, UK

2 Centre for Music Performance Research, Royal Northern College of Music, UK
3 Faculty of Technology, Southampton Solent University, UK
An important aspect of perceiving sung music is understanding the

words. Previous research has suggested several factors affecting the in-
telligibility of sung text. This study investigates two of those factors: the
number of singers and the singing expertise of the listener. We expected
more singers to cause greater variability in the acoustic signal and be
harder to comprehend. Listeners who are themselves experienced singers
are more likely to be attuned to factors affecting singers diction and were
expected to be better than non-singers at understanding the sung text.
Forty eight participants, half accomplished singers and half self-reported
non-singers, listened to four 8-bar unaccompanied songs twice (in order
to test for familiarity) and wrote out the texts as they heard them. Two
performances were given by a soloist, two by a trio of singers in unison.
Participants were significantly better at understanding the words on the
second hearing than the first, and singers significantly better than non-
singers overall. There was no effect of the number of singers. Hence fa-
miliarity and singing experience both benefited sung text understanding.
An effect of the number of singers may be more apparent when compar-
ing a soloist with a choir.
Keywords: singing; intelligibility; expertise; ensemble; lyrics
Much music is sung, whether accompanied or a cappella, and whether solo or

ensemble. An important aspect of perceiving sung music is the extent to
which the words can be understood. Much of the existing empirical research
on intelligibility has focused on isolated vowels sung by solo singers, at vari-
ous pitches (Sundberg 1987, Benolken and Swanson 1990, Hollien et al.
2000). Recently, however, Collister and Huron (2008) compared the intelli-
gibility of solo sung versus spoken whole words, investigating consonant
confusions as well as vowels. Their results showed many more identification
errors for sung words.
Intelligibility is clearly an issue, however, for groups of singers such as
choirs: Fisher (1991) developed a research-informed articulatory diction
development method in response to the finding that approaches to choral
dictionare based primarily on tradition and personal preference (p. 270);
Racette et al. (2000) suggest that choral singing may be more effective than
solo singing for improving word intelligibility (for aphasics) since it may
entrain more than one auditory-vocal interface (p. 2571), while Emmons and
Chase (2006) argue that, in the interests of better diction, consonants should
be approximated and vowels modified.
Two years ago, we reported the results of a survey of listeners views on
factors that might affect the intelligibility of sung text (Fine and Ginsborg
2007a). Almost 400 open-ended statements were provided by 94 respon-
dents, most of whom devoted much of their music listening to vocal and cho-
ral music, and stated that the intelligibility of sung text was important to
them when listening to lyrics in their own, or another language. We had sug-
gested four possible broad categories of factor, relating to performer(s), envi-
ronment, listener, and music and/or lyrics. As expected from the research on
intelligibility cited above, a third of all statements concerned performer-re-
lated factors including articulation, diction and enunciation, breathing and
phrasing, communicating text, expression and stage presence, voice quality,
and range. One additional factor suggested was choral ensemble.
We then explored the effect of expertise. Our respondents were roughly
equally divided between experts (professional, semi-professional, and student
singers, and some singing teachers) and non-experts (amateur singers and
non-singers) (Fine and Ginsborg 2007b). The singing teachersperhaps as
they are in the business of improving singers skills, including intelligibility
made the highest proportion of performer-related statements.
We are now investigating some of the factors nominated by our survey re-
spondents in subsequent empirical research, in order to increase our under-
standing of how to enhance the intelligibility of sung text and improve
singers diction through more effective vocal pedagogy. The present study
investigates two factors: (1) number of singers and (2) listeners experience of
singing. More singers are likely to cause more variability and noise in the
acoustic signal, thus making its decoding and understanding more difficult;
indeed, some questionnaire respondents reported that choirs are generally
harder to understand than solo singers. If listeners are themselves experi-
enced singers, they are more likely to be attuned to factors affecting singers
diction, and they may be better than non-singers at resolving the acoustic
signal into recognizable words.
Our aims were, therefore, twofold. First, we asked if sung text is harder to
understand when performed by a group of singers rather than a soloist. We
predicted that a single singer would be easier to understand than multiple
singers, even when singing in unison. Second, we investigated the extent to
which the listeners singing experience affects his or her ability to understand
sung text, even when the piece of music is unfamiliar. It was hypothesized
that expert singers would be better at comprehending sung text because of
their own experience of singing words.
METHOD
Participants
There were 48 participants (15 M, 23 F), with a mean age of 36.2 years (one
participant did not give her age). Half were self-reported singers (mean
age=36.9 years) and half self-reported non-singers (mean age=35.6 years).
Materials
Four songs, consisting of 8-bar melodies in duple time by Glinka and

Hundley, were used. Texts by Shanks and Purdy were set respectively to two
of these songs. A sequence of numbers interspersed with shorts words (and,
no, and the) were set to the same melodies to create the other two songs.
All four songs have been used in previous research on singers memory (Gins-
borg 2002). The songs were recorded both by a solo soprano and by a trio (F,
M, M) singing in unison. All stimuli were recorded in the same room at 48
kHz, 24 bit resolution onto a digital audio workstation, and then encoded as
mp3 files at a constant bit rate of 320 kilobits per second (kbit/s). The micro-
phone was a Neumann KM130 omnidirectional condenser placed about 30
cm from the singer(s). All songs were unaccompanied. Stimuli were played to
participants as mp3 files on a laptop using its internal speakers. All partici-
pants stated that the stimuli were loud enough.
Procedure
Participants first completed a short questionnaire asking about their singing

experience and how much they listened to sung music. They heard a short
practice stimulus (4-bar melody in duple time), and then listened to four of
the eight melodies, two solo and two ensemble. The exact melodies used were
counterbalanced, so each stimulus was heard by 24 participants.
Each test stimulus was presented twice and the participants were in-
structed to write down the words they heard sung. On the first hearing of each
song, a pause was given after each line to allow the participant to catch up (as
the focus of the study was perception rather than memorization ability). The
second time through, the stimulus was played without breaks. Different pens
were used on each hearing so that writing differences by hearing were appar-
ent during subsequent scoring.
The productions were then scored as follows. An error was counted for
every word missing or incorrectly heard, and for each additional word. Incor-
rect order of words or numbers was not counted as an error, as memory was
not being tested. Errors were then subtracted from the total number of words
in the song (varying from 22 to 27), and then the performance score was
transformed into a percentage, used in the analyses below.
RESULTS
The participants were asked for their singing experience on a 5-point Likert
scale (1=non-singer, 2=occasional singer, 3=keen amateur singer, 4=semi-
professional singer, 5=professional singer). They were split into two groups
(24 had answered 1-2, 24 answered 3-5). On average, the singers had sung for
22.8 years, tended to rehearse for 3.7 hours per week, and reported listening
to sung music 6.9 hours per week. On average, the non-singers had sung for
0.9 years, did not rehearse at all, and reported listening to sung music for
10.5 hours per week.
Mean performance data (understanding sung text) are shown in Table 1.
A mixed Analysis of Variance (ANOVA) was carried out. The within-subject
variables were number of singers (solo vs. ensemble) and hearing (first vs.
second). The between-subject variable was singing experience (singer vs. non-
singer). The dependent variable was % performance score. Hearing was
highly significant (F1,46=82.13, p<0.001), with better performance on the sec-
ond hearing (94.8%) than the first (90.1%). Singing experience was also sig-
nificant (F1,46=5.95, p<0.02), with singers scoring 94.1% and non-singers
scoring 90.8%. However, number of singers was not significant (92.9% solo,
92.0% ensemble).
The relationships between experience and performance factors were in-
vestigated. Overall performance significantly correlated with singing experi-
ence (r48=0.31, p<0.05), number of years singing (r48=0.33, p<0.05), hours of
singing per week (r48=0.45, p<0.01), and number of hours listening to sung
Table 1. Sung text understanding performance (mean percentage data).
Solo Ensemble
First hearing Second hearing First hearing Second hearing
Singers 92.1 96.4 91.3 96.6
Non-singers 89.3 93.9 87.6 92.5
music per week (r48=-0.31, p=0.05). Not surprisingly, singing experience also
correlated with years singing (r48=0.84, p<0.001) and number of hours sing-
ing per week (r48=0.77, p<0.001). When the singers and non-singers were
investigated separately, the only significant correlations were, for singers, age
and years singing (r23=0.84, p<0.001) and, for non-singers, overall perform-
ance and hours listening to sung music (r24=-0.53, p< 0.01), which was unex-
pectedly a negative correlation.
It was notable that certain words were more often misheard than others.
A preliminary survey shows the following words were misheard or omitted by
more than a quarter of the participants (n=12): run, fields, still, no,
the, and are. In particular, non-numeric words were often missed in the
number stimuli, perhaps due to the lack of semantic context. The patterns of
words misheard varied slightly between the solo and the ensemble stimuli.
DISCUSSION
Overall, participants could understand the sung words well (92% accuracy).
The hypothesis that experienced singers would find it easier to understand
sung text than inexperienced or non-singers was supported, both by more
accurate text identification for singers than for non-singers and a significant
correlation between singing experience and overall performance. This may be
because singers are used to producing sung text themselves, and therefore
better attuned to understanding text sung by others. They will also have ex-
perience of being in rehearsals and performances, and hearing others sing
while knowing the words themselves, forging a stronger connection between
the words and their sung acoustic signals.
However, the results do not support the hypothesis that ensemble singing
is harder to understand than solo singing, with no significant difference in
performance evident between the two conditions. This may be because the
three singers involved all concentrated on making diction as clear as possible
so there were few substantial differences between the stimuli in the acoustic
signal. Another possible reason is that ensemble effects only really become
apparent when there are many singers (even small choirs normally contain at
least eight singers). These results are considered inconclusive and are not
sufficient to reject the hypothesis; further study will be undertaken with lar-
ger ensembles, and also consider the effect of harmony versus unison singing.
Hearing the song for a second time significantly improved intelligibility and
underlines the importance of familiarity and repeated hearings on listeners
ability to understand sung text. However, there were certain words that were
still hard to understand on the second hearing. In future studies, it would be
interesting to investigate the acoustic properties of these words and also the
effects of the presence or absence of semantic context on intelligibility.
Philip Fine, Department of Psychology, University of Buckingham, Hunter Street,

Buckingham MK18 1EG, UK; Email: philip.fine@buckingham.ac.uk
References
Benolken M. S. and Swanson C. E. (1990). The effect of pitch-related changes on the

perception of sung vowels. J. of the Acoust. Society of America, 87, pp. 1781-1785.
Collister L. B. and Huron D. (2008). Comparison of word intelligibility in spoken and
sung phrases. Empirical Musicology Review, 3, pp. 109-125.
Emmons S. and Chase C. (2006). Prescriptions for Choral Excellence. Oxford: Oxford
University Press.
Fine P. and Ginsborg J. (2007a). Perceived factors affecting the intelligibility of sung
text. In K. Maimets-Volk, R. Parncutt, M. Marin, and J. Ross (eds.), Proceedings of
the Third Conference on Interdisciplinary Musicology (CIM07). Tallinn, Estonia:
Available online at www-gewi.uni-graz.at/cim07.
Fine P. and Ginsborg J. (2007b). How singers influence the understanding of sung text.
In A. Williamon and D. Coimbra (eds.), Proceedings of ISPS 2007 (pp.253-258).
Utrecht, The Netherlands: European Association of Conservatoires (AEC).
Fisher R. E. (1991). The design, development, and evaluation of a systematic method for
English diction in choral performance. Journal of Research in Music Education, 39,
pp. 270-281.
Ginsborg J. (2002). Classical singers learning and memorising a new song: An observa-
tional study. Psychology of Music, 30, pp. 58-101.
Hollien H., Mendes-Schwartz A.P., and Nielsen K. (2000). Perceptual confusions of
high-pitched sung vowels. Journal of Voice, 14, pp. 287-298.
Racette A., Bard C., and Peretz I. (2000). Making non-fluent aphasics speak: Sing along!
Brain, 129, pp. 2571-2584.
Sundberg J. (1987). The Science of the Singing Voice. DeKalb, Illinois, USA: Northern
Illinois University Press.
The PC-survey: How does use of performance

cues vary across musicians, instruments,
musical styles, and performances?
Roger Chaffin, Alexander Demos, and Mary Crawford
Department of Psychology, University of Connecticut, USA
Performance cues (PCs) are the mental landmarks that a musician

monitors to track the progress of a piece as it unfolds during perform-
ance. We describe a survey to determine how PC use is affected by ex-
perience, instrument, musical style, and by the goals of the performance.
We summarize results from longitudinal case studies in which PCs were
reported to suggest the kind of variation to be found. Understanding how
musicians use PCs should improve pedagogy by increasing our under-
standing of how musicians memorize.
Keywords: music performance; memory; performance cues
Performance cues (PCs) are the landmarks in a piece of music that a musician
thinks about during performance. They provide a mental map of the piece
that allows the performer to monitor the music as it unfolds and to recover
from mistakes and memory lapses. PCs are prepared during practice so that
they come to mind automatically on stage, giving the musician the ability to
focus on each aspect of the piece at the right moment, providing the flexibility
needed for musical spontaneity and to recover from disruptions.
We know that musicians use PCs from longitudinal case studies in which
experienced performers recorded their practice as they prepared new pieces
for performance and then reported the PCs that they used (Chaffin 2006,
Ginsborg et al. 2006, Chaffin et al. 2002, in press). The validity of the reports
was supported by behavioral evidence from practice, polished performance,
and written recall.
MAIN CONTRIBUTION
Here, we describe a survey of PC use designed to find out how the number
and type of PCs that a musician uses is affected by experience, instrument,
musical style, and type of performance. We plan to ask musicians to report
PCs for two pieces (one easy, one harder to prepare) that they have already
prepared or are in the process of preparing for performance. Although we call
it a survey, each musicians contribution is more like a case-study, but with
no recording of practice. Colleagues and their graduate students at a variety
of institutions will conduct an open-ended series of independent but related
studies, each composed of several of these case studies and each contributing
to a central database. In this way, we will build up a more comprehensive
picture of PC use. Meanwhile, the local investigators will answer specific
questions about factors affecting PC use such as, for example, effects of con-
servatory training.
We will describe the procedure to be used in the survey and then report
results from longitudinal case studies conducted to date to provide an indica-
tion of the kinds of results that we expect to obtain.
Types of PC
We have found it useful to distinguish five main types of PC: structure, ex-
pression, interpretation, basic technique, and shared. Structural PCs are criti-
cal places in the formal structure such as harmonic and melodic boundaries.
Expressive PCs represent turning points in the musical feeling (e.g. excited or
sad). Interpretive PCs represent the changes in tempo, dynamics, timbre, or
color that accomplish these expressive effects. Basic PCs represent details of
technique that must be implemented in order to be able to produce these
musical gestures as planned, such as a fingering required to set up the hand
for what follows. Shared PCs coordinate ensemble playing.
Basic PCs vary considerably across instruments. Many instruments re-
quire attention to fingering, while a singer might think instead about breath
control. String players must attend to left hand shifts and to right hand
changes in bowing direction. Some kinds of interpretive PCs appear to be
common across instruments (e.g. phrasing, dynamics, and tempo), while
others are instrument specific (e.g. pedaling on the piano, intonation for
strings, word meaning for singers).
Several PCs may refer to a single place in the music (see Figure 1). For ex-
ample, basic, interpretive, and expressive PCs at the same spot might indicate
a pianists decision to use the 1st finger (basic) in order to play forte (inter-
pretation) because this is the emotional climax (expression). The presence
Figure 1. One page of a PC report by pianist Gabriela Imreh for Claude Debussys Clair
de Lune showing expressive, interpretive, and basic PCs.
Figure 2. Excerpts from PC reports made by cellist Tnia Lisboa for J. S. Bachs Cello
Suite VI (Prelude) on four separate copies of the score for (1) structure (top left: har-
monic (H), melodic (M), and lower level (L3) boundaries), (2) expression and interpre-
tation (top right, interpretive PCs in parentheses), (3) hand position and intonation
(bottom left, red and black respectively), and (4) bowing and fingering (bottom right,
red and black respectively). (See full color version at www.performancescience.org.)
of the three different PCs would indicate that the pianist was prepared to
think of any or all of these aspects at this point.
The musician does not necessarily think of every PC in very single per-
formance. In reporting the presence of a PC, the musician is saying that s/he
is prepared to think about this feature of the music during performance, if
necessary. On a good day, the pianist in the above example may decide to
leave the fingering and the forte to take care of themselves while focusing on
the climax. On a bad day, the same pianist may be fighting to get the notes
right and let the forte and climax take care of themselves in order to focus on
the fingering. One benefit of well prepared PCs is that they allow the musician
to give very similar performances under very different conditions, including
changes in their own mental and emotional states.
Reporting PCs
In the initial research on PCs, pianist Gabriela Imreh reported PCs on a spe-
cially prepared version of the score (see Figure 1, Chaffin 2006). Subse-
quently, most musicians have marked PCs on multiple copies of a published
score. For example, cellist Tnia Lisboa used four separate copies of the score
(see Figure 2, Chaffin et al. in press).
Some preliminary comparisons across musicians and pieces
To show how PCs can be compared across musicians and pieces, we have
summarized PC reports from nine longitudinal case studies, using previously
published and unpublished data from our laboratory. Three pianists provided
reports for the Italian Concerto (Presto): a professional (Imreh) and two uni-
versity student piano performance majors, one MA and one BA-level. Imreh
also provided reports for a second piece, Clair de Lune by Claude Debussy.
We also have reports by three other professional soloists for cello (Lisboa),
voice (Ginsborg), and piano (Silva), and by two students: a BA-level trumpet
performance major and a 14-year old piano student.
Figure 3 shows the number of PCs of each type for each piece. To allow
comparison across pieces, frequencies were normalized by dividing by the
number of beats in each piece. Despite the small sample, there are suggestive
differences. The trumpet and grade-school students reported fewer PCs than
other musicians by an order of magnitude. For the Presto, the two students
reported fewer PCs than the professional. The professional pianist (Imreh)
reported fewer PCs for the easier (for her) Clair de Lune than for the chal-
lenging Presto. With the exception of the Presto, the four professionals all
reported similar numbers of PCs.
Figure 3. Number of performance cues per beat for different musicians and pieces. (See
full color version at www.performancescience.org.)
Figure 4. Percentage of performance cues of each type, normalized over number of

beats. (See full color version at www.performancescience.org.)
Figure 4 shows the number of PCs of each type as a percentage of the total
reported (normalized by number of beats). Experience mattered; the profes-
sionals used more basic PCs, the college students more interpretive PCs, the
grade school student (and trumpeter) more structural PCs.
IMPLICATIONS
All the longitudinal case studies of experienced soloists preparing new works
conducted by our laboratory have found that the musicians engaged in ex-
tended practice of PCs. This suggests that PCs are necessary for reliable per-
formance, perhaps because motor memory is unreliable (Chaffin et al. in
press). While the proposed survey is unlikely to discover whether some musi-
cians do not use PCs, it should identify the factors that affect PC use and es-
tablish the range of variability in the use of PCs.
We hope that PC surveys will become commonplace in music conservato-
ries and departments. We believe that the self-study involved provides musi-
cians with insight into their own learning and memorization. All of the
musicians who have participated in the longitudinal case studies that form
the basis for this work report that they found the process of self-study to be
beneficial (Chaffin et al. 2002, pp. 266-268; Chaffin et al. in press). In addi-
tion, we expect the conclusions will be of value to music pedagogy and also to
the psychology of music performance.
Materials for conducting PC surveys can be found on our website at www.
htfdcc.uconn.edu/psyclabs/musiclab.html.
Acknowledgments
Thanks to Alexandra Paberty Clarke and Kristen Begosh for contributing data for the
grade school student and trumpeter, respectively.
Roger Chaffin, Department of Psychology, University of Connecticut, 406 Babbidge

Road, U-1020, Storrs, Connecticut 06269-1020, USA; Email: roger.chaffin@uconn.edu
References
Chaffin R. (2007). Learning Clair de Lune: Retrieval practice and expert memorization.
Music Perception, 24, pp. 377-393.
Chaffin R., Imreh G., and Crawford M. (2002). Practicing Perfection. Mahwah, New
Jersey, USA: Erlbaum Associates.
Chaffin R., Lisboa T., Logan T., and Begosh K. T. (in press). Preparing for memorized
cello performance: The role of performance cues. Psychology of Music.
Ginsborg J., Chaffin R., and Nicholson G. (2006). Shared performance cues in singing
and conducting: A content analysis of talk during practice. Psychology of Music, 34,
pp. 167-194.
Thematic session:
Performance analysis
Finger motion in piano performance:

Touch and tempo
Werner Goebl1 and Caroline Palmer2
1 Department of Computational Perception, Johannes Kepler University Linz, Austria

2 Department of Psychology, McGill University, Canada
This study investigated movement properties of pianists fingers with

three-dimensional motion capture technology while pianists performed
melodic passages at a range of tempi. The main question was whether
finger motion dynamics change with performance tempo, an important
issue for practicing and training. Kinematic landmarks determined from
the finger trajectories changed considerably as the tempo became faster;
piano touch was under deliberate control only at slow tempi. Individual
differences in performance speed led to specific claims about desirable
finger dynamics for successful piano playing.
Keywords: motion capture; piano performance; finger dynamics; pro-

portionality; touch
Piano pedagogues disagree on how performers should develop the ability to

perform scale passages evenly and dexterously at very fast rates. One side
points out the importance of practicing fast sequences at very slow tempi,
while others hold that practicing at the intended fast tempo is more appropri-
ate. The main argument of the latter is that movement strategies change con-
siderably across different tempifor example, as human gait changes from
walking to runningand movements that are learned while practicing slowly
are not useful at fast tempi. We address here whether kinematic properties of
finger movements scale proportionately with performance tempo.
We investigated the movements of pianists fingers and hands as they
performed melodies at a wide range of tempi to test the proportionality hy-
pothesis. Furthermore, we examined how pianists touchthe way pianists
fingers approach the piano keysis affected by tempo by measuring key-
strokes containing a finger-key landmark, a marker for a pianists touch. This
work aims to generate potential recommendations for piano pedagogy, based

on observations of skilled pianists performing at different tempi.
METHOD
Participants
Twelve highly-trained pianists participated in the study. They were 20 to 33

years old (mean=27.0 years; NB. one participant was 61 years old with 40
years of experience in playing the piano) and had 10-25 years of piano lessons
(mean=18.7 years); most were piano performance students in Montreal.
Stimuli and design
One isochronous melody (the fast melody) was created that was easy to
perform with the right hand and could be continuously repeated (see notation
in Figure 2); it was designed to be performed at very fast tempi (tempo con-
ditions were 7, 8.4, 9.6, 10.7, 11, 7, 12.3, 14 and 16 tones/s, presented on dif-
ferent trials); the pianists decided at what tempo they stopped performing
(open-ended design). The tempo was indicated by a metronome in a synchro-
nization-continuation paradigm. For comparison, we include data from two
moderate melodies that contained 16 tones and were performed at moder-
ate to medium fast tempi (2, 4, 6, and 7 tones/s), as reported earlier (Goebl
and Palmer 2008).
Procedure
A passive motion capture system (Vicon 460) equipped with six infrared
cameras tracked the movements of 4 mm reflective markers glued on pianists
finger joints, hand, and wrist at a sampling rate of 250 frames/s. The motion
trajectories of the five finger-tip markers were smoothed with functional data
analysis techniques (Ramsay and Silverman 2005) and analyzed in the verti-
cal dimension (height above piano key surface).
Kinematic landmarks were extracted prior to each keystroke (see also
Figure 2): the key-bottom landmark (KB, the finger is stopped by the keybed)
and the maximum finger height (mxH, to be interpreted as the beginning of
the finger movement) for all keystrokes. An additional finger-key landmark
sometimes occurs when fingers strike the keys from a distance above the key
surface (struck touch, see Askenfelt and Jansson 1990) and the acceleration
peak is larger than a threshold of 10 m/s2 (for details, see Goebl and Palmer
2009). A pressed touch does not feature such a landmark. In addition, we
determined the peak velocity at which the finger arrives at the key surface.
Time before key-bottom (events)

2
24
mxH
17
1.5 Time of maximum height
17 24
1 17 FK
24
17
0.5 24 Time of FK
moderate melodies fast melody

0 KB
2 4 6 8 10 12 14 16
Rate (tones/s)
Figure 1. Time of maximum height (mxH, upper graphs) and time of finger-key contact
(FK, lower graphs) in number of events prior to note onset (key-bottom, KB) by per-
formed tempo. Thick line is the mean of 12 pianists; Pianists 17 and 24 are plotted sepa-
rately. (See full color version at www.performancescience.org.)
RESULTS
The open-ended design of the fast melody generated the following results:
all pianists were able to perform up to a rate of 11.7 tones/s (sixteenth notes at
176 bpm, beat=quarter note); 8 played up to 12.3, 7 up to 13.3, 6 up to 14.0,
and only 3 up to 16.0 tones per second (considerably faster, for example, than
the metronome markings of Chopins Op.25/11). The fast pianists had
similar amounts of piano lessons and years of playing, compared to the
slower pianists; the only difference was the weekly practice, significantly
higher for the fast players (25 vs. 13 hours). The following analyses attempt
to identify kinematic properties that distinguish the fast players from the slow
players.
Figure 1 shows the average timing values of the identified landmarks by
tempo condition for the moderate melodies (2-7 tones/s) and the fast melody
(7 tones/s and faster). Timing has been normalized on an event-to-event basis
relative to KB contacts to allow comparisons across tempo conditions. At
about 8 tones/s, the movement initiation (time of maximum height) toward
the current keystroke occurs more than one keystroke before (starts to over-
lap with) the previous keystroke (KB). Furthermore, the time of FK contact
approaches the time of the previous keystroke at the fastest rates. This over-
lap in movement landmarks between current and previous keystrokes may be
60
Medium fast (7 tones/s)
50
mxH
40
4
30 2
3
FK
Height above keys (mm)
20
10 4
3
KB 2
0
5.6 5.8 6 6.2 6.4 6.6 6.8 7 7.2
60
50 Very fast (14 tones/s)

40 4: mxH
30
3: FK
4
20
2 3
10 4 2: KB
3 2
0
4.3 4.4 4.5 4.6 4.7 4.8 4.9 5
Time (s)
Figure 2. Finger trajectories of index (2), middle (3), and ring finger (4) of Pianist 24
playing one cycle of the fast melody at a medium fast tempo (upper panel) and a very
fast tempo (lower panel). Three kinematic landmarks are labeled in the finger trajecto-
ries before each keystroke: the maximum height (mxH), the finger-key contact (FK),
and the key-bottom contact (KB). Vertical lines denote MIDI onset times. (See full color
version at www.performancescience.org.)
a speed-limiting feature in piano performance. This is shown further in Fig-

ure 2 for one pianists trajectories of index, middle, and ring fingers. In the
medium-fast condition (7 tones/s, top panel), the index fingers keystroke was
finished striking the second tone (B4) before the next keystroke (middle fin-
ger, C5) was initiated (mxH); at very fast rates (14 tones/s, bottom panel), the
index finger had not yet reached key-bottom for the same tone while the next
(middle) finger had already made key contact (FK), and the second-next
(ring) finger (D5) had started its descent towards the key.
To demonstrate that landmark overlap is important for performing at fast
rates, we contrasted two different pianists. Pianist 17 is a slow player who
mastered the minimal number of tempo conditions; she played the piano over
the past 15 years, but stopped a year ago; she practices 7 hours per week and
Pianist 17 Pianist 24
1 1
FK velocity (m/s) 0.8 0.8
0.6 0.6
0.4 fast melody

0.4
fast melody
0.2 0.2
moderate melodies moderate melodies
0 0
2 4 6 8 10 12 14 16 2 4 6 8 10 12 14 16
Rate (tones/s)
Figure 3. Finger velocity in meters per second at finger-key contact (FK) against per-
formance rate, separately for Pianist 17 (left) and Pianist 24 (right). Proportional scaling
with rate is plotted with solid lines.
performs occasionally in church. Pianist 24 was able to produce the very fast
tempo conditions; he played the piano for 20 years and still studies piano
actively; he practices 21 hours per week and performs in public 4-5 times per
year. The individual landmark timing for these two pianists is shown in Fig-
ure 1. Pianist 17 needed longer than Pianist 24 to perform a keystroke at a
given tempo. Thus, her keystroke landmark timing overlaps with those of
previous keystrokes at slower tempi than do Pianist 24s keystroke land-
marks. Moreover at each pianists fastest performed tempo condition, the FK
timing occurred close to the arrival (KB) of the previous keystroke
Goebl and Palmer (2008) identified two groups of pianists who differed in
their playing behavior across tempo conditions: a low-FK group who
showed a positive relation between FK landmark proportion and tempo (low
at slow tempo conditions) and a high-FK group with high (close to 1) FK
proportions for all tempo conditions. These two groups differed also at the
velocity with which their fingers arrived at the key surface (FK velocity): the
velocities of the low-FK group scaled proportionately with the tempo condi-
tions, but not those of the high-FK group. The fingers of the low-FK group
arrived at the key surface with twice the velocity when the performance
tempo doubled.
We analyzed the same FK measures for the pianists performing the fast
melody in this study; they all showed ceiling effects in their FK proportions at
all tempi faster than 7 tones/s and no proportionality overall in the FK finger
velocities. The FK measures were also contrasted for the two exemplary pi-
anists; their mean finger velocities at finger-key contact are plotted against
the full range of performed tempi in Figure 3 for the moderate melodies (2-7
tones/s) and the fast melody (7 tones/s and faster). The slow pianist (17)
showed approximate proportionality across the tempi, whereas the fast

pianist (24) did not. These findings suggest that the link between performed
tempo and FK velocity might be detrimental to fast playing because propor-
tionality cannot be maintained at extremely fast speeds.
DISCUSSION
Pianists finger movement dynamics changed across different performance

tempi; finger movements at key contact changed considerably as pianists
accommodated faster tempi. Overall, these findings suggest that practicing at
the final tempo is important from a motor control perspective. Furthermore,
individual differences indicated that dissociating different finger movement
properties from performance tempo may be essential for fast piano playing.
However, individual cases of finger velocity-tempo proportionality implied
that practicing slowly should be done very softly (that is with low FK velocity)
so that finger movements at slow rates become more similar to those typically
seen at final fast rates.
This research demonstrates how motion-based approaches could be es-
tablished more widely in the future to enhance our understanding of the
complex movement patterns executed by skilled pianists and to develop po-
tential recommendations for piano pedagogy.
Acknowledgments
This work was supported by the Austrian Research Fund (J2526) to the first author and
by the Canada Research Chairs program and NSERC Grant 298173 to the second.
Werner Goebl, Department of Computational Perception, Johannes Kepler University

Linz, Altenberger Strasse 69, Linz 4040, Austria; Email: werner.goebl@jku.at
References
Askenfelt A. and Jansson E. V. (1990). From touch to string vibrations. Timing in the
grand piano action. Journal of the Acoustical Society of America, 88, pp.52-63.
Goebl W. and Palmer C. (2008). Tactile feedback and timing accuracy in piano per-
formance. Experimental Brain Research, 186, pp. 471-479.
Goebl W. and Palmer C. (2009). Synchronization of timing and motion among per-
forming musicians. Music Perception, 26, pp. 427-438.
Ramsay J. O. and Silverman B. W. (2005). Functional Data Analysis (2e). New York:
Springer.
Quantitative multidimensional approach of

technical pianistic level
Paul Molin
LEAD CNRS, University of Burgundy, France
The purpose of this study is to examine whether it is possible to define

quantitatively the technical level of a pianist by summarizing the individ-
ual characteristics of the pianists technical ability in a fewat most
threeindependent variables. These variables could be used in competi-
tions and examinations as well as by piano teachers to measure students
progress, or by the students themselves in their daily work.
Keywords: pianistic level; principal component analysis; pianist motor

skill; correlation of difficulties; MIDI
A statistical approach to measuring the biomechanical characteristics of pi-

ano playing was used by Altenmller and Jabusch (2005) to measure pianists
focal dystonia, then by the same authors (2007, 2009) to measure the devel-
opment of adults and childrens motor skill, using the Musical Instrument
Digital Interchange (MIDI) recording format. An ordinary MIDI file contains
a succession of events, each event being characterized by three parameters:
occurrence time of the event (T), duration (D), and pitch (P). From these pa-
rameters, many characteristics (tone overlap, equality of produced sound
velocity, equality of rhythm, etc.) can be computed, each variable giving an
insight of the technical level of a pianist. The advantages and disadvantages of
MIDI coding are widely discussed by Clarke (2004). Keyboard technical skill
and how to improve it was studied early, for example by J. S. Bach (Kla-
vierbung) and D. Scarlatti (Exercizi), and C. P. E Bach wrote a treatise Essai
sur la vritable manire de jouer les instruments clavier in 1762. A system-
atic study of piano playing difficulties was completed by Alfred Cortot (1928),
with the author recommending a detailed daily work based on five-finger
exercises, scales, arpeggios, trills, double notes, and octaves, with daily key
transposition. This well known book is the basis of the daily training of pian-
ists. Our purpose is, from a sample of these typical training patterns, to ex-
tract the qualities and flaws of a pianist, and summarize them by synthetic
graphs and synthetic variables.
METHOD
Participants
Eighteen pianists (mean age 22.5 years, ranging from 12-59), from middle
level music conservatories to professional level participated in the experi-
ment. They had played piano for a mean of 14.5 years (range=4-52) and re-
ceived either course credit or payment for their participation.
Materials
The pianists performed on a classical upright piano (Yamaha U2) with MIDI
interface. MIDI data were collected on a DELL laptop PC, by using the
Cubase SX program. All statistic programs were written in the Matlab pro-
gramming language by members of the LEAD-CNRS laboratory.
Procedure
The pianists were submitted to a series of exercises containing the most

common classical difficulties in piano playing (scales, trills, and arpeggios).
In order to obtain a photograph of the pianist, a few attempts only (up to 5)
were authorized for each exercise. The tempo of each exercise was free. The
pianists were allowed 20 minutes for warming up and getting used to the
piano prior to the actual performances. Not all the pianists used this warm up
time. An operator was always present to guide the participants, verify the
fingering, and record the performances. The detailed exercises were as fol-
lows:
Four scales (Eb major, E major, Bb major, Eb minor) were requested

from participants, played on four octaves twice without interruption,
hands together at a pitch distance of one octave. The choice of scales was
motivated by the fact that these scales have a very natural fingering, so all
pianists used the same fingering. There were two easy scales (Eb major, E
major), one more difficult (Bb major), and a difficult scale (Eb minor).
All the possible combinations of fingers (1-2, 1-3, 4-5) were used for the
trill exercise.
Eight arpeggios (C major, C major 4/6, Db major, Db major 4/6, C# mi-
nor, C# minor 4/6, Bb minor, Bb minor 4/6), played on four octaves twice
without interruption, hands together at a pitch distance of one octave.
Fingering was verified by the operator to be 1-2-3 for the right hand and
1-4-2 for the left hand for the natural arpeggio state, 1-2-4 for right hand
and 1-3-2 for left hand in the upward direction, and the reverse for the
down direction.
RESULTS
Defining and computing the relevant variables
For all exercises, the first indices of quality are a high tempo and few false
notes. But these are not the only qualities of a good piano technique; for
scales and arpeggios, the main word characterizing the quality of playing is
homogeneity. This means rhythmic homogeneity for each hand, rhythmic
homogeneity between hands, homogeneity of velocity (intensity of key de-
pressing force) for each hand, velocity homogeneity between hands, and
overlap homogeneity (legato homogeneity). Of course, in an artistic perform-
ance of a piece of music, this homogeneity is not observed; all the discrepan-
cies in an artistic performance are voluntary acts, guided by the artistic
sensibility. But the homogeneity required in the exercises is the way to gain
the control of the movements, in order to play with expression.
We then retained seven variables: (1) tempo (TPO), (2) tempo coefficient
of variation (CVTPO), (3) number of false notes (FN), (4) intra-hand velocity
standard deviation (STDVEL), (5) intra-hand overlap standard deviation
(STDOVL), (6) lag time between hands (LAGLR), and (7) velocity difference
between hands (DVELLR). For trills, only the first five variables were used,
since the exercise was performed hand-by-hand.
Why a free tempo?
In the experiments of Jabusch et al. (2004, 2007, 2009), the authors fixed
the tempo of the exercises (scales) and allowed the subjects to repeat them
10-15 times. They then computed a mean scale for each subject. Our point
of view is slightly different. An important part of the technical level is the
pianists awareness of his own skille.g. I will not begin a scale at a metro-
nomic tempo of 160 if I know that I will not finish the scale. We then left the
subjects free to choose their tempos and considered the tempo variable as a
level variable among others.
Principal component analysis (PCA) (Hotelling 1933) for scales
We present complete results and graphs for the Eb major scale. Table 1 shows
the correlation matrix of the seven variables for the scale. It is not surprising
Table 1. Correlation matrix between the seven quality variables for Eb major scale.
Tempo False Inter hand Intra hand CV tempo Overlap Time lag
notes velocity velocity STD between
variations STD hands
(TPO) (FN) (DVELLR) (STDVEL) (CVTPO) (STDOVL) (LAGLR)
TPO 1 -0.179 -0.610 -0.165 -0.468 -0.503 -0.198
FN -0.179 1 -0.024 0.422 0.682 0.187 0.686
DVELLR -0.610 -0.024 1 0.291 0.128 0.516 0.134
STDVEL -0.165 0.422 0.291 1 0.220 0.242 0.724
CVTPO -0.468 0.682 0.128 0.220 1 0.540 0.432
STDOVL -0.503 0.187 0.516 0.242 0.540 1 0.159
LAGLR -0.198 0.686 0.134 0.724 0.432 0.159 1
that the tempo variable is negatively correlated with all other variables: intui-
tively, tempo is a positive index of quality, while the other variables are nega-
tives. Some other correlations, intuitively expected, can be noticed: for
example, the high correlation (0.682) between the false notes and CV. Tempo
variables can be understood as a false note induces a stress for the pianist,
then an irregularity of tempo. In the same way, the high correlation between
the CV tempo and the time lag can be understood as a false note in one hand
does not affect the other hand, and then increases the time lag between
hands.
Figure 1 shows the graphic results of the Eb major PCA, which is consis-
tent with respect to the correlation matrix. The left part of the figure is the
correlation circle, which represents the correlations between variables and
principal components. The first two principal components account for about
70% of the total variance of the data, and thus the bi-dimensional represen-
tation is reliable. As expected, the tempo variable is opposed to all others. It is
very interesting to see that the correlation between tempo and false notes is
low (-0.179). A good pianist can do many wrong notesa known joke of Al-
fred Cortot: if you picked all my false notes in a concerto, it could be a sec-
ond concerto.
In order to interpret the graph of subjects, let us describe some of them:
pianists 3, 4, and 9 are beginning the second cycle of studies in a regional
conservatory, pianist 18 is a concert pianist, pianist 2 is a piano teacher, and
pianist 1 is an amateur (the author). All other pianists are second or third
cycle students in a regional conservatory. Then the first principal component
Table 2. Metronomic and adjusted tempos (scores) for the Eb major scale.
Subject 1 2 3 4 5 7 9 10 12 13 14 15 16 17 18
Tempo 86.7 79.2 65.9 31.0 36.7 71.7 54.5 41.8 35.6 64.8 85.5 94.6 71.7 64.2 143.3
Score 63.0 59.8 26.7 1.7 16.6 50.0 13.8 25.1 7.1 30.1 62.7 64.0 49.2 38.7 125.0
Figure 1. Principal component analysis of the Eb major scale. The right panel represents
the variables and the left panel the subjects.
can be easily interpreted as the technical level axis. The second axis can de-
scribe more precisely the flaws of the pianist: for example, a pianist far above
the second axis is characterized by many false notes, a bad synchronism be-
tween hands, but a good legato (little overlap STD).
DISCUSSION
The PCA analysis appears to be a good visualization tool for a synthetic pres-
entation of the technical level of a pianist. The results for trills and arpeggios
are similar to those of scales. All the default variables are positively corre-
lated, and all negatively correlated with the tempo variable. As we stated in
the section on PCA above, the principal components cannot be used as they
are, but they suggest an overall evaluation in a form opposing the tempo with
all the other variables. For example, for the Eb major scale, the first principal
component has the following equation:
PC1=-0.3595*TPO+0.3991*FN+0.2860*DVELLR+
0.3682*STDVEL+0.4331*CVTPO+0.3668*STDOVL+ 0.4147*LAGLH
All the coefficients have the same magnitude. The idea is then to give the
results to the musician in the form of an adjusted tempo, very understandable
by musicians:
TPOADJ=PC1/-.3595=TPO-1.1102*FN-0.7956*DVELLR
-1.0244*STDVEL-1.2049*CVTPO-1.0205*STDOVL -1.1537*LAGLH
This formula gives the results in Table 2. Under this system, a pianist who
plays fast and has many flaws gets the same score as a pianist who plays
slower but with good legato, good rhythmic homogeneity, few false notes, etc.
Acknowledgments
I wish to thank Konrad Maria Engel for kindly accepting to lend his talent to our ex-
periments, to Eckart Altenmller and Hans Christian Jabusch for fruitful help and dis-
cussions, to Chlo Fourier and Michel Cara the indefatigable operators, and to
Vronique Assadas for her stylistic help.
Paul Molin, LEAD CNRS, University of Burgundy, Esplanade Erasme, Dijon, Cte dOr
21000, France; Email: paul.molin@u-bourgogne.fr
References
Bach C. P .E. (1762). Essai sur la Vritable Manire de Jouer les Instruments Clavier.
Paris: CNRS Editions 2002.
Clarke E. F. (2004). Empirical methods in the study of performance. In E. F. Clarke and
N. Cook (eds), Empirical Musicology (pp. 77-102).Oxford: Oxford University Press.
Cortot A. (1928). Principes Rationnels de la Technique Pianistique. Paris: Salabert.
Hotelling H. (1933). Analysis of a complex of statistical variables with principal compo-
nents. Journal of Educational Psychology, 24, pp. 417-441.
Jabusch H.C., Alpers H., Kopiez R., and Altenmller E. (2009). The influence of practice
on development of motor skills in pianists: A longitudinal study in a select motor
task. Human Movement Science, 28, pp. 74-84.
Jabusch H.C., Vauth H., and Altenmller E. (2004). Quantification of focal dystonia in
pianists using scale analysis. Movement Disorders, 19, pp. 171-180.
Jabusch H.C., Yong R., and Altenmller E. (2007). Biographical predictors of music-
related motor skills in children pianists. In A. Williamon and D. Coimbra (eds.),
Proceedings of ISPS 2007 (pp.363-368). Utrecht, The Netherlands: European As-
sociation of Conservatoires (AEC).
Evaluation of a scale performance on the piano

using spline and regression models
Shinya Morita1, Norio Emura2, Masanobu Miura3, Seiko Akinaga4, and

Masuzo Yanagida5
1 Graduate School of Science and Technology, Ryukoku University, Japan

2 National Institute of Advanced Industrial Science and Technology, Japan
3 Faculty of Science and Technology, Ryukoku University, Japan
4 Department of Education, Shukugawa Gakuin College, Japan
5 Faculty of Engineering, Doshisha University, Japan
Recently, many systems have been developed for supporting novice

and/or beginner pianists. Even though these systems use a set of per-
formance data, they cannot properly evaluate performance proficiency
because a procedure for calculating proficiency has not been proposed.
To solve this problem, we have developed an evaluation model by intro-
ducing a spline curve or regression curve, assuming it to be a standard
for evaluating proficiency for scale performances from an aesthetic view-
point. This paper introduces other models for evaluation, comprised of a
standard curve for evaluating a piano performance. Curves used are (1) a
spline curve and (2) regression curves using n-dimensional models
(1n10). Investigated here is the effectiveness of the curves, so as to de-
termine a better curve for automatically evaluating performance. As a re-
sult, correlation coefficients between scores predicted by the spline curve
model and evaluation scores by experts were 0.65, whereas the average
of correlation coefficients between scores predicted by regression curves
models and evaluation scores by experts was 0.58. In other words, a cor-
relation coefficient of the spline curve model is confirmed to be higher
than those of regression curve models. Therefore, it was confirmed that
the spline curve model is more effective than n-dimensional curves at
automatically evaluating performance.
Keywords: piano; scale performance; spline curve; n-dimensional curve;

MIDI
In recent years, a lot of people who have wanted to learn how to play the pi-
ano have been able to create easily an environment in which to learn due to
the development of the MIDI keyboard. However, most people who start of-
ten give up practicing the piano. There are many reasons for this: (1) special-
ists instructions cannot be given due to temporal and spatial restrictions, (2)
the piano performance cannot be objectively evaluated, (3) beginners cannot
plan how to teach themselves the piano, (4) beginners tire of practicing if they
only use textbooks, and (5) it takes a lot of time to master the piano. To over-
come these difficulties, many systems have been developed for supporting
novice players and/or beginner pianists. However, even though these systems
record and use a set of performance data, they cannot appropriately evaluate
a performance with the deviations from a metronome in artistic terms. To
solve this problem, we have developed a method to evaluate automatically the
skill level and proficiency for performing a one octave scale (e.g. Akinaga
2006).
METHOD
Materials
This study uses the following evaluation models: spline curves and regression
curves. The spline curve used here is the interpolated curve that passes
through the representative points. The representative points are determined
by finger crossing in three ways. Our previous study aimed to evaluate a per-
formance containing artistic expressions such as intonation. Therefore, spline
curves were introduced as a standard for evaluating performance proficiency.
An example of a spline curve is shown in Figure 1. Regression curves used in
this study are n-dimensional polynomial (1n10). When n=1, the model
means a straight line, or when n=15, the model means the actual performance.
When n increases, obtained regression curves become closer to the actual
performance.
Procedure
Investigated here is the effectiveness of the proposed models using regression

curves, so as to determine a better model for automatically evaluating per-
formance. Correlation coefficients between predicted scores proposed by
models and evaluation scores evaluated by experts were calculated in order to
compare spline curve models to regression curves models.
Figure 1. Example of a spline curve.
Figure 2. Representative points of condition 1.

Spline curve models
Condition 1. Fifteen inputted notes are categorized into four clusters. Centers
of each cluster are assumed as representative points. In these conditions, the
clusters are comprised of C-D-E-F, F-G-A-B-C, C-B-A-G-F, and F-E-D-
C, respectively. Representative points of condition 1 are shown in Figure 2.
Condition 2. Fifteen inputted notes are categorized into four clusters.
Centers of each cluster are assumed as representative points. In these condi-
tions, the clusters are comprised of C-D-E, F-G-A-B-C, C-B-A-G-F, and
E-D-C, respectively. Here, overlapping notes for crossing fingers belong to a
cluster, whereas in condition 1 they belong to both clusters. Representative
points of condition 2 are shown in Figure 3.
Condition 3. Fifteen inputted notes are categorized into three clusters.
Centers of each cluster are assumed as representative points. In these condi-
tions, the clusters are comprised of C-D-E, F-G-A-B-C-B-A-G-F, and E-
D-C, respectively. Unlike condition 2, here two clusters around fingerings of
returning are categorized into a cluster. Representative points of condition 3
are shown in Figure 4.
RESULTS
Determination coefficient
The determination coefficients between proposed evaluation models and ac-

tual performance were calculated in order to compare spline curve models to
regression curves models. The results of the determination coefficient are
shown in Figure 5.
Performance of evaluation models
The correlation coefficient between scores predicted by proposed evaluation

models and scores evaluated by experts were calculated in order to compare
spline curve models to regression curves models. The results of the determi-
nation coefficient are shown in Figure 6.
DISCUSSION
Figure 5 shows the determination coefficient between proposed evaluation

models and the actual performance. As a result, the determination coeffi-
cients for spline curve models are relatively lower than those for n-dimen-
sional curves (3n10). Figure 6 shows the correlation coefficients between
scores predicted by proposed evaluation models and scores evaluated by ex-
Figure 5. Results of calculating coefficients of determination between actual perform-

ance and spline/n-dimensional curves.
Figure 6. Result of calculating correlation coefficients for proficiency scores between

those given on average by experts and those given by spline/n-dimensional curves. (See
full color versions at www.performancescience.org.)
perts. As a result, the average of correlation coefficients between predicted

scores predicted by the spline curve model and scores evaluated by experts
was 0.65, whereas the average of correlation coefficients between scores pre-
dicted by regression curves models and scores evaluated by experts was 0.58.
Correlation coefficients between scores predicted by regression models and

scores evaluated by experts increased when n increased. In other words, when
the models are similar to actual performance, it was recognized to obtain
evaluation scores similar to those given by experts. Interestingly, the correla-
tion coefficient of the spline curve model and tenth-dimensional model are
almost the same in terms of correlation. In this study, the spline curve model
passes through six representative points. Therefore, essentially the correla-
tion coefficient of the spline curve model and six-dimensional model should
be almost the same value. However, the correlation coefficient of spline curve
model is higher than that of the six-dimensional model. These results show
that the spline curve model is thought to be better than n-dimensional models
for automatically evaluating performance.
Acknowledgments
This study was partly supported by the High-Tech Research Center project for private
universities, matching fund subsidy from the Japanese Ministry of Education, Culture,
Sports, Science, and Technology (MEXT) and Research for Promoting Technological
Seeds from the Japan Science and Technology Agency (JST).
Shinya Morita, Graduate School of Science and Technology, Ryukoku University, 1-5
Yokotani, Seta Oe-cho, Otsu Shiga 520-2194, Japan; Email: t08m079@mail.ryukoku.
ac.jp
References
Akinaga S., Miura M., Emura N., and Yanagida M. (2006). Toward realizing automatic
evaluation of playing scales on the piano. In M. Baroni, A. R. Addessi, R. Caterina,
and M. Costa (eds), Proceedings of the Ninth International Conference on Music
Perception and Cognition [CD-ROM] (pp. 1843-1847). Bologna, Italy: Bologna Uni-
versity Press.
Dannenberg R., Sanchez M., Joseph A. et al. (1990). A computer-based multi-media
tutor for beginning piano students. Interface, 19, pp. 155-73.
Dannenberg R., Sanchez M., Joseph A. et al. (1993). Results from the Piano Tutor Pro-
ject. In D. Smalley, N. Zahler, and P. Galvani (eds), Proceedings of the Fourth Bien-
nial Arts and Technology Symposium (pp. 143-150). New London, Connecticut,
USA: Connecticut College.
Symposium:
Performance science: Implications for
educational and professional practice
Profiling musicians health, wellbeing, and

performance
Aaron Williamon1, David Wasley2, Rosie Burt-Perkins1, Jane Ginsborg3,

and Wulf Hildebrandt4
1 Centre for Performance Science, Royal College of Music, London, UK

2 Cardiff School of Sport, University of Wales Institute Cardiff, UK
4 Institute of Anatomy and Cell Biology, Philipps University Marburg, Germany
This study profiles music students physical and mental fitness for per-
formance. Participants were recruited from the Royal College of Music
(RCM, n=59) and Royal Northern College of Music (RNCM, n=32), and
standardized measures of health promoting behaviors, anxiety, perfec-
tionism, cardiovascular fitness, and physical strength and flexibility were
employed to assess students performance-related health and wellbeing.
The resulting profile indicates that (1) students tend to fall outside of
their target BMI, with more students falling below their target than above
it, (2) cardiovascular fitness is most frequently below average or average,
with under 40% of students achieving above average cardiovascular fit-
ness, (3) student fatigue correlates variously with aspects of perfection-
ism, trait anxiety, health promotion, and self-regulated learning, and (4)
pain that is reported to stop performance is most often linked to the up-
per arm/elbow, left and right hands, and the back. The value of such
profiling exercises in educational contexts is discussed, with examples of
implementation drawn from a UK conservatoire.
Keywords: health profiling; physical fitness; mental health; health pro-

moting behaviors; music education
Having the capacity to assess musicians skills and abilities is of particular

benefit for researchers, musicians, and those involved in musicians educa-
tional and professional development. The knowledge generated through
physical and mental profiling exercises can be applied, for instance, as the
basis for refining current training approaches, tailoring and implementing
novel performance enhancement initiatives, and raising awareness of impor-

tant issues of musicians health and wellbeing.
This article highlights the value of skills profiling in educational and pro-
fessional contexts. Specifically, we profiled music students fitness for per-
formance through a wide range of self-report and objective physical measures
of health and wellbeing. The application of the resulting profile within a con-
servatoire context is used to illustrate the value of such work to the training of
performers.
METHOD
Participants
A total of 91 students (57 female, 34 male; mean age=21.88 years, SD=4.63

years) from the Royal College of Music (RCM, n=59) and the Royal Northern
College of Music (RNCM, n=32) took part in the study. The participants rep-
resented the following instrumental specialisms: strings (n=42), keyboard
(17), woodwind (14), voice (8), brass (8), composition (1), and percussion (1).
Materials
Music students fitness for performance was profiled through the following
self-report measures:
Questions on musical/personal background (e.g. age, sex, instrument)

Short Warwick Edinburgh Mental Wellbeing Scale (WEMWBS)
Frost Multidimensional Perfectionism Scale (Frost MPS)
Health Promoting Lifestyle Profile (HPLP) II
Self-efficacy for Musical Learning Scale
Self-regulation Scale
Spielberger Trait Anxiety Inventory (TAI)
Ratings of musculoskeletal health and fatigue
Descriptions of the above measures, including methods for scoring each

and their relevant subscales, are provided by Kreutz et al. (2008, 2009) and
Ginsborg et al. (2009).
Procedure
The profiling procedure was divided into three stages. Stage 1 introduced the
participant to the profile study, secured written consent, and screened health
suitability for the physical assessment. Each participants height (cm), weight
(kg), hand spans (cm), and finger spans (angle) were recorded.
Stage 2 was conducted online using the environment provided by Sur-
veymonkey, to which the self-report measures above were adapted. The
questionnaire was identical at both institutions, and students were able to ask
a member of the research team for clarification on the questions presented as
necessary.
Stage 3 comprised the physical assessment, conducted by an exercise sci-
entist. The assessment included measures of balance (dominant balance),
grip strength (kg), core flexibility (cm), body fat/lean percentages, body fat/
lean mass (kg), body mass index (BMI), shoulder flexibility internal and ex-
ternal (dominant balance), and sub-maximal cardiovascular fitness (heart
rate recovery, in bpm). Body composition measurements were taken using a
bioelectrical impedance meter and sub-maximal fitness by a 3-min step test
with bpm measured using a Polar heart rate monitor. Participants were given
a printed summary of their results and were fully debriefed.
Data analysis
Preliminary analyses of the data revealed no significant differences between

the student cohorts from each institution, except on the spiritual growth sub-
scale of the HPLP II (t89=3.07, p<0.01, where RCM>RNCM) and the Self-
regulation Scale (t89=1.99, p=0.05, where RCM>RNCM). For the purposes of
this article, the two cohorts have been combined in all analyses.
Here, we focus on the physical measures of fitness, as well as Pearson cor-
relations between physical measures and self-report measures, including
ratings of musculoskeletal health and fatigue. Descriptive statistics for each of
the measures employed are available from the corresponding author by re-
quest.
RESULTS
Descriptive statistics for BMI and sub-maximal cardiovascular fitnessas

representative of key profiling measures of physical fitnessare illustrated in
Figures 1 and 2, respectively.
Significant correlations between self-reported musculoskeletal health and
fatigue, psychological self-report measures, and physical fitness are listed in
Table 1. For these analyses, ratings of musculoskeletal health and fatigue were
grouped into six categories, following the factor analysis reported by Kreutz et
al. (2008).
100
80
60
Percentage
men
women
40
20
0
Below target Target Above target
Figure 1. Distribution of BMI for male and female participants (n=89).
100
80
60
Percentage
men
women
40
20
0
Below average Average Above average
Figure 2. Cardiovascular fitness. Below average includes below average, poor, and
very poor. Above average includes above average, good, and excellent (n=84).
DISCUSSION
The profile arising from these results can be summarized as follows: (1) stu-
dents tend to fall outside of their target BMI, with more students falling below
their target than above it, (2) cardiovascular fitness is most frequently below
Table 1. Significant Pearson correlations (p<0.05) between ratings of pain in the elbows,
hands, spine, mouth and feelings of fatigue with other self-report measures and BMI,
where + indicates a significant positive correlation and - a negative correlation.
Upper Left Right

arms/elbows hand hand Spine Mouth Fatigue
General health rating - - -
Pain stops performance + + + +
HPLPII sm+ sg- sm-
WEMWBS - -
Frost MPS pec+ ps- o- cmd+ pec+
Self-regulation -
TAI +
BMI -
Note. sm=HPLPII subscale for stress management, sg=HPLPII subscale for spiritual
growth, pec=Frost MPS subscale for parental expectations and criticism, ps=Frost MPS
subscale for personal standards, o= Frost MPS subscale for organizational skills,
cmd=Frost MPS subscale for concern over mistakes and doubt above actions.
average or average, with under 40% of students achieving above average car-
diovascular fitness, (3) student fatigue correlates variously with aspects of
perfectionism, trait anxiety, health promotion, and self-regulated learning,
and (4) pain that is reported to stop performance is most often linked to the
upper arm/elbow, left and right hands, and the back.
Given the physicality of musicians work, such a trend toward poor fitness
is concerning. With the injury and pain problems frequently reported in con-
servatoriesand in light of profiling exercises such as reported herethere is
clearly a need for educational programs to address this issue. At the RCM, for
example, the profile is used to inform a compulsory seminar series for all
incoming undergraduate students, designed to raise awareness of the impor-
tance of musicians fitness and wellbeing. To enable students to follow this up
with practical steps, poor fitness demonstrated in the profile is targeted
through a fitness awareness scheme, and aspects of student fatigue and injury
addressed through Peak Performance Workshops, introducing students to
Alexander technique, yoga, Pilates, and tai chi, among other interventions.
Further work to extend the sample is underway, with new intakes at the
RCM and RNCM, and at other UK conservatoires, in order to refine and
streamline the profiling procedure. The tools to emerge from this process will
be able to inform educators of students fitness for performance, allowing

them to take necessary steps to safeguard students health and wellbeing.
Acknowledgments
This work was supported by a grant from the Royal College of Music, London.
Aaron Williamon, Centre for Performance Science, Royal College of Music, Prince Con-
sort Road, London SW7 2BS, UK; Email: awilliamon@rcm.ac.uk
References
Ginsborg J., Kreutz G., Thomas M., and Williamon A. (2009). Healthy behaviours in
music and non-music performance students. Health Education, 109, pp. 242-258.
Kreutz G., Ginsborg J., and Williamon A. (2009). Health-promoting behaviours in con-
servatoire students. Psychology of Music, 37, pp. 47-60.
Kreutz G., Ginsborg J., and Williamon A. (2008). Music students health problems and
health-promoting behaviors. Medical Problems of Performing Artists, 23, pp. 3-11.
Dance science: Scientific investigations into

the effect of dance specific fitness training and
its impact upon pedagogic practices and dance
performance
Emma Redding, Sarah Irvine, Edel Quin, and Sonia Rafferty
Department of Dance Science, Trinity Laban Conservatoire of Music and Dance, UK
Dance training has developed eclectically to serve the different ap-

proaches to dance performance and making; however, there is a discrep-
ancy between the physiological demands of training and dance
performance. It is no longer acceptable to train dancers without due re-
gard for physiological concerns if they are to be prepared appropriately to
meet the demands of current choreographic work. Research over the last
two decades has examined the physical fitness status of professional and
student dancers, but there is still debate about how fit dancers should be.
There is a discrepancy in the physical intensity level between training,
rehearsal, and performance, and the idea of supplementary fitness
training has been debated, albeit untested longitudinally with large
groups of dancers. The purpose of this study was to examine the effect of
a one-year dance specific fitness program on undergraduate contempo-
rary dance students undertaking full-time vocational training and to ob-
serve any impact the findings may make upon dance pedagogic practices.
Results from the pre- and post-screening assessments show that, follow-
ing the intervention, there was a significant decrease (p<0.05) in mean
heart rates across all five stages of the Dance Aerobic Fitness Test
(DAFT), indicating improvement in the dancers aerobic capabilities.
Findings impacted upon the school curriculum, in that the timetable now
comprises weekly fitness classes.
Keywords: fitness; dance; physiological; intervention; aerobic
Research over the last two decades has examined the physical fitness status of
professional and student dancers (Chatfield et al. 1990, Cohen et al. 1982,
Dahlstrom et al. 1996, Novak et al. 1978, Rimmer et al. 1994), but there is
still debate about how fit dancers should be. One study (Wanke 2001) as-
sessed the effect of an integrated endurance training program for ballet
dancers (n=16) across a ballet season and found positive results. Some stud-
ies have suggested a discrepancy in the intensity level between training, re-
hearsal, and performance (Rist 1994, Wyon and Redding 2005, Redding and
Wyon 2001), and the idea of supplementary fitness training has been de-
bated, albeit untested longitudinally with large groups of dancers.
The applicability of laboratory tests and training regimes from sports is
questionable (Schantz and Astrand 1984, Redding and Wyon 2003), and it is
becoming increasingly necessary to gather relevant data and qualitative ob-
servationsphysiological and psychologicalin order to develop specific
methods of promoting and assessing dance fitness. The purpose of this study,
longitudinal and experimental in design, was to examine the effect of a one-
year dance specific fitness program on undergraduate modern dance students
undertaking full-time training.
METHOD
Participants
Screening information from first year students (n=86) provided initial data
on a series of physiological tests, including the Dance Aerobic Fitness Test
(DAFT; Wyon et al. 2003), the Wingate Anaerobic bike Test (WAnT), and the
vertical jump height test, which were repeated at the end of the one-year in-
tervention. Students also completed an evaluation questionnaire regarding
their personal appreciation of the program. Existing screening information on
the previous student cohort (n=85) provided an opportunity to compare
physiological changes across the year without the fitness intervention.
Procedure
A weekly 90 minute fitness class was developed across the year according to
principles of periodization and specificity (Bompa 1999). The primary aim
was for the structure and content to be responsive to curriculum needs and to
engage in a dialogue with technique teachers and therapists. Intensity and
duration of exercises were considered. The dancers heart rates were regularly
monitored to ensure that the intensity level was appropriate to elicit a train-
ing response. Functional fitness training preceded more dance-based move-
ment that increasingly replicated vocabulary from technique classes. For
example, plyometric training was introduced, initially using basic parallel foot
250
200
b.min-1
Pre
150
Post
100
50
1 2 3 4 5
DAFT 5 stages
Figure 1. Mean heart rates of dancers undertaking dance specific fitness training before
and after the one year intervention.
positions, and later modified to include turned out positions that more closely
mimicked the type of jumps seen in dance. Upper body strength exercises
gradually progressed to incorporate partner lifting of varying speeds and
complexities. During the final phase, a circuit type structure reflected the
variety of activity and speed of succession that would be encountered in a
dance class.
Data treatment and analysis
All data were entered into SPSS (v.12). Means and standard deviations were
calculated for all variables. Dependant t-tests were used to monitor changes
in heart rate and jump height data collected from the pre- and post-testing.
RESULTS
Results from the pre- and post-screening assessments show that, following
the intervention, there was a significant decrease (p<0.05) in mean heart
rates across all five stages of the DAFT test (Figure 1), indicating improve-
ment in the dancers aerobic capabilities. In addition, it appears that this par-
ticular group of students displayed lower heart rates at every stage, compared
with the group of first year students from the previous year who did not en-
gage in regular fitness training (Table 1).
The females in the intervention group improved their jump height
(p<0.05), while the males did not. Participant responses to the evaluation
Table 1. Mean heart rates across each stage of the DAFT for three different groups of
dance students. There was a significant difference between the two groups at every stage
(p<0.05).
Stage 1 Stage 2 Stage 3 Stage 4

(b.min-1) (b.min-1) (b.min-1) (b.min-1)
First year students with no fitness 122 13 153 15 179 14 190 13
at end of year 1 (n=38)
First year students post fitness 115 12 129 23 162 13 177 11
end of year 1 (n=50)
questionnaire suggest that the students are motivated to participate in their

fitness classes. Responses focused on students increased knowledge with
respect to physical preparation for their dance skills training, the importance
of application of the theoretical aspects of their fitness course, and the new
perspective they have of dancers as athletes as well as creative artists.
DISCUSSION
Results from the DAFT test show that aerobic fitness improved over the one-
year period, supporting findings of previous studies (Wanke 2001). When
comparing the DAFT heart rate data to that of the dancers who did not par-
ticipate (from the previous Year 1 cohort), there appears to be an overall re-
duction at each stage, indicating that it is the fitness training intervention
that has contributed to this effect.
Comments from the evaluation questionnaire show that students have gained
knowledge and skills that they do not acquire at any other point in their dance
training and are capable of applying this knowledge and understanding to
their wider physical development. The most frequent physiological self-
observed effects were reductions in fatigue, improvement in general energy
levels, and improved capacity in dance classes to sustain technique and
jumping ability. The importance of warm-up and cool down was commonly
cited and the recognition of the relationship between fitness and injury
prevention was highlighted. In psychological terms, preliminary analysis of
the comments reveals that students have found fitness valuable as a relaxa-
tion tool. They explain that they feel more positive and confident in general
and find a reduction in stress, tension, and frustration. Reasons include the
holistic approach, the absence of competition in the fitness class, and the
increased knowledge, awareness, and value of relaxation techniques as an
element of fitness. Motivation appears to be increased, although it is still de-

batable whether this is specific to fitness classes or can be transferred to other
areas of training.
This study was the first of its kind in that it assessed the effect of supple-
mentary fitness training in dance across one year. Findings impacted upon
the school curriculum. The timetable now comprises weekly fitness classes.
The dance-specific fitness team aimed to counteract some possible injury risk
factors, such as lack of strength and jumping ability, through the introduction
of dance-specific strength work using partnering and plyometric jump train-
ing. Also specific short term demands have been addressed. For example,
groups of students have followed a prescribed training plan to prepare for a
three week historical project, dealing with specific pieces from Limon, Gra-
ham, and Cunningham repertory. It is hoped that this dance-specific fitness
training will have a more far-reaching effect than functional sports fitness
and provide a stimulus for students that can enhance the complete training
experience. It is also envisaged that other dance training institutions will con-
sider adopting a similar model so that findings can be shared and more effec-
tive training programs for dancers devised.
Emma Redding, Department of Dance Science, Trinity Laban Conservatoire of Music

and Dance, Creekside, London SE8 3DZ, UK; Email: e.redding@laban.org
References
Bompa T. O. (1999). Periodization. Champaign, Illinois, USA: Human Kinetics.

Chatfield S. J., Byrnes W. C., Lally D. A., and Rowe S. E. (1990). Cross-sectional physi-
ologic profiling of modern dancers. Dance Research Journal, 22, pp.13-20.
Cohen J. L, Segal K. R., Witriol R., and McArdle. W. D. (1982). Cardiorespiratory re-
sponses to ballet exercise and VO2 max of elite ballet dancers. Medicine and Science
in Sport and Exercise, 14, pp. 212-217.
Dahlstrom M., Inasio J., Jansson E., and Kaijser L. (1996). Physical fitness and physical
effort in dancers: A comparison of four major dance styles. Impulse, 4, pp. 193-209.
Novak L. P., Magill L. A., and Schutte J. E. (1978). Maximal oxygen intake and body
composition of female dancers. European Journal of Applied Physiology, 39, pp.
227-282.
Redding E. and Wyon M. A. (2001). A comparative analysis of the physiological re-
sponses to training and at the end of a performing period of two dance companies.
Paper presented at the Eleventh Annual Meeting of the International Association
of Dance Medicine and Science, Madrid, Spain.
Redding E. and Wyon M. A. (2003). Strengths and weaknesses of current methods for
evaluating the aerobic power of dancers. Journal of Dance Medicine and Science, 7,
pp. 10-16.
Rimmer J. H., Jay D., and Plowman S. A. (1994). Physiological characteristics of trained
dancers and intensity level of ballet class and rehearsal. Impulse, 2, pp. 97-105.
Rist R. A. (1994). Children and exercise: A dance medicine perspective. Sportcare Jour-
nal, 1, pp. 1-7.
Schantz P. G. and Astrand P. O. (1984). Physiological characteristics of classical ballet.
Medicine and Science in Sport and Exercise, 16, pp. 472-476.
Wanke E. M. (2001). Breathless: Results of a long term study of an integrated endur-
ance training in professional dancers. Paper presented at the Eleventh Annual
Meeting of the International Association of Dance Medicine and Science, Madrid,
Spain.
Wyon M. A. and Redding E. (2005). Physiological monitoring of cardiorespiratory ad-
aptations during rehearsal and performance of contemporary dance. Journal of
Strength and Conditioning Research, 19, pp. 611-614.
Wyon M. A., Redding E., Abt G. et al. (2003). Development, reliability and validity of a
multi-stage dance specific aerobic fitness test (DAFT). Journal of Dance Medicine
and Science, 7, pp. 80-84.
Developing evidence-based interventions to

enhance performance
Terry Clark and Aaron Williamon
Centre for Performance Science, Royal College of Music, London, UK
Increasingly, a broader range of training programs and evidence-based

interventions are being employed within music education contexts. It is
important that these programs are empirically tested to provide an unbi-
ased assessment of their impact and efficacy. This article highlights
methods for developing and testing intervention programs. A nine-week
musician-specific mental skills training program was developed and de-
livered to performance students from the Royal College of Music (RCM).
Pre- and post-testing involved a battery of questionnaires, public per-
formances, and performance-related tasks. Feedback from these students
was collected throughout concerning their views on the relevance and
usefulness of the program. In comparison with a control group, this ex-
perimental group demonstrated significant changes in their views toward
practice activities and specific practicing behaviors. A significant increase
in self-efficacy for performing was also found for the experimental group,
as was an increase in imagery vividness. These results are further sup-
ported by comments from students in the experimental group revealing
greater levels of self-awareness and confidence, facilitative views toward
and heightened control over anxiety, and healthier perspectives toward
music making. Feedback from the participants also provided insight into
issues pertaining to content and delivery.
Keywords: evidence-based training; imagery; mental skills; performance

science; music education
Training methods within the performing arts have typically been based upon
tradition and personal experience. In addition to these, within many areas of
performance, a broader range of programs are becoming increasingly em-
ployed with the aim of equipping performers with skills necessary for effec-
tive performance preparation and career management. However, some of
these programs are developed and implemented with little empirical support
for their efficacy. In dance, for instance, educators are being encouraged to
review and assess how training is provided (Kimmerle and Cte-Laurence
2003). From the viewpoint of researchers, performers, and those who train
performers, it is important that these programs be subjected to empirical
testing in order to provide an unbiased assessment of how they can be em-
ployed to enhance performance, not to mention a level of quality assurance
regarding their efficacy.
This paper highlights methods for empirically developing and testing in-
tervention programs with a focus on evidence-based outcomes. In particular,
a musician-specific mental skills training program is evaluated qualitatively
and quantitatively in terms of its potential to enhance learning and perform-
ing skills.
METHOD
Participants
Twenty-three undergraduate and postgraduate performance students were

recruited from the RCM. Of these, 14 were assigned to an experimental group
while 9 formed a control group. Ages ranged from 20-28 years (mean=22.3,
median=22.0, SD=2.2). In terms of the participants year of study, 4 were
year 1 undergraduates, 6 were in year 2, 5 were in year 3, 3 were in year 4,
and 5 were postgraduates. For their instrument groupings, 7 were pianists, 4
were vocalists, 7 were string players, and 5 were woodwind or brass players.
Materials
A battery of quantitative and qualitative methods was employed to assess the

training program. Quantitative measures addressed the participants practice
attitudes and behaviors (adapted from an interview schedule on self-regu-
lated learning by Zimmerman 1986), assessing how important the partici-
pants felt it was to engage in a variety of self-regulated learning behaviors as
well as how often they actually employed the different types of learning be-
haviors. The participants were also asked to self-rate their proficiency on a
selection of musical skills. In addition, imagery ability was assessed using the
randomized short version of Betts Questionnaire upon Mental Imagery (Betts
QMI, Sheehan 1967). Personality characteristics were assessed, including
self-efficacy for performing, in reference to both a live performance and a
hypothetical performance. The Revised Competitive State Anxiety Inventory
(CSAI-2R, Cox et al. 2003) was employed to measure self-confidence and
state anxiety prior to performing, while trait anxiety was measured using the
trait anxiety index from the Spielberger State-Trait Anxiety Inventory (TAI,
Spielberger et al. 1970). Lastly, each participant gave a 15-minute public per-
formance comprising two contrasting pieces of their choice in order to meas-
ure performance ability.
Qualitative methods involved feedback from the experimental group col-
lected during and following the training phase concerning students views on
the relevance and usefulness of the program, focus groups conducted with the
experimental group following the training and final testing phase, and case
notes kept by the first author throughout the project.
Procedure
Participants from both the experimental and control groups first attended a
session during which they completed most of the questionnaires. Following
this, each participant performed their chosen pieces in a public performance.
Immediately prior to performing, the participants completed the self-efficacy
for performing questionnaire and the CSAI-2R. The performances were
videoed and assessed for quality by two external adjudicators. Upon comple-
tion of the first round of testing, a musician-specific mental skills training
program was delivered to the experimental group. The topics covered fell in
three main categories: (1) motivation and effective practice, (2) relaxation and
arousal control, and (3) performance preparation and enhancement. The
training phase involved one 60-minute group session and one 30-minute
individual session per week for nine weeks. The control group, meanwhile,
received no additional training. Following the training phase, the experi-
mental and control groups again completed the questionnaires and public
performances, using a procedure identical to the first round of testing. De-
scriptive statistics for the questionnaires were initially calculated from the
two rounds of testing, following which repeated-measures analyses of vari-
ance (ANOVAs) were employed to elucidate changes occurring from one
testing phase to the next.
RESULTS
Quantitative results
The assessment protocol employed in this study yielded a number of signifi-

cant findings (see Table 1). In comparison with the control group, the experi-
mental group demonstrated significant changes in their views toward practice
activities and in their specific practicing behaviors. This was evidenced by a
Table 1. Significant results from repeated-measures ANOVAs. The mean scores (and
standard deviations) for the experimental (EG) and control groups (CG) for both phases
of testing are provided, as are the resulting F values.
Pre-test Post-test F
Self-regulated learning EC 48.69 (6.21) 48.69 (4.51) F1,21=6.22, p<0.01
CG 51.18 (6.34) 51.18 (4.90)
Quantity of practice EC 4.39 (1.15) 4.39 (1.35) F1,21=9.64, p<0.01
CG 4.27 (1.35) 4.27 (1.06)
Technical proficiency EC 4.81 (0.66) 4.81 (0.73) F1,21=5.59, p<0.01
CG 5.18 (0.75) 5.18 (0.83)
Self-efficacy EC 41.94 (5.09) 41.94 (5.03) F1,21=6.88, p<0.01
CG 45.00 (6.18) 45.00 (6.24)
Betts QMI: Total EC 105.44 (26.13) 105.44 (26.22) F1,21=9.26, p<0.01
CG 78.36 (16.63) 78.36 (21.92)
Betts QMI: Sensations EC 16.00 (6.09) 16.00 (4.80) F1,21=5.66, p<0.01
CG 11.09 (3.18) 11.09 (3.89)
Betts QMI: Touch EC 15.63 (4.65) 15.63 (5.43) F1,21=6.75, p<0.01
CG 10.64 (4.06) 10.64 (4.51)
significant increase in the experimental groups scores on the self-regulated

learning questionnaire compared with the control group. The increase in
practicing behaviors over the control group also emerged on the musical skills
items of quantity of practice and technical proficiency. Increased practice
effectiveness appeared to enhance the experimental groups self-efficacy as
well, as demonstrated by their significant increase beyond that of the control
group. The experimental groups imagery ability also increased significantly
over the control groups, specifically within the total score from the Betts
QMI, along with the scores for the subcomponents of touch and interoceptive
sensations.
Qualitative feedback
The results in Table 1 were further supported by comments from the partici-
pants revealing that, as a result of taking part in the training program, they
were experiencing improved practice efficiency, greater levels of self-aware-
ness and confidence when performing, facilitative views toward and a height-
ened sense of control over anxiety, and healthier perspectives toward music
making.
Feedback from the participants also provided insight into issues pertain-
ing to the content and delivery of the program. The participants felt that the
training program could have been strengthened with the inclusion of more
case studies and less information on research findings, more examples of
practical application, and a greater use of class discussion and activities to
facilitate learning from one another.
The case notes, as well as information derived from the follow-up focus
group, suggest that people may respond to a training program individually,
particularly in terms of their willingness to engage with some of the content
areas. Given this, prior to program implementation, it would seem instructive
to develop a sense of participants willingness to engage in change so that the
program could best be designed and delivered based on the participants
wants and needs.
DISCUSSION
While evidence-based training programs are commonplace within sport,

similar empirical backing is largely lacking for many programs within the
performing arts. Partly due to this, there is still little scientific evidence to
indicate the range of benefits performers stand to gain from such programs,
as well as the most efficacious means of providing training and education.
The mixture of quantitative and qualitative investigation employed in the
present study provides one means of addressing these questions.
In addition, a few issues arose as a result of the employment of this par-
ticular assessment protocol that are worth noting. Quantitative results de-
rived from questionnaires could potentially be somewhat misleading given
that the questionnaires employed may not be adequately sensitive, or even
appropriate, to pick up changes. This could account for the fact that while
anxiety intensity scores did not change significantly in the present investiga-
tion, the experimental group participants reported frequently how their per-
ceptions toward anxiety had changed. Moreover, while assessment of
performance quality is often important to many within performers training,
using it as a dependent measure to gauge program effectiveness could be
challenging, particularly given how slowly performance quality improves and
the somewhat subjective methods available for measuring improvement.
When employing new training methods and techniques, changes may be slow
to occur. This refers to both the participants willingness to engage in new
methods, as well as their abilities to engage with the methods presented. This
should be borne in mind by those implementing new interventions and
training programs.
In the present investigation, the importance of employing a mixed-meth-

ods approach to generate a more complete understanding of the effects of a
training program was reaffirmed. Through the knowledge gained, this study
aims to impact on performers training and education and, subsequently, on
their performance experiences. In doing so, it contributes to the growing body
of evidence-based research supporting novel programs for enhancing per-
formance.
Acknowledgments
We wish to thank Christopher Connolly, Caroline Marlow, and Melanie Gregg for their
assistance with the development of the mental skills training program.
Terry Clark, Centre for Performance Science, Royal College of Music, Prince Consort
Road, London SW7 2BS, UK; Email: tclark@rcm.ac.uk
References
Cox R., Martens M., and Russell W. (2003). Measuring anxiety in athletics: The Revised
Competitive State Anxiety Inventory-2. Journal of Sport and Exercise Psychology,
25, pp. 519-533.
Kimmerle M. and Cte-Laurence P. (2003). Teaching Dance Skills. Andover, New Jer-
sey, USA: Michael Ryan Publishing.
Sheehan P. (1967). A shortened form of Betts questionnaire upon mental imagery.
Journal of Clinical Psychology, 23, pp. 386-389.
Spielberger C. D., Gorsuch R. L., and Lushene R. F. (1970). STAI: Manual for the State-
Trait Anxiety Inventory. Palo Alto, California, USA: Consulting Psychologists
Press.
Zimmerman B. J. (1986). Becoming a self-regulated learner: Which are the key sub-
processes? Contemporary Educational Psychology, 11, pp. 307-313.
How can neuroscience help performers?
Eckart Altenmller1 and Hans-Christian Jabusch2
1 Institute of Music Physiology and Musicians Medicine, Hanover University

of Music and Drama, Germany
2 Institute of Musicians Medicine, Dresden University of Music
Carl Maria von Weber, Germany
Many neuroscientists are interested in musicians and in the neurobiology

of music perception and performance. This interest is usually motivated
by the attractiveness of the topic (music as an art) and by the enormous
effects of music on brain networks and brain morphology, demonstrating
the powerful mechanisms of brain plasticity in the short- and long-term
range. Therefore, most neuroscientists consider music as an excellent
paradigm to study brain mechanisms related to sensorimotor or percep-
tive learning. However, it remains open as to whether the growing body
of research in this area has been made fertile for musicians, for example
with respect to improvement of practice or teaching strategies. We report
new results of brain imaging studies focusing on sensorimotor integra-
tion while novices learn to play the piano. Interestingly, auditory-sen-
sorimotor integration can be established in less than 20 mins of piano
practice, demonstrating the dynamics of brain plasticity in this specific
task. Implications include the usefulness of pure auditory stimulation for
the acquisition of skilled finger movements. Furthermore, we review re-
cently published work on error monitoring in skilled pianists. This re-
search demonstrates that errors are already identified by the brain 50
ms prior to their actual execution. We discuss this finding with respect to
practical consequences concerning error avoidance. Based on these
studies, we demonstrate the utility of some neuroscience research for
musicians, particularly when the researchers themselves are trained mu-
sicians and work diligently to translate their findings from one discipline
to the other.
Keywords: neuroscience; auditory-sensorimotor integration; pianists;

focal dystonia; translational science
Eckart Altenmller, Institute of Music Physiology and Musicians Medicine, Hanover

University of Music and Drama, Hohenzollernstrasse 47, Hanover 30161, Germany;
Email: altenmueller@hmt-hannover.de
Thematic session:
Expression and interpretation I
Expressive timing: Martha Argerich plays

Chopins Prelude op. 28/4 in E minor
Olivier Senn, Lorenz Kilchenmann, and Marc-Antoine Camp
Institute for Music Performance Studies, Lucerne School of Music, Switzerland
This paper scrutinizes the temporal organization of Martha Argerichs

interpretation of Chopins Prelude op. 28/4 in E minor, recorded in 1975
for Deutsche Grammophon (DG 415 836-2). It proposes a method for
extracting the timing of the attacks from the audio signal, it visualizes the
data of bars 1-4, and maps Argerichs timing to Chopins composition in a
process of inverse interpretation.
Keywords: microtiming; interpretation; piano; Martha Argerich; Fr-

dric Chopin
When Western art music is performed, microrhythmic organization often

differs considerably from the notated rhythm of the score; these deviations
can be referred to as expressive timing (Clarke 1999) and have been studied
intensely in empirical rhythm research (see Pfleiderer 2006 for an overview).
There has been a considerable amount of research concerning performance
invariants in time organization; most of it analyzed performances of classi-
cal/romantic piano music on recordings. Several studies present evidence
that performers tend to slow down at the end of a melodic phrase (e.g. Repp
1998, Clarke 1999) and that melody voices anticipate accompanying voices by
a few milliseconds (melody lead, e.g. Palmer 1989, Goebl 2001). Most con-
clusions endorse a general assumption in music performance research that
expressive timing relates closely to structural features of a composition (e.g.
Clarke 1988, Palmer 1996).
Against this background of invariant performance behavior, we would like
to analyze one particular recording as an individual response to a composi-
tion: between 22-25 October 1975, Martha Argerich recorded Frdric Cho-
pins Preludes op. 28 in Munich for Deutsche Grammophon (catalogue
number DG 415 836-2). Our analysis in this paper concentrates on the tem-
poral organization of the E minor Prelude op. 28/4, bars 1-4 (although meas-
urements for the whole piece have been made). It divides the fabric of the
Prelude into two rhythmic layers: the solo voice in the right hand and the
sequence of accompanying chords in the left hand each have their own
rhythmic structure (see Figure 2) and are analyzed separately. Finally, the
timing phenomena are related to the score in a speculative process of inverse
interpretation: inspired by Timmers et al. (2000), we assume that there is a
great number of possible ways to analyze the structure of a composition. We
do not analyze the compositional structure first, and map it to the perform-
ance data later, but the other way around. We inspect the timing data and try
to separate expected phenomena from Argerichs individual approach. Then,
the score of the piece is analyzed in order to find the properties that are high-
lighted by Argerichs interpretation. This analysis is done strictly from a lis-
teners point of view: we are not trying to reconstruct Argerichs thoughts or
intentions. But we do try to articulate which sense or meaning weas listen-
ersmight discover in Argerichs interpretation of Chopins piece.
MAIN CONTRIBUTION
Method
Our method to secure tone onset times seeks to meet two requirements: it
must allow us to differentiate tone onsets in two separate rhythmic layers
(solo voice and accompaniment), and the accuracy of the measurements must
be within 10 ms. Our approach implies two steps. First, we marked the onsets
roughly in a sonogram (window size=4410 samples, hop size=441 samples);
this places the timing in a range of 20 ms near the attack time. Since the
notes of the accompaniment chords are not necessarily struck at the same
time, only the onset of the loudest note of each chord was measured. In a
second step we adjusted the measurements with the help of filters and two
loudness measurements: a narrow bandpass filter (bandwith=100 Hz) was
employed to isolate the fundamental tone (or, if necessary, one of the first
partials) of a note; the delay caused by the filtering was compensated. Then,
the behavior of two loudness measurements was interpreted in order to de-
tect the precise timing of the onset. It was set to the moment when a fast peak
level (window size=1 ms) overshoots an RMS level (27 ms) which reacts early
but slowly to an energy surge. First tests with monophonic piano examples
suggest that this method places the timing markers within a range of 2-12 ms
after the physical onset; more systematic tests need yet to be conducted in
order to evaluate this (rather conservative) estimate. All measurements and
the visualizations of timing data in Figure 2 were made with the Lucerne Au-
dio Recording Analyzer (software freely available at www.hslu.ch/lara).
Figure 1. Average distribution of bar duration on the 8 eighth notes in the accompani-
ment in % (calculated with timing data from bars 1-11 and 13-22).
Analysis
In bars 1-11 and 13-22, the accompaniment of Chopins Prelude op. 28/4 con-
sists of a rhythmically uniform sequence of chords, pulsating in eighth notes.
Argerich plays these eighth notes with flexible rhythm; nevertheless, a gen-
eral bar profile is discernible.
Figure 1 shows the mean distribution of bar durations among the 8 eighth
notes of a bar throughout the whole piece (in % of bar duration). Two ritar-
dando zones can be identified: one strong ritardando spans over the bar line,
from eighth note 7 (13.1% of bar time) and the longest 8 (15.2%) to 1 (14.2%).
We would like to call this formation bar line ritardando. Another, much
more subtle lengthening concerns eighth note 5 (12.7%). We call this the
mid-bar ritardando. Besides the performance invariants reported in the
research literature (slowing down on phrase endings, melody lead), this aver-
age bar profile can serve us as a background for identifying Argerichs indi-
vidual reactions to particular configurations within the composition.
Figure 2 presents Argerichs rhythmic disposition of bars 1-4 alongside
the respective excerpt of the score. The rhythm diagram just above the score
(L for left hand) represents the timing of the accompaniment: every vertical
line on the horizontal timeline marks the timing of a detected onset. The
width (and consequently the height) of the squares represent the inter-onset-
intervals (IOI) on different metrical levels of the score. The smallest squares
denote IOIs between neighboring eighth notes, the biggest squares denote
IOIs between successive downbeatsi.e. they represent bar durations. All
IOIs are given as a tiny numerical value (in ms) in the upper part of each
square. The top rhythm diagram in Figure 2 (R for right hand) represents
the timing of the solo melody: the smaller squares stand for the anacrusis, the
dotted half notes, and the quarter notes, respectively. The biggest squares
visualize bar durations analogous to the lower diagram. Both diagrams (R+L)
are adjusted horizontally to the same timeline. The numbers between the
Figure 2. Rhythm diagram and score (according to G. Henle edition, 1968), bars 1-4.
diagrams represent differences between onsets in the right and the left hand,
whichaccording to the scoreoccupy the same metric position. Most of the
time, Argerich plays the solo voices onsets earlier than the accompaniments:
in bars 1-4, the melody leads in seven of nine instances (78%), the average
melody lead is 28 ms.
Argerich takes up the slow tempo of the anacrusis in the accompaniment.
In bar 1, she starts with a quite long first eighth note (807 ms) which takes
17.2% of bar duration; then she speeds up. No mid-bar ritardando is discerni-
ble in bar 1, but the bar line ritardando (eighth notes 7 and 8) is distinct. In
bar 2, Argerich starts fast but slows down considerably in the second half of
the bar. This tendency is even stronger in bar 3: the first half of bar 3 is fast
(28.7 bpm) and takes only 43.6% of bar duration (2092 ms), whereas the
second half is considerably slower (22.2 bpm) and lasts for 56.4% of bar du-
ration (2704 ms).
Which properties in the score might trigger the rhythmic phenomena of

bars 1-3? Our attempt at inverse interpretation focuses on the repetitive pat-
tern of the solo voice: the piece receives a first impulse from the octave leap of
the anacrusis. Then the solo melody seems to be stuck in repetitionthe mo-
tif C5-B4 is repeated three times in an identical way. With her fine sense of
dramaturgy, Argerich seems to make a new effort to gain some speed with
every new bar (hence the faster first halves), but this effort is lost as the bar
progresses, the tension relaxes and the music seems to stagnate.
The situation changes completely in bar 4: the bar line ritardando is ab-
sent, but the mid-bar ritardando appears for the first time and very distinc-
tively on bar 4, eighth note 5 (704 ms, 15.8%). Why is that? At the end of bar
4, the solo melody progresses from the repeated melodic pattern C5-B4it
descends one tone to Bb4-A4. Following the research literature, we would
expect the tempo to slow down at the end of bar 4, which is the end of the first
4-bar period. But the contrary is the case: after the mid-bar ritardando of bar
4, Argerich speeds up again. This development makes sense, if we take into
account that the quarter notes of the melody are most likely heard as ana-
cruses to the dotted half notes on the downbeats. The Bb4 quarter note in the
solo melody at the end of bar 4, thus, does not belong to the first 4-bar period
but to the next formal entity. With the pronounced mid-bar ritardando of bar
4, Argerich slows down at the end of the first period just before this anacrus-
tic Bb4, and then gathers speed for the second period: the melodic anacrusis
to bar 5 gives a new impulse to the piece, and the rhythm of the accompani-
ment supports this impulse. Compared with bars 3 (25.1 bpm) and 4 (27.0
bpm), bar 5 speeds up considerably (31.8 bpm).
IMPLICATIONS
Using time measurements, we described the microrhythmic fabric of the first

4-bar passage in Martha Argerichs 1975 recording of Chopins Prelude op.
28/4. The specificity of her rhythmic interpretation was established against
the background of rhythm performance invariants (as reported in the re-
search literature) and an averaged profile of the eighth note lengths in her
recording. Then, in a process of inverse interpretation, we have tried to figure
out constellations in the score that might have triggered Argerichs specific
approach. This procedure casts an analytical light on the composition through
the lens of a concrete performance. It is no news that we can learn a great
deal about structural features of a composition by listening to an eminent
performer. Inverse interpretation attempts to transfer some of this intuitive
knowledge into the explicit domain of music performance studies.
Acknowledgments
We would like to thank John Rink for giving the first idea to do this paper, Claudia
Emmenegger for typesetting the music example, and Richard Beaudoin for his careful
proofreading and inspiring collaboration. Many thanks go to the faculty and the stu-
dents of the Department of Music at Harvard University for their constructive com-
ments on an early draft of this paper.
Olivier Senn, Institute for Music Performance Studies, Lucerne School of Music, Zen-
tralstrasse 18, 6003 Lucerne, Switzerland; Email: olivier.senn@hslu.ch
References
Clarke E. F. (1988). Generative principles in music performance. In J. A. Sloboda (ed.),

Generative Processes in Music (pp. 1-26). Oxford: Clarendon Press.
Clarke E. F. (1999). Rhythm and timing in music. In D. Deutsch (ed.), The Psychology
of Music (pp. 473-500). San Diego, California, USA: Academic Press.
Goebl W. (2001). Melody lead in piano performance: Expressive device or artifact?
Journal of the Acoustical Society of America, 110, pp. 563-572.
Goebl W. and Palmer C. (2009). Synchronization of timing and motion among per-
forming musicians. Music Perception, 26, pp. 427-438.
Palmer C. (1989). Mapping musical thought to musical performance. Journal of Ex-
perimental Psychology: Human Perception and Performance, 15, pp. 331-346.
Palmer C. (1996). On the assignment of structure in music performance. Music Percep-
tion, 14, pp. 23-56.
Pfleiderer M. (2006). Rhythmus: Psychologische, Theoretische und Stilanalytische
Aspekte Populrer Musik. Bielefeld, Germany: Transcript Verlag.
Repp B. H. (1998). Obligatory expectations of expressive timing induced by perception
of musical structure. Psychological Research, 61, pp. 33-43.
Timmers R., Ashley R. D., Desain P., and Heijink H (2000). The influence of musical
context on tempo rubato. Journal of New Music Research, 29, pp. 131-158.
Controlling the pacing of retards and

accelerandos in piano performance: A roller
coaster solution?
Cristine MacKie
Department of Music, Royal Holloway, University of London, UK
Performers of a musical work generally well understand how to control

steady tempos carried by the musical beat. More problematic for the
performer, however, is controlling the graduated pacing of flexible tem-
pos such as retards and accelerandos that depart from, yet link, the es-
tablished tempos. According to Epstein (1995), the pacing of retards and
accelerandos in performance is among the least accomplished events in
our concert halls; indeed it is often left to chance. In this article, I pro-
pose that the performer may control the pacing of the ascent and descent
of the melodic curves which characterize Bruyres, a prelude from Book
2 for piano by Debussy, by using a subtle application of retards and ac-
celerandos following the principle of the roller coaster. However, the re-
tards and accelerandos, like the roller coaster, also require some means
of support. I shall suggest too that Debussys deployment of cadences is
intended to give support to that pacing. But how may the performer
achieve this?
Keywords: piano performance; accelerations; retards; cubic curve; roller

coaster
There are few systematic studies on either retards or accelerations. While

Sundberg and Verillo (1980) have dealt with the retard as a gesture during
the closure of a work, perhaps more relevantly Epstein (1995) has proposed in
an empirical study that retards or accelerations may be best performed if they
fit within a mathematical equation known as the cubic curve. However, this is
a pilot study that does not investigate how retards and accelerations, which
are inverse forms of each other, may be paced smoothly one to the other. Ki-
netic and potential energywhich are exemplified in the pacing of the retar-
dations and accelerations of the roller coastermay provide, in part, a
simpler solution, as will certain aspects of the rethink which has taken place
in recent years in musical academic circles, notably Berry (1989) and Rink
(2002) who have looked afresh at the relationship between performance and
analysis. These changes have made possible the informal analysis at the end
of this article, which seeks to demonstrate that the graduated pacing of the
ascent and descent of the melodic curves in Bruyres requires a subtle appli-
cation of retards and accelerations following the principle of the roller
coaster. Informal tools of analysis such as dcoupage, diagrams, and histori-
cal evidence are used to support this approach.
MAIN CONTRIBUTION
The cubic curve
Retards and accelerations are governed by the principle of graduated change

in the pacing of motion. As such, they do not proceed toward ad hoc goal
tempos but rather link tempos that are already predetermined. Describing
the retard/acceleration process, Epstein (1995) states that:
The initial tempo from which the change starts, the number of steps
though which the change must move, a related tempo to be achieved at
the end of the time series: these elements set the temporal framework
within which an retard/acceleration must be shaped (Epstein 1995, p.
418).
He suggests, also, that the cubic curve, while mathematically complex, can
describe an act of timing such as the retard and acceleration which not only
feels natural and seemingly effortless to effect, but is compatible with our
own musical intuitions (Epstein 1995; see Figure 1 below). From this
mathematical concept, Epstein extrapolates the idea that the shape of the
curve could represent a possible timing trajectory in the moulding of a retard
and accelerando (p. 419).
In this pilot study, Epstein transferred performances of retards by Szell
and Stravinsky and accelerations by Karajan and Herrara de la Fuente onto
an analog audio tape and timed the durations of their successive beats with
the tape measure technique (p. 417). This enabled him to assemble a highly
correlated model of a retard and accelerando performance from which he
observed that their trajectory fitted within the model of the cubic curve with
striking similarity. This, he suggests, means that they all had a good compre-
hension of the boundary tempos that frame the timing path of their re-
tards and accelerations. Finally, a computer programme revealed that the
y z
w x
Figure 1. A reproduction of Epsteins (1995) diagram which shows: first, two horizontal
lines between w-x and y-z. These lines represent the steady, or predetermined, tempos.
Second, an s-shaped ascending curve links the horizontal line w and x to the horizontal
line at the uppermost point of the curve marked y and z.
Retardation Acceleration
Figure 2. A general representation of the curve of a roller coaster. The arrow on the left
shows the path of the retard of the roller coaster as it ascends toward the peak of the
curve. The arrow on the right shows the path of acceleration of the roller coaster as it
descends on the downward side of the curve.
retards and accelerandos measured in the study consistently fitted within the
cubic curve model with less than 7% variance. Epstein concluded that the
cubic model does about as well as it is possible to do (p. 423).
The roller coaster
Since the time constraints on a performers life may discourage the study of
the models described above in detail, it may be more effective for him or her
to observe the similarities between the shape of a melodic curve in Bruyres
and the path of the roller coaster as it rises and falls (this being closely related
to the cubic curve). The energy of the roller coaster, during its fall, is trans-
formed from potential energy to kinetic energy (energy in motion) and, dur-
ing its rise, from kinetic energy to potential energy (stored energy). Thus,
during the ascent of the roller coaster there is a retardation as it draws toward
the top of the curve, and an acceleration as it descends on the other side of
the curve (see Figure 2).
Figure 3. An informal analysis of bars 9-13 of Bruyres by Debussy. (1) Above the score,
the contour of the melodic curve is highlighted with a black line to expose it more
clearly, (2) level 1 shows the prolonged cadential memory device, (3) level 2 shows the
deployment of the cadences irregulires at an intermediate level of hearing, (4) the
dotted lines on both sides below level 2 are an approximation of the area of the prede-
termined tempo, 66 to the quarter note.
Figure 4. The curve represents the path of the roller coaster as it ascends and descends.
The vertical lines below represent the hypothetical cadential points of support.
The analysis
This informal analysis seeks to expose elements that will support the pacing
of the ascent and descent of the melodic curves in Bruyres. Since the aim is
to apply a subtle application of retards and accelerations following the princi-
ple of the roller coaster, it will, first, expose the shape of the melodic curve
between bars 9-13, since curves not angles are the very nature of this compo-
sition (Schmitz 1950, p. 172), and second, demonstrate that the role of the
perfect cadence and the cadences irregulires support the pacing of the
curve. Dcoupage and a diagram will be employed as part of the analytical
process. For example, in bars 9-13, the vertical bar lines have been removed
to expose a complete melodic curve. Its contour is highlighted above with a
black line (see Figure 3) to reveal the similarity of the rise and fall of the mu-
sical curve with the ascending and descending path of the roller coaster. Thus,
just as the energy of the roller coaster is transformed from kinetic to potential
energy as it draws toward the peak of the curve, bars 9-10 may be performed
with a subtle departure from the predetermined tempo of approx 66 to the
quarter note, and a retardation applied toward the Eb at the top of the me-
lodic curve.
In the same way in which the roller coaster uses kinetic energy during its
descent of the curve to fuel the rest of the ride (Epstein 1995, p. 27), per-
formers may plan their descent of the curve with a graduated accelerative
pacing through bars 10-13 and gradually rejoin the predetermined tempo of
66 to the quarter note smoothly in bar 13.
However, the pacing of the ritardando and accelerando, like the roller
coaster, needs some means of support or stability. This article suggests that
this is provided by Debussys deployment of a hierarchical arrangement of
cadential points (see Figure 4).
At level 1 (see bars 10-13, Figure 3), the function of the perfect cadence
may be seen as a prolonged cadential memory device which serves to main-
tain a sense of unity in a work that otherwise might be heard as a series of
fragmentary and isolated melodic curves. The dominant chord of the cadence
is placed under the highest point of the melodic curve on the first beat of bar
10, and it does not resolve onto the tonic, Ab major, until the first beat of bar
13. This, alongside the added energy of the accacciatura in the base, contrib-
utes stability to the pacing of the retardation, pushing the music up and for-
ward to the top of the melodic curve like the car on the roller coaster. The
positioning of the tonic chord Ab on the first beat of bar 13 enhances the
sense of closure as the car descends to the end of the melodic curve.
At level 2 (see Figure 3), the cadences irregulires are identifiable by the
harmonic progression using the subdominant with an added sixth, which
then resolves onto the tonic. These may be seen first between the third beat of
bar 8 and the first beat of bar 9 (see Figure 3) and similarly between bars 12
and 13. In both instances, their function is to support the pacing of a subtle
retard toward the end of the descent of the melodic curve.
Conclusion
The similarity of the melodic curves in Bruyres with the roller coasters track
and the degree of slope and corresponding retardation and acceleration with
its coefficient of friction at all points along the way (Epstein 1995, p.27),
which in this case is supported by two different types of cadences, is clear to
see. If the performer plans and controls this recurring accumulation and re-
lease of tension, which together affect the pacing of the retardation and accel-
eration during the ascent and descent of the melodic curve, this will, as
Epstein points out lead the car to its designated end, its motion controlled
throughout (1995, p. 27). If not, the pacing of the ascent and descent of the
melodic curve, like the car of the roller coaster, may be spent too soon and the
intended musical effect, which is to control a graduated pacing of the ascent
and descent of the melodic curve from the predetermined tempo of approxi-
mately 66 to the quarter note, to the same related tempo at the downward
end of curve may be dissipated.
IMPLICATIONS
Future research may consider associating the graduated pacing of retards and
accelerandos in piano performance with the kinetics of body movements. For
example, bodily movements may be organized to reduce any angular move-
ments in performance by planning ideal trajectories of bodily motion that
take into account the demands of the musical score.
Acknowledgments
I wish to acknowledge Euan MacKie and Bruce Paterson for their invaluable comments
and proof reading.
Cristine MacKie, Department of Music, University of London, Egham, Surrey TW20

OEX, UK; Email: c.d.mackie@rhul.ac.uk, mackie_cristine@hotmail.com
References
Debussy C. (1913). Prelude Book 2, No. 5 Bruyres. London: Boosey and Hawkes.
Epstein D. (1995). Shaping Time. New York: Schirmer Books.
Rink J. (1995). The Practice of Performance. Cambridge: Cambridge University Press.
Schmitz R. (1966). The Piano Works of Claude Debussy. New York: Dover Publications.
Sunberg V. and Verillo V. (1980). On the anatomy of the retard: A study of timing in
music. Journal of the Acoustical Society of America, 68, pp. 772-779.
Saltzer F. (1982). Structural Hearing (volumes 1 and 2). New York: Dover Publications.
Maintaining skill across the life span:

Magaloffs entire Chopin at age 77
Sebastian Flossmann1, Werner Goebl1, and Gerhard Widmer1,2
1 Department of Computational Perception, Johannes Kepler University Linz, Austria

2 Austrian Research Institute for Artificial Intelligence, Vienna, Austria
The study is based on a corpus containing the entire works of Chopin

performed by Nikita Magaloff at the age of 77, precisely measured and
fully annotated with score information. On this data, we test a model of
successful aging including selection, optimization, and compensation hy-
potheses (SOC). We identify performance errors, compare Magaloffs
etudes with recordings by 14 other renowned pianists, and investigate
specific age effects in a selected nocturne in 14 different recordings.
Keywords: performance errors; symbolic data; SOC model; aging virtu-

osity; piano performance
Many renowned pianists perform with great success up to old ages (e.g.
Backhaus played his last concert at 85, Horowitz at 84, Arrau at 88). The de-
mands posed by performing publicly are enormous (motor skills, memory,
physical endurance, stress factors; see Williamon 2004). Theories of human
life-span development identify three factors to be mainly responsible for
successful aging: selection, optimization, and compensation (SOC model,
Baltes and Baltes 1990). Applied to piano performance this would imply that
older pianists play a smaller repertoire (selection), practice these few pieces
more (optimization), and hide technical deficiencies by reducing the tempo of
fast passages while maintaining tempo contrasts between fast and slow pas-
sages (compensation) (Vitouch 2005).
In this study, we examine a unique corpus of Chopin performances by
Nikita Magaloff, recorded on stage at age 77. We test whether Magaloff actu-
ally used strategies identified in the SOC model to master this unprecedented
project. First, we assess his performance by quantifying performance errors.
Second, we analyze recordings of the etudes by other renowned pianists to
test whether Magaloffs performance tempi were slower than those of the
others. Finally, we examine whether tempo contrasts are maintained when

fast sections are performed slower at older ages by analyzing recordings of the
Nocturne Op.15 No.1 (Andante cantabile), which contains a fast, technically
demanding middle section (con fuoco).
METHOD
Materials
In Spring 1989, Magaloff performed the entire work of Chopin for solo piano
that was published during Chopins lifetime (Op.1-64) in six public appear-
ances at the Vienna Konzerthaus. These concerts were recorded with a
Bsendorfer computer-controlled grand piano that provides a huge set of
symbolic performance data with highest precision156 pieces over 320,000
performed notes; about 10 hours of performed music.
To put Magaloffs etudes performances into context, recordings of the
etudes by the following performers were also analyzed (a total of 289 per-
formances): Arrau (recorded 1956), Ashkenazy (1975), Backhaus (1928), Biret
(1990), Cortot (1934), Gavrilov (1985), Giusiano (2006), Harasiewicz (1961),
Lortie (1986), Lugansky (1999), Magaloff (1975), Magaloff (1989), Pollini
(1972), Schirmer (2003), Shaboyan (2007), and Sokolov (1985).
The 14 recordings of the Nocturne Op.15 No.1 were by Argerich (1965),
Arrau (1978), Ashkenazy (1985), Barenboim (1981), Harasiewicz (1961),
Horowitz (1957), Leonskaja (1992), Maisenberg (1995), Magaloff (1975),
Perahia (1994), Pires (96), Pollini (68), Richter (68), and Rubinstein (1965).
Procedure
To make Magaloffs performances accessible for analysis, the entire Chopin

scores were scanned (946 pages) and subsequently converted into a digital
format (musicXML) using a commercial optical music recognition software
and custom-made post-correction steps. The data from Magaloffs perform-
ances were then semi-automatically matched to the symbolic scores, building
a huge corpus with precise performance information for all score notes and
vice-versa. Based on the alignment, performance errors were categorized as
insertion, deletion, or substitution errors. We extracted basic tempo values
(see Note) of Magaloffs performances of the etudes Op.10 and Op.25 in order
to compare them with recordings by the other famous pianists. These audio
recordings were semi-automatically beat-tracked using the software Beatroot
(Dixon 2007) to determine the expressive timing at the beat level; tempo
values were then extracted as before.
Table 1. Error % by piece category and error type (i.e. insertion, deletion, substitution).
Ins. Del. Sub. Ins. Del. Sub.

Rondi 1.86 2.40 2.50 Polonaises 5.74 4.09 1.54
Sonatas 4.20 3.63 1.82 Preludes 3.38 2.97 1.56
Mazurkas 2.44 3.41 1.00 Impromptus 1.36 2.12 0.89
Nocturnes 2.22 2.46 0.99 Scherzi 6.15 2.97 1.63
Etudes 3.90 3.94 1.33 Ballades 5.00 2.33 1.23
Waltzes 2.48 3.53 1.26 Pieces 4.36 3.49 2.27
RESULTS
Performance errors
Overall, Magaloffs data contained 3.73% insertion, 3.28% deletion, and

1.52% substitution errors. This is slightly higher than Repps (1996) account
for other pianists (1.48%, 0.98%, and 0.21%, respectively), but comparing the
particular piece used by Repp (Op.28/15), the error percentages were similar.
With a percentage higher than 5%, the scherzi, ballades, and polonaises stand
out in terms of insertion errors (see Table 1). The Allegro de Concert Op.20 in
the category pieces shows an exceptionally high insertion percentage
(6.77%). With an insertion percentage below 2.3%, the nocturnes, rondi, and
impromptus constitute the low-insertion categories. The impromptus are also
the category with the lowest percentage of deletion errors (2.12%), while the
etudes and polonaises exhibit the highest percentage of deletions.
Performance tempo of etudes
Table 2 shows the tempo modes obtained for all pianists. Each performance is
named by the first two letters of the pianist, followed by the pianists age at
the time of the recording. For the sake of comparison the metronome indica-
tions from the Henle Edition (Zimmermann 1983) were added (HEN). In 12
of the 18 pieces, Magaloffs tempo (MA) is within a 10% range of the Henle
indications. Three pieces are more than 5% slower and three pieces more
than 5% faster compared with the metronome markings. Compared with the
performances of 14 other recordings (including an earlier performance by
Magaloff in 1975) Magaloffs performances of the Op.10 etudes are on average
1.2% slower than the average over all other recordings. The Op.25 etudes are
on average about 5.6% slower than the average performance.
Comparing Magaloffs recordings at the age of 63 and 77, the tempi vary
to a surprising degree, but no systematic tempo decrease in the latter can be
Table 2. Tempo modes of different pianist for selected pieces from Op.10 and Op.25.
Entries are named by the first two letters of the pianists name and age at recording.
Op.10/1 Op.10/2 Op.10/4 Op.10/10 Op.10/12 Op.25/1

BI49 157 BI49 129 HA29 157 BI49 426 PO30 64 HA29 77
HA29 159 MA77 139 BI49 157 BA44 450 LO27 64 AS38 84
SH32 163 SH32 140 AR53 161 MA63 467 MA63 65 LO27 91
CO56 164 HEN 144 SC31 165 SC31 471 SC31 66 LU27 93
MA63 165 HA29 145 MA63 166 HEN 480 LU27 66 SO35 94
SC31 169 MA63 145 SH32 169 SH32 480 AS38 66 GA30 102
AS38 170 CO56 149 LO27 169 AR53 483 HA29 68 MA63 102
MA77 170 AR53 152 PO30 169 LU27 487 BA44 71 BI49 103
HEN 176 SC31 152 MA77 170 HA29 505 SH32 71 HEN 104
PO30 178 PO30 152 GI33 174 GA30 508 MA77 72 MA77 104
LO27 179 LO27 156 AS38 174 AS38 512 BI49 74 AR53 104
BA44 179 AS38 157 CO56 175 PO30 513 CO56 75 GI33 105
LU27 180 LU27 159 HEN 176 LO27 529 HEN 76 BA44 109
GA30 190 GI33 165 LU27 179 CO56 542 GI33 77 PO30 111
GI33 191 GA30 173 BA44 191 MA77 550 GA30 87 CO57 118
AR53 196 BA44 176 GA30 197 GI33 574 AR53 88
Op.25/6 Op.25/8 Op.25/9 Op.25/10 Op.25/10 Op.25/12

HEN 69 BI49 64 BI49 94 MA77 64- 90- 65 HA29 51 HA29 58
MA63 70 HA29 66 HA29 104 BI49 64- 106- 68 BI49 53 MA77 62
BI49 71 HEN 69 AR53 107 LO27 67- 86- 70 MA63 58 MA63 69
AR53 71 GA30 69 MA77 107 BA44 71- 112- 70 GI33 59 AS38 70
CO57 73 MA63 69 LU27 107 AR53 71- 96- 68 MA77 60 LO27 73
PO30 74 AR53 70 CO57 110 AS38 71- 84- 70 LO27 61 CO57 73
BA44 74 LO27 71 HEN 112 MA63 71- 100- 70 CO57 61 BI49 74
MA77 75 MA77 71 PO30 113 CO57 71- 127- 71 AS38 62 GI33 74
AS38 75 CO57 73 MA63 115 HEN 72- 126- 72 LU27 63 SO35 76
HA29 75 GI33 73 GI33 117 PO30 72- 104- 74 PO30 63 LU27 76
LO27 77 AS38 73 LO27 118 GI33 74- 129- 73 AR53 63 PO30 76
GI33 78 PO30 76 GA30 120 HA29 74- 112- 76 SO35 66 AR53 77
LU27 83 LU27 77 AS38 125 LU27 75- 96- 71 HEN 69 HEN 80
GA30 84 BA44 78 SO35 125 SO35 83-86- 87 BA44 69 BA44 82
SO35 85 SO35 81 BA44 131 GA30 86-117- 81 GA30 71 GA30 83
Nocturne Op. 15 No. 1

100 MA77 2
ARG24
Tempo Ratio (II to I)

90 PO26 1.8
Tempo II (bpm)
HEN
MA77
1.6 PO26
80 AR75
RU78
1.4
RU78
70 HEN
1.2
ARG24
60 AR75 1
r=0.621* n=14 p=0.0178 r=0.086 n=14 p=0.7707
50 0.8
20 40 60 80 20 40 60 80
Age (years) Age (years)
Figure 1. Nocturne Op.15 No.1 by 14 pianists and Magaloff: basic tempo of middle sec-
tion (left) and tempo ratio between middle and first section (right) against performers
age. Dashed lines indicate given tempo (left) or tempo ratio (right) by Henle edition.
found. On the contrary, in 12 pieces out of 18, the recording at age 77 is faster,
sometimes to a considerable degree (up to 17% in Op.10 No.10). On the
whole, no significant correlation of age and tempo could be established.
Age effects and tempo contrast in a nocturne
For an exemplary piece containing tempo contrasts, we examined the tempo

values in performances of the Nocturne Op.15 No.1 by 14 other pianists. We
found a significant correlation between the performance tempo of the middle
section and the age of the performer (the older, the slower; see Figure 1).
However, the tempo ratios between the contrasting sections of the piece
showed no overall age effect, confirming Vitouchs (2005) interpretation of
the SOC model. Age seemed to have no effect on Magaloffs nocturne; he
played faster than the youngest of the performers while keeping a comparable
tempo ratio. The same tendency could be found in Op.25 No.10; however, the
negative correlation was not significant.
DISCUSSION
Based on the fact that Magaloff performed the entire piano works by Chopin,
we can refute the selection part of the SOC model. Due to missing informa-
tion about his practice regime before and during the performance period, we
cannot make a statement about optimization processes. Magaloffs tempi do
not point to compensation processes, which were indeed found with other
famous pianists. However, his relatively high error rates may indicate that
Magaloff aimed at realizing his musical ideas of Chopins work rather than at
error-free performances. In sum, Magaloffs data does not seem to corrobo-
rate the SOC model. This study is the first of its kind to examine a huge cor-
pus of symbolic performance data of the entire work of a composer and to put
it into context of a substantial number of other recordings.
Note
A basic tempo value was estimated by the mode value, the most frequent bin of an in-
terbeat interval histogram with a bin size of 4% of the mean inter-beat interval.
Acknowledgments
Funded by the Austrian National Research Fund (FWF), project no. P19349-N15.
Sebastian Flossmann, Dept of Computational Perception, Johannes Kepler University

Linz, Altenberger Strasse 69, Linz 4040, Austria; Email: sebastian.flossmann@jku.at
References
Baltes P. B. and Baltes M. M. (1990). Psychological perspectives on successful aging:

The model of selective optimization with compensation. In P. B. Baltes and M. M.
Baltes (eds.), Successful Aging (pp. 1-34). Cambridge: Cambridge University Press.
Dixon S. (2007). Evaluation of the audio beat tracking system BeatRoot. Journal of
New Music Research, 36, pp. 39-50.
Goebl W., Pampalk E., and Widmer G. (2004). Exploring expressive performance tra-
jectories: Six famous pianists play six Chopin pieces. Proceedings of the Eighth In-
ternational Conference on Music Perception and Cognition (pp. 505-509),
Adelaide, Australia: Causal Productions.
Goebl W., Flossmann S., and Widmer G. (2009). Computational investigations into
between-hand synchronization in piano playing: Magaloffs complete Chopin. Pro-
ceedings of the Sixth Sound and Music Computing Conference (pp. 291-296),
Porto, Portugal: Casa da Msica.
Repp B. H. (1996). The art of inaccuracy: Why pianists errors are difficult to hear. Mu-
sic Perception, 14, pp. 161-184.
Vitouch O. (2005). Erwerb musikalischer Expertise [Acquisition of musical expertise].
In T. H. Stoffer and R. Oerter (eds.), Allgemeine Musikpsychologie (Enzyklopdie
der Psychologie) (vol. D/VII/1, pp. 657-715). Gttingen, Germany: Hogrefe.
Williamon A. (2004). Musical Excellence. Oxford: Oxford University Press.
Zimmermann E. (1983). Chopin Etden, Urtext. Munich, Germany: G. Henle Verlag.
Thematic session:
Expertise development
Strategies for achieving performing excellence

of twentieth and twenty-first century art song
Cathy Aggett
School of Communication Arts, University of Western Sydney, Australia
Strategies professional singers use in the selection and preparation of

twentieth and twenty-first century art song were gathered from an inter-
national study carried out by email during 2006-07. These strategies
were applied in the preparation of 34 Australian art songs for a recital in
2008 by three professional singers in a practice-led research project in-
volving the author. The 84 resulting performance and learning strategies
from the professional singers study are discussed in relation to the lit-
erature and their application by the three singers who strove for per-
formance excellence in their preparation and performance of a recital of
these Australian art songs. The application of how these strategies can
ultimately benefit singers and vocal teachers learning and performing
twentieth and twenty-first century art song is discussed, along with the
categorization of the strategies with reference to the literature.
Keywords: practice strategies; learning strategies; art song; practice-led

research; performing excellence
The challenges singers face in relation to performance excellence of twentieth

and twenty-first century art song are seldom discussed in the literature. Sing-
ers perform with a partneran accompanistand must together interpret the
text of the song they perform (Fine and Ginsborg 2007). The demands mod-
ern composers have placed on performers involve challenges in rhythm, new
vocal techniques, and often new notation approaches in scores, to name but a
few (Mabry 2002). While each performer will rely to a certain extent on their
intuition and innate musical talent (Rink 2002), various skills involving the
preparation of the music can be learned and may go towards enhancing the
singers path in achieving perceived excellence.
Within the scope of practice (Hallam 1997) and learning strategies (Niel-
sen 1999), a performer has the possibility to develop and maintain their skills
in many areas. Skills such as sight-reading/sight-singing (i.e. being able to

sing notation) is one many singers struggle to master despite its importance
(Killian and Henry 2005). The skill of memorizing the words and melody
together was found to be effective for singers (Ginsborg and Sloboda 2007),
with the formal structure of a song providing the basic framework for mem-
ory. Communicating emotion through both psychological and physiological
means is essential for a singer, with strategies involving both speech and
singing (Sundberg et al. 1994).
Other strategies musicians use in their pursuit of performance excellence
(Williamon 2004) include the areas of analysis, practice (Jrgensen 2004),
imagery (Emmons and Thomas 1998), rehearsal, memory (Barry and Hallam
2002), self regulation (McPherson and Zimmerman 2002), technique, goals,
organization, and time management (Jrgensen 2004).
This paper presents the findings of (1) a study investigating strategy use
by professional singers of the selection and preparation of twentieth and
twenty-first century art song and (2) the practical application of those strate-
gies in the preparation and performance of twentieth and twenty-first century
Australian art song.
METHOD
Participants and data collection procedures
Data were collected from two sources, the first being a group of n=14 profes-
sional singers from America (2), Australia (7), Canada (1), New Zealand (3),
and Spain (1) in a study from 2006-07. Six were male, seven female, and their
voice types were soprano (6), mezzo soprano (2), tenor (4), baritone (1), and
bass/baritone (1). The professional singers completed a series of three email
questionnaires, the first of which sought responses about the selection and
subsequent preparation of twentieth and twenty-first century art song reper-
toire from a program for a recent performance. The second email question-
naire sought information on learning and preparing one twentieth and
twenty-first century art song contained in the submitted program. Partici-
pants were requested to send a score of the song, which they could annotate
to illustrate their answers. Specific and generic questions (relating to any
twentieth and twenty-first century art song) about learning strategies, proce-
dures (i.e. order of learning), and the use of vocal techniques and challenges
were asked in the third and final email questionnaire. The resulting qualita-
tive data was categorized into a list of either performance or learning strate-
gies. The draft list of strategies (n=50) was informed by a list of strategies
drawn from the initial literature review, combined, and made available to
three professional singers to apply as they prepared a recital of Australian art

songs.
The second source of data collection came from three professional Aus-
tralian singers in a project that involved the author (a soprano), a mezzo, and
a baritone who all applied the collected strategies in their preparations for a
2008 recital of 34 Australian art songs. The recital was recorded and the
singers and accompanists were interviewed following the performance. This
practice-led research project (Aggett 2008) required the three singers to re-
cord their practice sessions and keep and reflect on practice diaries, choosing
and adopting appropriate strategies as a means of improving their perform-
ances. Their new strategies resulted in further expansion to the existing list of
strategies.
Materials
The 34 songs for the recital were chosen as a result of an extensive search of
all the Australian art songs in the Australian Music Centre (submitted from
2005-09) and other sources including the Mitchell (Sydney, New South
Wales) and National (Canberra, Australian Capital Territory) libraries, in an
attempt to select a repertoire of songs suitable for all voice types and ad-
dressing a range of pedagogical aspects.
RESULTS
Eighty-four strategies drawn from the literature review and first group of
participants were categorized and coded as being musical (M), performance
(P), or contextual (C) according to themes evident in the literature. Many of
these strategies can be viewed as performance (self-regulatory) strategies for
the singer and/or learning strategies (for a teacher or singer). Performance
strategies (n=55) included overall strategies to help prepare a piece (10), pitch
strategies (12), rhythm strategies (5), textual strategies (7), vocal techniques
(7), and ensemble and accompaniment strategies (7). Learning strategies
(n=29) included general strategies (14), order of learning (8), pitch strategies
(1), aural strategies (1), and strategies for preparing a whole song cycle or
program (4).
The second group of participants adopted n=118 performance strategies
42 musical, 49 performance, and 27 contextual; a total of n=43 learning
strategies were adopted, with 21 being musical, 11 performance, and 11 con-
textual, with singers introducing 13 new strategies to the list (see Table 1).
Table 1. Strategy use in relation to recurring musical (M), performance (P), and con-
textual (C) themes existing in the literature on and about twentieth and twenty-first
century solo vocal music and Australian song. The performance and learning strategies
used by professional singers as reported in the study conducted in 2006-07 were sorted
into M, P, and C themes, some of which were adopted by the three professional singers
in a 2008 recital.
M P C M P C
Occurrence in literature 163 70 269
2006-07 prof singers study

Performance strategies (PS) 14 41 13 Learning strategies (LS) 12 14 21
2008 recital: Application of strategies
Soprano (PS) 24 29 14 (LS) 4 6 1
Mezzo soprano (PS) 3 6 5 (LS) 11 - 9
Baritone (PS) 15 14 8 (LS) 6 5 1
Strategy use total (PS) 42 49 27 (LS) 21 11 11
DISCUSSION
Most frequently encountered performance strategies in the first study were

pitch and overall strategies to learn a song (see Figure 1). Because of the
breakdown of tonality in the twentieth and twenty-first centuries, vocal rep-
ertoire of this period can be musically challenging, and it is not surprising to
note the emphasis on pitch. However, few participants had definite learning
approaches to tackle the repertoire, and their responses demonstrated that
many had a lack of understanding of current trends in song composition.
General strategies were the most frequently employed learning strategies
reported by singers; for example, both a mezzo and tenor found it helpful to
use a sound file from a composer to help learn a melody when no recording
was available. Many of the strategies can be viewed as performance/self-
practice strategies for the singer and/or learning strategies for the teacher or
singer. The number of composer biographies and articles included in the re-
view of literature consulted explains the high reference to contextual strate-
gies found (see Table 1).
Strategies that second study participants found most useful related to
performance rather than learning strategies. From the practice journals, it
was apparent each singer had tailored their learning, practice, and strategy
use to suit each song, their own personal challenges, and the technical de-
Figure 1. Distribution of performance and learning strategies collected from responses

by 14 professional singers in the 2006/7 international study conducted by email ques-
tionnaires.
mands specific to the repertoire. Both an ensemble and pitch learning strat-
egy used with the sixth song of Peggy Glanville Hicks 13 Ways of Looking at
a Blackbird involved the accompanist playing the notes of each bar as a block
chord to familiarize the singer with the tonality, gradually adding in the
rhythm.
The use and application of performance rather than learning strategies,
drawn from literature and the experience of professional singers, were found
to be effective in striving for performing excellence when preparing twentieth
and twenty-first century Australian art song. Categorizing strategies by the
themes apparent in the literature and cross-referencing them with the experi-
ences of professional singers can offer singers and singing teachers a way of
tackling the challenges offered by twentieth and twenty-first century vocal art
song.
Acknowledgments
This paper was written with the assistance of a Graduate Student International Travel
Assistance Award from the Australian Music and Psychology Society.
Cathy Aggett, School of Communication Arts, University of Western Sydney, Locked Bag
1797, Penrith South DC, New South Wales 1797, Australia; Email: c.aggett@uws.edu.au
References
Aggett C. (2008). Singers strategies for performing and learning 20th Century Austra-
lian art songs. Paper presented at the Twenty-eighth World Conference of the In-
ternational Society for Music Education, Bologna, Italy.
Barry N. H. and Hallam S. (2002). Practice. In R. Parncutt and G. E. McPherson (eds.),
The Science and Psychology of Music Performance (pp. 151-165). Oxford: Oxford
University Press.
Emmons S. and Thomas A. (1998). Power Performance for Singers. Oxford: Oxford
University Press.
Fine P. and Ginsborg J. (2007). How singers influence the understanding of sung text.
In A. Williamon and D. Coimbra (eds.), Proceedings of ISPS 2007 (pp. 253-258).
Ginsborg J. and Sloboda J. A. (2007). Singers recall for the words and melody of a new,
unaccompanied song. Psychology of Music, 35, pp. 421-440.
Hallam S. (1997). The development of memorisation strategies in musicians: Implica-
tions for education. British Journal of Music Education, 14, pp. 87-97.
Jrgensen H. (2004). Strategies for individual practice. In A. Williamon (ed.), Musical
Excellence (pp. 85-103). Oxford: Oxford University Press.
Killian J. N. and Henry M. L. (2005) A comparison of successful and unsuccessful
strategies in individual sight-singing preparation and performance. Journal of Re-
search in Music Education, 53, pp. 51-65.
Mabry S. (2002). Exploring Twentieth-Century Vocal Music. Oxford: Oxford University
Press.
McPherson G. E. and Zimmerman B. J. (2002). Self-regulation of musical learning. A
social cognitive perspective. In R. Cowell and C. Richardson (eds.), The New Hand-
book of Research on Music Teaching and Learning (pp. 327-347). New York: The
National Association for Music Education.
Nielsen S. (1999). Learning strategies in instrumental music practice. British Journal of
Music Education, 16, pp. 275-291.
Rink J. (2002). Analysis and (or?) performance. In J. Rink (ed.), Musical Performance
(pp. 35-58). Cambridge: Cambridge University Press.
Sundberg J., Iwarsson J., and Hagegrd H. (1994). A singers expression of emotions in
sung performance. Speech, Music and Hearing Quarterly Progress and Status Re-
port, 35, pp. 81-92.
Williamon A. (2004). Musical excellence. Oxford: Oxford University Press.
Developing the ability to perform:

Investigating the field of higher education and
expertise development for learning and
performing the double bass
Florian Pertzborn1,2,3, Daniela Coimbra1,2, Susan Hallam4, and

Ana Cristina Braga5
1 Superior School of Music, Porto Polytechnic Institute, Portugal

2 School of the Arts, Portuguese Catholic University, Portugal
3 National Symphony Orchestra of Porto, Casa da Msica, Portugal
4 Institute of Education, University of London, UK
5 Department of Production and System Engineering, University of Minho, Portugal
The aim of the present study was to find out if there were significant dif-
ferences between the learning student, the supervising professor, and the
expert performer working in a symphony orchestra regarding the devel-
opment of their ability to perform. Survey results demonstrate that all
groups showed a secure level of understanding of the concepts of deliber-
ate practice and self-awareness, while those of metacognition were prac-
tically unknown. While students reported the most practice of all groups,
they rarely record themselves or use computer technology. All partici-
pants were highly supported by their parents especially at the time when
they decided to become a professional. The instrumental diversity of size,
scordatura, playing postures or bowing schools was seen as an advantage
rather than a disadvantage, overruling statements of previously under-
taken research. The study highlights the importance of early pre-profes-
sional learning for the university student.
Keywords: double bass; skill; practice; pre-professional learning; expert

performance
Over the past decade, researchers have taken numerous approaches to

studying how musicians acquire and refine their skills as performers (e.g.
Hallam 1997, Jrgensen 2000). While many of these studies have provided
new perspectives on instrumental learning (e.g. Hallam 1997), it remains

unknown whether the findings found their way back to benefit the population
(Bastian 1998). What has been noticed is that the results of studio research
only very slowly establish themselves as the learning to learn concepts to
professional practice in higher education (Pertzborn 2007). If these links
remain disconnected to instrumental practice, they may not give the student
the means to achieve maximum proficiency. However, and also in some way
confirmed in this investigation, most double bass students are still advised to
study from method books published in the nineteenth century. It seems
somehow awkward that more recent method books have excluded an in-
depth exploration of the processes of skill acquisition as guidelines to profes-
sional instrumental practice.
METHOD
Participants
Fourteen professors (PROF), 82 students (STUD) from 13 universities, 33

double bassists (ORCH) from 7 symphony orchestras representing seven
countries in three continents, participated in the DAP survey (see Figure 1).
Principals from symphony orchestras collaborated in handing out the ques-
tionnaires to their section colleagues, and university professors to their stu-
dents, as well as sending back the completed questionnaires. The total
number of participants was 129; the response rate was 85%.
Materials
Together with the empirical approach adapted to this study, literature was
reviewed comprising the areas of (1) the learner and (2) the instrument. The
first part focused on practice, expertise, task requirements, and biographical
issues. A second part reviewed the history of the double bass, its performer,
and their influences on the development of expertise of the instrument and its
role in the repertoire. Many of the findings that came out of the historical
reviewsuch as the diversity of at least two bow schools, different playing
postures and tuningsare still subject to controversial discussion today.
These concerns were included in the survey to examine the impact of history
and tradition on the present generation. The blueprint of the survey (see Fig-
ure 2) was based on a multidimensional learning model to instrumental
practice conceived by Hallam (1997).
Figure 1. Field of investigation and groups of participants.
Figure 2. Blueprint of the DAP survey.
Thirty-four questions were organized in presage, process, and outcome

factors, assembling a wide range of topics: learner characteristics, population,
music education, task requirements, environment and instrument, processes
of practice, performance experience and expertise, methods, biographic in-
formation, professional experience, health and wellbeing, and lifestyle. Four
different types of responding were used: (1) rating from 1-5, (2) multiple
choice, (3) selecting between yes/no/rarely, and (4) numeric indication of, for
example, years of playing. Five open-ended questions were used to assem-
ble individual opinions or different views. The surveys were translated to
English, German, and Portuguese.
Procedure
The data collected from the survey were processed through the statistical
program SPSS 17.0, taking into account the most adequate statistical tech-
nique to the involved variables. The data were organized according to the
nature of variables, most of them qualitative and descriptive techniques.
Principal component analysis was used to reduce the dimensionality and to
identify profiles considering biography, education, etc. To compare the re-
sults, parametric Analysis of Variance (ANOVA) or non parametric Kruskal
Wallis techniques were used. For the other topics, some independent chi-
squared tests were performed to evaluate the association between the differ-
ent groups of PROF, ORCH, and STUD and the correspondent item. The de-
cision rule used to detect significant statistical evidence was a probability (p)
value less than 0.05. In addition, the survey provided 204 individual com-
ments to open-ended questions which were assembled for further discussion
according to the topics and group of population.
RESULTS
Results from the survey revealed that the double bass is still a male domi-
nated instrument. Male participants were dominant in the group of the PROF
(79%), STUD (72%), and ORCH (100%). No female orchestra bassist partici-
pated in the survey. The mean age of PROF was 50 years male and 41 years
female, ORCH 41 years (only male), and STUD 23 years male and 21 years
female. The average age for starting the double bass for all groups was be-
tween 14-15 years. However, the starting age ranged from 3-20 years. A high
percentage of all groups previously gained experience on other instruments,
which they learned for 4-7 years before moving to the double bass at the age
of 13-15. The late and most diverse starting age reveals a relative inconsis-
tency when comparing to other instruments such as the violin. These findings
confirm those from a similar study undertaken by Langner (2003).
The majority of the ORCH group (52%) practices regularly only when they
have important performances, while this is done less by PROF (36%) and
STUD (25%). This may be because PROF and STUD have a different time
frame to build up repertoire in advance. They may also have more freedom in
choosing their repertoire, while the ORCH group normally has to deal with an
intense and pre-given performance schedule. Additional comments were
made on these issues, which confirm these views. The PROF (78%) and
ORCH (76%) reported that they practice between 1-3 hours daily, while this
was assumed by 60% of the STUD; 35% of STUD reported practicing 4-6
hours per day. In all groups, between 54-67% of the participants tended to
maintain a practice free day through the week. STUD (73%), unlike PROF and
ORCH (38%), consider studying etudes as a key point of their practice. The
landscape of practice contents was fairly even and consistent in all groups
and showed a high percentage in learning new repertoire, maintaining or

re-learning already performed material, maintaining or improving technical
issues, and preparing performance.
PROF obtained the highest scores for factual knowledge of practice issues
(80-100%), followed by ORCH (50-85%), and STUD (35-84%), with the ex-
ception of the concepts of metacognition. Here, the groups PROF (58%),
ORCH (50%), and STUD (64%) reported that they were not familiar with
these concepts. This fact may well indicate that concepts of practice are
widely known to most of the participants and that topics of instrumental re-
search have the potential to find their way back to the practicing musician.
Only a minority of all groups recorded themselves on audio: PROF (43%),
ORCH (18%), and STUD (30%). The usage of video was still much lower than
the usage of audio: PROF (21%), ORCH (14%), and STUD (7%). The usage of
computer software as a practice aid was indicated by PROF (50%), ORCH
(26%), and STUD (20%). A much lower score was attained for the usage of
MIDI accompaniment: PROF (29%), ORCH (11%), and STUD (25%). These
results illustrate a low rate of usage of audio, video, and computer technology
in all groups. In the century of almost unlimited access to recording and mu-
sic technology, one may have expected a higher usage of these media. Listen-
ing to CD audio recordings and instructional DVDs received a higher score of
appreciation: PROF (57% for DVD, 71% for CD), ORCH (44% for DVD, 59%
for CD), and STUD (35% for DVD, 69 % for CD). It is noted that CD re-
cordings received a higher score than DVDs.
Indications on lifestyle revealed that all groups highly enjoyed their pro-
fession as a double bassist (95-100%). Nevertheless, ORCH (48%), STUD
(27%), and PROF (22%) already thought at some point in their careers about
doing something else. The PROF was the most active in performing sports
regularly, while the ORCH and the STUD showed less activity in this domain.
In addition to this, 43% of the PROF, 32% of the ORCH, and 19% of the STUD
reported to have another professional qualification, activity, or expertise.
DISCUSSION
Results of this study revealed a strong relation and influence between teach-
ers and their students. The group of the PROF was the most experienced in all
performance domains and had also the highest academic qualification when
compared to the ORCH and the STUD. The latter, however, seemed to follow
their PROF in being active in all domains, although professional opportuni-
ties such as performing in festivals are yet rather more limited for them. Early
instrumental learning on the double bass has been identified as one of the
potential key factors to a prospective professional career. This has obvious

implications in early learning methodologies and in providing adequate in-
struments for the young learner. The low use of technology, while making
such a high investment of practice time to achieve a competitive level of per-
formance, is a second point that needs further investigation. The proposal is
that a wider perspective of learning, practice, and self-evaluation in combina-
tion with the use of technology as a support for practice and feedback should
be taken into account to form efficient venues to move forward on the road to
excellence for the double bass. By providing better and more challenging con-
cepts to expertise and (early) instrumental learning, future generations of
double bassists may breakdown more rapidly the technical barriers of the
instrument and will use the whole range of the instrument as an advantage
rather than a limitation.
Acknowledgments
This work was supported by the Fundao para a Cincia e Tecnologia and Programa
Operacional Cincia e Inovao 2010 (POCI 2010), co-funded by the Portuguese Gov-
ernment and European Union by the FEDER Program.
Florian Pertzborn, Escola Superior de Musica ESMAE/IPP, Rua de Alegria 504, 4200
Porto, Portugal; Email: florianpertzborn@esmae-ipp.pt
References
Bastian H. G. (1989). Leben fr Musik. Mainz, Germany: Schott Mainz.

Ericsson K. A. (1996). The Road to Excellence. Mahweh, New Jersey, USA: Erlbaum.
Hallam S. (1997). What do we know about practicing? In H. Jrgenson and A. C. Leh-
mann (eds.), Does Practice Make Perfect? (p.179-231). Oslo: Norwegian Academy
of Music.
Langner D. (2003). Flawed expertise: The case of the string players. In J. W. Davidson
(ed.), The Music Practitioner (pp. 251-259). Aldershot, Hampshire, UK: Ashgate.
Jrgensen H. (2000). Student learning in higher education: Who is responsible? British
Journal of Music Education, 17, pp. 67-77.
Pertzborn F.(2007). Motor control and learning: The basics to skilled instrumental
performance. In A. Williamon and D. Coimbra (eds.), Proceedings of ISPS 2007
(pp. 115-120). Utrecht, The Netherlands: European Association of Conservatoires
(AEC).
Pathways of excellence in science and dance:

Lessons learned from a Portuguese case study
Liliana S. Arajo1, Jos Fernando A. Cruz1, and Leandro S. Almeida2
1 School of Psychology, University of Minho, Portugal

2 Institute of Education, University of Minho, Portugal
Many researchers are studying the early identification of exceptional in-

dividuals, expertise acquisition, personality, and the role of external
catalysts, aspiring to explain their influence on the process of talent and
expertise development. Researchers seek a deeper understanding of
these exceptional individuals, using multiple methodologies and tech-
niques, studying multiple contexts and domains, and focusing on multi-
ple dimensions and constructs associated with excellence. Moreover,
much psychological research on excellence takes a predominantly out-
sider perspective favoring quantitative inquiry, but little is really known
about the experiences and personal meanings of exceptional individuals.
In this study, we explored the insider perspective of being excellent. Us-
ing a case study approach, two dancers and two scientists were inter-
viewed. A theoretically-oriented content analysis grid was constructed to
assist data analysis. Overall, participants stressed personal characteris-
tics such as extreme curiosity, persistence, and passion for their work as
central features in nurturing and sustaining their motivation for practice,
and also as key features of being an excellent performer. Additionally,
intensive and hard work, discipline, and strong commitment were em-
phasized and other personal and contextual dimensions highlighted. Sin-
gularities and some reflections that emerged from this study are
discussed.
Keywords: excellence; dance; science; performance; qualitative
The interest and curiosity about the lives of exceptional scientists, artists, or
athletes has always encouraged researchers to search for the reasons and
determinants of being creative and attaining high levels of performance. In
science and performing arts, pioneering work by Roe (1946), Bloom (1985),
Gardner (1993), and Feist (2006) has inspired many researchers. For in-
stance, Lubinski et al. (2001) tracked intellectually precocious youth over 25
years, reaching conclusions about stimulating early educational experiences,
mentoring relationships, early development of interests, and strong commit-
ment to career and work as significant attributes in developing exceptional
expertise in science. Recently, Feist (2006) conducted and reviewed several
studies on personality and development of scientists, arguing for the need of a
new psychology of science for a complete understanding of scientific
thought and behavior. Conversely, research in dance seems to be still devel-
oping, focusing essentially on classical and non-professional dancers, and
more negative topics, such as eating behavior disorders, injury, stress, or
physical and biomechanical issues (see Krasnow and Kabanni 1999). Still,
some efforts have been developed towards a more complete dance science,
with studies focusing on professional and elite performing artists and psy-
chological issues, such as personality profiles, expertise, or motivation (e.g.
Hays 2002, Kogan 2002, Urea 2004).
Moreover, several theoretical models have been developed to explain the
process of performing excellence in different domains (see Arajo et al.
2007). The theoretical models of Renzulli (2002) or Gagn (2004), explaining
talent development, Ericssons deliberate practice and expert performance
approach (Ericsson and Charness 1994), as well as the Berlin school para-
digm on the study of wisdom (Baltes and Staudinger 2000) represent some of
the strongest approaches that influence current empirical research. Overall,
findings suggest that singularity of exceptional individuals seems to emerge
from a dynamic combination of contextual and personal factors.
Despite the vast amount of research on excellent performance, several is-
sues remain ambiguous. Definitions of excellence are still vague and multiple
criteria are used to identify excellent performance (Weiss et al. 2002, Trost
2000, Urea 2004, Zannotto 2006). Additionally, little is known about the
trajectories of excellent individuals from their own thoughts and meanings.
Quantitative inquiry has been favored in the study of performance excellence,
even if the singularity of successful trajectories and the complexity associated
with performing excellence in adulthood is assumed. Nevertheless, qualitative
methods are being gradually accepted as a valuable way to investigate singu-
lar and complex topics in-depth. Hence, a qualitative case study was con-
ducted aiming to analyze the singularities and commonalities behind talent
and success trajectories. In this paper, we present general findings and dis-
cuss some lessons learned from this research.
METHOD
Participants
Two female dancers (contemporary dance) and two scientists (life sciences,
female and physics, male) were consensually identified by a panel of experts
in each specific domain. Participants were all nominated for revealing excel-
lence in performance and being actively engaged in their respective field. The
ages of participants were 36, 41, 38, and 43, respectively.
Materials
The interview protocol was created after reviewing relevant literature and
interview guides successfully used in previous studies with exceptional indi-
viduals in different domains (e.g. Connaughton et al. 2007, Kiewra and Cres-
well 2000). A semi-structured protocol covering the following main areas in a
fluid and flexible sequence was used: (1) career path, (2) past achievements
and actual performance, (3) expertise acquisition, (4) personal characteris-
tics, (5) role models and significant others, and (6) relationships within each
professional community. In addition, participants academic and professional
pathway and productivity were analyzed through their actualized curricula
vitae.
Procedure
A nomination strategy was conducted to select participants. Although criti-

cized, this strategy seems to be effective, particularly in domains where ob-
jective and rigorous criteria are difficult to establish and gold standards do
not exist. In addition to peers recognition, some quantitative criteria were
considered (Weiss et al. 2002, Uren, 2004, Zanotto 2006). Criteria to select
scientists included: number of publications on refereed journals, number of
citations, highly cited papers, awards, members of editorial boards, and
grants. Awards, participation on international/European dance compa-
nies/projects, and professional certification were the criteria used to select
dancers. Identified participants were contacted by email or telephone, study
aims were presented, and data anonymity and confidentiality was guaran-
teed. Subsequent to participants agreement, interviews were scheduled ac-
cording to their time and location. Interviews were recorded integrally and
transcribed verbatim and sent to participants for verification. A theoretically-
oriented content analysis grid was designed to assist data content analysis.
Since the interview protocol had a semi-structured and flexible format,
themes in the analysis grid had a different organization reflecting that unan-
ticipated issues could arise during interviews. Transcripts were sorted into
three main dimensions: (1) contextual factors, (2) personal factors, and (3)
answers to a specific question: what makes you an excellent professional.
Texts were coded and analyzed using Maxqda qualitative analysis software
and following the proposals of Schilling (2006) and Mayring (2000) on
qualitative content analysis. Some validity procedures were used such as tri-
angulation, member checking, and peer-debriefing (Onwuegbuzie and Leech
2007).
RESULTS
An overview of participants productivity through their curricula vitae shows

some evidence for their superior performance. The scientists had several
publications in refereed journals, some of which are highly cited papers. We
found them working within the most influential international research net-
works and heading research teams with national and international grant
funding. The dancers were also distinguished by their creative and out-
standing performance. They work with the most creative and renowned Por-
tuguese choreographers and are frequently selected to perform on
choreographic projects in Europe.
Interviews content analysis highlighted several features confirming some
general research findings (e.g. Feist 1996, Lubinski et al. 2001, Kogan 2002).
Describing some aspects of macro context (social, historical, economical,
cultural), all the participants mentioned that, at the time they started to be-
come involved in domain-specific activities, Portuguese society was not sen-
sitive to their domain (dance/research), and opportunities and choices were
limited. Nevertheless, participants reveal that, currently, it is possible to find
a stimulating and optimal professional environment. Furthermore, partici-
pants clearly identified significant people in their lives who provided impor-
tant socio-emotional support but also an inspiring role in their career choices
and opportunities. Transcript analysis also indicates a strong presence of
individuality and singularity on participants trajectories. Overall, partici-
pants stressed personal characteristics such as curiosity, perseverance, open-
ness to new experiences, adaptability, and passion as central in nurturing and
sustaining their motivation. Additionally, intensive and hard work, discipline
and sustained commitment were considered important factors in performing
excellence. Finally, all the participants affirmed their strong concern with
society and a sense of social responsibility and also the importance of family
and social networks as core life values.
DISCUSSION
In this study, the role of positive emotional states and strong psychological
features are clearly associated with performing excellence, in addition to
contextual factors. However, we observed a slight difference between artists
and scientists, mainly on working strategies and career trajectories. Whereas
careers in science are more organized and sequential, trajectories in dance are
irregular and unstable. Scientists seem to plan more, following rigid sched-
ules, while dancers have more flexible agendas and engage in multiple activi-
ties other than physical training, which seems to contribute to their technical
and performance skills (Kogan 2002). Furthermore, this study provided a
rich description of individuals trajectories, showing several important condi-
tions needed to excel in a small and peripheral country such as Portugal. Par-
ticipants critical picture of the Portuguese educational and professional
system questions its role in the identification and encouragement of potential
talents, suggesting the importance of mentorship and awareness of creative,
sometimes unconventional, accomplishments. Not withstanding some well
known limitations of qualitative case studies, we believe that more qualitative
research is needed in order to promote a deeper understanding of performing
excellence. A vast amount of research with scientists exists, but contemporary
dancers are as yet under studied. Hence, the present case study suggests that
scientists and dancers personal wellbeing, as well as their career develop-
ment patterns, are interesting topics for further research efforts.
Liliana S. Arajo, School of Psychology, University of Minho, Campus de Gualtar, Braga

4710-057, Portugal; Email: liliana.araujo@iep.uminho.pt
References
Arajo L. S., Almeida L. S., and Cruz J. F. (2007). Excellence in achievement contexts:
Psychological science applications and future directions. In A. Williamon and D.
Coimbra (eds.), Proceedings of ISPS 2007 (pp. 17-22). Utrecht, The Netherlands:
European Association of Conservatoires (AEC).
Baltes P. B. and Staudinger U. M. (2000). Wisdom: A metaheuristic (pragmatic) to
orchestrate mind and virtue toward excellence. American Psychologist, 55, pp. 122-
136.
Bloom B. (1985). Developing Talent in Young People. New York: Ballantine Books.
Connaughton D., Wadey R., Hanton S., and Jones G. (2008). The development and
maintenance of mental toughness: Perceptions of elite performers. International
Journal of Sports Sciences, 26, pp. 83-95.
Ericsson K. A. and Charness N. (1994). Expert performance: Its structure and acquisi-
tion. American Psychologist, 49, pp. 725-747.
Feist, G. (2006). The Psychology of Science and the Origins of Scientific Mind. London:
Yale University Press.
Gagn F. (2004). Transforming gifts into talents: The DMGT as a developmental theory.
High Ability Studies, 15, pp. 119-147.
Gardner H. (1993). Creating Minds. New York: Basic Books.
Hays K. F. (2002). The enhancement of performance excellence among performing
artists. Journal of Applied Sport Psychology, 14, pp. 299-312.
Kiewra K. A. and Creswell J. W. (2000). Conversations with three highly productive
educational psychologists: Richard Anderson, Richard Mayer, and Michael
Pressley. Educational Psychology Review, 12, pp. 135-161.
Kogan N. (2002). Careers in the performing arts: A psychological perspective. Creativ-
ity Research Journal, 14, pp. 1-16.
Krasnow D. and Kabbani M. (1999). Dance science research and the modern dance.
Medical Problems of Performing Artists, 14, pp. 16-20.
Lubinski D., Benbow C., Shea D. et al. (2001). Men and women at promise for scientific
excellence: Similarity not dissimilarity. Psychological Science, 12, pp. 309-317.
Mayring P. (2000). Qualitative content analysis. Forum: Qualitative Social Research, 1,
available at http://nbn-resolving.de/urn:nbn:de:0114-fqs0002204.
Onwuegbuzie A. J. and Leech N. L. (2007). Validity and qualitative research: An oxymo-
ron? Quality and Quantity, 41, pp. 233-249.
Roe A. (1946). Artists and their work. Journal of Personality, 15, pp. 1-40.
Schilling J. (2006). On the pragmatics of qualitative assessment: Designing the process
for content analysis. European Journal of Psychological Assessment, 22, pp. 28-37.
Shanteau J., Weiss. D. J., Thomas R. P., and Pounds J. C. (2002). Performance-based
assessment of expertise: How to decide if someone is an expert or not. European
Journal of Operational Research, 136, pp. 253-263.
Trost G. (2000). Prediction of excellence in school, higher education and work. In K.
Heller, F. Mnks, R. Sternberg, and R. Subotnik (eds.), International Handbook of
Giftedness and Talent (2e, pp. 317-330). Oxford: Pergamon.
Urea C. (2004). Skill Acquisition in Ballet Dancers: The Relationship between Delib-
erate Practice and Expertise. Unpublished doctoral thesis, Florida State University.
Zanotto E. D. (2006). The scientists pyramid. Scientometrics, 69, pp. 175-181.
Thematic session:
Expression and interpretation II
Diversity and homogeneity in contemporary

violin recordings of solo Bach
Dorottya Fabian
School of English, Media, and Performing Arts,

University of New South Wales, Australia
Over 20 recent commercial recordings of Bachs solos for violin were

studied to examine performance trends of the last 25 years and to test the
widely held opinion that style has become fairly homogenous and lacking
in individuality. The main trends found were the influence of historically
informed practices (e.g. in bowing, fingering, vibrato) and a move toward
greater flexibility in phrasing and rhythmic delivery. Both strong and
subtle individual differences were observable, but these seem less suit-
able for quantitative reporting.
Keywords: J. S. Bach; violin; vibrato; performance; recordings
Research into performance styles on sound recordings has been growing

steadily in the last decade or so. The standard narrative that has emerged
regarding trends over a century of recordings is a claim for a continued ti-
dying up of performance (Philip 2004, p. 232). Commentators lament the
increased uniformity of interpretations as we move from the early decades of
recording to mid-century and beyond (e.g. Day 2000). Yet studies of specific
repertoires show just as much individuality in recent as in old recordings (e.g.
Ornoy 2008, Repp 1990, 1992, Fabian and Schubert 2008). The contradiction
may stem from the fact that investigations tend to focus on earlier recordings.
Few published research exists that examines in any systematic way the art-
istry of contemporary musicians. It is also likely that in solo or chamber mu-
sic there is more room and need for an individual voice than in works
involving large ensembles such as symphonies. Given the growing amount of
information available regarding the history of violin playing on record (Katz
2003, Milsom 2003, Fabian 2005, Leech-Wilkinson 2009) and yet a lack of
detailed information on current practices, this study aims to provide data to
fill this hiatus. To this end, it examines contemporary recorded performances

of Bachs Sonatas and Partitas for Solo Violin (BWV 1001-1006).
METHOD
Materials
Over 20 commercially available studio and concert recordings of the Bach

solos released since 1976 were analyzed. Nine of the violinists play on period
instruments, and at least five others are known to have been influenced by the
achievements of historically informed performance (HIP) to some extent.
Four players made more than one recording of the works during the desig-
nated period. Selected performers and dates are listed in Tables 1-3; a full
discography is available from the author.
Procedure
Aural analyses of bowing, fingering, and phrasing together with software as-
sisted analyses of vibrato, tempo, and rhythmic flexibility were conducted.
The analysis aimed to distinguish between mainstream and period violin-
ists and examined their techniques and artistic approach. Only general trends
and aspects of the quantitative data are reported.
RESULTS
Bowing differed considerably across most violinists; earlier and mainstream

players choosing a more seamless bowing (e.g. Perlman, Hahn) while younger
players, especially those using period apparatus, a more articulated style with
faster note decay. Fingering choices involving lower positions and use of open
strings were more common among HIP-inspired players.
Phrasing showed great diversity reflecting the continuum from literalistic
to flexible approaches. Of those with multiple recordings, Kremer seems to
have changed his interpretation the most: it became much more strongly ac-
cented with slower tempi and a narrower and less continuous vibrato. The
many different ways the works are delivered can be illustrated by comparing
the beginning of the G minor Adagio. Ehnes and Kremer play it in a literal
manner, while others with greater degree of freedom. Huggett plays the
opening chord with emphasis on the low G followed by arpeggiation and then
a pause on the high G. Wallfisch and others also arpeggiate but faster, reach-
ing the top G in one gesture. When played chordally, it is broken 2+2.
Vibrato also provided variety. HIP violinists on recordings from after 1995
tended to use less and narrower vibrato. Among mainstream violinists, Bar-
Table 1. Vibrato rate, width, and frequency of use measured on selected notes in differ-
ent movements and averaged across each selected violinist. (NB. Rate in cycle/second,
width in semitones; frequency refers to occurrence of vibrato on the selected pitches.)
Performer, date Vibrato rate (c/s) Width (semitones) Frequency of use

mean SD mean SD %
Huggett 1995 (HIP) 5.8 0.50 0.2 0.06 46.3
Kuijken 1983 (HIP) 5.9 0.60 0.2 0.07 63.3
Kuijken 2001 (HIP) 5.9 0.80 0.1 0.90 60.0
Luca 1977 (HIP) 6.8 0.10 0.1 0.08 39.5
Podger 1999 (HIP) 6.0 0.70 0.2 0.09 27.0
Schroeder 1985 (HIP) 6.5 0.60 0.2 0.06 55.0
Van Dael 1996 (HIP) 5.5 0.80 0.1 0.04 23.0
Wallfisch 1997 (HIP) 5.8 0.50 0.1 0.14 45.5
Barton-Pine 1999 6.9 0.90 0.2 0.12 62.0
Barton-Pine 2007 6.1 0.90 0.1 0.03 14.0
Hahn 1999 6.6 0.30 0.2 0.10 100.0
Kremer 1980 6.3 0.36 0.3 0.07 91.0
Kremer 2005 6.2 0.80 0.2 0.12 45.3
Mintz 1984 5.9 0.40 0.3 0.04 90.0
Perlman 1987 6.6 1.70 0.3 0.15 76.0
Ricci 1981 6.0 0.50 0.3 0.17 76.0
Szenthelyi 2002 6.4 0.28 0.3 0.10 82.0
Tetzlaff 1994 6.2 1.10 0.2 0.12 64.5
Tetzlaff 2005 6.5 0.43 0.3 0.48 91.0
Tognetti 2005 5.9 0.30 0.1 0.03 36.0
Zehetmair 1983 4.8 0.50 0.4 0.16 14.0
ton-Pine used vibrato the least and Hahn the most. Multiple recordings of
violinists indicate a decreasing use of vibrato. Table 1 provides details.
Overall tempo choices showed no clear trend but varied most in fast and
fugal movements, as well as in the C major Adagio. Table 2 provides an ex-
ample.
Variety was discovered in relation to ornamentation as well. Luca, Podger,
Huggett, Gringolts, and Tognetti added graces and embellishments (e.g. A
minor Andante, E major Gavotte). Their solutions differed radically from
each-other both in terms of quantity and type of added ornaments. Rhythmic
flexibility was more obvious in the most recent and HIP versions, where the
articulation of metrical groups and the inflection of hierarchically important
Table 2. Tempo choices in selected G minor sonata recordings. Mean metronome values
were calculated from movement duration. SDs marked as negative indicate variance
that is slower than the mean. SD>2 is italicized.
Date, performer Adagio Fuga Siciliano Presto

mean SD mean SD mean SD mean SD
1977 Luca 27.0 1.28 69.0 -0.16 19.0 -2.61 65.0 -1.22
1980 Kremer 20.0 -1.16 86.0 2.31 27.0 0.26 87.0 1.54
1981 Kuijken 27.0 1.14 70.0 -0.06 24.0 -0.69 75.0 0.11
1983 Zehetmair 28.0 1.33 80.0 1.43 31.0 1.76 81.0 0.82
1984 Mintz 19.0 -1.38 63.0 -1.06 24.0 -0.79 73.0 -0.15
1986 Perlman 20.0 -1.14 69.0 -0.16 26.0 -0.14 80.0 0.62
1993 Tetzlaf 27.0 1.16 69.0 -0.18 29.0 1.01 95.0 2.52
1995 Huggett 22.0 -0.65 56.0 -2.10 24.0 -0.92 63.0 -1.44
1996 Poulet 23.0 -0.15 77.0 1.03 28.0 0.79 74.0 -0.11
1996 van Dael 27.0 1.01 69.0 -0.25 29.0 0.92 66.0 -1.06
1997 Wallfisch 24.0 0.13 65.0 -0.77 28.0 0.48 66.0 -1.06
1997 Podger 23.0 -0.21 68.0 -0.34 27.0 0.19 72.0 -0.27
1998 Ghler 24.0 0.31 64.0 -0.91 25.0 -0.64 72.0 -0.31
1999 Ehnes 17.0 -2.06 67.0 -0.52 27.0 0.08 80.0 0.62
2001 Szenthelyi 28.0 1.43 73.0 0.34 30.0 1.25 67.0 -0.89
2005 Tetzlaff 24.0 0.13 74.0 0.55 25.0 -0.45 83.0 0.98
2005 Kremer 22.0 -0.44 82.0 1.78 26.0 0.02 80.0 0.72
2005 Tognetti 21.0 -0.96 68.0 -0.40 23.0 -1.27 76.0 0.20
2001 Kuijken 25.0 0.55 67.0 -0.49 24.0 -0.69 67.0 -0.95
1999 Barton-Pine 23.0 -0.18 72.0 0.20 28.0 0.72 69.0 -0.74
2004 Barton-Pine 23.0 -0.15 69.0 -0.25 28.0 0.72 75.0 0.06
Mean 24.0 70.0 26.0 75.0
notes created easily detectable rhythmic irregularities. The analysis of dotting

ratios revealed under-dotting in the D minor Corrente movement. This re-
flects the view that dotted rhythms within fast triplet motion are to be as-
similated into long-short swings. In contrast, ratios in the C major Adagio
were found to be universally over-dotted (see Table 3).
DISCUSSION
The analyses indicate great diversity in approach, along the continuum from
literalistic to flexible. Several mainstream violinists are influenced by current
beliefs about HIP. This can be seen in the declining use of vibrato; the more
Table 3. Dotting ratios calculated from IOIs in selected D minor Corrente and C major
Adagio recordings. A ratio of 0.75 indicates literal dotting (3:1).
Performer, date Dotting ratio

D minor Corrente C major Adagio
Barton-Pine 1999 0.76 n/a
Barton-Pine 2004 0.74 n/a
Barton-Pine 2007 0.70 0.82
Brooks 2001 (HIP) 0.72 0.88
Ehnes 1999 0.80 0.81
Hahn 1997 0.78 0.81
Huggett 1995 (HIP) 0.69 0.79
Kremer 1980 0.76 0.80
Kremer 2005 0.76 0.85
Kuijken 1983 (HIP) 0.71 0.86
Kuijken 2001 (HIP) 0.71 0.86
Luca 1977 (HIP) 0.74 0.80
Mintz 1984 0.78 0.81
Perlman 1987 0.65 0.80
Podger 1999 (HIP) 0.72 0.77
Poppen 2001 (HIP) 0.76 n/a
Schroeder 1985 (HIP) 0.72 0.78
Szenthelyi 2002 0.74 0.77
Tetzlaff 1994 0.77 0.78
Tetzlaff 2005 0.78 0.82
Tognetti 2005 0.72 0.81
van Dael 1996 (HIP) 0.73 0.81
Wallfisch 1997 (HIP) 0.72 0.85
Zehetmair 1983 0.78 0.85
articulated and lifted bowing style, and the use of lower positions and open
strings. Multiple recordings of artists, especially those of Kremer, show this
tendency clearly. Apart from this broad stylistic fashion no real homogeneity
was found. Rather, the examination showed diversity in tempo choices, or-
namentation, and approaches to phrasing and expressive flexibilities. These
qualitative differences do not lend themselves easily to quantitative reporting
and can be better explained through detailed individual comparisons.
Acknowledgments
This project has been supported by an Australian Research Council discovery grant
(DP0879616). I would like to thank Bridget Kruithof for help with data collection and
Rachel Barton-Pine for her generosity in providing me with recordings of her concerts.
Dorottya Fabian, School of English, Media and Performing Arts, University of New
South Wales, Webster Building, Sydney, New South Wales 2052, Australia; Email: d.
fabian@unsw.edu.au
References
Day T. (2000). A Century of Recorded Music. New Haven, Connecticut, USA: Yale Uni-
versity Press.
Fabian D. (2005). Towards a performance history of Bachs Sonatas and Partitas for
solo violin. In L. Vikrius and V. Lampert (eds.), Essays in Honor of Lszl Somfai
(pp. 87-108). Lanham, Maryland, USA: Scarecrow Press.
Fabian D. and Schubert E. (2008). Musical character and the performance and percep-
tion of dotting, articulation and tempo in recordings of Variation 7 of J. S. Bachs
Goldberg Variations (BWV 988). Musicae Scientiae, 12, pp. 177-203.
Katz M. (2003). Beethoven in the age of mechanical reproduction: The Violin Concerto
on record, Beethoven Forum, 10, pp. 38-54.
Leech-Wilkinson D. (2009). The Changing Sound of Music. London: Centre for the
History and Analysis of Recorded Music (CHARM).
Milson D. (2003). Theory and Practice in Late Nineteenth-Century Violin Perform-
ance. Aldershot,UK: Ashgate.
Ornoy E. (2008). Recording analysis of J. S. Bachs G minor adagio for solo violin (ex-
cerpt): A case study. Journal of Music and Meaning, 6, available online at
www.musicandmeaning.net.
Philip R. (2004). Performing Music in the Age of Recording. London: Yale University
Press.
Repp B. (1990). Patterns of expressive timing in performances of a Beethoven minuet by
nineteen famous pianists. Journal of the Acoustical Society of America, 88, pp.
622-641.
Repp B. (1992). Diversity and commonality in music performance: An analysis of timing
microstructure in Schumanns Trumerei. Journal of the Acoustical Society of
America, 92, pp. 2546-2568.
Breathing in classical singing: Linking science

and teaching
Sally Collyer
Melbourne, Australia
This article addresses three issues. Firstly, breathing studies of trained

classical singers have found high inter-singer variability, which raises
many questions about the nature and role of breathing in singing; these
study findings and their implications are discussed from a pedagogical
perspective. Secondly, the article examines the resulting limitations in-
herent in a segmented model that divides voice into respiration, phona-
tion, and articulation and that excludes key components from a narrow
concept of a breathing system. Finally, the article examines how this
wider paradigm of singing-breathing inevitably positions breathing at the
confluence of technique, emotion, and musical line.
Keywords: singing; breathing; pedagogy; lung volumes; kinematics
When classical singers and singing teachers are asked what do you need to
sing well? the list almost always starts with good breathing. Yet, despite
hard work on all sides there remains a large gulf between research into
breathing for singing and its influence in the teaching studio. There are three
main reasons for the lack of interaction. First, singing teachers often find the
concepts, techniques, and equipment used in breathing studies to be difficult
to access. This lack of familiarity makes it difficult for teachers to interpret
and apply research findings in the studio. Second, the scope of what is con-
sidered to be the breathing system is often too narrow, neglecting the respi-
ratory contribution of elements not immediately associated with respiration.
Last, conceptualizing breathing as simply air supply for phonation negates its
role as the link between phonation, emotion, and musical line, a role that is
crucial to singing and in teaching singing. This has led some pedagogical ap-
proaches to reject studies as unrealistic or irrelevant.
The article introduces, from a pedagogical perspective, some of the con-
cepts and methods used in the study of breathing. The first section explains
certain assumptions, terminology, and techniques employed in the measure-

ment of breathing generally and with respect to singing. The second section
discusses problems that arise when paradigms of breathing employ a narrow
definition of what constitutes the breathing system. The third section ex-
plores how a wider paradigm supports a view of breathing more relevant to
the role of breathing in singing and more consistent with historical vocal
pedagogy. The paper is not intended to be exhaustive or prescriptive but to
assist the teacher in accessing findings and exploring paradigms in other dis-
ciplines which have implications for the discipline of vocal pedagogy. The
paper outlines the major points to be presented in the second and third sec-
tions, but particularly takes the opportunity to describe the measurement of
lung volumes and chest-wall kinematics to assist the teacher in understand-
ing implications for pedagogy.
MAIN CONTRIBUTION
Breathing studies
This section briefly describes the techniques used in breathing measurement.

Interpreting any research requires an understanding of the background con-
cepts and the methods used. Teachers who happily discourse on partials, the
singers formant, and the relationship between the thyroarytenoid and crico-
thyroid go silent at the mention of recoil, rib-cage volume contribution, or
resting expiratory level. There is nothing inherently more complex about res-
piratory analysis than acoustical analysis, or about the physiology of one part
of the human body versus another. The difference is familiarity. Powerful
personal computers and cheap (and more recently free) software offer singing
teachers the chance to experience firsthand how these relate to their teaching
and to the singer standing before them. By contrast, equipment for measuring
breathing is currently specialized, cumbersome, and expensive, and few
teachers have the opportunity to play, one of the most powerful aids to
learning. However, advances in sensor tracking technology, fuelled by the
enormous entertainment and virtual reality industries, make it increasingly
probable that breathing measurement will become as accessible in the future
as acoustic measurement is today. The chance to play with respiratory re-
cording will provide an impressive learning tool but will require that teachers
acquire an appreciation of the discipline of respiratory research, just as many
have acquired an appreciation of acoustics. Fortunately, a basic understand-
ing of breathing analysis is not difficult to acquire and offers singing teachers
access to studies that speak directly to their questions.
Lung volumes
The most common assessment of breathing is lung volume (LV). Key meas-
urements are LV at the start and end of each phrase, with the difference be-
tween these being the amount of air used (to normalize comparisons, LV
measures are expressed as a percentage of the individuals vital capacity
[%VC], measured by having the singer breathe in to full capacity and then
exhale as far as possible). On average, singers generally start phrases at 70-
80%VC and end phrases at 30-50%VC (Hixon 1991). The individual singer is
highly consistent in LV measurements (Thomasson and Sundberg 1999), but
singers are very different (Collyer et al. 2008, Collyer et al. in press). Intra-
singer consistency versus inter-singer variability can be seen in Figure 1. The
question immediately arises whether a particular LV behavior is associated
with better singing, but no study has yet been able to find such a link. The
next question, then, is how is LV measured?
Chest-wall kinematics
It is possible to measure LV directly with a spirometer or indirectly by meas-

uring airflow over time. However, these require a mouthpiece (and nose clip)
or a mask, which tend to inhibit singing. Konno and Mead (1967) developed
an alternative approach which estimated LV change from changes in the sur-
face dimensions of the ribcage and abdomen (known together as the chest
wall, and kinematics is simply the study of motion). Generally, ribcage (RC)
and abdomen (AB) dimensions have been measured by encircling the RC and
AB (strain gauges, inductance bands) or by applying sensors which measure
the anteroposterior (front-to-back) distance. In brief, changes in RC and AB
dimension are recorded while the singer performs various respiratory ma-
neuvers on a spirometer (with mouthpiece and nose clip). Known changes in
RC and AB dimension are therefore matched with known changes in LV, so
changes in RC and AB can be used to estimate LV when the singer is no
longer encumbered by the spirometer. The left panel in Figure 2 shows how
LV is the weighted sum of RC and AB, weighted because it takes less change
in RC than in AB dimension to move a given LV. The right panel shows a
CWK (or volume contribution) plot. The key to understanding these plots is
to remember that, since LV change is the sum of RC and AB change, changes
in LV are at 45o angles, as indicated by the dotted arrows. So, the line A-B
moves toward 100%VC and thus is inhalatory; the line also tells us that the
inhalation was due almost entirely to an AB increase (see Figure 2, left panel).
Line B-C shows that an RC increase and an AB decrease cancelled each other
out, so there was no net change in LV; we know this because line B-C is par-
100%
Lung volume at the start
Vital Singing (= exhaling)
Capacity
Lung volume at the end
0%
1 2 3
1 2 3 4 5 6 7 8 9 10
Figure 1. Lung volumes (LV) used in the messa di voce on B4. At left: consistency of one
singer performing three messe di voce. At right: variability amongst 10 female singers.
Data are taken from studies published in Collyer et al. (2008) and Collyer et al. (in
press). (See full color version at www.performancescience.org.)
Figure 2. Relationship between LV estimation and chest-wall kinematics (CWK). See

main text for description.
RIBCAGE
1 2 3 4 5

ABDOMEN
6 7 8 9 10
Figure 3. CWK plots for the LV changes shown in Figure 1. (See full color version at
www.performancescience.org.)
allel to the 100%VC line. Line C-D-E follows the singer performing a messa di
voce. Note that at first (C-D) this singer expands the RC (paradoxical behav-
ior), but greater AB decrease gives net exhalation and then RC also decreases
(D-E).
CWK plots have provided a great deal of information about singers be-
havior. The prephonatory adjustment (B-C) is very common (Thorpe et al.
2001). So too are paradoxical movements (where RC or AB dimension de-
creases during inhalation or increases during singing). But the most striking
feature is that, while a trained singer shows a highly consistent CWK pattern
on the same task, singers are very different from each other. Figure 3 shows
the CWK plots underlying the LV changes in Figure 1.
What is the breathing system?
The second section of the paper addresses the definition of a breathing sys-
tem. Voice is often conceptualized as consisting of three isolated stages: res-
piration, phonation, and articulation. This segmented model is inconsistent
with human anatomy and physiology because it models the larynx as an inde-
pendently functioning entity rather than as an intrinsic and inseparable part
of the respiratory system at all stages of development. For example, the role
of the larynx in modulating airflow for singing derives from its role in modu-
lating airflow for respiratory needs, which in turn originates with modulation
of lung fluid efflux by the larynx, which is essential for lung development in
the womb (Harding et al. 1986). Even when protecting the airway during
swallowing, a role often considered to be the larynxs primary function, the
larynx also assumes an expiratory modulation role to assist aspiration (Ced-
borg et al. 2009). The conference presentation considers the limitations of
the segmented model, which requires that laryngeal adjustments can occur
independently of respiratory adjustments.
Breathing, emotion, and musical line
The final section of the paper extends this wider breathing system model to
incorporate emotional influences of and on breathing and then to relate these
to the nature of musical line. This pivotal role of breathing as the intermedi-
ary between phonation, emotion, and musicwhich is at the heart of histori-
cal pedagogys reverence for breathingcannot be sustained under a
segmented or larynx-driven paradigm. The conference presentation discusses
how a wider concept of the nature of breathing is consistent with the nature
of singing at all levels: the technical, the interpretative, and the human.
IMPLICATIONS
LV and CWK data raise many questions for the teacher. What determines a
singers breathing pattern? How susceptible to training is breathing? What
represents an improvement for a singer? How does the teacher match train-
ing with singer? In the conference presentation, findings of the studies are
presented from a pedagogical perspective, including the implications of study
limitations such as technical constraints.
Sally Collyer, PO Box 156, Box Hill, Victoria 3128, Australia; Email: sallycollyer@yahoo.
com.au
References
Cedborg A. I. H., Sundman E., Bodn K. et al. (2009). Co-ordination of spontaneous

swallowing with respiratory airflow and diaphragmatic and abdominal muscle ac-
tivity in healthy adult humans. Experimental Physiology, 94, pp. 459-468.
Collyer S., Thorpe C. W., Callaghan J., and Davis P. J. (2008). The influence of funda-
mental frequency and sound pressure level range on breathing patterns in female
classical singing. Journal of Speech, Language, and Hearing Research, 51, pp. 612-
628.
Collyer S., Kenny D. T., and Archer M. (in press). The effect of abdominal kinematic
directives on respiratory behaviour in female classical singing. Logopedics Phoniat-
rics Vocology.
Harding R., Bocking A. D., and Sigger J. N. (1986). Influence of upper respiratory tract
on liquid flow to and from fetal lungs. Journal of Applied Physiology, 61, pp. 68-74.
Hixon T. J. (1991). Respiratory Function in Speech and Song. San Diego, California,
USA: Singular Publishing.
Konno K. and Mead J. (1967). Measurement of the separate volume changes of rib cage
and abdomen during breathing. Journal of Applied Physiology, 22, pp. 407-422.
Thomasson M. and Sundberg J. (1999). Consistency of phonatory breathing patterns in
professional operatic singers. Journal of Voice, 13, pp. 529-541.
Thorpe C. W., Cala S. J., Chapman J., and Davis P. J. (2001). Patterns of breath support
in projection of the singing voice. Journal of Voice, 15, pp. 86-104.
Synchronization of motion and timing in

clarinet performance
Caroline Palmer1, Erik Koopmans1, Christine Carter2, Janeen D. Loehr1,

and Marcelo Wanderley2

2 Schulich School of Music, McGill University, Canada
We examined the effect of expressive intent on timing and movement in

clarinet performance. Clarinetists performances were recorded with mo-
tion capture while they performed with three expressive intents: expres-
sive (normal) performance, exaggerated performance, and inexpressive
performance. Acoustic measures (intensity, pitch height, duration) were
compared with ancillary gestures (bell movement). The more expressive
performances contained larger expressive timing measures and bell
movement. Clarinetists marked phrase boundaries with increases in both
expressive timing and clarinet motion. Neither pitch height nor sound
intensity accounted for bell movements beyond expressive timing. These
findings suggest that ancillary bell gestures are rule-governed and corre-
late with some acoustic features of musical expression.
Keywords: expressive timing; clarinet performance; motion capture;

movement analysis; ancillary gestures
How do performing musicians express a range of musical intentions? Pat-

terns of expressive timing have been linked to performers expressive inten-
tions (Palmer 1989), and various auditory cues provide sufficient information
for listeners to distinguish between levels of expressive intent (Kendall and
Carterette 1990). Visual cues resulting from pianists head and upper torso
movements have been shown to be more effective than auditory cues in con-
veying performers intended expressivity, suggesting that movement best
conveys expression (Davidson 1995). In piano performance (on which most
research is conducted), expressive movement is limited to the hands and the
upper torso/head region (Davidson 2007); a larger variety of movements may
be found in the performance of wind instruments, including the clarinet
(Wanderley 2002). Limited research to date combines acoustic and move-

ment features of expressive performance. The current study examines the
effect of expressive intent on both timing and movement in clarinet perform-
ance.
Expressive rubato timing in music performance is characterized by in-
creased tone durations as performers approach cadences or phrase endings
(Gabrielsson 1987, Palmer 1989), and additional lengthening of tones notated
with long durations and shortening of tones notated with shorter durations
(Sundberg et al. 1991). Inexpressive performances tend to exhibit less range
of rubato timing than expressive performances (Gabrielsson 1987, Palmer
1989). These studies, however, did not investigate performers movements or
their relation to musical intent.
Wanderley (2002) analyzed clarinetists movement gestures while they
performed the same music in expressive and inexpressive ways. Following
Delalandes (1988) classification of musicians ancillary gestures as those
visible body movements that are not directly linked to sound production,
Wanderley (2002) recorded the ancillary movements of the clarinet bell with
a motion tracker system. Surprisingly, clarinetists vertical and horizontal bell
movements were highly consistent across their repeated expressive perform-
ances. However, Wanderleys instructions to clarinetists to move the clarinet
as little as possible in the inexpressive condition precluded comparisons of
expressive timing with movement.
The current study examines the relationship of clarinetists ancillary
movements to their expressive goals by combining Palmers (1989) expressive
performance conditions with Wanderleys (2002) measures of bell motion in
clarinet performance. Skilled clarinetists performed the same musical piece
under three different expressive instructions: normal expressive, exaggerated,
and inexpressive performance. No instructions were given concerning move-
ment. We tested the hypotheses that increased expressive intent results in
both increased variation in timing (Palmer 1989) and larger amplitudes of
ancillary movement (Davidson 2007, Wanderley et al. 2005).
METHOD
Participants
Eight adult clarinetists from the Montreal community participated in this

study (with 8-20 years of performing experience, mean=12 years).
Materials
The first eight measures of Mozarts Clarinet Concerto in A Major (K. 622),
second movement, were used. The excerpt (Figure 1) was chosen for its high
expressive content and the fact that it is well known. All clarinetists per-
formed on an A clarinet. All clarinetists were familiar with the excerpt before
participating. An active motion capture system (NDI Optotrak Certus) meas-
ured the clarinetists movements at a 250 Hz sampling rate. Markers were
placed on each fingernail tip of the clarinetists hands and on the clarinet: one
marker on the barrel (below the mouthpiece), one centered just above the
bell, and two on the bell. These markers provided information about the ori-
entation of the clarinet. An AKG C414 B-XLS standing microphone, as well as
a microphone integrated into the clarinet barrel, recorded the audio at 44.1
kHz. The clarinet movements are reported only up to the notated ornament in
measure seven, which permitted a variety of temporal interpretations and
was therefore excluded.
Procedure
The clarinetists were instructed to perform the melody as they would in a

normal concert setting (normal), with an exaggerated version of their inter-
pretation (exaggerated), and with a flat and inexpressive interpretation (inex-
pressive). The orientation of the clarinet was computed from the marker data
in spherical coordinates, eliminating translational position differences that
arise between participants of different heights and standing positions. Clari-
net elevation (vertical rotation) was measured as degrees of rotation around
the mouthpiece marker, relative to a rest position perpendicular to the floor.
Tone onsets were detected by identifying large negative peaks in the first de-
rivative of sound intensity, taken from the integrated barrel microphone.
Audio and motion data were smoothed using functional data analysis (Ram-
say and Silverman 2005) and registered by tone onsets.
RESULTS
Figure 1 shows the mean clarinet elevation in each expressive condition;

positive values indicate the bell was raised relative to the mouthpiece. A
functional Analysis of Variance (ANOVA) on the continuous elevation data
indicated significantly lower elevations at the beginnings of phrases as clari-
netists performed more expressively (significant regions are shown in hori-
zontal lines at the bottom of Figure 1). Overall, clarinetists were more variable
Figure 1. Mean clarinet bell elevation (degrees) in each expressive condition. Horizontal
lines indicate significance regions (grey and black indicate p<0.05 and p<0.01, respec-
tively); vertical shaded bars indicate locations of rests at phrase boundaries.
Figure 2. Mean timing deviation, bell elevation, and intensity (from standing mic) for
each expressive condition. Gaps indicate locations of ends of phrases.
in clarinet elevation (measured by standard deviation) from inexpressive to

exaggerated conditions (F2,14=33.04, p<0.01).
Next we measured expressive timing, in terms of the difference between
observed and expected interonset intervals (in %) relative to the notated du-
rations in the score ([observed-expected]/expected); positive values indicate
lengthened tones. Figure 2 shows clarinetists mean expressive timing devia-

tions, which indicated slowing from the beginnings to the ends of each
phrase. Furthermore, the variability of timing deviations (measured in stan-
dard deviation) increased with expressive intent (F2,14=5.3, p<0.01). Figure 2
(bottom) shows the peak intensity for each tone, which reflected the phrase
structure, but in a different pattern (inverted U-shape) than that of the eleva-
tion and timing deviation measures.
To examine whether the ancillary gestures were related to the acoustic
features of performance, regression analyses were conducted to predict the
bell elevations for each tone from expressive factors: timing deviations, sound
intensity (correlated with loudness), and sounded pitch (as measured by mo-
dal frequency). Individual correlations among variables indicated that sound
intensity and pitch correlated with each other (mean r=0.82, p<0.01). In ad-
dition, sounded pitch correlated with elevation in expressive performances
(mean r=0.51, p<0.05); however, only expressive timing predicted elevation
significantly when other acoustic features were controlled in each expressive
condition [semipartial correlations: standardized regression coefficient ()
for timing=0.65 in exaggerated condition, p<0.01; =0.58, p<0.01 in expres-
sive condition; =0.44, p=0.06 in inexpressive condition].
DISCUSSION
Expressive timing and clarinet bell motion increased across each phrase with
clarinetists expressive intent: performers used more bell motion and more
expressive timing when performing with greater expressive intent. Clarinet-
ists may use the bells elevation to help shape phrases, by physically repre-
senting change in musical features such as tension/relaxation. Together with
previous findings that documented performers reduced expressive timing
when musicians were asked to perform without moving (Wanderley et al.
2005), these findings suggest a bi-directional relationship between expres-
sion and movement in performance. The clarinet elevations and the expres-
sive use of rubato increased from phrase beginnings to endings. Clarinet tone
intensities instead showed an inverted U-shape throughout each phrase.
This study extended several findings on expressive performance to the
study of ancillary gestures in a wind instrument that permits considerable
degrees of motion. Our findings suggest that ancillary bell gestures are rule-
governed; they correlate with some but not all acoustic features of musical
expression, and in particular with phrase structure. Future research may de-
rive musical variables from performers movements by creating a taxonomy of
gestures that correspond to specific musical intentions.
Acknowledgments
This work was supported in part by Canada Research Chair and NSERC Grant 298173 to
the first author, by an NSERC-USRA fellowship to the second author, by a SSHRC fel-
lowship to the third author, and by an NSERC-CGS fellowship to the fourth author.
Caroline Palmer, Department of Psychology, McGill University, 1205 Dr Penfield Ave-

nue, Montreal, Quebec H3A 1B1, Canada; Email: caroline.palmer@mcgill.ca
References
Davidson J. W. (1995). What does the visual information contained in music perform-
ances offer the observer? Some preliminary thoughts. In R. Steinberg (ed.), Music
and the Mind Machine (pp. 105-114). Berlin: Springer.
Davidson J. W. (2007). Qualitative insights into the use of expressive body movement in
solo piano performance: A case study approach. Psychology of Music, 35, pp. 381-
401.
Delalande F. (1988). La gestique de Gould. In L. Courteau (ed.), Glenn Gould Pluriel
(pp. 85-111). Qubec City, Canada: Lousie Courteau Editrice.
Gabrielsson A. (1987). Once again: The theme from Mozarts piano Sonata in A Major
(k.331). In A. Gabrielsson (ed.), Action and Perception in Rhythm and Music (pp.
81-104). Stockholm: Royal Swedish Academy of Music.
Kendall R. A. and Carterette E. C. (1990). The communication of musical expression.
Palmer C. (1989). Mapping musical thought to musical performance. Journal of Ex-
perimental Psychology: Human Perception and Performance, 15, pp. 331-346.
Ramsay J. O. and Silverman B. W. (2005). Functional Data Analysis (2e). New York:
Springer.
Sundberg J., Friberg A., and Fryden L. (1991). Common secrets of musicians and listen-
ers: An analysis-by-synthesis study of musical performance. In P. Howell, R. West,
and I. Cross (eds.), Representing Musical Structure (pp. 161-197). London: Aca-
demic Press.
Wanderley M. M. (2002). Quantitative analysis of non-obvious performer gestures. In I.
Wachsmuth and T. Sowa (eds.), Gesture and Sign Languages in Human-Computer
Interaction (pp. 241-253). Berlin: Springer Verlag.
Wanderley M. M., Vines B., Middleton N. et al. (2005). The musical significance of
clarinetists ancillary gestures: An exploration of the field. Journal of New Music
Research, 34, pp. 97-113.
Thematic session:
Memory and performance
Memory for tactus and musical tempo: The

effects of expertise and speed on keeping time
Philip Fine1 and Stephen Bull2

2 Department of Early Music, Lisbon Higher School of Music, Portugal
The ability to keep time and remember speeds is important in musical

performance. Eight musicians and eight non-musicians heard a metro-
nomic beat at three musical speeds35, 110, and 185 beats per minute
(bpm)and clapped in time to it as accurately as possible. They were
then asked to recall the three speeds by clapping alone. After another set
of trials clapping to the metronome, they recalled the three speeds once
again. Results showed that the participants tended to clap too slowly for
the 110 bpm and 185 bpm tempi and that the medium speed was less ac-
curately remembered than the slow or fast. There was no effect of musi-
cal experience, but there was a trend of better recall of speed on the
second than the first recall trial.
Keywords: tactus; absolute; tempo; memory; clapping
One important aspect of much musical performance is the ability to keep a

beat or tactus going generally at a fairly steady rate, expressive variations in
tempo aside. In larger groups a conductor will often act as an external time-
keeper by dictating a self-paced beat, trying to make it as steady and constant
as possible. It is the responsibility of all players to follow the conductors beat
and play at the same tempo. Their behavior can be seen as a form of sensori-
motor synchronization (SMS), where they synchronize their motor behavior,
whatever that may be, to the conductors visual beat. Thus both the conduc-
tors internally paced beat and the players synchronization behavior are
forms of rhythmic time keeping. Research on SMS and self-pacing has shown
that there are a number of factors that affect the accuracy and variability of
the beat, and also that there are upper and lower tempo limits (e.g. Repp
2005, London 2004, Aschersleben 2002).
Keeping in time is clearly relevant to musical performance, and respond-

ing regularly (e.g. tapping or clapping) is a commonly used method of meas-
uring SMS ability. The asynchrony between the given beat and the
participant's response indicates their accuracy. Previous research has sug-
gested this SMS accuracy is affected by various factors, including the tempo of
the tactus and the participant's musical expertise (Repp 2005). Research
when the beat disappears but the participant has to keep tapping at the same
tempo has also been conducted. This is called continuation tapping, and typi-
cally appears to show an approximately 12 s period of maintained control
(Chen et al. 2002).
Another important skill in performing music in most cultures is the ability
to set and control the tempo: too fast and the piece may be unplayable and
detail glossed over; too slow and perceptual and structural coherence and
performers breathing (if relevant) may suffer. Conductors and unconducted
performers (often soloists) must be able to start a piece at the right tempo,
and better performances tend to show less evidence of tempo drift or un-
planned corrections or fluctuations in speed. Although there is much evidence
concerning absolute pitch (AP), the ability to recall at will a pitch from long
term memory (e.g. Ward 1999), there is less on absolute tempo (AT), the
ability to remember a particular beat speed (i.e. tactus) and recall it when
required for performance.
Previous research has tested tempo recall for pieces in long term memory.
Levitin and Cook (1996) had participants sing popular songs with which they
were familiar, and 72% of the productions were within 8% of the original
tempo. Trainor et al. (2004) also showed that infants preferred a familiar
piece provided the tempo was not changed by more than 25%, suggesting a
long-term tempo memory, though probably less accurate than that of adults.
The aims were, therefore, twofold. First, we investigated the effect of mu-
sical expertise on people's accuracy for carrying a tactus and remembering a
tempo in the absence of a specific piece of music, both measured by clapping
in time. It was hypothesized that expert musicians would be more accurate
both in their tapping accuracy and in their long term memory for tempo than
non-musicians. Secondly, we investigated the effect of different speeds of a
given tactus on tapping in time and, in particular, on tempo recall.
METHOD
Participants
There were 16 participants, split into two equal groups of expert musicians
(professional, semi-professional, and student) and those whose professional
life did not involve music. The musicians (5 M, 3 F) were aged between 20
and 48 years (mean=30.8). Some were specialists either in conducting or in
percussion playing. Non-musicians (6 M, 2 F) were aged between 23 and 74
years (mean=49.9), and had no musical training beyond school, though two
had choral singing experience, and one was a self-taught drummer.
Materials
Metronomic recordings of three different tempi: quarter note=35 bpm corre-

sponding to interstimulus interval (ISI) of 1714 ms, quarter note=110 bpm
corresponding to ISI of 545 ms, and quarter note=185 bpm corresponding to
ISI of 324 ms, were made on a MacBook 4.1 (2.1 GHz processor) in the soft-
ware Finale, using the sound of a woodblock. This was transferred to a CD,
along with instructions, and the recordings were played to participants in
various testing locations. Attention was given to minimizing ambient sound
and to avoiding the presence of aural or visual rhythmic stimuli in the room.
Participants productions together with the metronomic stimuli were re-
corded on the same MacBook computer and converted to AIFF format files.
These were then analyzed using the software program Praat (version 5.1.04)
to measure times between claps (Boersma and Weenink 2009).
Procedure
There were two tasks. First, participants were asked to clap, synchronizing as
closely as possible with a recorded beat (the beats on the recording disap-
peared and reappeared, see below). In the second task, participants were
asked to reproduce from memory the speeds of the original recording. The
three speeds of 35, 110, and 185 bpm (slow, medium, and fast, respec-
tively) were chosen so as not to be related to each other by any simple pro-
portion. Tempo order was counterbalanced between participants.
In the first section of each test, participants were asked to listen to the re-
cording of beats, to join in after hearing the first four beats, and to continue
clapping until told to stop. Some beats on the metronomic stimulus were ab-
sent and the participants were informed of this but asked to continue clap-
ping nonetheless. In all tests, there were 8 beats present, 4 absent, 4 present,
4 absent, 8 present, 8 absent, 8 present, and 8 absent. Thus, clapping regu-
larity in both the presence of a beat (SMS) and its absence (continuation)
were tested. Participants performed the above test at the three speeds and
then repeated them in the same order. A gap of about 30 s was left between
tests. They were then asked to fill in a questionnaire covering age, gender,
and musical experience. Approximately 4 mins were given for this.
Figure 1. Mean, minimum, and maximum recorded clapping tempi (bpm) as a function
of the given tempo. The target tempo is shown by the solid line. (See full color version at
Next, participants memory for the three speeds was tested. They were
asked to clap at the slow, medium, and fast speed, in the same order they had
already been tested on, both beginning and ending when indicated. A total of
about 32 claps were recorded for each trial. After the recall test of each speed,
the first test (clapping with the recorded beats) was repeated, followed, after a
1 min pause, by the test of memory once again. A complete testing session
took approximately half an hour per participant.
RESULTS
The present article focuses on the participants tempo memory data. Overall,
participants were fairly good at recalling the slow tempo but tended to under-
estimate both the medium and fast tempi (i.e. clapping too slowly) quite sub-
stantially (see Figure 1). However, their recall for these tempi was generally
better on the second recall trial, underestimating less (see Table 1). When the
musicians and non-musicians were analyzed separately, the musicians were
generally found to recall the medium and fast tempos more accurately than
the non-musicians, underestimating the speeds less (see Table 2). A 3-way
Analysis of Variance (ANOVA) with tempo, trial, and musical experience as
factors was conducted on the mean error as a percentage of the target tempo.
Table 1. Clapping tempo (bpm) in the recall tests (SD in brackets.)
Tempo Target tempo Recall 1 Recall 2

Slow 35 34.7 (4.86) 38.5 (13.36)
Medium 110 75.0 (23.06) 87.8 (28.52)
Fast 185 156.9 (35.02) 166.0 (23.75)
Table 2. Clapping recall tempo (bpm) as a function of musical training (SD in brackets).
Target tempo Musicians Non-musicians

Recall 1 Recall 2 Recall 1 Recall 2
Slow (35) 32.0 (3.24) 32.9 (3.02) 37.0 (5.01) 43.5 (17.05)
Medium (110) 74.5 (24.91) 97.5 (34.43) 75.4 (24.85) 80.5 (15.86)
Fast (185) 165.9 (21.46) 175.0 (22.95) 146.7 (45.90) 157.1 (23.75)
Tempo was highly significant (F2,22=10.96, p<0.01): Bonferroni post-hoc tests

showed that the medium tempo was recalled significantly more poorly than
either the slow or fast tempi (as can be seen from the mean and target tempo
lines in Figure 1). Trial showed a non-significant trend (F1,22=4.14, p<0.07), in
that the tempo error was smaller in the second trial than the first. Musical
experience was not found to be a significant factor.
DISCUSSION
Overall there was large variation in remembered tempo, particularly showing

underestimation (see Figure 1), but the slow and fast tempi were recalled
better than the medium tempo, at least in terms of percentage errors. It is
unclear why fast speeds were better remembered than the medium speed,
although future tests may study the relevance of remembering fewer tempi in
a test. However, musical experience was not found to affect tempo recall,
there being no difference between the two groups. This may be because three
of the non-musicians in fact had some musical experience (see Partici-
pants), diluting the difference between groups. Tempo memory was gener-
ally better on the second trial, after the participants had carried out more
SMS/continuation tasks. The continuation claps can be thought of as testing
short-term memory, with in effect immediate feedback, so this was not sur-
prising.
In future research, we intend to ask participants to clap to tempi identi-
fied by name (e.g. quick march and slow dirge) or bpm (e.g. quarter note=
65 or 140), rather than the terms slow, medium, and fast. We will vary
the number of tempi included in tests and accurately control the time elaps-
ing before participants are required to reproduce a tempo from memory. We
also intend to include distracter tasks, which may affect mental rehearsal, and
compare the use of a beat pad with clapping, to reveal the relevance of the
choreographic element of clapping.
Acknowledgments
Thanks to Derrick Walford for assistance with the data measurement using Praat.

References
Aschersleben G. (2002). Temporal control of movements in sensorimotor synchroniza-

tion. Brain & Cognition, 48, pp. 66-79.
Boersma P. and Weenink D. (2009). Praat: Doing Phonetics by Computer (version
5.1.04). Computer program, available online at www.praat.org.
Chen Y., Repp B H, and Patel A. D. (2002). Spectral decomposition of variability in
synchronization and continuation tapping: Comparisons between auditory and vis-
ual pacing and feedback conditions. Human Movement Science, 21, pp. 515-532.
Levitin D. J. and Cook P. R. (2006). Memory for musical tempo: Additional evidence
that auditory memory is absolute. Perception & Psychophysics, 58, pp. 927-935.
London J. (2004). Hearing in Time. Oxford: Oxford University Press.
Repp B. H. (2005). Sensorimotor synchronization: A review of the tapping literature.
Psychonomic Bulletin and Review, 12, pp. 969-992.
Trainor L. J., Wu L., and Tsang C. D. (2004). Long-term memory for music: Infants
remember tempo and timbre. Developmental Science, 7, pp. 289-296.
Ward W. D. (1999). Absolute pitch. In D. Deutsch (ed.), The Psychology of Music (pp.
265-298). San Diego, California. USA: Academic Press.
Moving backwards and forwards in time:

Recalling dance from long-term memory
Catherine Stevens1, Jane Ginsborg2, and Garry Lester3
1 MARCS Auditory Laboratories, University of Western Sydney, Australia

3 NAISDA Dance College, National Aboriginal and Islander
Skills Development Association, Australia
Recalling complex non-verbal motor sequences, such as contemporary

dance, provides insights into cues and associates in long-term memory
(LTM). Four dancers recalled material that they had not performed for
between 3-31 years; they initially recalled the material in silence, al-
though the material had been associated with particular musical ex-
cerpts. Eight exercises were recalled either immediately or after an un-
filled delay and using self-motion or a mannequin. Recall was greater
when immediate rather than delayed; self- versus mannequin-motion
had little impact on the length of material recalled. Qualitative data indi-
cated that music and dancer movement were important cues to LTM,
transitions were sometimes forgotten, and images associated with
movement recall were kinesthetic, verbal, visual, and auditory. The re-
sults suggest an interplay of procedural and declarative knowledge and
activation of multimodal images.
Keywords: movement; dance; procedural memory; declarative memory;

non-verbal behavior
At once non-verbal, communicative, and expressive, contemporary dance is

also visual, spatial, temporal, auditory, kinesthetic, affective, and dynamic.
This study investigates four dancers use of cues and associations when re-
calling complex dance sequences from LTM after intervals from 3-31 years.
Retrieval cues and associates in long-term memory
Jazz and classical musicians use knowledge of formal musical structure to aid
retrieval (Chaffin and Imreh 2002, Noice et al. 2008). Motor cues aid actors
recall of text and contextual cues significantly aid retrieval (Noice and Noice
1997, 2002). Music prompts increased recall for prose and verse, and the
recall of such material tends to begin and end at breath pause locations
(Rubin 1997).
In contemporary dance, there is no set repertoire of steps or verbal asso-
ciates. Contemporary dance and choreographic traditions are passed on not
in words but by showing and doing (Grove 2005). In the absence of notation
and scores the dancers are dynamic archives of works that they have re-
hearsed and performed. Smyth and Pendleton (1994) proposed that move-
ment involving configuration of body parts is encoded kinesthetically. Overby
(1990) has suggested that, while novice dancers prefer a visual mode of
thought, experienced dancers are inclined to use both verbal and visual
modes of thinking.
Margaret Barr dance-drama
Margaret Barr had a choreographic career that spanned more than 60 years
and three continents. Barr described her work as dance-drama because she
was interested in both the inherent drama of the body in motion and its abil-
ity to create and sustain the human drama of her primarily narrative based
works. Barrs technique classes were a series of set studies performed to the
driving rhythms of Carl Orffs Carmina Burana. This choreographed material
explored a broad range of movement dynamics including the elegant simplic-
ity of her lyrical studies to the extremes of strong, direct, thrusting sequences.
Throughout her classes great attention was given to the rhythm of the breath
cycle as the dancer moved. For Margaret the dramatic moment in a breath
cycle was the breath pausethe moment of stasis between the inhalation of
breath and its exhalation.
Long term recall of dance: Choosing a method of investigation
The opportunity to observe four mature dancers recalling movement material

that they have not performed for between 3-31 years is unique and challeng-
ing. There is a need to exercise some experimental control by systematically
manipulating potentially informative key variables although variables such as
dancer age, years of exposure, years since previous recall/performance, cur-
rent dance involvement, and body flexibility were out of our control. Cues to
recall were brought under partial experimental control with the dancers
asked to recall a specific dance exercise prompted initially by its verbal label,
and to recall using either their own body or a small mannequin, either imme-
diately or after a 20 s delay. Delayed rather than immediate recall gives time
for mental rehearsal and visual and/or motor imagery that should enhance
recall. The body versus mannequin conditions permitted the investigation of
procedural recall in the form of self-motion versus motion of another figure
without use of words. We anticipated greater recall using self-motion.
METHOD
Participants
Four female dancers who had all trained with the same choreographer vol-
unteered for the study (mean age=55.38 years, SD=10.13, range=44.0-68.5).
Two had performed the choreographers dance material within the past three
years, whereas the oldest participant had not performed it for 22 years
(dancer 2, C) and the second youngest participant for 31 years (dancer 3, F).
Materials and procedure
The eight exercises to be recalled were: Headwork Series, Simple Arms,

Complex Arms, Egyptian, Spiral, Thursday night class Lyric: Assyrian
Bull, Figure 8 Arm Rolls, and The Falls. Each had a descriptive label and was
associated with a particular piece of recorded music. Participants recalled
four exercises using their own body and four using the mannequin, either
immediately or 20 s after the label had been given. The mannequin was a 30
cm wooden artists doll as used in life drawing and art. A questionnaire asked:
(1) would you describe how you felt performing under each of these condi-
tions: (a) immediate body, (b) immediate doll, (c) delayed body, (d) delayed
doll?, (2) how do you feel that music, others in room, social, and emotional
factors affected recall of this movement material from your long-term mem-
ory?.
RESULTS
The duration of recalled material performed by individual participants was

calculated as a proportion of the music excerpt duration. Quantitative recall
data have been calculated from the 3 participants who had not performed the
material for 3 years (dancer 1), 22 years (dancer 2) or 31 years (dancer 3).
Table 1 shows the duration (in seconds) of material recalled by exercise and
by recall condition. The proportion of recalled material duration as a function
of the musical (veridical) excerpt is also shown. In Table 2, proportion of
recall duration is presented according to recall condition. It was hypothesized
that recall is greater in the delayed than in the immediate recall condition.
However, the mean proportions indicate the reverse with proportion recall of
Table 1. Duration of recall by dancer and seven exercises showing raw scores (in sec-
onds) and as proportions of the duration of the associated musical excerpt. (Proportions
shown in brackets; 0 refers to zero recall.)
Headwork Series Simple Arms Complex Arms Egyptian

199 s 50 s 50 s 150 s
Dancer 1 (K) 68.00 (0.34) 34.00 (0.68) 45.00 (0.90) 27.00 (0.18)
Dancer 2 (C) 29.00 (0.15) 78.00 (1.00) 31.00 (0.62) 54.00 (0.36)
Dancer 3 (F) 23.00 (0.12) 36.00 (0.72) 34.00 (0.68) 10.00 (0.06)
Mean 40.00 (0.20) 49.33 (0.80) 36.66 (0.73) 30.33 (0.21)
Spiral Fig. 8 Arm Rolls The Falls

cont. 50 s 135 s 50 s
Dancer 1 (K) 36.00 (0.72) 30.00 (0.22) 53.00 (1.00)
Dancer 2 (C) 25.00 (0.50) 0 27.00 (0.54)
Dancer 3 (F) 67.00 (1.00) 15.00 (0.11) 42.00 (0.84)
Mean 42.66 (0.74) 15.00 (0.11) 40.66 (0.79)
0.55 (SD=0.38) in the immediate recall condition and 0.44 (SD=0.31) in the
delayed recall condition. There was no difference between proportion recall
duration when the dancers own body was used (mean=0.50, SD=0.29) com-
pared with when the mannequin was used for recall (mean=0.49, SD=0.40).
Dancers reflected on their recall of each exercise and completed a short
written questionnaire. Their responses are organized below according to in-
dependent variables, cues, memory associates, and images.
Recall using the mannequin: immediate doll condition required thinking
as I moved it, i.e. feeling it in my own body as I moved the doll (K), doll
totally unable to movedisjointedresults lacking flow and staccato posi-
tioning which [is] what all/most dancers try to avoid (C), could not visualize
movement on the doll. Perhaps a real body would have triggered more mem-
ory (F).
Immediate versus delayed conditions: immediate and delayed body
didnt make much difference more that it immediately came to mind or not
(S). W commented that having time to rehearse the exercises in the delayed
recall condition led to some confusion, while F wrote that in the delayed body
recall condition she had a lot of trouble remembering anything first without
moving.
Table 2. Mean duration recall proportions collapsed across seven exercises and three
dancers and shown as a function of recall condition. (SD shown in brackets.)
Recall Immediate Immediate Delayed Delayed

condition body mannequin body mannequin
Mean proportion 0.53 (0.35) 0.56 (0.40) 0.47 (0.24) 0.41 (0.39)
Auditory imagery: During different exercises (Headwork, Simple Arms,

Spiral), all four dancers noted that during recall an inner voice was active:
forward, back, bring wrist forward, bring wrist back (W), forward, back,
forward, back, side, side, then around (K), forward and right, up and over
(C). All commented on their recall of Barrs voice as they performed the exer-
cises and the rich intonation and drama of her voice in their imagery.
All dancers commented on the importance of music. For W, recalling Spi-
ral music is such a key; it tells you theres nine swings, and more generally
absence of music meant that I had to hum the music to myself as I was
dancing/recalling, the music for me is essential to conveying the right se-
quence. C commented that she did not think she knew some exercises with-
out the music; the music rather than a label would trigger recall. For F, the
rhythm of the music was important. For K, the music timed the movement,
lack of the music made remembering more difficult.
What is forgotten? Two of the four dancers reported forgetting a transi-
tion: at the opening of Spiral (W), forgetting how to get up after a particular
phrase (F), and finishing an exercise suddenly having forgotten the middle
section (F). This pattern is consistent with schema theory junction errors
and switchesa kind of structural performance cue (Chaffin et al. 2002).
DISCUSSION
For an artform such as contemporary dance, where formal notation is more

the exception than the rule, dancers brains and bodies are repositories of
works of art that have been learned and memorized. The present study dem-
onstrates that multi-modal cues elicit and accompany long-term memory for
movement. Music, with its inherent sequential and temporal order, is a pro-
vocative and effective cue. Imagery is not only visual and verbal but also kin-
esthetic and auditory. Transitions between movement sequences may be
susceptible to forgetting, especially in the absence of sequential cues from
music. The dancers opportunity to move is important for dance recollection.
Future studies might focus on the learning and memorization of new mate-
rial. Interference paradigms could also be used effectively to shed light on the
nature of the memory code.
Acknowledgments
This research was supported by the Australian Research Council (LP0562687) and the
Australia Council for the Arts, Ausdance, and QL2 Centre for Youth Dance. We thank
Mark Gordon and our dancer participants for their generous contribution of time and
energy.
Catherine Stevens, MARCS Auditory Laboratories and School of Psychology, University

of Western Sydney, Locked Bag 1797, South Penrith DC, New South Wales 1797, Aus-
tralia; Email: kj.stevens@uws.edu.au
References
Jersey, USA: Erlbaum.
Grove R. (2005). Show me what you just did. In R. Grove et al. (eds.), Thinking in Four
Dimensions (pp. 37-49). Melbourne, Australia: Melbourne University Press.
Noice H., Jeffrey J., Noice T., and Chaffin R. (2008). Memorization by a jazz musician:
A case study. Psychology of Music, 36, pp. 63-79.
Noice H. and Noice T. (1997). Long-term retention of theatrical roles. Memory, 7, pp.
357-382.
Noice T. and Noice, H. (2002). Very long-term recall and recognition of well-learned
material. Applied Cognitive Psychology, 16, pp. 259-272.
Overby L. Y. (1990). A comparison of novice and experienced dancers imagery ability.
Journal of Mental Imagery, 14, pp. 173-184.
Rubin D. C. (1997). Very long-term memory for prose and verse. Journal of Verbal
Learning and Verbal Behavior, 16, pp. 611-621.
Smyth M. M. and Pendleton L. R. (1994). Memory for movement in professional ballet
dancers. International Journal of Sport Psychology, 25, pp. 282-294.
Memory span in dance: Influence of age and

experience
Sofia Carvalheiro and Lus Xarez Rodrigues
Faculty of Human Kinetics, Technical University of Lisbon, Portugal
The retention and recovery of movements are capacities that characterize

and determine the life of dancers and dance students. The application of
certain memory strategies is necessary and leads to an improvement in
motor performance. Usually, the lowest levels of retention are associated
with younger dancers when their performances are compared with the
oldest and more experienced in dance. We aim in this study to evaluate
capacity of the short-term motor memory of children and adults, with
and without experience in dance, when asked to reproduce immediately
(after presentation) sequences of movements. Results show that experi-
ence is crucial when comparing the maximal number of actions and se-
quences recalled in the correct order. For these same variables, children
show lower results than adults, but these differences are not significant
when we compare children with experience with adults without experi-
ence.
Keywords: motor memory; memory span; retention capacity; experience

in dance; motor actions
The quantification of the capacity to retain information is a topic of interest

since the 1950s, especially after the famous Millers article of 1956. The first
studies used exclusively words (Longstaff 1998), but after the end of the cen-
tury, motor memory started to be studied as movements were used as stimuli.
In the research of learning and motor control, motor memory is one of the
crucial topics of study.
In dance, the capacity to memorize is very important, particularly in its
artistic dimension which is very selective and in which professionals have to
perform huge sequences of motor actions, sometimes for more than an hour.
As actors have to retain/memorize a text, dancers are distinguished in many
cases for their good ability to memorize long passages of movement.
The amount of information and the time available for its retention are two
important aspects when we talk about memorization. According to several
authors (Marteniuk 1976, Perlmutter and Hall 1992, Weiss 1995) Millers
work has established a referential: the short-term memory has an overall
capacity of 72 items. The recovery of more information is dependent on the
use of the long-term memory.
The short-term memory increases considerably with age, so the capacity
to memorize is generally lower in children when compared with adults. This
difference is due to the lack of knowledge and strategies of memory that chil-
dren present (Ille and Cadopi 1999). As children grow older they acquire a
repertoire of strategies that will help to solve problems: from age seven they
begin to use strategies of rehearsal/memorization spontaneously, and until
age eleven, a rapid increase in the number and quality of strategies is evident
(Bouffard and Dun 1993, Weiss and Klint 1987). The acquisition of the label-
ing, repetition, and cluster strategies increases progressively up to eleven
years old, and during the next two years children acquire other strategies.
One problem that arises when studying motor memory is how to measure
itthat is, to find an appropriate unit of measurement to quantify this capac-
ity. It may be easier to identify the number of words in a spoken or written
discourse than the number of movements of a person who is dancing. The
tests that are used to measure the capacity of short-term memory are called
memory span. This type of testing identifies the longest sequence of non-re-
lated items that can be recalled immediately in the same order as they were
presented.
The measure established by Miller (72 chunks of information to the
short term memory capacity) means that the amount of information per
chunk varies. This leads the author to conclude that capacity of short-term
memory is a matter of organization of material to remember in units of
meaningful information, so that maximum information is retained (Mart-
eniuk 1976). But can the amount of recalled words be applied to, or similar to,
the ability to remember movements?
According to Bavelier et al. (2006) the capacity of the short-term memory
depends on the nature of the information, for example 72 items for language
and 41 for visual-spatial items. Furthermore, the ability to organize infor-
mation into chunks seems to depend more on familiarity and experience that
the individual has with the skill or knowledge area, than their pre-existing
capacity to store more information in memory (Rose 1997). This study tries to
identify the largest number of motor actions that subjects with and without
experience in dance are able to recall after the presentation of various se-
quences of motor actions in dance. We wanted to compare children with
adults, so participants were organized in four groups: children and adults

with and without experience.
METHOD
Participants
The children group consisted of 1o female subjects with experience in dance

and 6 female subjects without experience in dance. They were randomly cho-
sen from the following criteria: age, experience, and availability. Both groups
have an average age of 11 years. We considered that subjects have experience
in dance when they dance for more than a year in a regular, formal, and sys-
tematic way, as is the case in vocational teaching. Children with experience
had on average 6 years of experience, with 3 hours per day. The adult group
consisted of 12 female subjects, 6 with experience in dance, and 6 without
experience, mostly university students. They had on average 12 years of ex-
perience with 3 hours per day.
Materials
In the present study we have applied the memory span test. This is a test of
presentation/reproduction of items (motor actions) in which subjects are
classified/categorized by the number of items that they can identify and recall
in the correct order (Bouffard and Dunn 1993). The material to be reproduced
by subjects (motor actions and movement phrases), developed by the authors,
was created for this study taking into account the degree of difficulty of the
motor actions and movement phrases. We aimed to use sequences of motor
actions without complexity, considering that a task with more relationships
between the parts that compose it is more complex than a task with fewer
relationships. For example, a jump with a rotation during the aerial phase is
more complex than a vertical jump from two to two feet.
We created twelve sequences of non-related motor actions and eleven se-
quences of related motor actions. The transition between each sequence is
characterized by adding a motor action. Structure units were taken to be the
motor actions of the classification of movement in dance presented by Rodri-
gues (1999). The motor actions were our unit of measurement: posture, bal-
ance, gesture, turns, steps, displacement, jumps, and falls.
Procedure
The memory span test was done individually; only the participant and the
researcher, responsible for projecting the images and recording the partici-
pant performance, were in the studio. The test started with the projection of
the first group of non-related motor actions. The subject observed the se-
quence twice and then was asked to repeat it immediately once. The test fol-
lowed the same procedure until the last sequence (12 motor actions). Subjects
were asked to reproduce exactly what they saw, maintaining the overall form
and also keeping the order of the presented movements. Subjects were not
allowed to reproduce/rehearse the film during its projection. After a break of
10-15 minutes the second test was applied. The procedure was the same as the
first, but this time subjects were asked to recall eleven movement phrases
(related motor actions) of 2-12 actions per phrase.
RESULTS
The two types of sequences of our study (non-related and related motor ac-
tions) were studied according to the following variables: maximum number of
motor actions recalled in the correct order, number of sequences recalled in
the correct order, number of sequences recalled independently of order, total
actions recalled in the correct order, and total actions recalled independently
of order.
The maximum number of motor actions recalled in the correct order
ranges from 2-11 and seems to be the most interesting variable in quantifying
the number of actions per sequence either in future studies or in learning
applications. The results range from 3.50 actions in the children group with-
out experience, to 8.00 actions in the group of adults with experience. The
other two groups realized almost 5.00 actions sequentially in the correct or-
der. For this variable, the approximate values obtained for the group of chil-
dren with experience and the group of adults with no experience is consistent.
The number of sequences recalled in the correct order ranges from 0-5,
with values of 3.67 actions for adults with experience, and 2.00 actions in
adults with no experience. This variable focuses on the sequence; the number
of actions executed correctly before the first error. This means that there were
some subjects (2 adults without experience and 1 child with experience) who
failed the first sequence of 2 related motor actions and that there were some
(2 adults and 1 child with experience) who continued accurately until the fifth
motor action sequence, after having also correctly completed the earlier se-
quences. In this variable, children with experience presented results slightly
higher than adults with no experience.
The results obtained from the analysis of the number of sequences re-
called independently of orderthat is 12 and 11 sequences (related and un-
related, respectively)consider all of the sequences that were executed
without error because one subject can correctly complete 4 actions, not com-
plete 5 actions, and correctly complete 6 actions, etc. For this variable, the
values are between 1 and 6 activities, with averages ranging from 1.17 for chil-
dren with no experience to 4.17 for adults with experience. The other two
groups have similar values with a slight advantage for children with experi-
ence (3.20-2.20 actions) compared with adults without experience (3.17-2.17
actions).
The total of actions recalled in the correct order varies from 15.58% in
children without experience to 76.62% in adults with experience, which is the
equivalent to 12 and 59 actions from the 77 actions presented by the model.
The average values range from 33.76%-69.26%, with the highest values be-
longing to adults with experience and the lowest to children with no experi-
ence. Analysis of the total of actions recalled independently of order showed
mean values generally higher than the values of the previous variable, with
values ranging from 34.19%-70.30%. To study further the age and experience
independent variablesand because of sample size and according the type of
dependent variableswe used the nonparametric Mann-Whitney-Wilcoxon
or the t-Student test. To compare the related and non-related data we used
the Wilcoxon test and, for the paired data, the t-Student test. We considered a
result significant if p0.05.
DISCUSSION
The results indicate that there are significant differences when comparing the
groups with respect to experience. The number of sequences recalled in the
correct order depends on this factor as it appears in all possible comparisons
and cross-comparisons. The other four variables in the study also appear of-
ten, which reinforces the idea that years of practice in a particular activity
influence the ability to recall movement, which despite being generic (not
specific to a particular dance technique) have dance as frame of reference of
the motor repertoire. With regard to age, it is confirmed that for the variables
under study (which appear to have consensus in the literature) children are
less able than adults to memorize movements; this disadvantage is more ap-
parent if the sequences are particularly long (more than 5 motor actions),
common in many situations of dance learning. When we associated age and
experience, we found an aspect that seems worthy of development in future
study: that the group of children with experience have similar or sometimes
slightly higher scores than the group of adults without experience. This, and
the size of the sample for different age groups, should be considered in future
studies.
Sofia Carvalheiro, Faculty of Human Kinetics, Technical University of Lisbon, Estrada

da Costa, Lisbon, 1495-688 Cruz Quebrada, Portugal; Email: sofiacarvalheiro@gmail.
com
References
Bouffard M. and Dunn J. G. H. (1993). Childrens self-regulated learning of movement

sequences. Research Quarterly for Exercise and Sport, 64, pp. 393-403.
Bavelier D., Newport E. L., Hall M. L. et al. (2006). Persistent difference in short-term
memory span between sign and speech: implications for cross-linguistic compari-
sons. Psychological Science, 17, pp. 1090-1092.
Ille A. and Cadopi M. (1999). Memory for movement sequences in gymnastics: Effects of
age and skill level. Journal of Motor Behavior, 31, pp. 290-300.
Kimmerle M. and Ct-Laurence P. (2003). Teaching Dance Skills. Andover, New Jer-
sey, USA: J. Michael Ryan Publishing.
Longstaff J. S. (1998). Subjective organization in the recall of abstract body movements.
Perceptual and Motor Skills, 86, pp. 931-940.
Marteniuk R. G. (1976). Information Processing in Motor Skills. New York: Holt,
Rinehart and Winston.
Miller G. A. (1956). The magical number seven, plus or minus two: Some limits on our
capacity for processing information. Psychological Review, 63, pp. 81-97.
Perlmutter M. and Hall E. (1992). Adult Development and Aging (2e). New York: John
Wiley and Sons.
Rodrigues L. X. (1999). Morfologia do movimento danado: Gneros coreogrficos e
comportamento motor na dana teatral ocidental. Unpublished doctoral thesis,
Technical University of Lisbon.
Rose D. J. (1997). A Multilevel Approach to the Study of Motor Control and Learning.
Boston: Allyn and Bacon.
Weiss M. R. and Klint K. A. (1987). Show and tell in the gymnasium: An investigation
of developmental differences in modelling and verbal rehearsal of motor skills. Re-
search Quarterly for Exercise and Sport, 58, pp. 234-241.
Weiss V. (1995). Memory span as the quantum of action of thought. Cahiers de Psy-
chologie Cognitive, 14, pp. 387-408.
Graduate award paper
Fast feedforward error-detection mechanisms

in highly skilled music performance
Mara Herrojo Ruiz1, Hans-Christian Jabusch1,2, and Eckart Altenmller1

Expert performance in music or sport requires the ability to monitor on-

going behavior, detect errors in advance, and modify the performance
accordingly. In this context, errors have dramatic consequences, such as
playing the wrong note in a piano concert or mis-hitting a smash during a
tennis match. Detection and correction of errors in advance is possible
due to the fast functioning of the self-monitoring system. Surprisingly,
only sparse evidence about error monitoring has been published in the
music domain. Consequently, the present study investigated the electro-
physiological correlates of predictive (feedforward) error detection and
action control during piano performance. Pianists had to retrieve memo-
rized music pieces at a fast tempo in the presence or absence of auditory
feedback. Only wrong pitches were considered as erroneous actions. The
main outcome was a negative component elicited around 70 ms prior to
errors in the event-related potentials and which is generated by the ante-
rior cingulate cortex (ACC). This component was independent of the
auditory feedback and was assumed to reflect error detection processes.
Moreover, an interaction between the ACC and the lateral prefrontal
cortex predicted corrective mechanisms. The findings presented here
shed new light on the neural mechanisms of feedforward motor control.
Keywords: ACC; EEG; errors; motor control; performance monitoring
Playing tennis, performing a piece of music from memory, and even leaping
in a rhythmic gymnastics exercise are all complex multimodal tasks which
rely on predictive (feedforward) mechanisms acquired through extensive
training. All these refined sensory-motor tasks depend on time-based se-
quential behaviors and, as such, require accurate preparation in advance of

the events planned for production (Lashey 1951, Schmidt 1975, Pfordresher
and Palmer 2006). Moreover, the performance of complex overlearned motor
programs requires the perfect tuning of the action-monitoring system to the
extent that potential errors, which might otherwise interact with the goals,
must be predicted in advance (Bernstein 1967, Wolpert et al. 1995). Over-
learned motor tasks thus present an ideal paradigm for the study of brain
activity associated with feedforward mechanisms of error detection. However,
in the context of expert performance, there is sparse information concerning
the neural mechanisms responsible for predictive error detection processes.
A seminal finding in the context of the human action-monitoring system
was a negative deflection in the event-related potentials (ERPs) of the elec-
troencephalogram (EEG), termed error-related negativity (ERN), which
peaks about 70 ms after the incorrect key press in reaction-time tasks (Fal-
kenstein et al. 1990). The neural generators of the ERN have been located in
the anterior cingulate cortex (ACC; Dehaene et al. 1994). The ERN has been
hypothesized to reflect the error signal of a feedforward control mechanism
(Holroyd and Coles 2002, Bernstein 1967). Thus, during goal-oriented be-
havior, the ACC is assumed to monitor the ongoing performance and signal
the need for higher control in erroneous trials (Miller 2000). Further, the
ACC is hypothesized to interact with lateral prefrontal areas (lPFC), which in
turn implement performance adjustments (Botvinick et al. 2001, Ridderink-
hof et al. 2004). Surprisingly, no specific data about this interaction have
been reported during the monitoring of skilled overlearned performance. The
aims of the present study were the following:
To investigate with EEG the neural correlates of error-monitoring in hu-

mans by using piano performance as an example of a complex multimodal
task.
To study the interplay between auditory and sensory-motor information
in the processes triggered by an erroneous actionnamely, a wrong pitch.
To test whether the ACC and lPFC interact by means of phase synchroni-
zation in the feedforward mechanisms acting in advance of overt errors.
METHOD
Participants
Eighteen healthy pianists (10 males, 8 females; age range=20-29 years,

mean=22 years), who were students at or had graduated from the Hanover
University of Music and Drama, participated in this study. All participants
were professional pianists. Seventeen of the participants were right handed

and one was left handed, according to the Edinburgh inventory (Oldfield
1971). All participants reported normal hearing.
Materials
The stimuli were six sequences extracted from the right-hand parts of the
Preludes V, VI, and X of the Well Tempered Clavier (Book I) by J. S. Bach
and the Piano Sonata No. 52 in E Flat Major by J. Haydn. These pieces were
chosen because their parts for the right hand contain mostly single pitches of
the same value (duration)i.e. sixteenth notes. The number of notes per se-
quence was approximately 200. The tempo for each piece was selected so that
the interonset interval (IOI, the time between onsets of two subsequent
notes) was 125 ms (8 tons/s) in all cases. Participants were instructed to re-
hearse and memorize them before the experimental session.
Procedure
Participants had to perform the memorized pieces on a MIDI digital piano

(Wersi Digital Piano CT2, Halsenbach, Germany) in a synchronization-con-
tinuation paradigm (see Herrojo Ruiz et al. 2009 for more details). The key-
board and the right hand of the participant were covered with a board to
prevent participants from visually tracking hand and finger movements.
Playing the correct notes and maintaining accurate timing were stressed.
The experimental design consisted of two conditionsaudiomotor (AM),
motor (M)comprising 60 randomized trials (around 12,000 notes) each.
Both in AM and M, participants played the musical stimuli 1-6 from memory.
The only difference between both conditions was that the volume of the MIDI
keyboard was set to zero in M, thus cancelling out the auditory feedback.
During the performance, continuous EEG signals were recorded from 35
electrodes placed on the subjects scalps according to the extended 10-20
system referenced to linked mastoids. Data were sampled at 500 Hz; the up-
per cutoff was 100 Hz (software by NeuroScan, Inc., Herndon, Virginia, USA).
Visual trigger stimuli, note onsets, and metronome beats were automati-
cally documented with markers in the continuous EEG file. Performance was
also recorded as MIDI files using a standard MIDI sequencer program.
Data analysis
An error-detection algorithm was developed in MatLab, which compared

each MIDI performance with the pitch contents of a template (the score).
Only isolated pitch errors and correct notes, which were preceded and fol-
lowed by three correct notes, were included in the analysis. Two additional
constraints were set to all pre-selected errors and correct notes in order to
assure their temporal precision and to avoid overlapping of brain responses:
first, the time interval between MIDI note on and off was not accepted above
150 ms. Second, the minimal and maximal IOI prior to and after the error
were set to 100 and 300 ms.
We performed the following types of analysis of the EEG signal: first, the
standard time averaging technique was executed to analyze the ERPs of the
brain responses triggered by actions leading to pitch errors (wrong note was
played) as compared with actions leading to correct pitches. ERPs were de-
rived by averaging the raw epochs for each subject and condition and the
result was baseline-corrected (from 300 to 150 ms prior to the keystroke).
Second, we analyzed the bivariate (pair-wise) phase synchronization between
EEG signals (Lachaux et al. 1999, Varela et al. 2001), as a measure of the
functional interaction and integration between the underlying neural popula-
tions. This latter investigation focused on the electrodes F3-FCz and F4-FCz,
as a measure of the synchronization between areas located over the lPFC and
mPFC. To rule out an alternative explanation of volume conduction effects,
the bivariate synchronization index was additionally calculated between C3-
FCz and C4-FCz: these pairs have a similar distance as pairs F3-FCz and F4-
FCz, but a more posterior location; however, there are no a priori hypotheses
that posit a role of these brain regions in cognitive control. Three frequency
ranges were analyzed: (1) the theta band (4-8 Hz), based upon its modulation
of the ERN (Luu et al. 2004, Cavanagh et al. 2009), (2) the alpha band (8-13
Hz), as an indicator of attention-deficits and precursor of forthcoming
mistakes in monotonous tasks (Mazaheri et al. 2009), and (3) the beta band
(13-30 Hz), due to its sensitivity to movement-related changes in humans
(Pfurtscheller et al. 1997).
All statistical tests were performed by means of nonparametric univariate
permutation tests and multivariate synchronized permutations (Good 2005).
RESULTS
Performance data
The selection of isolated erroneous notes yielded a value of 80 (SD 30) in AM

and of 80 (SD 20) in M. In AM, the mean IOI was 121 ms (8 ms), whereas in
M the mean IOI was 123 ms (8 ms). The difference in mean IOI or number of
errors was not significant (p>0.05). These results confirmed that pianists
successfully performed the sequences with timing very close to the right IOI.
Moreover, these data indicated that pianists played with a similar timing and
error rate with or without auditory feedback.
Interestingly, the loudness (assessed by MIDI velocity) of errors was sig-
nificantly reduced as compared with the loudness of the corresponding cor-
rect notes in the same position on the musical score, both in AM (mean=-7,
SD=4, p=0.01) and in M (mean=-5, SD=4, p=0.01). The reduction in loud-
ness was similar in both conditions (p>0.05). These results indicated that a
corrective response had already been initiated by the time the subjects
pressed the erroneous key. Finally, a pre- and post-error slowing was ob-
served both in AM and M [pre-error slowing of 190 ms (SD=60) in AM and
170 ms (SD=60) in M; post-error slowing of 240 ms (SD=60) in AM and 200
ms (SD=60) in M].
Error signal preceding errors
The comparison between the waveforms of errors minus correct notes, both
in AM and in M, revealed a negative event-related potential triggered 70 ms
before the onset of errors. This negative component resembled the ERN and
was termed pre-error negativity (pre-ERN, see Figure 1A and 1B). Also, we
observed a final positive deflection between 240 and 280 ms (AM) and be-
tween 180 and 220 ms (M), resembling the Pe. The Pe was larger in AM. The
maximum of the pre-ERN was localized across frontocentral positions (Fz,
FCz, Cz) of the scalp in both conditions. Likewise, the topographic maxima of
the Pe in AM and M were localized across frontocentral electrode positions.
By means of multivariate statistical analysis, we established that the ERP
waveforms of the mesial frontocentral electrodes between errors and correct
notes differed depending on the task condition from 220 to 260 ms (signifi-
cant interaction of the factors event type x condition, p=0.008). A post hoc
univariate permutation test across subjects in AM revealed a significant en-
hanced negativity, the pre-ERN, between -70 and -20 ms before errors as
compared with correct notes (p=0.007) and a significant Pe (p=0.008) be-
tween 240 and 280 ms. A similar result was obtained in M between -50 and 0
ms for the pre-ERN (p=0.008) and between 180 and 220 ms for the Pe. The
comparison between ERPs in AM and M revealed a significantly larger Pe in
AM between 250 and 280 ms (p=0.008). However, the pre-ERN did not dif-
fer between conditions (p=0.3).
The source of activity generating the pre-ERN was localized by sLORETA
in the Brodmann area (BA) 32 of the rostral ACC (Figure 1C). Similarly, the
neural generators of the of the brain activity associated with the Pe were lo-
calized by sLORETA in the BA 24 of the rostral ACC (Figure 1D).
Figure 1. Note-onset ERPs depicted at electrode location FCz for erroneous notes
(dashed line), for correct notes (solid line), and for the difference (errors minus correct
notes, bold line) in the presence (A) and absence of auditory feedback (B). Note the
increased negativity just before the commission of the errors (pre-ERN, arrow) and the
large positivity after errors (Pe, arrow) in both conditions. Note the different scaling in
y-axes (C) sLORETA localized the source of activity associated with the pre-ERN in AM
and M in the BA 32 of the rostral ACC. (D) The neural generator for the Pe in AM and M
was localized by sLORETA in the BA 24 of the rostral ACC. (See full color version at
In summary, the pre-ERN was independent of the auditory feedback.

However, the auditory information did modulate the processing of the errors
after their execution, as reflected in a larger Pe in AM.
Phase synchronization as feedforward control mechanism
The lower beta band (13-15 Hz) phase coupling between FCz and F4 increased
robustly from -100 to 0 ms before overt errors compared with overt correct
notes (p=0.001, permutation test, Figure 2). Similar statistical tests were run
on the pairs F3-FCz, C4-FCz and C3-FCz, but no significant effects were
found. These findings give evidence of an increased right-lateralized phase
interaction between FCz and F4 preceding errors, which could be related to
Figure 2. Difference between erroneous and correct trials in the grand-averaged phase
synchrony between channels F4 and FCz. An increase in bivariate phase synchrony can
be observed starting 100 ms prior to the note onset and due to more phase coupling for
error trials. Significant differences are marked by the black contour (p=0.001). The left
panel illustrates the results of the correlation analysis. The right panel depicts the elec-
trodes (dark pink) FCz and F4, in addition to electrodes F3, C3 and C4 (light pink). (See
the feedforward mechanisms of error detection and correction. To investigate

the latter, we assessed the modulations by the F4-FCz phase coupling of the
corrective mechanisms by a regression analysis between the beta phase cou-
pling and the decrease in loudness of errors. The phase synchrony index aver-
aged in the time-frequency window -100-0 ms x 13-15 Hz was highly signifi-
cantly correlated with the reduction in the loudness of errors (Spearmans
rho=0.62, p=0.001). Such positive correlation suggests that in subjects with a
higher F4-FCz phase coupling there was a better correction mechanism that
results in the reduction of the loudness of errors.
DISCUSSION
In the present study, the main finding was a negative event-related potential
triggered 70 ms before the onset of errors, the pre-ERN. The pre-ERN sig-
naled error detection in advance and was generated by the rostral ACC. Fur-
thermore, this component was independent of the auditory feedback and had
a correlate at the behavioral level: the loudness of pitch errors was reduced.
This behavioral finding suggests that a corrective response had already been
initiated by the instant of pressing the erroneous key.
Furthermore, the auditory information did modulate the processing of the

errors after their execution, as reflected in a larger Pe around 250-280 ms in
AM, generated by the rostral ACC. Based on the evidence that associates the
rostral ACC with emotional processing (Luu et al. 2003), the previous result
might indicate an enhanced emotional evaluation of the errors with auditory
feedback.
Finally, in searching for a feedforward mechanism that might detect the
upcoming wrong note in advance and try to cancel the associated sensory
consequences, we found that the beta band phase synchrony between F4 and
FCz increased 100 ms before errors. This measurement positively correlated
with the reduction in loudness of errors. Thus, the degree of increased pre-
error bivariate synchronization between this pair of electrodeswith F4 rep-
resenting the lPFC and FCz representing the ACCwas associated with more
efficient correction mechanisms.
The present results demonstrate that the pitch accuracy and temporal
precision required in the production of fast complex musical sequences is
possible in part by the perfect functioning of feedforward mechanisms in
highly skilled pianists. Internal forward models can predict the next state of a
system from its current state and motor command (Bernstein 1967, Wolpert
et al. 1995). Further, they compare the actual motor outflow (efference copy)
with the motor command. In case of a mismatch, an error signal is triggered
to cancel the undesired sensory effects of the movement (reafference) and a
corrective response is initiated. This model is supported by the present re-
sults. In our paradigm, (1) the reported pre-ERN may be the neural correlate
of this error signal, and (2) the auditory feedback plays a fundamental role in
the emotional evaluation of errors. In addition, (3) the feedforward control
signal, which indicates the need of a corrective response and/or posterior
behavioral adjustments, is indexed by the pre-error increase in beta phase
synchrony between FCz and F4. Finally, (4) the corrective response triggered
to cancel the undesired sensory effects of the wrong movement leads to the
observed decrease in the loudness of errors and to the pre-error slowing
(~190 ms).
The evidence for our hypothesis that the auditory feedback does not me-
diate the detection of the pitch errors prior to the execution in piano perform-
ance is in agreement with previous studies. Lashley (1951) postulated that
auditory feedback could not control for the fast motor sequences of piano
performance at a high tempo. This statement was further strengthened by the
study of Finney and Palmer (2003), which demonstrated that the presence or
absence of auditory feedback in the retrieval of memorized music sequences
did not affect the error rate. Performance of rapid movements must thus be
prepared in advance (Schmidt 1975, Pfordresher and Palmer 2006). We be-

lieve that this outcome is characteristic not only of skilled piano performance
but also of other skilled time-based sequential sensory-motor behaviors, such
as those used in sports, dance, or speech. These behaviors share as common
aspect the preparation in advance of the events planned for production, pos-
sible through extensive training. Consequently, these behaviors rely rather on
forward predictive mechanisms than on the slow sensory and proprioceptive
loops (Wolpert et al. 1995).
Optimization in performance might be achieved by the interaction be-
tween the action-monitoring and the cognitive control systems (Botvinick et
al. 2001). According to this view, the action-monitoring system supervises
ongoing performance and signals the need for adjustments, which are in turn
implemented by the cognitive control system. Most of the previous investiga-
tions have located the neural activity associated with action-monitoring proc-
esses in the ACC, whereas the control system is ascribed to the lateral PFC
(Botvinick et al. 2001). However, a different theory posits that neural activity
in the ACC might mediate both monitoring and control mechanisms (Rush-
worth et al. 2004). Our results are in line with the former model, in which the
ACC monitors ongoing performance to detect unfavorable upcoming actions
and interacts with the lPFC, so that this brain region can implement the be-
havioral adjustments. To the best of our knowledge, out results are the first to
demonstrate that around 100 ms before errors are committed, increased beta
phase synchronization between electrode regions located over the ACC and
lPFC is associated with better corrective mechanisms. As a consequence, it
seems plausible that feedforward control mechanisms are triggered already
100 ms before errors are committed.
In summary, our findings provide compelling evidence for the hypothesis
that fast predictive mechanisms are required for the optimal execution of pre-
programmed temporal and spatial movement patterns that characterize piano
performance (Catalan et al. 1998). The present results can be generalized to
expert performance in sports, music, and dance.
Acknowledgments
This work was funded by the Center of Systems Neuroscience, Hanover, and the EU
through the Marie Curie Early Stage Training Contract (MEST-CT-2005-021014) to
Mara Herrojo Ruiz.
Mara Herrojo Ruiz, Institute of Music Physiology and Musicians Medicine, Hanover
Email: Maria.Herrojo-Ruiz@charite.de
References
Bernstein N. A. (1967). The Co-ordination and Regulation of Movements. Oxford: Per-

gamon Press.
Botvinick M. M., Braver T. S., Barch D. M. et al. (2001). Conflict monitoring and cogni-
tive control. Psychological Review, 108, pp. 624-652.
Cavanagh J. F., Cohen M. X., and Allen J. J. (2009). Prelude to and resolution of an
error: EEG phase synchrony reveals cognitive control dynamics during action
monitoring. Journal of Neuroscience, 29, pp. 98-105.
Dehaene S., Posner M. I., and Tucker D. M. (1994). Localization of a neural system for
error detection and compensation. Psychological Science, 5, pp. 303-305.
Falkenstein M., Hohnsbein J., Hoormann J., and Blanke L. (1990). Effects of errors in
choice reaction tasks on the ERP under focused and divided attention. In C. H. M
Brunia, A. W. K. Gaillard, and A. Kok (eds), Psychophysiological Brain Research
(pp. 192-195). Tilburg, The Netherlands: Tilburg University Press.
Finney S. A. and Palmer C. (2003). Auditory feedback and memory for music perform-
ance: Sound evidence for an encoding effect. Memory and Cognition, 31, pp. 51-64.
Holroyd C. B. and Coles M. G. (2002). The neural basis of human error processing:
Reinforcement learning, dopamine, and the error-related negativity. Psychological
Review, 109, pp. 679-709.
Lashley K. (1951). The problem of serial order in behavior. In L. Jeffress (ed.), Cerebral
Mechanisms of Behavior (pp. 112-136). New York: Wiley.
Luu P., Tucker D. M., Derryberry D. et al. (2003). Activity in human medial frontal
cortex in emotional evaluation and error monitoring. Psychological Science, 14, pp.
47-53.
Luu P. and Tucker D. M. (2004). Self-regulation by the medial frontal cortex: Limbic
representation of motive set-points. In M. Beauregard (ed), Consciousness, Emo-
tional Self-regulation and the Brain (pp. 123-161). Amsterdam: John Benjamin.
Mazaheri A., Nieuwenhuis I. L., van Dijk H., andJensen O. (2009). Prestimulus alpha
and mu activity predicts failure to inhibit motor responses. Human Brain Mapping,
30, pp. 1791-1800.
Miller E. K. (2000). The prefrontal cortex and cognitive control. Nature Reviews Neu-
roscience, 1, pp. 59-65.
Pfordresher P. Q. and Palmer C. (2006). Effects of hearing the past, present, or future
during music performance. Perception and Psychophysics, 68, pp. 362-376.
Pfurtscheller G., Stancak A., and Edlinger E. G. (1997). On the existence of different
types of central beta rhythms below 30 Hz. Electroencephalography and Clinical
Neurophysiology, 102, pp. 316-325.
Rushworth M. F., Walton M. E., Kennerley S. W., and Bannerman D. M. (2004). Action
sets and decisions in the medial frontal cortex. Trends in Cognitive Sciences, 8, pp.
410-417.
Schmidt R. A. (1975). A schema theory of motor skill learning. Psychological Review,
82, pp.225-260.
Wolpert D. M., Ghahramani Z., and Jordan M. I. (1995). An internal model for sensori-
motor integration. Science, 269, pp. 1880-1882.
Thursday
17 December 2009
Keynote paper
A balanced approach to excellence: Life-skill

intervention and elite performance
Deidre Anderson
Macquarie University, Australia
To become an elite performer in the modern world, individuals must dis-

cipline themselves to train and practice for many years. Usually they
need to dedicate most of every day in pursuit of their dreams. More and
more countries have been developing systems to identify this talent very
early in life and are finding new ways to nurture it. The challenge faced
by most elite performers is how to manage this intense focus and still
build a resilience and capacity to meet the many transitions and demands
required in both elite performance and life. These skills are more often
than not under developed as a result of the myopic environment typically
created for the elite performer. One of the most challenging transitions
often faced is retirement. The basis of much research in this area has
been undertaken in sport (Alison and Meyer 1988, Baillie 1993, Blann
and Zaichkowsky 1989, Blinde and Greendorfer 1985, Lerch 1984,
Lavallee et al. 1997). Other performance environments such as dance and
music have also captured the interest of researchers (Davidson and Bur-
land 2006, Burland and Davidson 2004, Patton and Ryan 2000, Sapos-
nek 1995, Patton and McMahon 1999, Wallach 1988). This growing body
of literature has led to some performance environments introducing early
intervention programs to broaden the life-skills of the performer. The
belief is that this will protect them from the uncertainty and anxiety
about their futures and will result in fewer traumas when they have to
deal with the transition from elite performance.
Keywords: transitions; life-skills; athlete identity; athlete foreclosure;

self identity
RETIREMENT AS A TRANSITION
Social gerontological models of aging and thanatological models of death and

dying have been used by researchers to better understand career termination
from elite sport (Atchley 1991, Lerch 1984, McPherson 1980, Rosenberg 1981,
Freidman and Havighurst 1947, Rose 1962, Kuypers and Bergston 1973).
These approaches can also be applied to a range of other elite performance
environments, including the arts, dance, music, and acting industries.
The social gerontological perspective takes into account the following
theories: Activity Theory, Disengagement Theory, Subculture Theory, Conti-
nuity Theory, Social Breakdown Theory, and Social Exchange Theory. Some
of these will be discussed below.
Activity Theory (Friedman and Havighurst 1954; Havighurst and Albrecht
1953) maintains that as various activities are lost, the individual establishes
new roles, and the maintenance of these contributes positively to self concept
and life satisfaction. An important difficulty with this theory when relating to
elite performance, according to Baillie and Danish (1992), is that it may not
always be possible to duplicate the type of environment created within the
elite performance context.
Disengagement Theory (Cumming and Henry 1961) was developed as a
response to Activity Theory and suggests that the ageing process sees indi-
viduals mutually agreeing to withdraw from society and thereby providing
them with the opportunity to enjoy life in the period after paid employment.
This theory proposes that it is a function of both the individual and society
that naturally occurs. When attempting to draw parallels with the elite per-
formance environment, this concept is a difficult one as it assumes that the
withdrawal is mutually agreeable between both parties, and this may not al-
ways be the case.
Continuity Theory represents a refinement of Activity Theory (Atchley
1977) and posits that, although activities and habits might alter or be replaced
by others throughout our lives, these may not always provide the same
meaningful experience. If individuals are capable of retaining the continuity
of meaning in their activities, retirement is said to be less traumatic (Ander-
son 1999). In relating this theory to elite performance the challenge will be to
find meaning in other activities when for so long many elite performers have
characterized their lives around one specialized activity.
Additional theories worthy of consideration in the present context are So-
cial Breakdown Theory (Kuypers and Bergston 1973) and Exchange Theory
(Dowd 1975). Kuypers and Bergston proposed that with any role loss, such as
retirement, the individual might become the recipient of negative external
labels, leading to a breakdown in their social framework. Therefore, any ma-
jor role loss is seen to have significant potential to impact adversely on self
image. In order to limit the downward spiral that often results, they argued
that a social reconstruction is typically needed.
Exchange Theory (Homans 1961) illustrated that retirement can be a

positive experience if social activities and networks are considered and possi-
bly redefined. He believed that if an individual can continue to maintain val-
ued social networks then retirement should be an easier process. Rosenberg
(1981) has suggested that this theory is the most applicable to elite sport as it
is often the inability of an athlete to create these new networks that contrib-
utes to a difficult adjustment.
Social Breakdown and Exchange Theories appear to be most useful in the
elite performance context as individuals who retire from these environments
often require a social reconstruction of themselves. Application of these theo-
ries to date has demonstrated useful parallels (Rosenberg 1981), but further
work is required. Continuity, Activity and Disengagement Theories may also
have some relevance, but again, more applied research in the elite perform-
ance setting is needed to understand their true value.
The Thanatological Models studied the process of death and dying. Intro-
duced by Park (1912) these theories may also have implications for retirement
from elite performance and incorporate Social Death (Kalish 1966), Social
Awareness (Lerch 1984), and the stages of death (Kubler-Ross 1969, 2005).
Social Death assumes that individuals experience a loss of social func-
tioning which results in those around them retreating from the relationship
as though they have already died (Lerch 1984). This is sometimes observed in
the elite performance environment when individuals feel a sense of isolation
and ostracism prior to their retirement or when they sustain a career ending
injury.
Why this may occur can be explained through Social Awareness Theory,
which proposes that individuals have different levels of awareness: closed,
suspected, mutual, and open. In closed awareness, terminally ill patients may
not be aware of their impending death whilst those around may know. The
reasons for this could include the doctors not informing the person and the
persons non-acceptance of the inevitable outcome. When applying this to
elite performance and in particular sport, it can be seen when an athlete does
not know that retirement is sometimes imminent because the coach, man-
agement, and even team mates have not discussed it with them.
The concept of suspected awareness exists when a dying patient believes
or suspects that death is a matter of time, and they engage others to either
confirm or deny their suspicions. In the elite performance environment,
many performers will pick up on changes in the way they are being treated
and seek to determine whether they are about to be retired.
Mutual awareness is when all parties are aware of the impending death
and everyone involved behaves as if nothing is going to happen. In elite per-
formance, the individuals career transition would not be discussed, but it is

possible that they may sense a degree of isolation but cannot explain why.
Open awareness exists when all parties acknowledge that the patient is
dying, and this provides everyone with a chance to discuss their feelings and
thus gives the patient a greater sense of control. In terms of elite perform-
ance, individuals can begin to plan their post performance career. According
to Rosenberg (1984), the most common awareness models seen in the sport-
ing environments are closed and suspected.
Kubler-Ross (1969) created a psychological reactions model in an attempt
to understand terminally ill patients. The stages of dying, as described by
Kubler-Ross, include denial, isolation, anger, bargaining, depression, and
acceptance. These psychological reactions have been used widely to describe
retirement from elite sport (see Figure1).
The initial stages of denial and isolation see the elite performer refusing
to acknowledge the decision that retirement is approaching. The anger stage
may see the performer becoming quite disturbed about the situation. Bar-
gaining will result in creating personal debates and negotiation in an attempt
to stay connected to the impending loss. Depression will see the performer
distressed and over anxious about the situation. Finally, the performer will
acknowledge and accept the situation and be able to transition.
Coakley (1983) challenged the use of gerontological concepts and as-
sumptions that retirement from sport is an inevitable source of stress. He
indicated that the problem is not retirement per se, but the personal and so-
cial characteristics of the individual that affect the ability to adjust to changed
circumstances.
The personal attributes of the elite performer and how much their self
identity is connected to the performance environment will have an impact on
the quality of the transition.
The time taken and the ability to transition through these stages, known
as the grieving process, will depend upon whether the decision was a volun-
tary (own choice) or an involuntary one (e.g. injury or deselection). The ath-
lete retiring by choice may go through the grieving process more quickly and
with greater ease than athletes who have retirement forced upon them (For-
tunato and Morris 1995).
These sentiments are also shared by Oglivie and Taylor, (1993) who be-
lieved that an elite athletes adjustment to retirement will be smoother if the
retirement is voluntary. In other words, the athlete makes the choice, is pre-
pared for life after sport, measures identity on more than just sport, and has
good relationships and a broad social support network.
Retirement from Elite Performance Self Disco very
Q ui ck F ix Wa y SEL F IDENT IT Y
DET OUR
Sel f man ag em e n t
Wh o is o n m y tr ai n?
A tt itu d es
Self Kn o wle dg e
In ter p er s on al s ki lls
C OPING G ROVE De cis io n m ak in g s kil ls
Mo n ey Ch as e
T h e Pil lar s VAL UES AN D

DEPRESSI ON PREF ERENC ES BL D
A CCEPT ANCE
BAR GAIN ING
GR EIF ING PROC ESS Self Discovery

ANG ER
INJ URY
PERF RO M ANCE
DESEL ECTION
DENIA L- I SOL AT ION

VAL UES AND Self ma n ag eme n t
PREF ERENCES BL D At ti tu d es
Se lf Kn o wle dge
I nt er p er so n al s kil ls
Dec isi on m ak in g sk ill s
Involuntary W ho i s o n m y t ra in ? SPECI ALI SED SUPPORT
Highway
G REIF ING PRO CESS

Voluntary
Hig hway
Figure 1. The process of dealing with both voluntary and involuntary retirement from
elite performance. (See full color version at www.performancescience.org.)
Kleber and Brock (1992) found that adjustment difficulties might well be
explained by how the athlete perceives the level to which sport has con-
stricted their development. What the athlete believes could be reflected in
several ways. They might see that engagement in sport prevented serious
orientation toward a career and life beyond. It is also possible that sport may
have promoted degrees of self-detachment and desensitization to the wider
world, thereby leaving the athlete with heightened feelings of vulnerability as
retirement approaches (Anderson 1999).
Whether or not the decision to retire from elite performance is voluntary
or involuntary, qualitative analysis reveals that performers are frequently
disadvantaged by a failure to plan for life outside the performance environ-
ment (McPherson 1980). If young performers can be socialized to perceive
the value of planning for retirement during their elite performance careers,
they might minimize or even avoid the trauma that is currently being experi-
enced by many individuals as cited in research and anecdotal reports (Danish

et al. 1991).
Retirement from elite performance is a process rather than a single step.
It often requires a change in how the performer views themselves and those
around them. Clearly each performer would deal with retirement in a very
individual manner.
Most elite performers do not prepare for retirement and lack the re-
sources to deal with it, especially when their performance aspirations are not
met, and generally experience some form of trauma as a result. Some re-
searchers (Baillie and Danish 1992, Fortunato 1996) believed that these diffi-
culties can be explained through understanding the concept of athlete
identity.
The elite performer is often blind to the fact that their career can be brief,
curtailed, or even unexpectedly terminated due to a wide variety of circum-
stances. It is therefore important that an early intervention plan is estab-
lished. The implementation of such a plan may result in a less stressful time
both emotionally, psychologically, and financially, both during and after their
involvement in elite performance.
Brewer et al. (1993) proposed that if an athlete measures their identity on
their sporting performance alone they will have difficulties in dealing with the
many transitions associated with both sport and life. Saposnek (1995) be-
lieved that dancers experience similar challenges as a result of their early and
enduring identification with their field.
One potential outcome of focusing on a particular element of life is that
the performer can be distracted from giving due attention to matters periph-
eral and external to the elite environment. Importantly, this may lead them to
foreclose on other aspects of their personal development.
Organizations responsible for elite performers should be encouraging
them to develop skills outside of their performance arenas. This will help
them minimize the risk of being left with feelings of insecurity and uncer-
tainty in their current and future lives. A course of action that equips them to
deal with a range of transitions they will face is a crucial element of both pro-
fessional and personal development.
THE NEED FOR LIFE-SKILL PROGRAMS
One way of supporting the performer is through the introduction of life-skill

programs which aim to up-skill the performer to deal with the many transi-
tions that they are expected to face in both life and elite performance.
Navigating the Transitions to Performance Excellence
PERSONAL
BEST
T ERM INAL AVE
RESPONSI BIL IT Y RD
Wh o i s o n m y tr a in
Per so n al
r es po n sib il ity fo r
t he jo u r ne y
F RUST RAT IO N R D
INFOR MA TI ON OVERL OA D
IN TEG RAT ED GO ALS
Wa nt s su cc es s to o
kn o win g wh at b a dly , to o m uc h
in f or m at io n is v ali d p r ess u re o n s elf
PRO FESSI ONA L APPRO ACH
GR EIF WAY
I NJUR Y M ANAG M ENT M ed ia ro le m o d el, b ra n d

M Y WAY Re tir em e n t , r eg r et s,
ne w be gi nn in g s
T ake s a lo ng t er m p e rs pe ct iv e Q ue st io ni ng Dr
PERF ORM ANC E AVE O VERT AINING T ERRACE

N OT WHER E T HE A THL ET E
W ANT S I T TO BE
Transi ti on RECO VERY W AY
Highway
PERSPECT IVE WAY DET OUR SELF IDENT IT Y
PR OGR ESSION C LO SE
DO MEST I C T O
NT REN ATI ONAL CO MPET I TI ON PRESSUR E PARA DE
Fr o m : fa ns , m e di a,
fa m ily , f r ien d s, wo r k, c lu b
REL OC AT ION CLO SE
F O CUS C LO SE
PERSIST ENC E L ANE Av oi d qu ic k- fir e

m ag ic al so lu ti on s PRO BLEM COU RT
Un re so lv ed
Per so n al is su es
Transition F EAR A VE
Highway NO T HRU RO AD
Fe ar o f su cc es s & fa ilu re
Figure 2. Transitions in elite performance. (See full color version at www.performance

science.org.)
More specifically, these transitions are defined as events resulting in

changes in assumptions about oneself and the world and which require a cor-
responding change in behavior and relationships (Schlossberg 1981). Figure 2
outlines most of the common transitions that elite performers will face at
some stage. Each transition requires a range of skills that must be learned if
the individual is going to sustain and meet their performance and personal
objectives.
Transitions are either voluntary or involuntary in nature and performers
need to develop a variety of skills so that they can successfully move through
the various transitions and issues they confront.
The ability to move through and cope with each of the above transitions,
which at times can be multi-phased, will depend on how well the situation
can be normalized. To be able to do this requires an awareness of both self
and the performance environment. The performer should therefore be en-
couraged to learn early in their careers about the real performance environ-
ment within which they will be expected to function. Such knowledge and
take up of early intervention programs that enable them to be better prepared

are highly recommended.
The quality of the adjustment to the many transitions will often be de-
pendent upon the following: individual characteristics, socio-economic envi-
ronment, type and nature of the transition, developmental issues, athlete
identity, and identity foreclosure.
The pressure placed on the performer to focus solely on their perform-
ance often means that when they need to deal with a transition they only have
the capacity to draw on a limited set of experiences. As many performers be-
gin their journeys at a young age, unless a broader perspective on life is de-
veloped, their true potential as an elite performer may never be realized.
Developing life-skills during the performance journey enables the athlete
to take control of the important aspects of life and provides an even more
intense ability to focus on the performance dreams. As well the transition into
retirement from elite performance is often smoother. This perspective is still
not fully understood by many coaches, teachers, and administrators, despite
the ongoing resources spent on helping performers who are unable to pro-
gress through the transition process.
Developing the self in a holistic manner creates a stronger base from
which to operate and enables the performer to draw on a broader range of
skills to cope with the transitions. This integrated approach is essential in
order to provide the right attitude, knowledge, and skills. The pillars that
make up the self may include performance goals, personal development, fam-
ily, friends, professional development, and social connections.
The performer should be encouraged to set goals in all areas of life and
know what their personal strengths and weaknesses are both within and out-
side their performance environment. It is also important that they seek, and
can interpret feedback from a range of sources, including coaches, teachers,
umpires, media, spectators, team members, opponents, officials, family,
friends, peers, and supporters.
The performer needs to understand that being assertive and developing
effective interpersonal communication skills in both their performance arena
and life will give them a greater capacity to influence their environment.
Effective time management is a skill that is often not picked up by per-
formers early enough in their careers. This can be quite common as most of
their time is structured and managed for them. In the modern environment,
performers need to embrace technology to better manage the often competing
demands.
Creating confidence in personal presentation and public interaction can

also assist the performer to deal with the public and to take full advantage of
what this can offer.
Managing personal income, financial planning, and budgeting can also
help the performer take full advantage of their financial situations and ensure
that their future has been planned for adequately.
Many performers may be fit ,but few can claim to also be healthy. Per-
sonal health and fitness knowledge is important so that they can sustain their
commitment to performance for as long as possible. Nutrition, recovery,
performance psychology, injury prevention, drugs, and an understanding of
the human body will give important knowledge that can sometimes be the
difference between success and failure.
THE ELITE PERFORMANCE ENVIRONMENT
Organizations responsible for elite performance should create an environ-

ment early that educates elite performers, and those aspiring to be at the elite
level, about what to expect. They should also drive a culture throughout the
environment that promotes a balanced approach to excellence. This is a criti-
cal factor in assisting the performer to process and adjust to transitions such
as retirement, injury, and poor performance and many others mentioned
above.
These organizations should foster an open door policy enabling informal
contact to be maintained after the performer retires and, where possible,
utilize their skills to assist others.
Organizational policy, practice, and culture must support the introduction
and implementation of an elite performers life-skill program. The program
should be created on the basis of identified need, be flexible enough to ensure
that it is individualized, incorporated into the overall assessment of perform-
ance, and is thoroughly consistent with the balanced approach concept.
Deidre Anderson, Macquarie University, North Ryde, New South Wales 2109, Australia;
Email: deidre.anderson@dvcc.mq.edu.au
References
Alison M. T. and Meyer C. (1988). Career problems and retirements among elite ath-
letes: The female tennis professional. Sociology of Sport Journal, 5, pp. 212-222.
Anderson D. K. (1998). Life Skill Intervention and Elite Performances. Unpublished

masters thesis, Victoria University of Technology.
Anderson D. K. (1996). Life skill intervention and elite performance. Paper presented at
the Annual Conference on Counseling Athletes, National Athlete Career and Edu-
cation Program, Springfield, Massachusetts, USA.
Atchley R. (1977). The Social Forces in Later Life. Belmont, California, USA:
Wadsworth.
Baillie P. H. F. (1993). Understanding retirement from sports: Therapeutic ideas for
helping athletes in transition. The Counseling Psychologist, 21, pp. 399-410.
Baillie P. H. F. and Danish S. J. (1992). Understanding the career transition of athletes.
The Sport Psychologist, 6, pp. 77-98.
Blann F. W. and Zaichkowsky L (1989). National Hockey League and Major League
Baseball Players Post-sport Career Transition Surveys. Toronto: National Hockey
League Players Association.
Blinde E. M. and Greendorfer S. L. (1985). A reconceptualization of the process of leav-
ing the role of competitive athlete. International Review for the Sociology of Sport,
20, pp. 87-93.
Burland K. and Davidson J. W. (2004). Tracing a musical life Transition. In J. W.
Davidson (ed.), The Music Practitioner (pp. 224-250). Aldershot, UK: Ashgate.
Cumming E. and Henry W. (1961). Growing Old. New York: Basic Books.
Davidson J. W. and Burland K. (2006). Musician identity formation. In G. E. McPher-
son (ed.), The Child as Musician (pp. 475-490). Oxford: Oxford University Press.
Dowd J. J. (1975). Ageing as exchange: A preface of theory. Journal of Gerontology, 30,
pp. 584-594.
Fortunato V., Anderson D., Morris T., and Seedsman T. (1995). Career transition re-
search at Victoria University. In R. Vanfraechem-Raway and Y. Vanden Auweel
(eds.), Proceedings of the Ninth European Congress of Sports Psychology (pp. 533-
543). Brussels: European Federation of Sports Psychology.
Homans G. (1961) Social Behavior. New York: Harcourt Brace.
Kalish R. (1966). A continuity of subjectivity perceived death. The Gerontolgist, 6, pp.
73-76.
Kubler-Ross E. (1969). On Death and Dying. New York: Macmillan.
Kubler-Ross E. (2005). On Grief and Grieving. New York: Simon and Schuster.
Kuypers J. A. and Bengston V. L. (1973). Social breakdown and competence: A model of
normal aging. Human Development, 16, pp. 181-220.
Lavallee D., Gordon S., and Grove J. R. (1997). Retirement from sport and the loss of
athletic identity. Journal of Personal and Interpersonal Loss, 2, pp. 129-147.
Lavallee D. and Wylleman P. (2000). Career Transitions in Sport. Morgantown, West
Virginia, USA: Fitness Information Technology.
Lerch S. (1984). Athlete retirement as social death: An overview. In N. Theberge and P.

Donnelly (eds.), Sport and the Sociological Imagination (pp. 259-272). Fort Worth,
Texas, USA: Texas Christian University Press.
Park R. (1912). Thanatology. Journal of the American Medical Association, 58, pp.
1243-1246.
Patton W. and McMahon M. (1999). Career Development and Systems Theory. Pacific
Grove, California, USA: Brooks/Cole.
Rosenberg E. (1981). Professional athletic retirement: Bringing theory and research
together. Paper presented at the Regional Symposium of the International Com-
mittee for the Sociology of Sport, Vancouver, Canada.
Rosenberg E. (1984). Athletic retirement as social death: Concepts and perspectives. In
N. Theberge and P. Donnelly (eds.), Sport and the Sociological Imagination (pp
245-258). Fort Worth, Texas, USA: Texas Christian University Press.
Saposnek S. A. (1995). After the Ballet: The Effects of Career Transition on Sense of
Identify, Self-concept, and Body Image. Unpublished manuscript, Reed College.
Wallach E. (1988). Life after performing: Career transition for dancers. Dance/USA,
February, pp. 5-11.
Poster session
Memory consolidation in musicians: The

effects of sleep, interference, and recall on
musicians performance of a keyboard melody
Sarah E. Allen
Meadows School of the Arts, Southern Methodist University, USA
Procedural memory consolidation has been shown to enhance a variety

of perceptual and motor skills during sleep, but only recently has this ef-
fect been investigated in trained musicians performing music. I tested
the extent to which a music performance skill improves over a night of
sleep and whether the consolidation process is inhibited when musicians
learn two melodies in juxtaposition during a single training session.
Fifty-five participants learned to perform either one or two 13-note piano
melodies during an evening training session and were retested on the
target melody the following morning. Participants showed evidence of
overnight performance gains in performance speed and accuracy; how-
ever, learning a second, similar melody seemed to block these overnight
gains in the melody learned first. These results indicate that experienced
learners performing a familiar type of task, and one that includes audi-
tory processing demands, benefit from overnight consolidation of proce-
dural memories but that these benefits may be inhibited when musicians
learn similar, competing tasks in juxtaposition.
Keywords: memory; music; consolidation; motor skills; sleep
Sarah E. Allen, Meadows School of the Arts, Southern Methodist University, P.O. Box
750356, Dallas, Texas 75214, USA; Email: sarahallen@smu.edu
Health and wellbeing education in British

conservatoires
Louise Atkins
Royal Welsh College of Music and Drama, UK
Musicians health and wellbeing is rapidly becoming an essential compo-

nent within the training of aspiring musicians all over the world. At the
forefront of tackling these issues are educational institutions, with con-
servatoires in particular leading the way in creating educational pro-
grams that promote the ideal of a sustainable technique and lifestyle for
the professional musicians of the future. Strategic frameworks for health
and wellbeing education are beginning to develop in other countries as
national strategies. However, development in British music schools is not
so cohesive, and each institution takes a different approach to the deliv-
ery of their health and wellbeing provision. This project investigates cur-
rent health and wellbeing education programs on offer in British
conservatoires and presents the students view of this area of their edu-
cation. In this investigation, a wide range of health and wellbeing activi-
ties were revealed, with conservatoires showing support to students
through a variety of activities, from practical classes to research opportu-
nities. However, despite this extensive range of services, a high propor-
tion of students revealed that they felt their institution was not providing
adequate support for their health and wellbeing, stating that they would
like a more proactive approach.
Keywords: health and wellbeing; education; music students; injury pre-

vention; conservatoires
Health and wellbeing is rapidly becoming a major consideration for profes-

sional musicians who are looking toward avoiding injury, prolonging their
careers, and performing to the best of their abilities. Studies with particular
instrumental groups of college level students have shown high incidences of
injury, with between 30% and 50% of students being affected at some point in
their education (Guptill et al. 2000). Musicians health and wellbeing has
therefore become a topic for growth and development within educational

institutions and is rapidly becoming an essential component within the
training of aspiring practical musicians all over the world. Educational pro-
grammes that promote the ideal of a sustainable technique and healthy life-
style for the professional musician of the future are beginning to become
integrated into education programs, aiming to raise awareness among stu-
dents and staff alike. This project examines health and wellbeing provisions
and their associated resources currently provided to students in British con-
servatoires, including how, where, and with what emphasis education in such
matters lies. Issues relating to student injury, including treatment pathways
and recovery support, are also explored. This is completed through a series of
case studies with individual conservatoires and a questionnaire study with
conservatoire students.
METHOD
Participants
The study was carried out across seven UK conservatoires, with members of
staff from each institution aiding with data collection. The study also ques-
tioned a total of 46 student participants from UK conservatoires: 26 male and
20 female participants, with a mean age of 22 years. The students were from
both undergraduate and postgraduate courses and played cello, clarinet, flute,
horn, oboe, percussion, piano (including two accompanists), saxophone (clas-
sical and jazz), trumpet, trombone, tuba, violin, and voice. Twenty-seven
participants had experienced an injury, 19 had not, and 15 were currently
dealing with an injury.
Materials
Case-studies were compiled on seven British conservatoires, giving a detailed

picture of health and wellbeing provisions, each being approached on an indi-
vidual basis and various departments being asked to describe the health and
wellbeing activities taking place.
A self-reporting questionnaire was used to gain an insight into the
thoughts and feelings of conservatoire students toward their experiences of
health and wellbeing provisions. The questionnaire explored topics such as
individual experiences of injury, experiences of health and wellbeing educa-
tion within their own institution, and the importance of health and wellbeing
as a topic.
Procedure
Each institution was approached on an individual basis and case studies were
completed with the cooperation of staff members. Since the personal data of
students was handled in a variety of different ways, procedure for finding
participants and completing the questionnaire was carried out through a vari-
ety of methods. Some questionnaires were filled out over the telephone with
the researcher taking notes, while others were completed by hand or via
email.
RESULTS
This investigation revealed a wide variety of activities, some of which are de-
scribed below, with conservatoires providing support to students on many
levels. One-off workshops were found to be the most common form of health
and wellbeing activity, giving students insight into issues ranging from injury
prevention to performance anxiety and from hearing health awareness to how
to create a healthier lifestyle. Specialist departmental workshops were noted
by 45% of students as being particularly successful, and one student com-
mented that there are fortnightly classes for accompanists where our specific
issues can be addressed; I find this really useful. In three conservatoires,
workshops have developed into a regular series of classes.
Three institutions present health and wellbeing issues as a first-year pro-
ject. As well as addressing practical skills, these classes also focus on aca-
demic knowledge, developing professional skills, and addressing
performance-practice issues. One participant commented: I had health and
wellbeing education when I was in the first year, but Ive forgotten everything
theyve said and I know others have too. Only 43% of participants recognized
that health and wellbeing was part of their curriculum.
In four conservatoires, health and wellbeing issues were also found to be
raised as a part of larger projects, such as outreach projects and teacher
training programs. Conservatoire students are increasingly given the oppor-
tunity to pursue academic study of the subject further, with five conserva-
toires noting an increase in the amount of students choosing health and
wellbeing as a topic for dissertation projects. Masters modules and doctoral
study in the area is also developing within conservatoires.
Conservatoires also showed diverse support for those who become in-
jured, with a wide range of services for psychological and physical wellbeing,
including counseling and physiotherapy and access to information, support,
funding, and treatment. However, only 35% of students suggested that they
had taken advantage of this. All seven conservatoires provide resources to
support health and wellbeing education, which includes literature, journals,

and web-based support. In three conservatoires, internet homepages provide
comprehensive information on health and wellbeing activities, while in four
conservatoires intranet sites contain notes, recordings, or videos from lec-
tures and workshops. One student commented: I think its good that there is
so much information on the intranet, and it would be good if all the stuff we
learned about was up there so you could look at it again. However, when
students were asked if they were aware of who was responsible for health
issues within their institution, 43% could name a point of contact.
When asked if they felt that their institutions were doing enough for their
health and wellbeing, only 22% responded yes. However, 93% of students
questioned stated that health and wellbeing was important to them, com-
menting: if you havent got your health, you havent got your career, when
you are healthy you find performing easier and can play to a higher stan-
dard, and awareness of health and wellbeing is essential in order to be able
to perform to your best abilities.
DISCUSSION
This investigation has shown that conservatoires provide a wide range of

health and wellbeing information to their students through both timetabled
provision but also through less formal individual education. Although a wide
variety of activities and support for health and wellbeing issues were found to
be offered to students, the level, content, and delivery of this provision varied
greatly between institutions. Each conservatoire uses the resources and fa-
cilities readily available to them and maximizes the use of local specialists and
passionate members of staff as driving forces to their work. The services
available are therefore reliant on these resources and are influenced by other
issues such as timetabling, funding, student support, and logistics. This re-
veals that, as yet, there is no united front within British conservatoires in
disseminating information on health and wellbeing issues to students. This
investigation has revealed that students do not feel that they are getting the
full benefit of health and wellbeing provisions within their institutions, pro-
voking the questions of whether or not the topic should become an integral
part of a curriculum, the extent to which conservatoires should take a proac-
tive role in preventing injury, and the importance of science in a musical set-
ting.
Creating a set of coherent and organized goals for health and wellbeing
education would offer both students and staff a more structured approach to
an essential topic. Ensuring that students have adequate training in how inju-
ries may be prevented, how they might detect the onset of injury, where they
go for treatment of an injury, and how they are supported back to recovery is
essential. Having consistency in comprehensive education and injury support
would allow students to study in an environment comparatively free of con-
cern over injury and leave their institutions with a set of skills central in a
professional career.
Students have expressed within this study their belief in the importance
of education in health and wellbeing matters. Students have also stressed the
importance of raising awareness of topics relating to injury, including both
the prevention and treatment of playing-related disorders. For this kind of
information to be successfully interpreted by music students, the role of the
one-to-one teacher should be carefully evaluated. Although creating effective
learning plans within the curriculum may be very beneficial, it is one-to-one
teachers who have the most influence on the practical usage of techniques.
Without the involvement of these teachers in health and wellbeing education,
the topic may struggle to be translated into useable, practical techniques.
Acknowledgments
This research project has been carried out with the kind support of the Musicians Be-
nevolent Fund and the British Association of Performing Arts Medicine. Both organiza-
tions have provided access to key contacts within the field, along with a range of other
resources and continuing help and support throughout this study.
Louise Atkins, Royal Welsh College of Music and Drama, Castle Grounds, Cathays Park,
Cardiff CF10 3ER, UK; Email: louise.atkins@rwcmd.ac.uk
References
Guptill C., Zaza C., & Paul S. (2000). An occupational study of physical playing-related
injuries in college students. Medical Problems of Performing Artists, 15, pp. 86-90.
Evidence of noise-induced hearing loss among

orchestral musicians
Bradford C. Backus1 and Aaron Williamon2
1 Ear Institute, University College London, UK

An assessment of hearing thresholds among student orchestral musicians

was carried out at the UCL Ear Institute in conjunction with the Royal
College of Music (RCM). Audiogram data taken from 162 students (86 F,
76 M; mean age=23.7 years, SD=4.8) showed a statistically significant
notch at 6 kHz in the left ear, indicative of noise-induced hearing loss
(NIHL), but no significant notch was found in the right. Noise exposure
asymmetry did not appear to account for notch asymmetry as trombone
and trumpet players showed evidence of the same left notch trend as lat-
eralized instruments such as violin and viola players. The earliest audio-
metric indicator of impending NIHL for musicians may be a developing
hearing threshold notch at 6 kHz in the left ear.
Keywords: noise exposure; hearing damage; musical performance; or-

chestral musicians; music students
Both musicians and non-musicians are susceptible to work-related hearing

damage from sound, but unlike other professions, the sound musicians create
is not a by-product of their workit is their product. This distinction makes
musicians a difficult special case when it comes to determining what noise
regulations should apply to them. Although studies have detailed noise-expo-
sure for many professions including musicians (e.g. Taylor et al. 1965), rela-
tively little is known about how the noise of music affects student
musicians hearing (for reviews, see Royster et al. 1991; Fearn 1976, 1993; Lee
et al. 2005).
In this article, we report our measurements of hearing thresholds in a
large cohort of young orchestral musicians.
METHOD
Participants
One hundred and sixty-two RCM students (86 F, 76 M; mean age=23.7 years,
SD=4.8) participated in the audiogram study over two years, from 2007 to
2009.
Procedure
Participants completed a consent form and then went through standard clini-
cal audiometric testing. They were subsequently asked to respond to a survey
detailing their noise exposure.
Audiogram measurements were made with participants comfortably
seated in a soundproof room. A recently calibrated (June 2006, June 2007,
and June 2008) Kamplex KC 50 audiometer with TDH-39 earphones were
used for the standard audiometric frequencies 125, 250, 500, 1k, 2k, 4k, 6k,
and 8 kHz; high frequency headphones were used to measure 10k, 12.5k, and
16 kHz. A manual Hughson-Westlake procedure was used with 5 dB resolu-
tion for most subjects, but some were done with finer (3 or 1 dB) resolution
around areas of particular interest.
Audiogram corrections
Audiogram data were age-corrected using ISO 7029:2000 in order to remove

age effects and leave only noise-related effects (in this case, the removal of
age effects did not have an impact on the qualitative results, as the cohort was
all young musicians of similar age, range=19-36 years). Accounting for age
effects above 8 kHz is not supported by the standard, so for 10, 12, 12.5 and 16
kHz age effects were first estimated by extrapolating the mathematical coeffi-
cients used in the ISO standard linearly and then calculating and removing
those extrapolated effects of age.
Audiometer calibrations were initially carried out according to ISO 389,
but since Lutman and Qasem (1998) reported that this calibration proce-
durespecifically the pairing of the THD39 or TDH 39P headphone with the
specified IEC 303 couplercreates a resonance that can artificially increase
thresholds at 6 kHz, we re-calibrated the audiometer using an IEC 318 cou-
pler. We indeed did discover an artifact that would have increased thresholds
at 6 kHz by 3.4 dB in the right and 2.4 dB in the left. This artifact was re-
moved from the data retrospectively. We compared our corrected data with
an independent data set taken from students from the Royal Academy of Mu-
sic (RAM) using headphones that do not have any known artifacts and found
Figure 1. Summary of hearing threshold levels (dB HL) from audiograms taken from
RCM students. Boxes extend from lower quartile to upper quartile with median hearing
level demarked as the center of a notch; whiskers show extent of the contiguous data
(minus outliers). The average of the data is plotted as a thick line. Notches in the inter-
quartile boxes display the variability of the median between samples. The width of a
notch is computed so that box plots whose notches do not overlap have different medi-
ans at the p<0.05 significance level. The 6 kHz threshold notch for the left ear was sta-
tistically significant. No significant notch was found for the right ear.
the two sets agreed in all statistical aspects, indicating that our correction was
successful. (The RAM data will be presented in conjunction with these data in
a subsequent paper.)
RESULTS
Audiogram data (Figure 1) from the young musician group showed a statisti-
cally significant threshold notch at 6 kHz in the left ear, indicative of noise-
induced hearing loss. A Wilcox signed rank test showed the median threshold
at 6 kHz was statistically higher than the median from either the 4 or 8 kHz
tests (Z=-6.37, p<0.001, and Z=-4.52, p<0.001). Further, paired t-tests indi-
cated that the mean threshold at 6 kHz was significantly higher than thresh-
olds taken at 4 kHz or 8 kHz (p<0.001 for both). The right also showed that
the 6 kHz threshold departed significantly from 4 kHz, but there was no sig-
nificant difference between 6 and 8 kHz thresholds and therefore no signifi-
cant notch.
A comparison of two subgroups, trumpet/trombone players versus vio-
lin/viola players, was carried out to see whether noise exposure asymmetry
with violin and viola players assumed to get more exposure in their left ears
could account for the asymmetry we saw in the overall audiogram data. We
found, however, that both the trumpet/trombone subgroup and the vio-
lin/viola subgroup showed the same trends as the entire group data (cf. Fig-
Figure 2. Summary of hearing threshold levels (dB HL) from audiograms taken from 20
trumpet/trombone players (top panel) and 37 violin/viola players (bottom panel), dem-
onstrating that trumpet/trombone players exhibited notches at 6 kHz preferentially in
the left ear. Boxes extend from lower quartile to upper quartile with median hearing
level demarked as the center of a notch; whiskers show extent of the contiguous data
(minus outliers). The average of the data is plotted as a thick line. Notches in the inter-
quartile boxes are centered at the median and display the variability of the median be-
tween samples. The width of a notch is computed so that box plots whose notches do not
overlap have different medians at the p<0.05 significance level.
Table 1. Average hearing threshold levels (dB HL) for 4, 6, and 8 kHz from audiograms
from (1) the entire musician group, (2) 20 trumpet/trombone players, and (3) 37 vio-
lin/viola payers. All groups show similar trends, with higher thresholds in the left ear at
6 kHz and perhaps higher thresholds in the right ear at 8 kHz. These factors combine to
form a more visible notch in left-ear audiogram traces, such as those in Figures 1 and 2.
Left ear Right ear

4 kHz 6 kHz 8 kHz 4 kHz 6 kHz 8 kHz
All musicians 1.320.63 8.020.74 4.250.75 1.460.72 6.410.82 5.480.92
Violin/viola 3.051.39 7.561.81 3.631.11 0.571.20 5.141.33 3.441.48
Trumpet/trombone 2.572.16 8.632.43 2.081.81 1.982.31 6.032.31 6.802.37
ures 1 and 2), with the left ear being more likely to display a notch at 6 kHz.
The notch was more likely to be salient in the left because: (1) hearing thresh-
old levels at 6 kHz were generally higher (but not statistically higher, p=0.15
in paired t-test) in the left than in the right and (2) hearing threshold levels at
8 kHz were generally higher (but not statistically significantly higher, p=0.29
in a paired t-test) in the right ear (see Table 1).
DISCUSSION
Our audiogram data reveal a statistically significant notch in the left ear at 6
kHz across the group of young musiciansa potential hallmark of noise-in-
duced hearing lossbut interestingly, no such notch appeared for the right
ear data. We initially suspected that this asymmetry, clearly apparent across
the whole group, may have been caused by a subgroup of violin/viola players
who receive higher levels of sound in their left ears. However, data for this
group alone did not show a preponderance of the asymmetry. An established
body of literature suggests that there may be differences in the physiological
susceptibility of left and right ears to NIHL (Watson 1967). Exactly what type
of differences could account for the asymmetry is still a source of speculation.
For example, it is possible that olivocochlear efferents are stronger on the left
side or that there is a difference in the middle ear reflex between the two ears
(Nageris et al. 2007). It is also interesting to note that many reports indicate
that TEOAEs appear to be stronger in the right than in the left ear in infants
(Keefe et al. 2008). It should be noted that in many of these studies the right
ear is tested first, which could account for the asymmetry by producing an
efferent response prior to the subsequent left ear test.
In conclusion, the earliest audiometric indicator of impending NIHL for
musicians may be a developing notch at 6 kHz in the left ear.
Bradford C. Backus, Ear Institute, University College London, 332 Grays Inn Road,
London WC1X 8EE, UK; Email: b.backus@ucl.ac.uk
References
Fearn R. W. (1976). Hearing-loss caused by different exposures to amplified pop music.

Journal of Sound and Vibration, 47, pp. 454-456.
Fearn R. W. (1993). Hearing loss in musicians. Journal of Sound and Vibration, 163, p.
372.
Keefe D. H., Gorga M. P., Jesteadt W., and Smith L. M. (2008). Ear asymmetries in
middle-ear, cochlear, and brainstem responses in human infants. Journal of the
Acoustical Society of America, 123, pp. 1504-1512.
Lee J., Behar A., Kunov H., and Wong W. (2003). Noise exposure of opera orchestra
players. Canadian Acoustics, 31, pp. 78-79.
Nageris B. I., Raveh E., Zilberberg M., and Attias J. (2007). Asymmetry in noise-in-
duced hearing loss: Relevance of acoustic reflex and left or right handedness. Otology
and Neurotology, 28, pp. 434-437.
Lutman M. E. and Qasem H. Y. N. (1998). A source of notches at 6 kHz. In D. Prasher
and L. (eds.), Advances on Noise Research: Biological Effects of Noise (vol. 1, pp.
170-176). London: Whurr.
Royster J. D., Royster L. H., and Killion M. C. (1991). Sound exposures and hearing
thresholds of symphony orchestra musicians. Journal of the Acoustical Society of
America, 89, pp. 2793-2803.
Taylor W., Pearson J., Mair A., and Burns W. (1965). Study of noise and hearing in jute
weaving. Journal of the Acoustical Society of America, 38, pp. 113-120.
Watson J. E. (1967). Bilateral asymmetry in noise induced hearing loss. Annals of Otol-
ogy, Rhinology, and Laryngology, 76, pp. 1040-1042.
Strategies for the pianist to enhance the artistic

quality of piano recording
Marija Bajalica and David Lockett
Elder Conservatorium of Music, The University of Adelaide, Australia
The dream of every pianist is to perform in an environment that enables

a work to unfold naturally and effortlessly. This, however, does not hap-
pen often. Much less so in a studio environment where, compared with
the regular concert setting, we have to consider additional sets of issues
related to the process of recording in order to achieve a recorded result
that meets as closely as possible the players artistic intentions. This pa-
per suggests some of the strategies that proved to be effective in enhanc-
ing the artistic quality of a piano recording.
Keywords: piano performance; piano recording; music interpretation;

artistic quality; classical music
When performing, pianists can listen and evaluate only from their position at
the keyboard. As for any further assessment and feedback, one can listen to a
recording after the event. But how faithful is it?
Pianists are often surprised on hearing a playback of their performances.
They frequently remark that the recording does not necessarily match their
perception of certain aspects of their interpretation. Vladimir Ashkenazy once
remarked in an interview:
In a recording studio, the difference is what finally gets onto tape, de-
pending on the properties of the acoustics and given piano. What hap-
pens very often is that what you hear on a playback is not exactly what
you thought you played, so you have to adjust. Then in the process of re-
recording, you have to have the results that will have the meaning that
you have inside you (Noyle 1987, p. 10).
It is also commonly recognized that an interpretation that proves suc-

cessful in a concert performance does not always have equal success in a re-
cording studio. Claudio Arrau put it this way:
Things that work in performance are sometimes not good on records and
vice versa. Recording has its own lawsthats why I am not very much
for the recording of live performances (Dubal 1997, p. 7).
From the pianists point of view, the dry acoustics often found in studios,
the nature of the piano, the lack of audience, and the recording set up are all
factors that impact upon the way a recorded performance is perceived. Al-
though the set of issues surrounding each performance in a recording session
is very complex, we have to learn to deal with them. But how can we adjust
our technical and musical approaches in order to create a recording that
represents a faithful image of our interpretative ideals?
This paper explores some of the factors that influence the quality of the
artistic product achieved in a recording studio. It considers technical, musi-
cal, and conceptual adjustments that a pianist can make in order to achieve a
recorded performance that represents a faithful image of his or her interpre-
tative ideals. Due to space limitations, only three case studies will be pre-
sented. These have been selected either because they are unlikely to be
encountered in a regular concert performance or because they are intriguing
with respect to the ways in which recording equipment processes certain de-
tails of interpretation.
METHOD
Participants
The project is being conducted with the assistance of Kevin Roper, Senior
Music Producer with the Australian Broadcasting Corporation (ABC) in Ade-
laide. The pianist was Marija Bajalica.
Materials
Recordings used for this research were made at the studios of the ABC in
Adelaide. The piano used was a Steinway Grand, model D, which was one
year old at the time of the first recording.
Procedure
The investigative method consists of professionally produced studio re-

cordings and a series of home recordings. The chain of recordings begins with
a studio recording followed by a number of home recordings prior to a second
studio recording. Each is used as a reference point for the next recording of
the same work. Recordings are evaluated against the following variables:
tempo, timing, agogics, dynamic range, articulation, pedaling, and voicing.
Evaluations identify the extent to which the recorded material speaks truly to
a desired interpretation. Once the evaluation has been made, strategies are
developed and implemented in preparation for the next recording session.
The home recordings have proved to be useful in facilitating a quicker and
better result during the studio sessions. Specifically, they have assisted in
resolving matters of tempo and tempo relationships. We noticed that certain
types of articulation (such as a specific way of creating legato) or particular
approaches to pedaling work for recording regardless of acoustics or the na-
ture of the instrument. This gives the pianist a clearer picture of what to do
while working on repertoire and preparing for a studio recording. It induces
greater confidence and a better chance of creating a successful recorded
product in a shorter period of time.
RESULTS
The first case study is the Allegro from Mozarts Sonata in B flat KV570. The
theme in the lower voice in Figure 1 is broad and flowing and needs to be
crafted into a single thought.
The first recording revealed inconsistencies in sound quality within this
phrase (specifically in bars 102 and 106). The sudden disjunction in tone be-
tween the D eighth notes and B quarter note in the lower voice made it sound
as if the theme was being performed on two different instruments.
To achieve the desired unity of expression, it was necessary to suppress
the multiple personalities found in this particular register of the instrument.
In order to achieve this, an unusually lazy legato was used, providing more
time to blend and disguise the different tonal characters. Since no pedal is
used in this section and microphones were positioned reasonably close to the
strings, there was a concern that longer overlapping of the fingers would be-
come noticeable on the recording, yielding distasteful and stylistically inap-
propriate results.
The recording made during the second studio session did not reveal this
to be an issue. Neither the acoustics nor the microphones recognized the lazy
legato technique as an inadequate one. The disjunctions in sound were mini-
Figure 1. Allegro from W.A. Mozarts Sonata Bb Major KV 570 (bars 101-108).
Figure 2. Allegretto from W.A. Mozarts Sonata Bb Major KV 570 (Bars 1 to 5).
Figure 3. Andante dolente section of Juliets Girlhood from Prokofievs Romeo and
Juliet Suite for Piano Op. 75 (bar 82).
mized and the legato that was achieved sounded rather like a cello. With
fewer disturbances in the lower voice, rests in the first voice became trans-
parent, providing a clear entry of the melody in the upper voice.
The second case study is from the third movement, Allegretto, of the
same Sonata, shown in Figure 2. The light-hearted, even irresponsible char-
acter of the theme dominates this movement. To complement its mood, the
pulse unit should be treated in a leisurely fashion, but with no delays.
The initial feeling after the first recording was that it had achieved the
right pulse and that the variety of detached notes were properly felt and exe-
cuted. Subsequent evaluation, however, concluded that the differences be-
tween finger staccato, hand staccato, and all different articulations from
portato to staccato could not be clearly distinguished. This compromised
some of the shades of expression that were being sought. Furthermore, a
certain amount of sostenuto can be heard in the theme that was not obvious
at the time of playing. This can be attributed to the splashing acoustics of the
hall that did not allow staccato notes to clear in time for the next to proceed.
Attempts to hear the end of one tone before producing the next ended in cre-
ating impression of sostenuto, which was undesirable for this movement.
The general conclusion from the first recording of the Allegretto was that
the attempt to engage with the acoustics and large proportions of the hall did
not result in the type of sound that was desired. The recording equipment
captured the initial moment that the sound was produced but very little of its
full journey. Therefore the decision was made that the next recording of the
same piece would explore the accustomed interpretation without any sponta-
neous adjustments provoked by the studio environment. There is a sense in
which this is counterintuitive as pianists routinely make fine adjustments in
response to what they are hearing.
The follow-up session proved that this was an extremely difficult and
awkward task. The second recording did, however, achieve something much
closer to a desired interpretation of the work. Its character is now supported
by an appropriate variety of articulation, and an effective pulse enables de-
tached notes to speak freely yet coherently.
The third case study is the Andante dolente section of Juliets Girlhood
from Prokofievs Romeo and Juliet Suite for Piano Op.75. It implies, for the
first time in the suite, a tragic conclusion to the opus, a shadow forming over
Juliets portrait. While shifting away from the rest of the piece, it speaks
through a dimmed tone of voice in the theme of the left hand (see Figure 3)
and a continuous vibration in the upper register.
In the first recording, the Andante dolente achieved its tranquillo state to
a certain extent. Some mild finger action was employed in executing the
thirty-second notes in the upper voice. The recording revealed that the mi-
crophones picked up every bit of that action so that the mechanics of the pi-
ano (the hammer action in particular) became overwhelmingly present in the
sound. Consequently, the right hand passage sounded more as a work of labor
and less as a continuous vibration. Along with it, the lower voice had lost its
peace. Dynamics in this section were very well balanced, so the main concern
for the re-recording session was the substance of the tone.
To project this section more successfully, the voices had to become more
specific in shades of expression. The lower voice required a somber quality of
tone, while the upper needed to achieve the effect of an inevitable and con-
tinuous occurrence.
Issues regarding the melody were resolved by voicing. The somber color
of tone can be obtained by feeling the key motion with greater sensitivity so
that any fast contact with the bottom of the key is avoided. By making a very
slow and controlled descent to the bottom of the key, the moment and the
place of the birth of the tone becomes uncertain. Supported with the right
amount of sound, the gloomy and somber effect is achieved.
To eliminate the intrusion of the hammer action in the right hand sound
and to accomplish the effect of vibration, finger action needs to be minimized.
The sound should be produced by a swaying of the weight over the keys, em-
ploying as little finger movement as possible.
This method was easier to implement than in the case of the Mozart work,
as it utilized only recognized pianistic skills. In fact, the acoustics of the hall
worked to the benefit of the piece and were of great help in the delivery of the
final product.
DISCUSSION
This paper addresses a set of issues that has been remarked upon by many
professional pianists but which has escaped serious, structured attention. It
identifies particular technical and artistic approaches that enable the pianist
to make the transition from concert hall to recording studio without com-
promising artistic outcomes. The first case study confirms that subtle changes
in articulation can be made to counteract tonal irregularities between regis-
ters. Such irregularities would go unnoticed in the concert hall but become
magnified in a recording. The second case study established that focusing
attention on the immediate sound feedback received from within the piano
(rather than from the acoustics of the hall) could be a key to a successful re-
cording. The final example concludes that studio recording requires much
greater sensitivity towards the key motion if traces of the hammer action are
to be cleared from the recorded sound. In cases where subtle shades create an
essence of the tone, this proved to be a crucial issue.
Marija Bajalica, Elder Conservatorium of Music, The University of Adelaide, Adelaide,

South Australia 5005, Australia; Email: marijabajalica@yahoo.com
References
Noyle L. J. (1987). Pianists on Playing. Metuchen, New Jersey, USA: Scarecrow Press.
Dubal D. (1997). Reflections from the Keyboard. New York: Schirmer Books.
Analyzing and representing Transylvanian

village music by using motion capture
Filippo Bonini Baraldi1, Emmanuel Bigand2, and Thierry Pozzo3
1 CREM-LESC, Ethnomusicology Research Center,

University Paris Ouest Nanterre, France
2 LEAD CNRS, Institut Universitaire de France, University of Burgundy, France
3 ERT Motricit-Plasticit, INSERM, University of Burgundy, France
Techniques based on motion capture can be useful to analyze and tran-

scribe a foreign musical system: Transylvanian village music. Two musi-
cal parameters were the object of study: rhythm and desynchronization
between two performers, a violinist and a viola player. Results showed
that rhythm is a local variant of the aksak system and is based on two du-
ration units (S=short, L=long), which respect the formula 2/3<S/L<3/4.
Performances are characterized by large deviations of the S/L ratio from
period to period, which have an expressive function. Deviations are re-
lated to a swinging interpretation, consisting of a voluntarily desynchro-
nization between the performers.
Keywords: ethnomusicology; Transylvania; gesture; rhythm; desynchro-

nization
Analysis and transcription of non-western music has always been a contro-

versial subject in ethnomusicology (England 1964). The Western musical
theory and notation system are often unadapted to describe musics of oral
traditions that rely on different concepts of rhythm, scale, interval, etc. which
are often implicit. For this reason, ethnomusicologists have progressively
developed specific techniques in order to analyze and represent formal pro-
prieties of the music they study (Rouget 1981). Emerging technologies based
on motion capture offer new possibilities for a deeper understanding of music
production and perception (Leman 2007). The main idea underlying the pre-
sent study is that these techniques can be particularly useful to analyze and
represent Transylvanian village music.
Two musical parameters of the de meseli or de jale repertoire (table

songs or songs of sorrow) were the objects of study: rhythm and desyn-
chronization between two performers, a violinist and a viola player. These
parameters are particularly apt to be studied by motion capture because they
are directly related to body movements. The metrical-rhythmical system of
these slow listening tunes belong to the family of aksak (from the Turkish
lame) rhythms, which have been documented in different parts of the world
and especially in Turkey and the Balkan region (Briloiu 1952). A rhythm is
an aksak when (1) it is periodical and (2) each period is composed by the
combination of two duration units (short=S and long=L), which are in a ratio
of S/L=1/1.5=2/3 (Briloiu 1952). Aksak rhythms are usually indicated by the
series constituting the period, for example 2.2.3, 2.3.2, 2.2.2.3, etc. The rep-
ertoire concerned here is characterized by a 2.3 period. Aksak has been at the
center of controversial debates in French ethnomusicology, and there is no
agreement on the way to theorize it (in terms of an irregular bichrone beat, or
of an underlying monochrone beat; see Arom 1992, Bout 1997, Cler 1994,
Cler and Estival 1997). Rather than with theoretical matters, the present
study deals with a concrete issue: precise measurement of the S/L ratio, which
is an important indicator of how the aksak model is locally conceived by mu-
sicians. A related issue concerns deviations of S/L ratio from norm values in
live performance. Our hypothesis is that aksak deviations are related to a
swinging effect, consisting of a voluntary desynchronization between per-
formers.
METHOD
Participants
Two professional gypsy musicians from Ceua, a small village of the Trgu
Mure region, were invited to the INSERM Laboratory of the University of
Burgundy. They are among the best (and last) interpreters of Transylvanian
village music, were born in 1953 and 1951, and have played together since
they were children. Csanyi plays the melody on the violin and Csangalo the
rhythmic-harmonic accompaniment on the contr, a specially prepared viola
with three strings tensed on a bridge with a flattened curve, tuned A-D-G.
Materials
The movements of 38 retroreflective markers (15 mm in diameter), placed at

various anatomical locations on the body, were measured using an optolec-
tronic device, Smart (BTS, Milan, Italy). Six infrared-emitting cameras were
attached to six tripods, 2 m from the ground on each side of the subject, at a
distance of 3 m from each subjects body. In this experiment, the motion of
only two markers was analyzedthat is, the ones located on the top of the
bows. Kinematic parameters in three dimensions (X, Y, and Z) were calcu-
lated from successive frames taken at 10 ms intervals. Kinematic variables
were low-pass filtered using a digital second-order Butterworth filter at a cut-
off frequency of 5 Hz. Sound was recorded separately and the session was
filmed with two additional standard cameras. Synchronization between sound
and image was obtained by using a clapper board equipped with two addi-
tional retroreflective markers.
Procedure
Musicians were standing at the center of a circular region surrounded by the

six infrared cameras. They were asked to play 30 s fragments of six tunes
issued from the local de meseli repertoire. One tune (Duo14, test melody) was
played 3 times (Duo12, Duo13, and Duo14). The session lasted about 3 hours
and musicians, who are used to much longer musical performances, felt
comfortable in the experimental setting.
RESULTS
Figure 1 shows the rhythmical pattern obtained by tracking the marker posi-
tioned at the top of the viola bow on the Y dimension. The periodical cycle of
the aksak rhythm is clearly recognizable. The short (S) and long (L) durations
for each period were obtained by measuring the time interval between one
pick and the next. The picks indicate a change in the bow direction, corre-
sponding to the beginning of a short rhythmic unit (superior picks) and of a
long rhythmic unit (inferior picks). This measure was obtained with a margin
of error of 7 ms. For the test melody (Duo14), short and long durations vary
as follows: 1.05 sS1.127 s, and 1.541 sL1.817 s. The mean value of S/L
ratio for Duo14 is 0.7 (corresponding to S/L=2.2/857, if S=2 as in aksak conven-
tion), and varies largely from period to period in the same performance
(2/3.044S/L2/2.448). The mean values of S/L ratio for the three performances of
the same melody are very similar: S/L=0.695 for Duo12 and 0.696 for Duo13.
Finally, for all 8 melodies the proportion S/L respected the formula:
3/4S/L2/3 (see Table 1).
Desynchronization was analyzed by superposing the movements of the

violin bow (melody) with those of the viola bow (accompaniment) (Figure 2
top). These patterns were projected on a stave to obtain a musical transcrip-
tion (Figure 2 bottom). Musicological analysis linked the notes of the melody
Figure 1. SMART viewer window. Top: Musicians virtual silhouettes (violinist left,
violist right). Bottom: Rhythmical pattern of Duo14, obtained by tracking the marker
positioned at the top of the viola bow on the Y dimension. (See full color version at
Table 1. S/L mean values for eights de meseli tunes (Duo12, Duo13, and Duo14 are dif-
ferent performances of the same tune). Aksak (2/3) and 3/4 ratio are included in the table
for comparison.
Ratio S/L (mean) Rhythm (S, L)

Aksak norm 0.667 2.3
Duo12 0.695 2.2, 878
Duo13 0.696 2.2, 874
Duo14 0.700 2.2, 857
Duo18 0.735 2.2, 721
Duo19 0.679 2.2, 946
Duo20 0.689 2.2, 903
Duo21 0.687 2.2, 911
Duo22 0.729 2.2, 743
3.4 0.750 2.2, 667
Mean value on 8 tunes 0.701 2.2, 853
with the expected chords of the accompaniment (Figure 2 bottom). What

emerges is a desynchronization between the two performers at almost each
period: melody both anticipates and follows the relative chords.
Figure 2. Top: Superposition of violin and viola bow movements for Duo14 (extract).
Bottom: Transcription of melody (violin bow) and harmonic-rhythmic accompaniment
(viola bow) obtained from changes of bows directions. Lines linking melody to accom-
paniment highlights desynchronizations between the two performers.
DISCUSSION
Motion capture technologies have been used here to study two parameters of
Transylvanian de meseli repertoire in live performances: rhythm and swing,
defined as the desynchronization between melody and harmonic-rhythmical
accompaniment. The tracking of only two markers (positioned at the top of
the bows) allowed us to obtain easy readable representations of (1) aksak
rhythmical patterns and (2) type and degree of desynchronizations. These
gesture-based representations served here both as a measuring tool and as a
support for musicological analysis and transcription.
While measures have been done for Turkish village music (using a sound
analyzer, Cler and Estival 1997), to our knowledge Transylvanian aksak rep-
ertoire has been analyzed only qualitatively (Bout 1997). In relation to a
similar repertoire as the one concerned here, Bout (1997) raised the problem
of determining if musicians conceive the rhythm in terms of an orthodox
aksak (2.3) or of a heterodox aksak (3.4). Results showed that the rhythm
concerned here is between the two: the de meseli repertoire is based on a
local variant of aksak characterized by the function 2/3<S/L<3/4.
Deviations from norm durations, which are generally linked to expressiv-
ity in live performances (Gabrielsson 1995), were found. While in the case of
turkish aksak, Cler and Estival (1997) described a stability of S/L deviations
along the entire performance, our results showed that they vary widely from
period to period. This difference may be attributed to the different social
function of the repertoire studied. Transylvanian listening tunes, as opposed
to Turkish dance tunes, may leave to the performers more liberty in rhythmi-
cal deviations. We advanced the hypothesis that these deviations in aksak

proportions are related to a swinging interpretation, intended as a desyn-
chronization between the two performers. Motion capture, associated to
musicological analysis, attested the large presence of desynchronizations
between melody and accompaniment, which suggests that they are intro-
duced voluntarily. Further research is needed to confirm if aksak deviations
and desynchronizations are systematically linked and in which manner they
depend from the musical structure of the de meseli tunes.
Acknowledgments
The first author has conducted field research in Transylvania from 2004-06 thanks to a
CNRS doctoral grant. This study was realized thanks to a financial support of the CNRS
Music, Cognition, Society network.
Filippo Bonini Baraldi, CREM-LESC Maison de lEthnologie et de lArcheologie, Univer-

sity Paris Ouest Nanterre, 200 Avenue de la Rpublique, Nanterre Cedex 92001, France;
Email: filibb@gmail.com
References
Arom S. (1992). la recherche du temps perdu: Mtrique et rythme en musique. In J.-

J. Wunen-Burger (ed.), Les Rythmes. (pp. 195-205). Paris: L'Harmattan.
Bout J. (1997). Pulsations retrouves: Les outils de la ralisation rythmique avant l're
du mtronome. Cahiers de Musiques Traditionnelles, 10, pp. 107-125.
Briloiu C. (1952/1984). Problems of Ethnomusicology. Cambridge: Cambridge Univer-
sity Press.
Cler J. (1994). Pour une thorie du rythme aksak. Revue de Musicologie, 82, pp. 181-
210.
Cler J. and Estival J. P. (1997). Structure, mouvement, raison graphique: Le modle
affect. Cahiers de musiques traditionnelles, 10, pp. 37-80.
England N. M. (1964). Symposium on transcription and analysis: An Hukwe song with
musical bow. Ethnomusicology VIII, 3, pp. 223-277.
Gabrielsson, A. (1995). Expressive deviations and performance. In R. Steinberg (ed.),
Music and the Mind Machine (pp. 35-47). Berlin: Springer-Verlag.
Leman M. (2007). Embodied Music Cognition and Mediation Technology. Cambridge,
Massachusetts, USA: MIT Press.
Rouget G. (1981). Ethnomusicologie et reprsentations de la musique. Le Courrier du
CNRS, hors srie 42, pp. 10-11.
Studying a score silently: What benefits can it

bring to performance?
Anabela Bravo and Philip Fine
Department of Psychology, University of Buckingham, UK
Mental practice and analysis can be considered as efficacious and useful

for performers, as they help them to develop a mental representation of
music. This paper discusses what information in the score is potentially
useful in developing performance expertise, and how performers can and
do utilize it. First, we are concerned with the process of learning and
performing a new score and, in particular, in which stages of this process
performers find mental practice and analysis with the score useful. Sec-
ond, we explore what information can be learned from the score before it
is performed, i.e. what sort of cognitive representations the performer
can obtain from the score. Third, we ask how performers can organize
and use that information, i.e. the benefits of mental rehearsal in the at-
tainment of performance excellence. These questions will be addressed
through the statements of psychologists, teachers, performers, and musi-
cologists in the relevant literature.
Keywords: expert performance; notated score; cognitive schemas; men-

tal rehearsal; analysis
Mental rehearsal is a cognitive process that complements physical rehearsal

and helps lead to the development of performance expertise. For the per-
former, musical ideas are primarily registered in written notation, and mental
rehearsal can take place with or without the score. Studying a score silently
has benefits for musicians, and can enhance the quality of the eventual per-
formance. Mental rehearsal with the score enables the musician to gain an
understanding of the structure of the piece and to form coherent mental rep-
resentations of it, including cognitive, auditory, and motor representations.
As musicians become more experienced, they become better at interpreting
and communicating musical ideas, and also better able to self-evaluate and to
consider strategies of learning.
Mental rehearsal with the score is clearly connected with analytical study,
and analytical insight no doubt interacts with the performance plan. Under-
standing levels of structure and hierarchical organization or categorization
within various musical parameters, for instance, can help the performers
interpretation. Analysis connotes conceptual skills as well as allowing the
improvement of perceptual and motor skills, in particular those related to an
expressive performance.
MAIN CONTRIBUTION
Our focus is on exploring how we map the conceptual and perceptual struc-
tures in the music to cognitive mental representations through score reading.
We will also explore the structure of the learning/performing process of a
new score, in connection with the implied cognitive models, to understand
better how it can inform performance through feedback.
Mental rehearsal as a learning strategy
Mental imagery has similarities to perception but occurs in the absence of an

external stimulus, instead based on information in memory. It relies on in-
ternal representations, to which we ascribe specific meanings (Zbikowski
2007). Auditory representations are important in all stages of musical per-
formance preparation, including sight-reading, recalling music from memory,
and polished performance (e.g. Repp 2001). Lehmann and Davidson (2002)
subdivided mental representations in performance into goal representations,
production representations, and representations of the current performance.
Williamon et al. (2002) investigated how mental representations in pian-
ists affected their recall of music. They found that pianists used longer prac-
tice segments and recalled larger sections of music in the later stages of
practice, when they also became more able to shift their attention between
different levels of the musical structure.
Through the process of learning contemporary music, Hill (2002) experi-
enced the value of earlier mental practice, which he found was a way of un-
derstanding the musical implications of the piece: he writes, as much as
possible is learned before we take a work to the instrument. But the main aim
of mental study is to liberate our musicality, to make sure that musical goals
not technical constraintscome first (p. 143).
Hultberg (2008) described a qualitative and collaborative study with two
instrumental students. The two students were able to identify certain complex
strategies for exploring musical meaning, which were related to their previous
musical education. Two different learning strategies were revealed, one giving
Table 1. The stage(s) of the learning/performing process of a new score when perform-
ers find mental practice and musical analysis with the score useful (n=75).
Mental Analysis of
Stage rehearsal the score
Sight-reading the score for the first time 33 25
Exploring and getting to know the music 35 41
Becoming fully familiar with the piece 33 34
Playing the whole piece fluently and musically 26 21
Memorizing the piece 32 22
Trial performances 15 9
Other 2 2
priority to exploring structural aspects in the printed score, the other focusing
on exploring the music while playing, i.e. auditory rather than visual.
An exploratory survey on performers experience of mental rehearsal and
the importance of music analysis in learning and performing a new piece is
currently being conducted. Two questions are pertinent to this paper, and
asked at which stages the performers tended to find these strategies most
useful. Current results (data are still being collected) for mental practice and
analysis are shown in Table 1.
As is apparent from the Table, mental practice is particularly useful right
from before initial sight-reading up to memorizing the piece for performance.
Score analysis was thought to be most useful while the performer is exploring
and becoming fully familiar with the piece, but less so before sight-reading.
What we can learn from the score
At a basic level, we need to observe the global parameters of musical notation,

such as clef, key signature, meter, and tempo. At an expert level, performers
use knowledge of music theories (explicitly or implicitly) as techniques for ob-
taining information. According to Williamon et al. (2002), music theories
present valid descriptions of the cognitive processes in music, they explain
how sound is structured in musical contexts, they lead to understanding the
underlying processes of mental representations, and they are hierarchically
sophisticated representational systems to describe human cognition (p. 516).
Some relevant music theories are outlined below.
Models of musical perception and cognition
Reductive theories (e.g. Schenker 1979, Lerdahl and Jackendoff 1983) reduce
the piece to its underlying contrapuntal-harmonic structure. They can explain
how listeners represent what is aurally understood in a piece of music. They
may clarify a pieces structure, reveal how it was composed, and draw atten-
tion to its hierarchical levels. Harmonic region based theories (e.g. Longuet-
Higgins 1987, Lerdahl 2001) evaluate harmonic distance, according to pitch
relationships, and map the harmonic progressions along the entire musical
piece. They account for sequential and hierarchical harmonic tension, voice-
leading, and harmonic attraction. Like reductive theories, they can represent
whole pieces of music graphically and succinctly.
Categorization theories (e.g. Delige and Mlen 1997, Ockelford 2005) ac-
count for surface level events and their similarities, transformations, and
derivations, and how these events can be categorized. Mathematically based
theories (e.g. pitch-class set, see Forte 1973) allow the discovery of rules and
their manipulation, explain permutations and combinations, and account for
levels of structure, particularly in contemporary music. Descriptive theories
explain and describe musical symbols and events along the score in a narra-
tive way. They promote insight into the referential meaning of the piece and
guide the aesthetic experience (e.g. Agawu 2009).
We can suppose that all the above theories, which talk mainly about how
we represent the music we hear, can be applied to how we analyze the printed
notation when we see a score. When expert musicians see a score, the analysis
they do may involve hearing the music internally and understanding its
structure, therefore creating an association between seeing and hearing.
How performers can utilize that information
The views of music psychologists, music teachers, expert performers (e.g.

pianists and conductors), and musicologists can all shed light on the contri-
bution of analysis to music performance.
Aiello and Williamon (2002) describe how mental rehearsal can allow
performers to: describe and analyse music in terms of macrostructure and
microstructure; learn the landmarks of the piece; and identify melodic and
rhythmic patterns, closures, and points of tension and resolution.
In terms of pedagogy, Provost (1994) suggested that teachers should en-
courage students to sing new pieces they are learning and use ear training
exercises, complemented by music theory, enabling more advanced students
to develop all the skills needed to play a range of music at a professional level.
Mental rehearsal in expert pianists (Chaffin et al. 2002) and conductors

(Battisti 2007) has been investigated and shown to contribute to the under-
standing of musical events, such as form, texture, and other musical dimen-
sions. It also improves interpretation and memorization and provides
knowledge of the pieces phrase structure.
According to Howells (1996) musicological perspective of analysis, analy-
sis enables us to go behind the surface detail and get an overview of the score.
Through analysis, performers can become familiar with musical styles, pro-
duce informed and aesthetically satisfying interpretations, and better plan,
execute, and evaluate general musical performances, including questions of
emphasis, articulation, and technique (e.g. bowing and fingering). Analysis
also enables performers to discover how different interpretations affect the
listeners perception and understanding of the musical work.
IMPLICATIONS
The usage of mental rehearsal and analysis helps various aspects of perform-
ance practice, including sight-reading, audiation, memorization, and tech-
nique. It also has implications for pedagogy and teaching, by stimulating
intellectual inquiry, creative interpretation, and improvisatory skills in the
student. A good level of analysis implies expertise and helps performers to be
versatile, not least by enabling them to create their own learning strategies,
and to acquire a capacity for evaluation of their own (self-evaluation) and
others performances. It enables musicians to generate and use mental repre-
sentations efficiently, and this aspect is a fundamental requirement to be an
expert performer (Lehmann et al. 2007). Future studies will investigate the
effects of expertise on strategy development.
Acknowledgments
The authors would like to thank the survey respondents (from the UK, Portugal, and
Spain), and the Portuguese Government (MCTES, FCT) for financial support.
Anabela Bravo, Department of Psychology, University of Buckingham, Hunter Street,

Buckingham MK18 1EG, UK; Email: anabela.bravo@buckingham.ac.uk
References
Agawu K. (2009). Music as Discourse. Oxford: Oxford University Press.
Aiello R. and Williamon A. (2002). Memory. In R. Parncutt and G. McPherson (eds.),

The Science and Psychology of Music Performance (pp. 167-182). Oxford: Oxford
University Press.
Battisti F. (2007). On Becoming a Conductor. Galesville, Maryland, USA: Meredith
Music Publications.
Jersey, USA: Erlbaum.
Delige I. and Mlen M. (1997). Cue abstraction in the representation of musical form.
In I. Delige and J. Sloboda (eds.), Perception and Cognition of Music (pp. 387-
412). Hove, UK: Psychology Press.
Forte A. (1973). The Structure of Atonal Music. New Haven, Connecticut, USA: Yale
University Press.
Hill P. (2002). From score to sound. In J. Rink (ed.), Musical Performance. (pp. 129-
143). Cambridge: Cambridge University Press.
Howell T. (1996). Musical analysis. British Journal of Music Education, 13, pp. 123-134.
Hultberg C. (2008). Instrumental studies strategies for finding interpretations. Psy-
chology of Music, 36, pp. 7-23.
Lehmann A. and Davidson J. (2002). Taking an acquired skills perspective on music
performance. In R. Colwell and C. Richardson (eds.), The New Handbook of Re-
search on Music Teaching and Learning (pp. 542-560). Oxford: Oxford University
Press.
Lehmann A., Sloboda J., and Woody R. (2007). Psychology for Musicians. Oxford:
Oxford University Press.
Lerdahl F. (2001). Tonal Pitch Space. Oxford: Oxford University Press.
Lerdahl F. and Jackendoff R. (1983). A Generative Theory of Tonal Music. Cambridge,
Longuet-Higgins H. (1987). Mental Processes. Cambridge, Massachusetts, USA: MIT
Press.
Ockelford A. (2005). Repetition in Music. Aldershot, UK: Ashgate.
Provost R. (1994). The Art and Technique of Performance. Chester, UK: Music Sales
Corporation.
Repp B. (2001). Expressive timing in the minds ear. In R. Godoy and H. Jorgensen
(eds.), Musical Imagery (pp. 185-200). Lisse, The Netherlands: Swets and Zeitlin-
ger.
Schenker H. (1979). Free Composition (English translation). New York: Longman.
Williamon A., Valentine E., and Valentine J. (2002). Shifting the focus of attention
between levels of musical structure. European Journal of Cognitive Psychology, 14,
pp. 493-520.
Zbikowski L. (2007). Aspects of meaning construction in music: Toward a cognitive
grammar of music. Almen Semiotik, 17, pp. 43-72.
The learning cultures of performance:

Applying a cultural theory of learning to
conservatoire research
Rosie Burt-Perkins1,2

2 Faculty of Education, University of Cambridge, UK
As educational institutions that play a role in the training of many per-

formers, conservatoires have increasingly become the focus of research.
Researchers have explored the one-to-one lesson context, identified and
tested means of achieving musical excellence, and worked to enhance
musicians health. There remains, though, little research that investigates
the conservatoire as a learning site, characterized by a set of interactions
between performer, institution, and music profession. Students learning
at conservatoires, as well as teachers and researchers attempting to em-
bed new pedagogical approaches or curricula, must negotiate an educa-
tional and musical system embedded in years of history: a system that
has, in other words, a learning culture. This paper explores the central
tenets of learning culture as a theoretical approach, arguing that conser-
vatoire research could benefit from a lens that views learning as inher-
ently cultural. The methodological implications of learning culture are
discussed, including the need for qualitative methods that seek interpre-
tive understandings and in-depth, rich data. The paper concludes by of-
fering implications for the application of learning culture within the field
of performance science, addressing Jrgensens (2009) call for increased
research on the institutional culture of conservatoires.
Keywords: conservatoire; institution; learning culture; social practice;

qualitative methodology
As educational institutions at the centre of the quest for performing excel-

lence, conservatoires have increasingly become the focus of research. Given
its unique position within the higher education system, the one-to-one lesson
has been the subject of plentiful enquiry, with focus given to student-teacher
relationship, pedagogical approaches, and the master-apprentice model of

learning. Recognizing the need to look beyond the individual classroom, re-
searchers have more recently begun to identify and test means of achieving
musical excellence (Williamon 2004), including strategies for enhancing
practice (Jrgensen 2004) and enhancing the health and wellbeing of music
students (Kreutz et al. 2009). Others have focused on assessment methods
within conservatoires, pioneering new approaches that include peer assess-
ment (Lebler 2008) and group assessment (Barratt and Moore 2005).
There remains, though, little research that investigates the conservatoire
as a learning site, addressing what Jrgensen (2009) terms the under-re-
searched institutional culture. While Nettl (1995) and Kingsbury (2001)
both consider aspects of institution and of culture in their work in the USA,
their ethnomusicological stance steers them away from exploration of the
interrelationships between institution, culture, andcruciallylearning.
Conservatoire students, as well as teachers and researchers attempting to
embed new pedagogical approaches, must negotiate an educational system
embedded in years of musical, social, and institutional practices; what I will
term learning culture. Such practices will influence both the types of learn-
ing that are promoted or inhibited within conservatoires, and also the ways in
which learners become socialized during their time at a conservatoire (see
James et al. 2007). Knowledge of these practices offers scope for identifying
challenges to effective learning and barriers to institutional change.
How, though, can research access practices (or cultures) that are often
hidden, tacit, or deeply embedded in day-to-day life? This paper outlines a
theory of learning culture, arguing that the conceptual and methodological
angles that it brings to the fore offer scope for new understandings of the
cultures of institutions that train students for performing excellence.
MAIN CONTRIBUTION
The term learning culture is not new to educational writing, yet there are
relatively few attempts to conceptualize the term or to operationalize it as a
theoretical tool. The exception is the work of James et al. (2007), who used
learning culture as a means of accessing the complexity of further (post com-
pulsory) education in the UK. At the center of their thinking is the definition
of learning culture as the social practices through which people learn (p.
23). So defined, it is not simply the contexts within which people learn but
rather the practices through which people learn, practices which promote,
inhibit or rule out certain kinds of learning (p. 28). The key assumption here
is that learning takes place in, and through, a cultural setting (Hodkinson and
James 2003). That is, if we are to meet the oft-posed challenge of changing
the culture, we need to look beyond the one-to-one lesson, practice studio, or
concert platform to discover the practices that belie what can and is learned
at conservatoires; how different learning cultures enable or disable different
learning possibilities for the people that come into contact with them (James
et al. 2007, p. 28).
As yet, learning culture has not been adapted for theoretical use in higher
education institutions or in music education. While space limits a full con-
ceptualization here, there are two central tenets to the theory as we consider
its application to performance-based institutions:
Learning culture incorporates, but is not bound by, institutional culture.

Rather, it is bound by the broader fields (see Bourdieu 1984) in which the
institution and its students and teachers operate. That is to say, knowl-
edge of the practices at play in a conservatoire will come from research
that seeks understanding of the lived experiences of learners as they move
into, within, and beyond the institution. This includes learners perform-
ance history, previous education, family, friends, gender, specialism, and
so forth, as well as social and cultural interactions within the institutional
setting. Operationally, this means taking a holistic approach to research-
ing culture, recognizing that the institution plays a large but not exclusive
part in constructing the practices through which performers learn.
Learning culture is a complex amalgamation of agency and structure,
assuming that individuals both shape and are shaped by the learning
cultures of which they are part (James et al. 2007). Social space is not
seen as an equal playing field but rather one in which people compete for
resources (Bourdieu 1984). In the context of performance, for example,
learners compete for recognition, the best teachers, performance op-
portunity, publicity, and so on. These sometimes obscured struggles occur
within, and interact with, a complex web of power relationships and
practices that both shape learning possibilities within a learning culture
and that shape the way in which the learning culture itself develops. In
order to uncover these interactions, researchers must embrace the com-
plexity of culture, searching for knowledge that may be obscured or hid-
den in order to reach in-depth understandings that have the potential to
inform institutional change.
Conceptually, then, learning culture focuses researchers on the practices

through which performers learn, assuming that these practices will be both
institutionally-mediated and constructed within the performers own experi-
ences and values. In uncovering these practicesor forms of learningit be-

comes possible to (1) highlight complex enablers and barriers at play for
young performers and the role that the institution plays in alleviating or con-
solidating these, (2) highlight institutional priorities and assumptions, and
their fit with those of performance students and teachers, and (3) highlight
cultural assumptions within a particular community or institution (such as a
conservatoire), which can illuminate barriers to institutional change. Knowl-
edge of what these practices are, and the roots of their construction, forms an
important first step in understanding the culture of learning to perform, and
the potentials for transformation of this culture.
Where does learning culture lead us methodologically?
Learning culture brings with it a host of methodological assumptions, all of

which work to capture the complexity needed to research culture in this man-
ner. First, it demands an epistemology that reflects the nature of knowledge
as constructed in social space, shaping and shaped by cultural factors such as
gender, class, instrument type, or institution. That is to say, post-positivist
paradigms that seek explanation (often through support or rejection of a hy-
pothesis) are rejected in favor of broadly constructionist stances that seek
understanding of how people interact within, and with, their surroundings.
This steers the researcher towards a qualitative methodology, with its empha-
sis on capturing the nuance and complexity of the social situation under
study (Janesick 2000, p. 380). In taking a qualitative approach, attention is
paid to the social world as it is understood by participants, actively seeking
different interpretations of the learning culture in order to capture its com-
plexity.
In particular, learning culture commits researchers to spending time in
the field, working with ethnographically-informed methods to reassemble
the parts into the wholes from which they were originally taken (Bloomer
2001, p. 430). The process of knowledge-construction includes a range of
methods, most prominently interviews, participant documentation, observa-
tion, and document analysis, all of which span across learners, teachers, and
management staff. In sum, taking learning culture as a theoretical lens re-
quires the researcher to embrace the messiness of social life, using qualita-
tive methods to make the hidden visible, and thereby shedding new light on
the cultures in, and through which, performance students learn.
IMPLICATIONS
This paper has proposed learning culture as a theoretical lens for tackling the
under-researched area of conservatoire culture. At the time of writing, field-
work has been recently completed at a UK conservatoire, limiting this paper
to a discussion of theoretical, rather than empirically-based, implications.
However, the potential for new understandings is, I argue, two-fold. First,
learning culture offers understanding of the hidden skills of performance;
those that accompany the more frequently researched motor skills, psycho-
logical training, and physiological preparation. In assuming that learning
occurs through social practice, learning culture emphasizes the skills involved
in mediating the (often blurred) transition from student to professional,
capturing ways of knowing and doing that are integral to the process of be-
coming a musician but that remain remarkably under-researched. Institu-
tional practices such as criteria for orchestral selection, for example, or social
practices such as peer networking, play a large but often sidelined part in
performers success. Knowledge of these practices has the potential to inform
both student experience and institutional curricula, as conservatoires strive to
offer competitive and world-leading preparation for performance-based ca-
reers.
Second, learning culture answers Jrgensens (2009) call for further re-
search focusing on conservatoires institutional culture. Identified as an area
where research is lacking, refrains such as we need to change the culture are
not uncommon. Yet such statements are notoriously slippery, referring to
practices that are historically located, influenced by wider policy agendas, and
often fraught with divided opinion and inaction. In viewing learning as occur-
ring through cultural practice, learning culture provides a theoretical lens
that grounds cultural discussions in the core task of educating young per-
formers. That is, institutional culture and the learning of performers are con-
sidered as inextricably intertwined and best understood as such. The first step
to transforming culture may well be to understand it; indeed, in establishing
performance science and its associated curriculum development within the
conservatoire landscape, knowledge of the learning culture at play may prove
a valuable tool.
Acknowledgments
I would like to thank Pamela Burnard, University of Cambridge, for her thoughtful,
challenging and supportive supervision of the ongoing PhD on which this paper is
based.
Rosie Burt-Perkins, Centre for Performance Science, Royal College of Music, Prince
Consort Road, London SW7 2BS, UK; Email: rperkins@rcm.ac.uk
References
Barratt E. and Moore H. (2005). Researching group assessment: Jazz in the conserva-
toire. British Journal of Music Education, 22, pp. 299-314.
Bloomer M. (2001). Young lives, learning and transformation: Some theoretical consid-
erations. Oxford Review of Education, 27, pp. 429-449.
Bourdieu P. (1984). Distinction. Cambridge, Massachusetts, USA: Harvard University
Press.
Hodkinson P. and James D. (2003). Transforming learning cultures in further educa-
tion. Journal of Vocational Education and Training, 55, pp. 389-406.
James D., Biesta G., Colley H. et al. (2007). Improving Learning Cultures in Further
Education. London: Routledge.
Janesick V. (2000). The choreography of qualitative research design: Minuets,
improvisations and crystallization. In N. Denzin and Y. Lincoln (eds.), Handbook of
Qualitative Research (pp. 379-400). Thousand Oaks, California, USA: Sage.
Jrgensen H. (2004). Strategies for individual practice. In A. Williamon (ed.), Musical
Excellence (pp. 85-104). Oxford: Oxford University Press.
Jrgensen H. (2009). Research in and for conservatoires: An overview. Paper presented
at The Reflective Conservatoire, Guildhall School of Music and Drama, London.
Kingsbury H. (2001). Music, Talent, and Performance. Philadelphia: Temple University
Press.
Kreutz G., Ginsborg J., and Williamon A. (2009). Health-promoting behaviours in con-
servatoire students. Psychology of Music, 37, pp. 47-60.
Lebler D. (2008). Popular music pedagogy: Peer learning in practice. Music Education
Research, 10, pp. 193-213.
Nettl B. (1995). Heartland Excursions. Urbana, Illinois, USA: University of Illinois
Press.
Musical narrative deconstruction: Ritual and

transgression
Sara Carvalho and Helena Marinho
Department of Communication and Art, University of Aveiro, Portugal
Envisaging musical narrative as a sequence of ritual structures, this re-

search lies on the concepts of ritual, transgression, and narrative as ap-
plied in a case study of a musical theatre piece, Sound Bridges. In
traditional musical narrative, the focus is on construction, and one could
speak of fields around which hierarchies, systems, and rules of musical
language are built. In this piece, we find unpredictable transgressive mu-
sical gestures, acted out by performers, combined with conventional nar-
rative procedures, as successive musical suspensions in the piece
generate pauses in the musical discourse. This research aimed to demon-
strate that the concepts of musical narrative and ritual dimension cannot
be seen as isolated objects but as entities of transformation by compos-
ers, and how the trilogy composer/performer/listener (audience) is asso-
ciated with narrative and ritual. Ritual and transgression can thus be
linked to traditional concepts of musical narrative connecting composing,
performing, and listening activities. The concept of ritual can be suc-
cessfully manipulated by the composer and the performer, particularly in
the context of contemporary music. Transgression of ritual, as planned
by the composer, can act as a deconstructive factor; as mediators, per-
formers take a crucial role in the process.
Keywords: performance; narrative; transgression; ritual; deconstruction
The term narrative has been addressed differently in research fields such as
literary studies, linguistics, aesthetics, and anthropology. In musical research,
different approaches to narrative have been discussed (Almn 2008, Klein
2004, Maus 1988, Tarasti 1979), resulting in several reorientations of the
concept: new consensus is developing about musical narrative that is
aware both of the limitations of musical expression and of the rich potential
of music as a narrative medium (Almn 2008, p. 3).
All these studies shed light on an issue that can be observed through dif-
ferent angles, but common to virtually all approaches to musical narrative is
the recognition of a degree of similarity between musical and literary dis-
course (Almn 2008, p. 11), a similarity that can be extended to theatrical
discourse, since narrative mechanisms native to one mediumfrequently
cross-pollinate with other media, resulting in complex semantic hybrids (p.
38).
In music, the use of theatrical devices can enhance performance. They are
also linked to ritual practices, a powerful idea that has been, and still is, ap-
plied to compositional works. In the context of theatrical and performance
studies, this idea has been developed by authors such as Richard Schechner,
and it remains a key concept in ethnomusicology studies: Rituals are collec-
tive memories encoded into actions. Play gives people a chance to tempo-
rarily experience the taboo, the excessive, and risky. Thus, ritual and play
transform people (Schechner 2007, p. 52). If musical narrative is envisaged
as a sequence of ritual structures, which involve an emotional attachment
from the audience to the musical work, it then becomes one of the effects that
a composer intends to produce in the listener, and the performer, as media-
tor, plays an essential role in this process.
This paper aims to demonstrate that the concepts of musical narrative
and ritual dimension cannot be seen as isolated objects but as entities of
transformation by composers, and how the trilogy composer/performer/
listener (audience) is associated with musical narrative and ritual.
MAIN CONTRIBUTION
This research lies on the concepts of ritual, transgression, and musical narra-
tive, as applied in a case study of a musical theatre piece, Sound Bridges. The
study focuses on how musical gesture takes different meanings in the trilogy
composer/performer/listener and discusses different concepts connecting the
role of embodied ritual in performance, and its effects in the listener.
In traditional musical narrative, the focus is on construction, and one
could speak of fields around which hierarchies, systems, and rules of musical
language are built. In this piece, we find transgressive musical gestures, acted
out by performers, combined with conventional narrative procedures, as the
successive suspensions in the piece generate pauses in the discourse and
challenge conventional hierarchical values. An effective non-verbal communi-
cation by the performers can mediate a narrative deconstruction of the lis-
teners expectations, through the connection between musical gestures on the
one hand, and musical narrative and ritual on the other.
Deconstructing musical narratives
In the words of Livingston (2008, p. 363):
The content of the narrative includes not only a series of represented

events, but actions whereby these events are presented to an implicit au-
dience, as well as the agent(s) responsible for those actions. Narrative
entails narrating which entails a narrator.
Therefore, we could extrapolate that in a piece of music we may speak about

musical narrative, and the composer, the performer, and the listener are the
participating agents: the composer as narrator, the listener as the audience,
and the performer as the mediator. A composer sets up a narrative in an
analogous way to a literary/theatrical work by establishing a close relation-
ship between verbal and musical modes of perception.
Verbal and non-verbal events are often configured into various relation-
ships, establishing a network of values that leads to understanding in the
listener. If this network is broken, the awareness of the narrative processes
can have an impact on the listener, as narrative acts as a potential link to
important aspects of human experience (Almn 2008, p. 41).
Familiarity builds on our common understanding of things. When listen-
ing to music, our imagination constructs narrative contexts and/or discursive
trajectories. Nevertheless, the incomprehension of a musical experience also
creates barriers and splinters the construction of a continuous thought. The
deconstruction of the individual narrative can be achieved through abrupt
interruptions in the musical flow. In this paper, the concept of flow, taken
from Csikszentmihalyis (1988) concept of flow or optimal experience, is ap-
plied to the listeners perspective as an action that can be enjoyable and re-
warding, creating understanding and individual sense of control.
The main objective of the musical theatre piece Sound Bridges is the nar-
rative deconstruction of the listeners thoughts. Thus, we find unpredictable
musical gestures that challenge conventional concert-performance rules, and
cyclically return to traditional settings. The central idea is to create suspense,
disruption, discontinuity, and rupture in the way the musical narrative frag-
ments are perceived, breaking the traditional formal outline of the piece.
Non-verbal gestures and non-verbal communication clearly play an im-
portant role in music performance. Among the functions ascribed to non-ver-
bal behavior, Highlen and Hill (1984, p. 368) point out that behavior is a
primary means of expressing or communicating emotions. In relation to
verbal behavior, non-verbal behaviors can repeat, contradict, complement,
accent, or regulate meaning. Verbal and non-verbal communication helps to

configure musical events into different categories that may organize musical
narrative perception.
Verbal tools
The pieces title and the program notes are verbal communicative tools, which
function as a direct way of transmitting an intention from the composer to the
performer and to the listener, and engage them into a narrative strategy.
Tempo indications, agogic and dynamics, as well as music theatre also func-
tion as means of verbal communication between composer and performer,
and can act indirectly in the listeners perception of narrative.
Non-verbal tools
In non-verbal communication, we find both musical and physical gestures; in

the case of this piece, it is through the music theatre genre that the composer
develops a non-verbal form of communication. The use of several gestures (as
described below) generates abrupt cuts in the musical flow. Composer, per-
former, and listener play several roles in this (de)constructed world. The
composer challenges both performer and listener to accept an unconventional
sequence of events. The efficacy of the piece requires the performers en-
gagement and willingness to adopt alternative behaviors.
Resorting to cuts as a dominant style, suppressing and subverting the tra-
ditional rules of musical writing are ways of intervening in the musical mate-
rial, creating multiple sensations and experiences. The musical narration
becomes discontinuous, compromising the linearity of the listeners musical
thinking. The deconstructed musical narrative prevents the indifferent ac-
ceptance of the listener in an attempt to provoke reflection.
Ritual and transgression
Music theatre performances create ritual-like ways of expression. The trilogy

composer, performer, and listener becomes deeply embedded with narrative
and ritual. Sound Bridges was planned as a pre-ordered set of ritualized mo-
ments: six blocs of music, cyclically interrupted by compositional devices that
break musical flow, disrupting the expected musical syntax. These composi-
tional devices include: mobile ring tone, score pages out of order, players
cough, motionless and repetition, and leaving the stage. In explanation:
Mobile ring tone: In concerts, turning off phones is a common ritual. Nev-
ertheless the audience is willing to forgive the transgression.
Score pages out of order: An uncommon, but possible, mishap. One
friend, a member of the audience, comments: Such bad luck! Maybe they
should restart.
Players cough: At this point some members of the audience realize that
the interruptions may be intentional.
Motionless and repetition: The performers stop, motionless, for 15 to 25 s;
then they repeat the same phrase, over and over again. Most of the audi-
ence understands that the first three interruptions were planned and that
the piece includes an intended transgression.
Leaving the stage: One of the performers continues to play while the other
two leave the stage.
Three different perspectives can be distinguished in rituals involving

Western art music: performance, composition, and listening. It is possible to
apply Schechners (2007, p. 56) four perspectives of rituals and ritualizing
namely structure, function, process, and experienceto characterize the in-
tervening actors (composer, performer, and listener).
Performers have to think about the manner in which they are going to
present the ritual, how to use the given space, and most importantly how to
enact that same ritual. In Sound Bridges, while acting and playing a role,
performers momentarily become someone else, actors of the intentions in-
scribed in the score. Effectiveness is the prime concern, and entertainment is
relegated to a secondary plane.
The ritual process used in composition involves the organization of per-
formance concepts and imagined dynamics, namely meaning, modes of per-
formance, choice of physical space, and performers. Composers also deal with
the aspect of function and experience, projecting how the composition will
impact performer and listener.
Listeners are also participants in the ritual action. van Gennep proposed a
three-phased structure of ritual action: the preliminal, liminal, and postlimi-
nal (cited in Schechner 2007, p. 58). The liminal phase corresponds to a
period of time when a person is betwixt and between social categories or
personal identities (p. 66). As pointed out by Schechner (2007, p. 66), dur-
ing the liminal phase, participants in the ritual become nothing, put into a
state of extreme vulnerability where they are open to change. Turner (1969)
used the term liminoid to distinguish voluntary activities (including the arts
and popular entertainment) from liminal, which refers to rites of passage.
Applying this concept to Sound Bridges, we can assume that listeners lose
their individual voice in order to accept a new musical narrative deconstruc-

tion.
IMPLICATIONS
Music theatre can thus function as a relevant field for research of non-verbal
techniques. Pieces written for the theatre can then be seen as a theatrical ac-
tion that is generated and determined by the music; in Sound Bridges these
theatrical actions, previously called blocs of music and interruptions, are gen-
erated by the music itself and allow for the deconstruction of musical narra-
tive through ritualized forms of transgression. Ritual and transgression can
be linked to traditional concepts of musical narrative connecting composing,
performing, and listening activities. The concept of ritual can be successfully
manipulated by the composer and the performer, particularly in the context
of contemporary music. Transgression of ritual, as planned by the composer,
can act as a deconstructive factor; as mediators, performers take a crucial role
in the process, allowing the creation of multiple sensations and experiences in
the listener.
Sara Carvalho, Department of Communication and Art, University of Aveiro, Campus de

Santiago, Aveiro 3810-193, Portugal; Email: scarvalho@ua.pt
References
Almn B. (2008). A Theory of Musical Narrative. Bloomington, Indiana, USA: Indiana

University Press.
Csikszentmihalyi M. and Csikszentmihalyi I. (1988). Optimal Experience. Cambridge:
Cambridge University Press.
Highlen P. S. and Hill C. E. (1984). Factors affecting client change in individual coun-
selling: Current status and theoretical speculations. In S. D. Brown and R. W. Lent
(eds.), Handbook of Counselling Psychology (pp. 334-396). New York: Wiley.
Klein M. L. (2004). Chopins fourth ballade as musical narrative. Music Theory Spec-
trum, 26, pp. 23-56.
Livingston P. (2008). Narrative. In B. Gaut and D. M. Lopes (eds.), The Routledge
Companion to Aesthetics (2e, pp. 359-369). New York: Routledge.
Maus F. E. (1988). Music as drama. Music Theory Spectrum, 10, pp. 56-73.
Schechner R. (2007). Performance Studies (2e). New York: Routledge.
Tarasti E. (1979). Myth and Music. The Hague, The Netherlands: Mount de Gruyer.
Turner V. (1969). The Ritual Process. Chicago, Illinois, USA: Aldine Publishing.
The piano repertoire preparation: A research

method as a potential tool for reflective
instrumental practice
Regina A. T. dos Santos and Liane Hentschke
Graduate Music Program, Federal University of Rio Grande do Sul, Brazil
This paper presents a qualitative research method that was constructed

to investigate how undergraduate piano students at different stages of
their academic education prepared their repertoire during an academic
semester. Three undergraduate piano studentsa first semester, fifth
semester, and eighth semester studentswere followed during an aca-
demic semester using a phenomenological approach. For each of the
three case studies, three data collection stages took place: the presenta-
tion interview, the observation of the repertoire under preparation, and
finally, the observation of the students recall of his own study process.
Four complementary research techniques were employed: the semi-
structured interview, observation of the performance of the music pieces,
a non-structured interview about the repertoire under preparation, and a
recall stimulated interview, in which the student reflected about his own
performance recorded in video and audio in the previous sections. Dur-
ing the data collection, the methodology gave the undergraduate piano
students the opportunity to develop reflective thinking. Instrumental and
verbal testimonies have shown that the interviews afforded the students
moments to show their thoughts and actions about their piano repertoire
preparation. The students acted as active agents in their piano practice,
demonstrating awareness of their weaknesses and strengths through self-
knowledge.
Keywords: piano practice; musical knowledge; reflective thinking;

qualitative research; repertoire
The literature has discussed progress made in instrumental practice research

toward the understanding of concepts related to practice and the resources
frequently used in instrumental practice contexts (see Williamon 2004).
There is a consensus in the literature that instrumental practice depends on

the nature and context of the task, as well as on the interests and engage-
ments of the student (see Barry and Hallam 2002, Gabrielssohn 2003).
According to Elliot (2005), students must reflect critically on the various
levels of meaning in the musical works that they are interpreting and per-
forming before, during, and after they make and listen to their music. Never-
theless, most of the research in instrumental practice has neglected to
highlight critical and reflexive approaches toward epistemological issues that
students may face in daily musical practice. In formal education, undergradu-
ate instrument students are supposed to prepare a repertoire containing
works from different styles within a given time, and they are to interpret the
score while taking into account stylistic aspects, without losing creativity
within narrow constraints. As Burnard (2006) pointed out, we should develop
reflexive thinking more effectively in our professional practice, taking into
account the potential for the use of this thinking in processes as a source of
professional agency. Reflection involves forms of thinking that include: (1)
continuous evaluation of beliefs, assumptions, and hypotheses (Burnard
2006), (2) nurturing of opportunities for intersubjective interactions, and (3)
interdependent reflexive thinking to illuminate creative thought and action as
a condition of the creative process.
In Brazil, the academic formation of undergraduate instrumental (piano)
students comprises eight semesters. In each semester, the student must pre-
pare repertoire composed of about three to five pieces by composers from
different periods or styles. The present study took into account that under-
graduate students present an epistemic relationship with the systematized
experiences and with the normative patterns within the tradition of the occi-
dental classical music, and this musical knowledge influences their practice
through the preparation of their piano repertoire. The present paper is in-
tended to contribute to instrumental teaching and learning by presenting a
method constructed to investigate musical knowledge, which in turn is shown
to be a potential tool for encouraging students to adopt critical and reflexive
approaches toward their own practice.
METHOD
Participants
Three undergraduate students from the same institution (out of a total of 30

volunteers belonging to three universities in the south of Brazil) were se-
lected: a first semester, a fifth semester, and an eighth semester student.
Materials
The students piano repertoire in preparation during an academic semester

was examined, under a phenomenological perspective.
Procedure
In this research, phenomenology was employed as a philosophical methodo-

logical principle for the construction of the object. According to Bowman
(1998), the phenomenological method focuses on the experience. The phe-
nomenological fundamentals represent an aperture toward the delimitation
of the problem and the tools to set boundaries for data collection and analy-
sis. The research was constructed in a qualitative interpretative approach,
considering the following principles (Schawandt 1994):
human action is considered significant and intentional

the ethical vis--vis respect for the participants, as well as fidelity toward
the students own life experience
the epistemological perspective that emphasizes the contribution of hu-
man subjectivity to knowledge
The collecting procedures involved four research techniques: the semi-

structured interview, observation of the recorded performances (aimed at
revealing some indices of tacit knowledge mobilized by the student that were
not verbalized but could be demonstrated in the performance), a non-struc-
tured interview about the repertoire under preparation, and a recall stimu-
lated interview, in which each student reflected on his own performances
recorded on video and audio in the previous sections.
The observation of the repertoire under preparation was carried out dur-
ing a university semester in which the collecting sessions took place every 2.5
weeks, with a total of four or five meetings. Interviews combining verbal tes-
timonies and instrumental performances could be favorable for understand-
ing how the repertoire was prepared during an academic semester, as well as
for revealing consistent clues of mobilized musical knowledge during the
preparation. These meetings aimed at monitoring and observing the prepara-
tion of the repertoire, reported by the student himself. The observation of the
repertoire was carried out by combining two research techniques: a non-
structured interview about preparation and practice and the observation of
the performance of the musical pieces. In the non-structured interview, the
questions came from information reported by the student, without imposing
determined questions that did not deal with his personal experience at that
moment. This attitude was based on phenomenology as a philosophical

method. In the final part of the research, the stimulated recall interviews were
conducted, focusing on the performances recorded during the previous sec-
tions. This stimulated interview at the end of the semester was used to avoid
influencing the student during the repertoire preparation.
RESULTS
During the interviews, answers to questions regarding musical comprehen-

sion (verbal and instrumental) of the works under preparation revealed that
some facts were tacit in nature. Often, while explaining our thinking (tacit),
we identify and articulate several aspects in order to feel understood. In these
situations, some ideas, perspectives, and thinking that remain latent up to
that moment arise as possibilities for further reflections and decisions. This
situation occurred during the interviews, when the communication and ex-
planation of the students own decisions, doubts, and certitudes to the re-
searcher promoted moments of reflection. This happened, for instance, in the
first interview of the fifth semester student, and he himself confirmed, by the
end of the collecting period, one of our suspicions (NB. according to the fifth
semester student, having to explain aspects of the repertoire preparation
caused him to reflect after each interview):
Reporting to someone prompted me to think more. And by verbalizing,

new possibilities arose, which I continued to think about after the inter-
views. I thought on the things I said, if it was right or notor if I should
explore that idea more.
Along with the data collection, several other occasions demonstrated that
the interviews, since they implied moments of reporting their repertoire
preparation, functioned as a catalyst for reflection upon students strategies,
their problems, and their conquests. Listed below are some excerpts from the
students comments.
The first semester student regretted his approach toward Bachs Fugue in
the first recall stimulated interview. He said:
This Fugue! I prepared the first page for almost four months. In the
exam, it was clear that the first page was good, and the others were infe-
rior to the first one! This would be unacceptable now! I need to read the
whole piece, even though I cannot play it well. But I have to read it com-
pletely, to realize: Here I will need to spend much time!
The fifth semester student mentioned:
I am realizing something now. Something that is becoming visible. We

have much accumulated knowledge. How many pieces we have studied,
how many records we have listened to, read about music, and many think
we have inside the mind, and often, we do not use all this knowledge in
our favor. Because the first time we open a score, we already have an
idea, but we do not chase itor simply, diverge in the direction of other
questions, such as those which are technical, which are secondary.
The cases also provided some other interesting testimonials. The first se-
mester student in the third recall stimulated interview mentioned:
For me, it is always worthy to show for other people what I am doing.
And in this way, recording, it became worthier. Now I can watch. I
think that this self-critic is very important. It helps! Sometimes it is more
efficient than any other thing.
The fifth semester student, during the third recall stimulated interview
commented:
During the preparation, it was interesting, because it forced me to think

more about the works. Now, watching the video, I could have a retro-
spective visionof my process. It was very interesting. I have already
begun to modify something. One aspect that became very strong for
meis this stuff with timing, to which I have to be more devoted.
The eighth semester student, in the second recall stimulated interview

said:
At the beginning, I had a little dread. I thought it might consume much

time and attention. But thereafterI felt wonderfulto be able to watch
myself from far away and from that distance, acquire more clear-
nessregarding the ideas, the videoeven in listening to me.
DISCUSSION
According to these testimonials, the employed methodology provided oppor-

tunities for undergraduate students to develop reflective thinking. The strate-
gies employed during data collection were shown to act potentially as
effective tools for instrumental teaching and learning, since they allow the
students to expose their thoughts and actions regarding preparation in order
to value their own musical experience. During the process of data collection,
piano undergraduate interviewees were active agents in their preparation and
demonstrated awareness about their weaknesses and strengths in music self-
knowledge. Fostering situations, in which the student may be given room to
question and to reflect upon his/her own process and products may be a
valuable strategy in instrumental teaching. Videos of the students perform-
ance may provide opportunity for the piano teacher and student to reflect
upon the instrumental practice and performance.
Regina A. T. dos Santos, Graduate Music Program, Federal University of Rio Grande do
Sul, Rua Prof. Annes Dias 112, Porto Alegre, Rio Grande do Sul 90020-090, Brazil;
Email: jhsreg@adufrgs.ufrgs.br
References
Barry N. H. and Hallam S. P. (2002). Practice. In R. Parncutt and G. E. McPherson

(eds.), The Science and Psychology of Music Performance (pp. 151-165). Oxford:
Bowman W. D. (1998). Philosophical Perspectives on Music. Oxford: Oxford University
Press.
Burnard P. (2006). Rethinking the imperatives for reflective practices in arts education.
In P. Burnard and S. Hennessy (eds.), Reflective Practice in Arts Education (pp. 3-
12). Dordrecht, The Netherlands: Springer.
Elliot D. J. (2005). Praxial Music Education. Oxford: Oxford University Press.
Gabrielssohn A. (2003). Music performance research at the millennium. Psychology of
Music, 31, pp. 221-272.
Schwandt T. A. (1994). Constructivist, interpretivist approaches to human inquiry. In N.
K. Denzin and Y. S. Lincon (eds.), Handbook of Qualitative Research (pp. 118-137).
Thousand Oaks, California, USA: Sage Publications.
The role of the soft palate in woodwind and

brass playing
Alison Evans1, Bronwen Ackermann2, and Tim Driscoll1
1 School of Public Health, University of Sydney, Australia

2 School of Medical Sciences, University of Sydney, Australia
Expired air provides the basis for sound production for musicians playing
a wind instrument. This air stream must be controlled and directed into
the mouthpiece of the instrument via a reed mechanism at the mouth. To
be able to do this, firm velopharyngeal closure is required to prevent air
leaking out through the nose from the oral cavity. In some musicians im-
pairment of soft palate function may result in this air leak occurring,
which is known as velopharyngeal insufficiency (VPI). A review of the
functional anatomy of the soft palate and how it relates to wind and brass
playing is discussed with a brief review of VPI as it is reported in musi-
cians. A better understanding of the soft palate anatomy will assist stu-
dents and music teachers to optimize their performance and prevent
performance related medical problems.
Keywords: soft palate; wind musicians; functional anatomy; velopharyn-

geal closure; velopharyngeal insufficiency
To play a wind instrument requires a sustained mouth pressure and a con-

stant airflow through the instrument. As the air stream reaches the mouth-
piece of the instrument, it causes a variety of mouth pressures, depending on
the amount of resistance applied by the particular instrument. Bouhuys
(1964) found that the normal maximum pressures sustained in playing a wind
instrument can range from 10-126 mmHg, with the highest pressure of 158
mmHg measured in a muted trombone. These pressures are at least 30 times
greater than required for normal speech production (Dibbell et al. 1979).
These high levels of intra-oral pressures can result in a serious medical
condition that can terminate the career of a wind musician, known as stress
velopharyngeal insufficiency (Klotz et al. 2001, Schwab and Schultze-Florey
2004). VPI results in the seal of the soft palate between the oropharynx and
the nasopharynx being incomplete, allowing air to escape from the nose when
playing (Ingrams et al. 2000). Other than in musicians, this condition is most
frequently reported as a consequence of a structural cleft palate deformity,
hence is often termed stress VPI when it occurs as a result of the stresses
imposed by playing a wind instrument. It has been observed that this condi-
tion usually occurs in advanced students pursuing a professional career, due
to an increase in physically demanding practice and performance regimen
which imposes undue stress on the soft palate over a prolonged period of time
(Schwab and Schultze-Florey 2004).
The majority of research into VPI has been by speech pathologists or oto-
laryngologists, and the literature deals primarily with cleft palate or speech
disorders. Of only ten articles investigating VPI in musicians, fourteen cases
were reported (see Table 1).
A dissertation thesis presented a case series of clarinetists with VPI (Gib-
son 1995). The author later published two articles summarizing his research
and outlining recent trends in the treatment of VPI (Gibson 1998, 2008).
There have been three questionnaire studies conducted in London, Germany,
and the USA. The authors reported, respectively, that 7%, 31%, and 34% of
the participating student and professional wind musicians have experienced
VPI (Ingrams et al. 2000, Schwab and Schultze-Florey 2004, Malick et al.
2007). Due to the small sample sizes of these studies, these figures do not
accurately represent the significance of this condition, although informal
discussions with fellow musicians and music teachers suggest that this condi-
tion is a well-known phenomenon.
The purpose of this review is to present the functional anatomy of the soft
palate and how this anatomy potentially relates to wind and brass instrument
performance.
MAIN CONTRIBUTION
When playing a wind instrument, air passes from the lungs up into the upper
respiratory tract and is eventually channeled into the mouthpiece of the in-
strument via some form of reed mechanism. As the air meets this resistance
when flowing into the instrument there is an increase in pressure in the upper
respiratory cavity, requiring firm control of the soft palate.
The soft palate, also known as the velum, separates the nasopharynx from
the oropharynx. This closure of the two parts is known as velopharyngeal or
palatopharyngeal closure and is important for swallowing, speech, and blow-
ing. The soft palate extends posteriorly from the hard palate, and five muscles
control its movement. Each of the paired muscles can be divided into
Table 1. Review of published cases of musicians treated for VPI.
Authors Instrument Sex Age (years)

Weber and Chase (1970) oboe female 23
Massengill and Quinn (1974) bassoon/saxophone male 18
Dibbell et al. (1979) oboe female 20
Dibbell et al. (1979) trumpet male 23
Peterson-Falzone (1985) clarinet female n/a
Gordon et al. (1984) bassoon female 31
Conley et al. (1995) trumpet male 17
Wolff (1995) trumpet male n/a
Wolff (1995) trombone female n/a
Ingrams et al. (2000) trombone male 20
Ingrams et al. (2000) clarinet male 18
Klotz et al. (2001) french horn female 19
Klotz et al. (2001) oboe female 20
McVicar et al. (2001) clarinet male 18
their principal actions on the soft palate. The levator veli palatini and the
musculus uvulae are elevators, the palatoglossus and palatopharyngeus are
depressors, and the tensor veli palatini is a tensor.
The muscles of the soft palate
The tensor veli palatini is the only muscle of the soft palate that is innervated
by the cranial nerve V (Moore and Dalley 2006). The muscle bellies are at-
tached superiorly to the scaphoid fossa of the medial pterygoid plate, the
spine of the sphenoid bone, and the cartilage of the pharyngotympanic tube.
It is then inferiorly attached to the palatine aponeurosis. The main action of
the tensor is to tense the soft palate which opens the pharyngotympanic tube
during swallowing and yawning (Moore and Dalley 2006). In playing a wind
instrument, the tensor acts with the levator to maintain velopharyngeal clo-
sure which allows the airflow through the mouth.
The levator veli palatini is innervated by the pharyngeal branch of the
vagus cranial nerve X and is superiorly attached to the cartilage of the
pharyngotympanic tube and the temporal bone. Like the tensor, the levator is
attached inferiorly to the palatine aponeurosis (Moore and Dalley 2006). The
levator is the most important muscle in the elevation of the soft palate during
swallowing and yawning. As described above, the action of the levator is im-
portant in closing the oronasal cavity ensuring airflow is directed through the
mouth.
The palatoglossus and the palatopharyngeus are innervated by the vagus
cranial nerve X (Moore and Dalley 2006). The palatoglossus is attached supe-
riorly to the palatine aponeurosis and is attached inferiorly to the side of the
tongue. The main action of this muscle is to elevate the posterior part of the
tongue and may also depress the soft palate onto the tongue (Moore and
Dalley 2006). If the palatoglossus depresses the soft palateopening the na-
sal cavity and elevating the tongue, thereby closing the oral cavitythen it
may be these actions that assist in the breathing technique known in musical
pedagogy as circular breathing. This technique, occasionally used by wind
instrumentalists, involves the simultaneous expiration of air through the
mouth and inspiration through the nose, which allows the player to blow
continuously through the instrument. Although circular breathing has been
discussed in the music literature, the action of these muscles needs further
research.
The palatopharyngeus is superiorly attached to the hard palate and the
palatine aponeurosis and is inferiorly attached to the lateral wall of the phar-
ynx. The palatopharyngeus tenses the soft palate and pulls the pharynx walls
superiorly, anteriorly, and medially during swallowing (Moore and Dalley
2006). Huang et al. (1998) found that the inferior action of the palatopharyn-
geus acts against the inferior action of the levator. Additionally, upon con-
traction of these muscles the velum stretches to make contact with the
posterior pharyngeal wall, enhancing velopharyngeal closure (Huang et al.
1998).
The musculus uvulae is also innervated by the vagus cranial nerve X
(Moore and Dalley 2006). The muscle is attached to the posterior nasal spine
and the palatine aponeurosis superiorly and is attached to the mucosa of the
uvula inferiorly. The main action of this muscle is to shorten the uvula and
pull it superiorly (Moore and Dalley 2006). It also assists the levator veli
palatini in palatopharyngeal closure by adding bulk to the soft palate, which
seals off the nasopharynx (Kuehn 1979). When the soft palate is elevated, the
uvula can be seen drawing upwards and backwards towards the back of the
throat.
Two other muscles that belong to the pharynx are also associated with the
soft palate. Both the superior pharyngeal constrictor and the salpingo-
pharyngeus are innervated by the pharyngeal branch of the vagus cranial
nerve X (Moore and Dalley 2006). The superior pharyngeal constrictor,
which is attached superiorly to the pterygoid hamulus, and the pterygoman-
dibular raphe, inserts into the pharyngeal aponeurosis forming the sides and
back walls of the nasopharynx and the back of upper oropharynx. The salpin-
gopharyngeus is attached to the cartilaginous part of the pharyngotympanic
tube and inserts inferiorly to the palatopharyngeus muscle (Moore and Dalley
2006). The superior constrictor, assisted by the salpingopharyngeus which
elevates the pharyngeal wall, pulls the pharyngeal wall forward to help gain
velopharyngeal closure.
IMPLICATIONS
As can be seen from this review, the function of the soft palate is essential for
maintaining upper respiratory tract structure under pressure and hence al-
lowing optimal airflow. It is crucial for wind and brass players to be able to
maintain firm velopharyngeal closure for optimum performance. Much more
research needs to be conducted into the mechanisms behind the performance
related problems, which may be associated with the soft palate. It is impor-
tant for musicians and music teachers to increase their understanding of the
functional anatomy and physiology for optimizing performance and injury
prevention.
Alison Evans, School of Public Health, Edward Ford Building A27, University of Sydney,
Sydney, NSW 2006, Australia; Email: evansa@health.usyd.edu.au
References
Bouhuys A. (1964). Lung volumes and breathing patterns in wind-instrument players.

Journal of Applied Physiology, 19, pp. 967-975.
Conley S. F., Beecher R. B., and Marks S. (1995). Stress velopharyngeal incompetence in
an adolescent trumpet player. Annals of Otology, Rhinology and Laryngology,
104, pp. 715-717.
Dibbell D. G., Ewanowski S., and Carter W. L. (1979). Successful correction of velo-
pharyngeal stress incompetence in musicians playing wind instruments. Plastic and
Reconstructive Surgery, 64, pp. 662-664.
Gibson C. (1995). The Soft Palate Air Leak in Clarinetists: A Multiple Case Study of
Stress Velopharyngeal Insufficiency. Unpublished doctoral thesis, University of
Missouri.
Gibson C. (1998). What on earth is that noise? The palatal air leak in clarinets. The
Clarinet, 25, pp. 30-32.
Gibson C. (2008). Current trends in treating the palatal air leak (stress velopharyngeal
insufficiency). The Clarinet, 35, pp. 42-44.
Gordon N. A., Astrachan D., and Yanagisawa E. (1994). Videoendoscopic diagnosis and
correction of velopharyngeal stress incompetence in a bassoonist. Annals of Otol-
ogy, Rhinology and Laryngology, 103, pp. 595-600.
Huang M. H. S and Lee S. T. (1998) Anatomic basis of cleft palate and velopharyngeal
surgery: Implications from a fresh cadaveric study. Plastic and Reconstructive Sur-
gery, 101, pp. 613-627.
Ingrams D. R., McFerran D. J., and Graham J. M. (2000). Velopharyngeal incompe-
tence in wind players. Physical and Emotional Hazards of a Performing Career:
Special issue of Music Performance, 2, pp. 105-112.
Klotz D. A., Howard J., Hengerer A. S., and Slupchynskj O. (2001). Lipoinjection aug-
mentation of the soft palate for velopharyngeal stress incompetence. Laryngoscope,
111, pp. 2157-2161.
Kuehn D. P. (1979). Velopharyngeal anatomy and physiology. Ear, Nose and Throat
Journal, 58, pp. 316-321.
Malick D., Moon J., and Canady J. (2007). Stress velopharyngeal incompetence: Preva-
lence, treatment, and management practices. Cleft Palate-Craniofacial Journal, 44,
pp. 424-433.
Massengill R. and Quinn G. (1974). Adenoidal atrophy, velopharyngeal incompetence
and sucking exercises: A two year follow-up case. Cleft Palate Journal, 11, pp. 196-
199.
McVicar R., Edmonds J., and Kearns D. (2002). Sphincter pharygoplasty for correction
of stress velopharyngeal insufficiency. Otolaryngology-Head and Neck Surgery,
127, pp. 248-250.
Moore K. L. and Dalley A. F. (2006). Clinically Oriented Anatomy (5e). Baltimore,
Maryland, USA: Lippincott Williams and Wilkins.
Peterson-Falzone S. J. (1985). Velopharyngeal inadequacy in the absence of overt cleft
palate. Journal of Craniofacial Genetics and Developmental Biology, Supplement
1, pp. 97-124.
Schwab B. and Schultze-Florey A. (2004). Velopharyngeal insufficiency in woodwind
and brass players. Medical Problems of Performing Artists, 19, pp. 21-25.
Weber J. and Chase R. A. (1970). Stress velopharyngeal incompetence in an oboe player.
Cleft Palate Journal, 7, pp. 858-861.
Wolff (III) E. A. (1995). Medical corner (stress velopharyngeal insufficiency). Interna-
tional Trombone Association Journal, 23, pp. 14-15.
The Vincenzo Vitale piano school: Famous

school but little known
Viviana Nicoleta Ferrari
Melbourne Graduate School of Education, University of Melbourne, Australia
This paper introduces some aspects of the Vincenzo Vitale piano school;
a twentieth century Italian school generated from the intersection of the
Neapolitan piano school and other European musical realities. Although
the School is recognized to be important by many, there is very limited
specific literature about either the School or Vincenzo Vitale (1908-84).
The thesis entitled The Piano Teaching of Vincenzo Vitale, undertaken
toward the completion of a Bachelors degree and specialization at the
University of Bologna in 2005, is still to my knowledge the only in-depth
scholarly study on the topic so far. The focus of the thesis was on the
principles of Vitales teachings, his theories, the formulation of these
theories, and their background.
Keywords: Vitale piano school; pedagogy; technique
Alongside the German, Russian, and French schools of piano playing during
the 1800s in Naples, the Neapolitan piano school, founded by Francesco
Lanza (1783-1862) and subsequently by Sigismund Thalberg (1812-71), was
flourishing. Well into the first part of the 1900s, Naples with its conservatory
was the center of piano playing in Italy. Pianists such as Michelangeli, Cic-
colini, Pollini, and most of the Italian pianists have their roots in this school.
Furthermore, the Neapolitan schools prerogatives of piano playing have been
exported outside of Italyfor example, Beniamino Cesi (1845-1907) taught
piano in St. Petersburg, Vincenzo Scaramuzza (1885-1968) in Buenos Aires,
and Vincenzo Vitale (1908-84) in the US. During the second half of the twen-
tieth century, the Vincenzo Vitale piano school was considered the apex of the
Neapolitan piano school and Vitale one of the most important Italian piano
teachers (e.g. Campanella 1994, Di Benedetto 2004, Valori 1981). Although
Vitale died in 1984, his name appears in many curricula vitae, some of whom,
in many cases, were just auditors to Vitales lectures; it would seem that his
name stands as a sort of brand of good piano playing and professionalism.

Because Vitale himself did not write his method of teaching in a volume, the
important legacy risks to be misinterpreted or even become extinct. Since
2002, I have addressed this problem by completing a 64,000 word thesis
entitled The Teachings of Vincenzo Vitale (2005) at the University of Bolo-
gna, and I am continuing the research into the application of Vitales teaching
principles toward a PhD at the University of Melbourne. The focus of the the-
sis completed in Bologna in 2005 was on the principles of Vitales teachings,
his theories, the formulation of these theories, and their background. Sche-
matically, now we know that the Vitale school was generated from Vitales
teaching dating back into the 1930s. The school is informed by two main
principles: (1) technique and interpretation cannot be separated and (2) pi-
ano playing is manifested though sound production. Furthermore, sound
production is at all times resulting from a combination of two fundamental
techniquesweight technique and percussive techniquethat originate from
the only two antithetic actions of the finger on the keyboard (sustainment of
weight or percussion action). We also know that the school is based on a spe-
cific system of training and Vitales teachings are propagated today through
the teaching of Vitales disciples and their students. This paper aims to report
on the results of the Bologna thesis as well as on the aims of the current re-
search into the evolution of the Vitalian system of teaching.
MAIN CONTRIBUTION
The Vitale piano school is predominantly an Italian phenomenon that is

centered on the Neapolitan figure of Vincenzo Vitale (Naples, 1908-84).
Vitales piano teaching took place mainly in conservatories, academies, and
summer schools, in Italy, in many European cities, as well as in Buenos Aires,
Bloomington (Indiana, USA), and Cairo. Students from many parts of the
world have come into contact directly or indirectly with Vitales teachings.
Vitale was also active as a historian, writer, journalist, and promoter in the
field of musicactivities that significantly enriched his teaching. Vitale devel-
oped by intuition, experimentation, and rational process a pedagogical sys-
tem of training intended as a tool in piano teaching, in its technical/inter-
pretative aspects, from preliminary to the highest level. His pedagogy can be
intended as an amalgamation of elements derived from the finger school
(Clementi/Lanza and Thalberg), through the Neapolitan piano school, as well
as the weight school (Liszt, Steinhausen, Deppe, Breithaupt, Matthay) fil-
tered through the Neapolitan Attilio Brugnolis treaties on the dynamics of
piano playing published in 1926 by Ricordi: Dinamica Pianistica, Trattato
sullInsegnamento Razionale del Pianoforte e sulla Motilit Muscolare ne

suoi Aspetti Psico-Fisiologici.
The Vitale piano school does not refer to a finite institute because such an
institution never existed. It came about spontaneously in the Italian pianistic
entourage (pedagogues, students, journalists, music critics) as a way to iden-
tify a specific and prolific phenomenon. Its specificities were denoted by the
technical drill that was part of the Vitalian training, the particular typology of
a fundamental sound (intense, full, round, direct, very clear), and a signifi-
cant number of skilled and/or successful professional musicians (pianists,
conductors, musicologists, pedagogues, journalists) who have referred to his
teaching, such as Michele Campanella, Bruno Canino, Carlo Bruno, Riccardo
Muti, Renato Di Benedetto, and Paolo Isotta.
Vitale seemed to have embraced and officialized the entity of his school
of teaching by releasing, in collaboration with Fonitcetra, three sets of re-
cordings between 1974 and 1981 with annexed booklets: La Scuola Pianistica
di Vincenzo Vitale (1974), La Scuola Pianistica di Vincenzo Vitale [2] (1980),
and Muzio Clementi, Gradus ad Parnassum: Incisione Integrale dei 100
Studi (1981). In the opinion of the critics, the recordings appeared as a show-
case of both the sound production and the principle traits of the school. If the
1974 booklet was an introduction to the school, the 1981 booklet was intended
as lessons on how to approach pianistic interpretative/technical issues. Today
Vitales concise comments can be of little practical use in piano pedagogy in
general without a more expanded knowledge of Vitales teaching. This seems
to confirm that Vitale believed that knowledge can be transmitted between
teacher and student only directly and that any written suggestions can be
misunderstood or non-productive due to the incapacity to adjust the expla-
nation to the real needs of the student. The three sets of recordings with the
annexed booklet are the only historical reference points authorized by Vitale
himself, so therein lays their historic importance as a basis for further study
in Vitales teachings.
Rattalino (1981), reviewing the Clementi album, placed Vitales teaching
as a successful modern realization of the concept of the virtuoso pianistic
sonority. The term pianistic sonority does not refer to the timbre by itself
but rather the timbre as an integral part of the expressive lexicon of pianis-
tic composition. More precisely, this lexicon is formed by the specific pianistic
use of an extensive use of scales, arpeggios, double thirds, double sixths, oc-
taves, and pianistic modulus that seem to originate from the interaction of the
pianists physiology of the arm complex and the specific sound qualities gen-
erated by the mechanism of the instrument (Rattalino 1983). Furthermore,
Rattalino suggests that the piano compositions of Clementi, Liszt, and Ravel
provided Vitale with an axis that exemplified this concept of pianistic lexicon,
and that axis stands as a referential typology of pianistic sonority. Interest-
ingly enough, Vitale sourced the fundamental ingredients of the specific ty-
pology of sound, the cantabile and the brillante sound, from the brief
introduction in Thalbergs LArt du chant appliqu au piano volume pub-
lished in 1850 by Girard in Naples. As Rattalino (1983) observed, Thalberg
modeled these two sonorities on the singing voice (cantabile) of the bel canto,
specifically on the distinction between cantar di petto and cantar di grazia.
Schematically, Vitales interpretation toward the realization of the two
fundamental ideas of sound on the instrument concerned the amount of
weight of the arm complex (shoulder, forearm, and hand) released on the
keypad. The cantabile sound requires a major amount of weight to be released
on the keypad and the brillante sound requires a major amount of suspension
of the weight to be released on the keypad. The following was deduced in this
line of thought: (1) the more weight released on the keypad, the more the
finger must sustain the weight which limits its percussiveness; and (2) the
more weight that is suspended (in other words, the weight of the arm, fore-
arm, and even hand is sustained by the appropriate muscles that have this
function), the more the finger has the right muscular/articular condition to
solve its percussive action on the keypad. These observations brought Vitale
to assert that sound production on the piano can be categorized from the
point of view of touch mechanisms in only two fundamental techniques: the
weight technique (tecnica di peso) and the percussive technique (tecnica per-
cussiva).
The two different technique typologies were inferred from what Vitale be-
lieved the only two fundamentally different possibilities of physiological ac-
tion of the finger on the keypad. The finger can either sustain, as a point of
support, the weight of the arm (necessary in delivering the expressive effects
in legato, cantabile, and in polyphonic passages), or it can execute a percus-
sive action (what is commonly known as articulation) necessary in the per-
formance of fast passages. In Vitales theory, the two technical categories
(weight and percussive) serve as a referential point of piano playing. In real-
ity, Vitale spoke openly of a compromise of the two mechanisms, as piano
performance (lesecuzione pianistica) necessitates a continuous technical
constructthe hybrid technique.
Vitales necessity to find practical solutions in piano playing made him re-
alize that there is no separation between interpretation and technique of exe-
cution, as one informs the other constantly. In fact, Vitale in his notes
(undated notes transcribed by the author) writes that the Greek terms techne,
which stands for art, and technicos, which means serves art, clarify the
inseparable dichotomy: imagination and concretization. The first generated

from Sehensucht, the other from the means of expression (Vitale, Document 9
in Ferrari 2005). Vitale intended technique as the functionality of interpreta-
tive means: it can be velocity, precision, or timbre. Technique represents the
means of appropriate sound production in order to deliver the musical piece.
The executors expressive capacity is manifested through the pianistic execu-
tion by achieving varied sounds in intensity, giving each note its appropriate
accent with the appropriate touch dictated by the pianists interpretation of
the composers indications. The more the technical means are controlled and
precise, the more the interpretation can be exteriorized with precision of in-
tent. A virtuoso pianist is one who achieves control over all the means in-
volved.
Thus, Vitale insisted that the first thing a student must know is the use of
the physical element that is in direct contact with the keys: the fingers. The
third phalanx of the fingers constituted the focal point of attention: its action
is the result of complex muscular processes that involve the entire organism.
For this reason one must know the physiological mechanisms that preclude
the motor activity. According to Vitale, without a proper understanding of
fundamental physiology at least of the upper limb, there is a greater risk of
imprinting an erroneous foundation for technique and consequent interpre-
tation. Vitales aims were to give the student sharp, clear guidelines in how to
resolve interpretational problems without limiting the student by imposing
ones own interpretational views. The practicality of theories interested him,
not their abstract value. This is reflected in his considerations about the use of
concepts and terminology such as relaxation, weight, and muscular dissocia-
tion. Vitale stated that any description of the act of piano playing from the
point of view of muscular activity can be only approximate and any definition
on weight technique, arm technique, dissociation, relaxed arm, cannot be
considered differently.
The Vitale technical drill is probably the most famous element of the
school. It consists of a series of concise exercises set in a progression that
respects the assimilation of physiological difficulties. These range from the
single note mechanism (four notes and five notes and chromatic exercise,
scales, arpeggios, trills, and repeated notes) to the double notes mechanism
(double thirds, sixths, octaves) as well as the combination of the weight and
percussive mechanism (one, two, or three fingers sustain the weight over long
notes while the other fingers act percussively on notes of a shorter value). All
exercises are to be executed in all tonalities, with different graduations of
volume (from forte to piano, as well as in crescendo and diminuendo), at
different tempi, and with different accents determined by a fixed set of
rhythmic variations (groups of two, three, or four notes). Vitale did not com-
pose the exercises; these were sourced from traditional piano pedagogy and
were assembled by him within a system of rational criteria of determined
aims.
IMPLICATIONS
The underlying problem is: what is, or can be, the use of the Vitale system of
teaching in the larger panorama of piano pedagogy? Possible solutions stand
in presenting the findings to a larger audience, and thus encourage scholarly
attention and debate to a substantial pianistic phenomenon.
Acknowledgments
I am grateful to the following professors who have given me, and are giving me, the
skills in conducting the research on the Vitale School: Renato Di Benedetto, Maurizio
Giani, Neryl Jeanneret, and Rosalind Hurworth.
Viviana Nicoleta Ferrari, Melbourne Graduate School of Education, University of Mel-

bourne, Victoria 3010, Australia; Email: ferrariv@unimelb.edu.au
References
Campanella M. (1994). Un rapporto solare con la tastiera. Autunno Musicale al San

Carlo, Settembre 94 Gennaio 95 (23 January, brochure, pp. 41-51). Naples, Italy:
San Carlo Theatre.
Di Benedetto R. (2004). Difficilum docere facile difficilia (introduction). In V. Vitale,
Pianoforte: Martelletti e Smorzatori (2e, pp. XVI-XXII). Naples, Italy: Conserva-
tory S.Pietro a Majella.
Ferrari V. (2005). Linsegnamento pianistico di Vincenzo Vitale. Unpublished bachelors
thesis, University of Bologna.
Rattalino P. (1981). Una Provocazione Culturale, Musica Viva, 12, pp. 28-31.
Valori A. (1981). Incontro col Vincenzo Vitale. Prospecttive Musicali, 9/10, pp. 5-8/11-
13.
Expertise in cryptic crossword performance:

An exploratory survey
Kathryn Friedlander1 and Philip Fine2
1 Independent researcher, UK
This paper investigates the link between practice and level of expertise in
cryptic crossword solvers. A survey was conducted of 241 expert solvers,
which established that, even among crossword experts, the range and
frequency of cryptics solved and the time spent upon this varied greatly.
The link between practice and level of expertise was therefore not
straightforward, but seemed to be connected to the primary focus of the
solverwhether they aimed to speed-solve or solve advanced cryptics
of exceptional difficulty. A brief review of motivational drivers for solvers
of cryptics is also included.
Keywords: crossword; problem solving; expertise; motivation; practice
Puzzles of all standards and for all ages are routinely available in newspapers
and magazines, and crazes for particular types of brainteaserssuch as Su-
dokuregularly sweep the world. One particularly long-lived puzzle type is
the crossword, which exists in two basic forms: cryptic and non-cryptic. In the
non-cryptic form, a clue word or phrase prompts the solver to discover a
synonym word which will fit into the grid and intersect with cross-checking
letters. This type of crossword is found throughout the world and is the main
(virtually the only) type of crossword to be found in the USA.
The focus of this study, however, is the cryptic crossword, which may be
found in broadsheet newspapers and quality magazines throughout the UK
and Commonwealth. In these puzzles, the surface reading of the clue is
phrased to be deliberately misleading. The solver learns to ignore this read-
ing, and to identify and crack a grammatical set of coded instructions which,
if executed precisely, will lead to the correct (and only) answer.
Although cryptic crosswords are not audience-oriented, many solvers
strive for self-improvement and recognition within the crossword commu-
nity, by tackling successively harder crosswords, attempting personal best

times, entering public speed-solving challenges, or maintaining annual sta-
tistics of correctly completed crosswords. There is thus ample opportunity to
develop and demonstrate ones expertise in cryptic crossword solving.
In contrast to the many expert-novice studies of performers in domains
such as chess (e.g. Gobet and Campitelli 2007) and music (e.g. Waters and
Underwood 1998), very little research exists into either the processes and
aptitudes required to complete cryptics or the development of cryptic cross-
word solving expertise. A few small-scale studies at Nottingham University
explored the relationship between lexical abilities and cryptic crossword
solving skill and touched upon the role of practice and expertise development
in passing (Underwood et al. 1988, Underwood et al. 1994, Deihim-Aazami
1999); otherwise research has been primarily directed toward the mechanics
of individual clue elements such as synonym retrieval or anagram solving
(e.g. Novick 2004), and has generally been based upon US-style non-cryptic
crosswords (e.g. Hambrick et al. 1999).
This exploratory survey of regular cryptic crossword solvers with a high
level of expertise aimed to investigate the cognitive skills, motivation, and
development of expertise in this domain. The survey was wide ranging, and
only those results relevant to performance expertise are reported here.
METHOD
Participants
Crossword experts were sought to participate, defined as those able to

achieve one of the following:
Solve one standard block-style (i.e. latticed-grid) cryptic crossword, at

the more difficult end of the spectrum, in 30 minutes or less. Cryptic
crosswords of varying difficulty are found in many daily newspapers (see
Biddlecombe 2000, link to Puzzle Sources); of these the most famous is
probably the (London) Times crossword. Competitors in the annual
Times Speed-solving Championship are set the challenge of solving four
Times crosswords in 1 hour (i.e. 15 mins each).
Tackle advanced cryptics. These challenging barred-grid prize puzzles
appear in a number of weekend newspapers and magazines (Biddlecombe
2000); of these, the Listener Crossword (in the Times on Saturdays) is
the most notoriously difficult, employing a high degree of concealment in
the clue mechanics, obscure vocabulary, grids of startling originality, and
a thematic challenge, involving a number of tricky lateral thinking steps
on the basis of very little guidance. Speed is not important: competition

entrants are allowed at least a week to submit solutions. Only a handful of
Listener solvers submit all 52 puzzles in a year correctly; those who solve
42+ appear on a roll of honor.
Participants for the survey were sought by means of advertisements cir-

culated (1) on The Crossword Centre website (www.crossword.org.uk, a UK-
based cryptic crossword forum, membership approx. 950) in May 2007 and
May 2008; (2) on the Guardian newspaper crossword message board in June
2007; (3) at the Times speed-solving competition in Cheltenham, England, in
October 2007; and (4) with the annual Listener Crossword statistics, distrib-
uted (on request) in March 2007 by the official statistician.
The advertisements were deliberately released in both competitive (Chel-
tenham/Listener statistics) and non-competitive (website/message boards)
settings. Care was also taken to advertize the survey both on and off the web.
Materials
A questionnaire was developed, piloted, and released seeking responses to 90

wide-ranging questions relating to the cognitive skills, motivation, and devel-
opment of expertise involved in the completion of cryptics.
Procedure
Respondents were invited to complete the survey online using SurveyMon-

key. A paper copy of the questionnaire was also available.
RESULTS
In total, 256 people responded. Due to missing data, 241 questionnaires were
analysed (209 M, 32 F; mean age=53 years, range=23-83). The mean starting
age for solving cryptics was 17 years; solvers typically spent 10 years on
block-style cryptic crosswords before going on to more advanced ones and
were highly experienced (mean years solving=36, range=2-66). There were
221 Listener Crossword solvers, although the majority of respondents had
first seen the survey advertized on the Crossword Centre website (n=161).
Respondents tended to solve cryptics regularly (mean time spent=8 hours
per week, range=1-30) and from various newspapers (mean different cryptics
regularly attempted=4.5; range=0-28). There were, however, significant dif-
ferences within these ranges at the individual level.
Table 1. Percentage of participants, by expert sub-category, showing volume of cross-

words solved.
Expert sub-category Focused Moderate Broad Extensive Voracious

Crossword Pro 34 57 6 3 0
Speed/Listener Super Expert 28 39 17 6 11
Speed Super Expert 38 31 25 6 0
Listener Super Expert 70 25 5 0 0
Ordinary Expert Participant 61 32 5 2 0
Table 2. Percentage of participants, by expert sub-category, showing hours spent solving

per week.
Expert sub-category 0-5 6-10 11-15 16-20 21-25 25+

Crossword Pro 33 52 6 6 3 0
Speed/Listener Super Expert 22 56 11 11 0 0
Speed Super Expert 38 44 19 0 0 0
Listener Super Expert 27 50 15 7 2 0
Ordinary Expert Participant 49 31 14 5 0 1
Table 3. Mean number of years spent solving cryptics, by expert sub-category.
Pro Speed/List Speed List Ord Average

Years solving 38 36 40 36 35 36
In order to investigate these differences, respondents were divided into

five expert subcategories, on the basis of responses given in the survey:
Those who edited or composed cryptics professionally (n=35);

Those who could speed-solve a cryptic in <15 mins and who had com-
pleted 42+ Listener crosswords correctly in one year (n=18);
Those who could speed-solve a cryptic in <15 mins (n=16);
Those who had completed 42+ Listeners correctly in one year (n=60);
Other expert participants who had not achieved any of the above (n=112).
These groups were first analyzed by the volume of crosswords they tack-
led (number of different crosswords solved by each participant, weighted by
the frequency with which they tackled them; see Table 1 for results). The
groups were then analyzed by the hours spent tackling cryptic crosswords
each week (Table 2). The average number of years spent solving cryptic
crosswords was also calculated for each sub-category (Table 3).
Respondents were also asked to consider and rate, on a 1-5 scale, a series
of 26 statements compiled specifically for the survey exploring a range of
potential motivational drivers (both extrinsic and intrinsic) behind their
solving behavior. Respondents replies indicated high levels of intrinsic moti-
vation. Primary intrinsic drives were related to (1) the buzz of succeeding in
an intellectual challenge (68% rated this very/highly motivating) and (2)
preserving mental sharpness (61% very/highly motivating). Expert cross-
word solving was seen as a solitary hobby: 88% of respondents claimed that
participation in team solving was not motivating for them. Generally, extrin-
sic motivational drivers such as prizes or public acclaim were also seen as less
important; however, 33% had participated in public speed-solving contests,
and 73% of those who tackled the Listener crossword (n=221) routinely sub-
mitted it for the official personal annual statistics, with 54% aiming to appear
on the Listener roll of honor.
DISCUSSION
Becoming an expert in any domain requires practice, hard work, and strong
motivation, both intrinsic and extrinsic. However, this study highlighted that
even among crossword experts the range and frequency of cryptics solved
varied greatly. Super-Experts in speed solving (alone or in combination
with high Listener expertise) seemed to attempt routinely a much broader
range of crosswords than others. It is possible that the high volume of clues
covered enables more of the problem solving process to become automatic for
these speed experts, leading to an increasingly instantaneous recognition of
the clue architecture. By contrast, those who were solely Listener Super-Ex-
perts solved in a conspicuously more focused range. Despite this, the time
spent on crosswords per week was relatively constant across the Super-Ex-
pert categories, with 77-82% of participants spending 0-10 hours on cross-
word solving, predominantly in the 6-10 hours category: the rapidity of the
Speed experts may have contributed to their ability to cover a wider range
of material in relatively quicker time; whereas the Listener Super-Experts
may have deliberately focused on harder and more time-consuming puzzles.
Those falling into the Ordinary Expert category, however, generally
lagged behind the Super-Expert categories both in the range of crosswords
tackled and in the time spent, with 49% of respondents spending 5 or fewer
hours per week on solving activities. This suggests that practice may be one
key factor in achieving Super-Expert status.
Expert solvers were highly intrinsically motivated, and pursued cryptic
crossword solving primarily as a source of intellectual stimulation and pleas-
ure. Solving is a solitary pastime, however, and many experts seemed to use
various extrinsic goals, such as the maintenance of annual statistics, as a
source of extra extrinsic incentive to supplement their own internal drive.
Acknowledgments
Many thanks to John Green, Peter Biddlecombe, and Derek Harrison for helping us to
advertize the survey with the annual Listener statistics, at the Times speed-solving com-
petition and on The Crossword Centres message board.

References
Biddlecombe P. (2000). Yet Another Guide to Cryptic Crosswords (YAGCC), accessed

at www.biddlecombe.demon.co.uk/yagcc/index.html.
Deihim-Aazami C. (1999). Cognitive Expertise in Solving Crossword Puzzles. Unpub-
lished doctoral thesis, University of Nottingham.
Gobet F. and Campitelli G. (2007). The role of domain-specific practice, handedness,
and starting age in chess. Developmental Psychology, 43, pp. 159-72.
Hambrick D. Z., Salthouse T. A., and Meinz E. J. (1999). Predictors of crossword puzzle
proficiency and moderators of age-cognition relations. Journal of Experimental
Psychology, 128, pp. 131-164.
Novick L. R. (2004). Research on Expertise in Anagram Solution, accessed at
www.vanderbilt.edu/peabody/novick/anagrams.html.
Underwood G., Deihim C., and Batt V. (1994). Expert performance in solving word
puzzles. Applied Cognitive Psychology, 8, pp. 531-548.
Underwood G., MacKeith J., and Everatt J. (1988). Individual differences in reading
skill and lexical memory. In M. M. Gruneberg, P. E. Morris, and R. N. Sykes (eds.),
Practical Aspects of Memory (pp. 301-308). Chichester, UK: Wiley.
Waters A.J. and Underwood G. (1998). Eye movements in a simple music reading task.
Psychology of Music, 26, pp. 46-60.
Relationship between playing strategy and

surface electromyograms in playing drums
Takuya Fujisawa1, Naoki Iwami1, Masafumi Kinou2, and Masanobu Miura2

Skills in controlling drumsticks correctly are required to play rhythms

without making any mistakes at suitably dynamic levels of sound. Results
obtained by analyzing drummers movements using visual information,
such as motion capture or camera recordings, have been reported in past
studies, but analysis using biological information has been the focus of
fewer studies. To play rhythms without making mistakes at suitably dy-
namic levels of sound throughout a piece of music, trained drummers are
assumed to use different playing strategies: a short-term playing strategy
(e.g. every stroke) and a long-term playing strategy (e.g. every piece of
music). The aim of this study was to investigate the relationship between
trained drummers playing strategies and muscle movements by re-
cording their surface electromyograms (EMGs) when playing the drums.
Three trained drummers and three non-drummers participated in an ex-
periment to record their surface EMGs when playing single-strokes. As a
result, it was confirmed that trained drummers play the drums under
both short-term and long-term playing strategies.
Keywords: electromyogram; drums; strain and laxity; movement analy-

sis; playing strategy
In drumming, the skill of controlling drumsticks correctly is required to play

rhythms without making any mistakes at suitable dynamic levels of sound.
Results of analysis concerning drummers movements from visual informa-
tion, such as motion capture or camera recordings, have been reported in past
studies (e.g. Dahl 2006), but analysis using biological information has been
studied much less. Movements of arms and hands when playing the drums
are understood to be complex, so the actual way muscles control these move-
ments has not been clarified. To play rhythms without making mistakes at
suitable dynamic levels of sound throughout a piece of music, a drummer

needs to consider how to perform. Such consideration is called a playing
strategy. Trained drummers are thought to have different playing strategies:
a short-term playing strategy (e.g. every stroke) and a long-term playing
strategy (e.g. every excerpt). In addition, a playing strategy is based on a
drummers skill and experience, so it is implemented with or without a
drummer being conscious of it. The aim of this study was to investigate the
relationship between trained drummers playing strategies and muscle move-
ments by recording their surface electromyograms (EMGs) when playing the
drums. The amount of EMG signals is proportional to the amount of muscle
strain. Therefore, by analyzing the EMGs of arms and hands when playing the
drums, it is possible to investigate the way arms and hands control drum-
sticks, in terms of muscle laxity and strain. In addition, by recording the
EMGs, information can be obtained about drummers conscious and uncon-
scious movements, so analyzing the EMGs is thought to be suitable for inves-
tigating drummers playing strategies.
METHOD
Participants
Three trained drummers and three non-drummers participated in an experi-

ment to record their surface EMGs when playing single-strokes under three
different tempi of 80, 100, and 120 bpm. All the participants were right-
handed.
Procedure
They were asked to play four beats for three minutes under the tempi denoted
above. The musical score played in this experiment is shown in Figure 1. All
the participants were allowed to rest for a minimum of one minute after each
performance. The flexor carpi ulnaris muscle of both arms, extensor carpi
ulnaris muscle of both arms, metronome signal, and oscillation of drumhead
were all measured. The movement of flexor carpi ulnaris muscle corresponds
to the movement of bending the wrist forward, and the movement of extensor
carpi ulnaris muscle corresponds to the movement of bending the wrist back-
ward.
Figure 1. The musical score.

Figure 2. Example of one-stroke mean EMG.
RESULTS
To investigate the relationship between playing strategies and muscle move-

ments, the short-term and long-term playing strategies were individually
analyzed.
Short-term playing strategy
To analyze the short-term playing strategy, mean EMG of the range of around
500 ms centered at the time a drumstick strikes a drum is called one-stroke
mean EMG and is compared among each measured muscle and each drum-
mer (see Figure 2). The results of the analysis of one-stroke mean EMG are
shown in Tables 1 and 2. Table 1 shows the results of a comparison between
the one-stroke mean EMGs of the flexor carpi ulnaris muscles and those of
extensor carpi ulnaris muscles for all participants. Table 2 shows the styles of
arm movement in terms of muscle for each participant. Table 2 confirms that
significant differences in the one-stroke mean EMG of all participants exten-
sor carpi ulnaris muscles are less/more than those of flexor carpi ulnaris
muscles. Flexor dominance is when the one-stroke mean EMG of the
extensor carpi ulnaris muscle is less than that of the flexor carpi ulnaris mus-
cle. Extensor dominance meanwhile is when the one-stroke mean EMG of
the flexor carpi ulnaris muscle is less than that of the extensor carpi ulnaris
muscle.
Table 1. Results of comparison between one-stroke mean EMG of flexor carpi ulnaris
muscle and one-stroke mean EMG of extensor carpi ulnaris muscle.
Player Tempo (bpm) Left arm Right arm

D1 80 flexor>extensor** flexor>extensor**
100 flexor>extensor** flexor>extensor**
120 flexor>extensor** flexor>extensor**
D2 80 flexor>extensor** flexor>extensor**
100 flexor>extensor* flexor>extensor**
120 - flexor>extensor**
D3 80 flexor<extensor** flexor<extensor**
100 flexor<extensor** flexor<extensor**
ND1 80 flexor<extensor** flexor>extensor**
120 flexor<extensor** -
ND2 80 flexor<extensor** flexor<extensor**
120 flexor<extensor** -
ND3 80 flexor<extensor** flexor<extensor**
Note. **=p<0.01, *=p<0.05, - =not significant.
DISCUSSION
Short-term playing strategy
Tables 1 and 2 confirm that four players have the same style of motor control
in both arms. Specifically, all trained drummers have this feature, whereas
just one non-drummer has it, indicating that trained drummers use the same
style of motor control in both arms when playing the drums. Conforming
right arm movements to left arm movements is thought to be essential to play
the drums without making mistakes throughout a piece of music. In short,
trained drummers play drums with the same style of motor control in both
arms under the short-term playing strategy.
Long-term playing strategy
Table 3 confirms that non-drummers mean EMGs during the last one minute
were significantly higher than those of the first one minute. Such results were
Table 2. Style of arm movement (flexor dominance or extensor dominance).
Player Left arm Right arm

D1 flexor dominance extensor dominance
D2 flexor dominance extensor dominance
D3 extensor dominance extensor dominance
ND1 extensor dominance -
ND2 flexor dominance extensor dominance
ND3 extensor dominance extensor dominance
Table 3. Results of comparison between mean EMGs during first and last one minute.
Player Tempo (bpm) Flexor of Extensor of Flexor of Extensor of

left arm left arm right arm right arm
D1 80 ** ** ** -
100 ** - - -
120 - * * *
D2 80 - * - **
100 - * - **
120 ** ** ** **
D3 80 - - - -
100 - ** - -
120 - - ** -
ND1 80 * ** * **
100 - ** - *
120 ** * * **
ND2 80 * ** - **
100 - ** - **
120 ** ** * *
ND3 80 * ** - *
100 ** - ** -
120 - - ** *
Note. **=p<0.01, *=p<0.05, - =not significant. Ratio of significant differences for D1,
D2, and D3=17/36 (42%). Ratio of significant differences for ND1, ND2, and ND3=
26/36 (72%).
obtained only for the non-drummers. This demonstrates that only non-drum-
mers mean EMGs during the last one minute were significantly higher than
those of the first one minute, indicating that trained drummers are suffi-
ciently relaxed when playing the drums throughout a piece of music. That is
to say, trained drummers play the drums with less strain in terms of muscle
control under the long-term playing strategy.
Acknowledgments
This study was partly supported by the High-Tech Research Center Project for Private
Universities matching fund subsidy from the Japanese Ministry of Education, Culture,
Sports, Science, and Technology (MEXT) and Research for Promoting Technological
Seeds from the Japan Science and Technology Agency (JST).
Takuya FUJISAWA, Graduate School of Science and Technology, Ryukoku University,

1-5 Yokotani, Seta Oe-cho, Otsu Shiga 520-2194, Japan; Email: t08m076@mail.
ryukoku.ac.jp
References
Dahl S. (2006). Movements and analysis of drumming. In E. Altenmller, J. Kesselring,

and M. Wiesendanger (eds.), Music, Motor Control, and the Brain (pp. 125-138).
Oxford: Oxford University Press.
Between hedonism and atomism:

Discrepancies in the performers musical
perception
Veronica Gaspar
Musical Performance Faculty, National University of Music Bucharest, Romania
The performers double hypostasisreceiver and music maker, not just

transmitterentails different approaches, sometimes acting divergently,
as well at the global level as at the sub-compounds of the musical text.
Music is perceived mainly through its supra-segmental level, while the
performance routine calls for a syntagmatic approach at a strict phone-
mic-morphemic level. Over the technical relationship between sounds,
the Western music performer has to build the emotional expression
through logic unities equivalent to words of a discourse undergoing sev-
eral temporal and spatial hypostases on the way from the composer to
the last participant of the musical process, the listener. Numerous trials
aiming at an objective analysis comprise, beside traditional musical
structures, also ad hoc meaning units extracted from imagery, memory
cues, or gesture. However, the internal holistic representation of the
Western performer is contradicted by both the concrete steps of his
practical action and the detailed analytical features. This research tries to
identify those musical ideas susceptible to play the role of intermediate
links between perceptual meaning units and concrete gesture. An at-
tempt aiming at bringing nearer analysis and living performance not only
unifies the segmental elements, but ensures a necessary space for their
psychological resonance.
Keywords: performers; analysts; meaning units; segmental; supra-seg-

mental
Veronica Gaspar, 5-7 Biserica Amzei, Bucharest 010391, Romania; Email: gasparveron
ica@yahoo.com
Effect of music in a healing room for recovery

from mental fatigue: A psychological
experiment of the relation between relaxing
music and listening space design
Yasuhiro Goto
Faculty of Psychology and Applied Communication, School of Humanities,

Hokusei Gakuen University, Japan
Two experiments were performed in order to examine an interaction of

relaxing music and listening space design. In Experiment 1, two types of
music were prepared, and a change of room evaluation was investigated.
The background music (BGM) used was either relaxing or non-relaxing.
Participants were asked to rate the degree of harmony between music
and room interior design. The results showed that (1) relaxing music was
judged more appropriate for room interior design than non-relaxing mu-
sic and (2) the room interior design was estimated as more calm when
the BGM was relaxing. In Experiment 2, the change in the impression
and the likes and dislikes concerning this room were examined by using
two types of lighting: calm lighting and ordinary lighting. Sixty partici-
pants were asked to rate the impression concerning this room. The re-
sults showed that (1) the change in the impression from the lighting did
not change the atmosphere formed by the interior design and (2) the
good impression of the room improved when the calm lighting was used.
This study showed that BGM, as well a room acoustics and lighting, can
play an additional role in a highly-designed indoor room.
Keywords: music; healing room; mental fatigue; relaxing music; listen-

ing space design
Yasuhiro Goto, Faculty of Psychology and Applied Communication, School of Humani-

ties, Hokusei Gakuen University, Nishi 2-Chome 3-1 Ohyachi, Sapporo 004-8631, Ja-
pan; Email: goto@hokusei.ac.jp
Performing proportion: Crux awareness in

Scarlatti interpretation
Nancy Lee Harper1, Toms Henriques2, and Rosalind Halton3
1 Department of Communication and Art, University of Aveiro, Portugal

2 Department of Music, Buffalo State College, USA
3 Department of Music, University of Newcastle, Australia
This study aims to determine if crux occurrence in Scarlattis K. 159 so-

nata conforms to a golden section or other proportionate position as
found in other sonatas and, if so, is interpreted thus by performers.
Keywords: Scarlatti; crux; golden section; performance; background/

foreground tempi
Performer-scholar Ralph Kirkpatrick identified a compositional phenomenon

in the majority of the Domenico Scarlatti sonatas, which he termed crux:
the point in each half where the thematic material at the ends of both halves
establishes the closing tonality (Kirkpatrick 1953, p. 255). Thus, the crux has
a triple function: melodic, harmonic, and structural. In previous research that
compared Scarlattis Essercizi and Cantabile sonatas, a very high incidence of
golden section proportion (GS) was found at the cruxs occurrence (Harper
2007), regardless of the character (Halton 2002). Also, in a sonata with GS
proportion (K. 380), it was determined that performers consciously or uncon-
sciously react to this proportion by expressing it in both foreground (surface)
and background (structural) tempi (Harper and Henriques 2008). Using
digital audio editors, the current study obtains data from a control group and
an experimental group in performance of the Sonata in C Major, K. 159 La
caza, in order to measure performance consistency and tempo-crux percep-
tions.
MAIN CONTRIBUTION
Scarlattis Sonata in C Major, K. 159 was chosen due to its structural propor-
tions and easy accessibility. In 6/8 and with a range of D-d, it is marked
allegro in various manuscripts. This bi-partite sonata is 64 measures long (26
Table 1. Crux proportions.
Crux Length Phi or GS Crux Differential Conclusions

Proportions (L) L x .618034 Crux-phi
A1 26 16, 068884 ms. 13 16-13= mirrored to
(1st half) measures =16 3 measures second half
B1, A1 38 23, 485292 ms. 52 (26) 26-23= mirrored to
(2nd half) measures =23 3 measures first half
A 21 12, 978714 ms. 9 12,9-9= close to mirror
Recap. only measures =12,9 3,9 measures proportion
Figure 1.The crux in A and B sections.
bars+38 bars) or 128 measures with the repetitions and no first or second
endings. In the overall AB form, the B part consists of B+A (or rounded bi-
nary form with developmental aspects and recapitulation), which may be seen
with repetitions as: [A1] [A2] [B1, A3,] [B2, A4]: A2=repeat of A1, B2=repeat
of B1, A3=variant of A1 in recapitulation, A4=repeat of A3. The crux appears
in measure (ms.) 13 (A section) and in ms. 52 (=ms. 26 of that section or ms.
9 of the recapitulation).
Before measuring the placement of the crux, a one measure discrepancy
was found in various manuscripts and editions in the first half (ms. 16); this
had to be considered with the logical conclusion that Scarlattis copyist
accidentally omitted it because of its repetitious nature. Gilberts Urtext
edition was chosen because of the logical symmetry of the sonatas halves.
The phi or golden section measurement of each half was taken in order to
compare with the crux position (Table 1). Although GS proportion is not
present, a mirrored symmetrical proportion is found: the crux occurs three
measures before phi in the first half and three measures after phi in the
second half. A proportion of +3 measure-differential (almost 4 measures)
occurs when exclusively measuring the crux position of the recapitulation.
Table 2. Real-time comparisons at key moments (in seconds).
Real time Crux Crux End End End End Crux Crux End End
A1 A2 A1 A2 B1 B2 A3 A4 A3 A4
ms. ms. ms. ms. ms. ms. ms. ms. ms. ms.
13 13 26 26 43 43 52 52 64 64
(39) (52) (95) (107) (88) (116) (90) (128)
Total
duration
Asperen 11,294 35,357 23,713 47,790 65,435 102,401 73,318 110,197 84,748 125,991
MM=128
Crudelli 11,379 35,861 24,403 48,774 66,685 105,434 74,537 113,202 87,077 128,896
MM=126
Nicolson 11,834 36,628 24,883 50,000 68,579 108,02876,761 116,186 87,077 131,788
MM=120
Halton 13,506 42,483 28,587 57,613 79,386 123,419 88,519 132,452102,235 149,432
MM=113
Pogorelich13,304 41,196 27,942 56,550 77,080 122,082 86,063 131,248 101,065 149,316
Model
MM=112
Harper 13,542 42,468 28,742 57,543 78,539 124,424 88,085 134,061103,112 153,162
MM=110
Tilney 14,252 44,176 30,048 59,786 81,062 125,922 90,541 135,561 104,998153,074
MM=110
Performers proportion
Technical problems prevented the comparison of equal live (experimental

group) and commercial recordings (control group): five commercial re-
cordings (harpsichord, fortepiano, modern piano) were used and two live
studio performances (harpsichord, modern piano). The performers of the live
recordings knew of the position of the crux, although it is unknown whether
the performers in the commercial recordings were aware of the position of the
crux and chose deliberate interpretations.
Real time and equalized time timings
Using audio digital editors, time readings of performances of all performers

were taken at several key moments (Table 2): crux A1, A2, end of A1, end of
A2, end of B1 (end of modulatory part in ms. 43), end of B2, crux A3, crux A4,
Table 3. Equalized time comparisons at key moments (in seconds).
Real time Crux Crux End End End End Crux Crux End End
A1 A2 A1 A2 B1 B2 A3 A4 A3 A4
ms. ms. ms. ms. ms. ms. ms. ms. ms. ms.
13 13 26 26 43 43 52 52 64 64
(39) (52) (95) (107) (88) (116) (90) (128)
Total
duration
Asperen 13,326 41,814 28,166 56,64077,502 121,351 86,779 130,556 100,412 149,316
length
ratio=0,843
Crudelli 13,08641,423 28,26656,665 77,274 122,08286,290131,081 100,861 149,316
ratio=0,863
Nicolson 13,256 41,323 28,156 56,607 77,641 122,370 86,962 131,547 101,000149,316
ratio=0,882
Halton 13,506 42,48328,587 57,613 79,386 123,419 88,519 132,452 102,235 149,316
ratio=1,007
Pogorelich 13,304 41,196 27,942 56,550 77,080122,08286,063131,248 101,065 149,316
MODEL
ratio=1,000
Harper 13,784 41,333 28,00155,983 76,540 121,289 85,861 130,693100,523149,316
ratio=1,0257
Tilney 13,784 42,95029,296 58,32479,053 122,823 88,286132,262 102,393 149,316
ratio=1,0251
Figure 2. Real time and equalized timings; Pogorelich and Halton were the same in real
time; all recordings were adjusted to Pogorelichs model for equalized timings. (See full
color version at www.performancescience.org.)
end of A3, and end of A4. This information gave foreground or surface tempo
measurements. Tempo and real time timings ranged from MM 110 to MM128
and from 125.991 to 153.074. To determine the internal or background
Table 4. Real time/reference model time: table ratio comparisons at key moments.
Real time/ Crux Crux End End End End Crux Crux End End
reference A1 A2 A1 A2 B1 B2 A3 A4 A3 A4
model time
Asperen 0,84900,8582 0,8486 0,8450 0,8489 0,8387 0,8589 0,8396 0,8385 149,
length >0,843> > > > > > - > 316
ratio=0,843
Crudelli 0,8553 0,8704 0,8733 0,8624 0,8651 0,8636 0,8660 0,8625 0,8615 149,
ratio=0,863 <() > > <() > > > -<() <() 316
Nicolson 0,8895 0,8891 0,8905 0,8841 0,8897 0,8848 0,8919 0,8852 0,8820 149,
ratio=0,882 > > > > > > > > > 316
Halton 1.0157 1,0312 1,0230 1,0187 1,0299 1,0109 1,0285 1,0091 1,0115 149,
ratio=1,007 316
Pogorelich 13,304 41,196 27,942 56,550 77,080 122,082 86,063 131,248 101,065 149,
MODEL 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 316
ratio=1,000
Harper 1,0178 1,030 1,0286 1,0159 1,0189 1,0191 1,0234 1,0214 1,0214 149,
ratio=1,0257 <! >() >() <! <! <! <! <! <! 316
Tilney 1,0712 1,0723 1,0753 1,0572 1,0516 1,0314 1,0520 1,038 1,0389 149,
ratio=1,0251 >() >() >() >() >() >() >() >() >() 316
tempi of the performances, all recordings needed to be equalized to the same

tempo-length. Pogorelichs recording was chosen as the model (Table 3) due
to the steady and musical consistency of the performance (MM=112;
149,316). Recordings longer than Pogorelichs were compressed; if shorter,
expanded. When compressed, the time modification ratio is a number smaller
than the model ratio of 1.000, and when expanded, it is larger than the model
(Table 4).
Table 4 shows that when one compares the length-ratios of the recordings
that were time-modified with the analysis ratios at the chosen key points that,
in most cases, the performers tend to modify their overall tempo at those key
points so as to stay closer to an underlying temporal framework that favors a
clear, well-measured tempo. In the expanded recordings (Asperen, Crudelli,
Nicholson), > indicates a convergence toward the reference model, or rather
the analysis ratios that are larger than the length ratios indicate that the ex-
pansion of time is happening slightly faster around the analysis points (crux,
etc.) and is going faster in order to catch up to the reference. The indication of
() means that the result is contrary to the prediction. Thus, comparisons
and analysis of length ratios reveal that Nicholsons performance has the
highest rate of convergences toward the timings of the model at 100%: As-
peren=88%, Harper=77%, Crudelli=55%, and Tilney=0%.
IMPLICATIONS
It is concluded that: (1) although the crux does not fit into direct GS propor-
tions, the symmetrical mirrored relationship between the two halves of the
sonata of three measures is significant, (2) when repeats are made by the
performers, the length of the A section corresponds to the occurrence of the
crux in the B section (52 measures), making a strong case for the necessity of
the repeats, and (3) regardless of live or commercial recordings or choice of
instruments, there occurred two performances of the same duration: Po-
gorelich (piano, commercial/model performance), Halton (harpsichord, live
performance). We may infer that there is a proportional sense of crux by
composer and by performers, which is evident in both foreground (surface)
and background (structural) tempi.
Nancy Lee Harper, Department of Communication and Art, University of Aveiro, Cam-
pus Santiago, 3810-193 Aveiro, Portugal; Email: harper@ua.pt
References
Halton R. (2002) Domenico Scarlatti and his Cantabile Sonatas. Musicology Australia,
25, pp.22-48.
Harper N. L. (2007) Golden section in the sonatas of Domencio Scarlatti: A re-exami-
nation of Kirkpatricks crux. In A. Williamon and D. Coimbra (eds.), Proceedings
of ISPS 2007 (pp.239-244). Utrecht, The Netherlands: European Association of
Conservatoires (AEC).
Harper N. L. and Henriques T. (2008). Performing musical structure: Crux-phi percep-
tions in Domenico Scarlattis sonata K. 380. Paper presented at the fourth Confer-
ence on Interdisciplinary Musicology (CIM08), Thessaloniki, Greece.
Scarlatti D. (1978). D. Scarlatti. Sonates-vol. IV, K. 156-205, le pupitre, Collection de
Musique Ancienne Publie sous la Direction de Franois Lesure. Paris: Heugel.
Recordings: van Asperen B. (1992). Domenico Scarlatti Sonate. Amersterdam: EMI
Classics; Crudelli M. (2002) Domenico Scarlatti. Rome: Radio Vaticana MK 17112
00; Nicolson L. (2004) Scarlatti Sonatas. Cologne: Capriccio; Pogorelich I.
(1993). Scarlatti Sonatas. Hamburg: DGG B100001GGV; Tilney C. (1988). Scarlatti
Sonatas. Toronto: Dorian Recordings DOR 90103.
Piano playing skills in a patient with

frontotemporal dementia: A longitudinal
case study
Sharpley Hsieh1,2, Eneida Mioshi1, Felicity Baker3, Olivier Piguet1,2, and

John R. Hodges1,2
1 Prince of Wales Medical Research Institute, Sydney, Australia

2 School of Medical Sciences, University of New South Wales, Australia
3 School of Music, University of Queensland, Australia
Patients with dementia, such as Alzheimers disease, can continue to play

the piano skillfully despite profound cognitive impairment. It has been
suggested that this may be because these skills have been well-rehearsed
and become automatic motor movements. Less is known about how
these musical skills may be related to the performance of everyday func-
tional abilities and, also, how the ability to play a musical instrument
may be affected in other dementia types. Recordings of a patient diag-
nosed with behavioral-variant frontotemporal dementia, a dementia syn-
drome clinically characterized by marked behavioral and cognitive
changes, playing the piano was taken 12 months apart. Aspects of musi-
cal performance (accuracy, tone quality, dynamics, rhythm, tempo, and
interpretation) were rated by professional musical teachers. The physical
and mental skills required to complete activities of daily living were as-
sessed. Tests of cognitive functioning and brain imaging were also con-
ducted over this period. Results showed that over one year, significant
declines were observed in the areas of cognition, the mental abilities re-
quired for everyday skills, as well as brain atrophy on imaging. Physical
skills for the performance of activities of daily living were relatively pre-
served, as was the ability to play the piano. These findings confirm previ-
ous reports and demonstrate the relative independence of procedural
skills in the context of significant cognitive impairment in patients with
dementia.
Keywords: frontotemporal dementia, piano playing; activities of daily

living; cognitive assessment; neuroimaging
Dementias are progressive neurodegenerative brain diseases characterized by

multiple deficits in cognition (e.g. memory, language, etc.), changes in be-
havior, and everyday functional impairment. Clinical profiles tend to be dis-
tinct for the different dementia types, although overlaps do exist. In contrast
to cognition, musical skills may be well-preserved in dementia (Baird and
Samson 2009), although this is an area that has not been well studied. Sev-
eral case reports have described patients with Alzheimers disease who re-
main capable of playing musical instruments skillfully over a number of years
despite increasing cognitive deficits (Beatty et al. 1999).
Patients with dementia who have shown an ability to play musical in-
struments typically play pieces that have been well-rehearsed prior to the
onset of disease. It is speculated that these skills may therefore reflect auto-
matic motor movements which are reliant on the implicit memory system
(Crystal et al. 1989). Some of these patients perform normally on tests of im-
plicit skills (e.g. speeded mirror reading, Crystal et al. 1989).
The neural basis of preserved musical performance skills in dementia has
been thought to rely on the motor circuits in the brain (i.e. basal ganglia,
cerebellum, motor areas of the thalamus, and cortex; Beatty et al. 1999) as
these remain relatively intact until the end stages of the Alzheimers disease
process.
No studies to date have compared musical performance skills with the
formal examination of performance on activities of daily living (ADLs), abili-
ties that can also be conceptualized as skills that have been learned and well-
rehearsed prior to the onset of dementia. In addition, little is known about
how musical performance skills may be affected in other dementia syn-
dromes.
The aim of the study was to investigate change in piano playing skills to-
gether with assessments of cognitive functioning, everyday skills, and brain
imaging over a 12-month period in a patient diagnosed with behavioral-vari-
ant frontotemporal dementia (FTD; Neary et al. 1998). FTD is a dementia
syndrome clinically and pathologically different to Alzheimers disease. It is
characterized by marked behavior change and executive dysfunction. Brain
pathology is most pronounced in the prefrontal and anterior temporal corti-
ces.
METHOD
Case description
RD is a 70-year-old, right-handed man with 18 years of education and an

amateur pianist. He has played the piano since childhood. In mid-2007, he
was diagnosed with behavioral-variant frontotemporal dementia. Disease

duration from the symptom onset to the time of the assessment was 3 years.
Procedure
Video recordings of RD playing Rachmaninoffs Prelude in C# minor (Edition

by Jean St John) with the music score were obtained a year apart in 2007 and
2008. This piece was well known to RD prior to the diagnosis of dementia. At
the time of the recordings, RD was playing it about once a week.
Each recording was subdivided into six sections of 30 to 60 s duration,
and each section was rated for the following: accuracy, tone quality, dynam-
ics, rhythm, tempo, and interpretation. Five independent raters who were
professional music teachers and blind to RDs condition rated each aspect on
a Likert scale (1-5, where 1=poor and 5=excellent) analogous to that used by
Beatty et al. (1999). Intra-class coefficients calculated to examine agreement
among the raters were statistically significant (p<0.05).
Cognitive ability and ADLs were also measured on these two occasions.
Cognitive domains of orientation, memory, fluency, language, and visuospa-
tial skills were examined using the Addenbrookes Cognitive Examination-
Revised (ACE-R; Mioshi et al. 2006). Physical and mental skills required to
complete ADL activities were measured using the Assessment of Motor and
Process Skills (AMPS; Fisher2003). In addition, high resolution structural T1
images were obtained on a 3T MRI scanner in 2007 and 2008. Percentage
brain volume change was estimated using SIENA (Structural Image Evalua-
tion, using Normalisation, of Atrophy; Smith et al. 2002), which is part of
FSL (Smith et al. 2004, www.fmrib.ox.ac.uk/fsl).
RESULTS
At baseline, cognitive performance on the ACE-R was 70/100, which is below

the cut-off of 82 typically used to indicate the presence of dementia. RD lost
points primarily on the orientation, memory, fluency, and language sections
of the test. Visuospatial skills were found to be preserved. At the 12-month
follow-up, RDs score had dropped to 50/100; again, scores remained within
the normal range for visuospatial skills.
On the AMPS, RDs motor and process scores at baseline were 1.61 and
0.64, respectively. Both these scores were below the level expected for his age,
indicating impairment in physical and mental skills required to complete
ADLs. At follow-up, AMPS motor and process scores were 1.32 and -0.09,
respectively. The relative decline in the AMPS scores over this period was
Table 1. Mean ratings for eight attributes of RDs piano playing skill.
Year 1 Year 2 F1,8

Accuracy of notes 2.63 2.16 2.27
Tone quality 2.90 2.70 <1
Dynamics 2.67 2.37 1.31
Accuracy of rhythm 2.17 2.10 <1
Tempo 1.87 1.67 <1
Interpretation 2.23 2.10 <1
Figure 1. Coronal MRI T1 images obtained in 2007 (left) and 2008 (right).
smaller for the motor than process scales and suggests that the physical as-
pects of ADL performance were relatively preserved in RD.
Mean ratings for RDs piano playing for both recordings are summarized
in Table 1. While all ratings of piano playing attributes at Year 2 were lower
than at Year 1, none of the changes reached statistical significance.
Structural MRIs showed significant orbitofrontal and anterior temporal
atrophy (see Figure 1), which became worse over 12 months. The estimated
percentage total brain volume loss over one year was 3.17%.
DISCUSSION
This case study examined piano-playing skills in a patient diagnosed with

behavioral-variant frontotemporal dementia over a 12-month period in con-
junction with assessment of cognition, ADL abilities, and brain imaging.
Over the interval of one year, deterioration on formal assessment of cog-
nition, functional abilities, and brain imaging was observed. In contrast, RDs
ability to play the piano remained consistent over this period. These findings
are similar to case reports in the literature of patients with Alzheimers dis-
ease who have relative preservation of musical performance skills despite
impairment in other domains (Beatty et al. 1999).
Although physical abilities required for the performance of ADLs re-
mained relatively stable in RD over 12 months, the mental skills necessary for
the same ADLs had worsened. This finding suggests that motor skills were
reasonably preserved in RD and may have supported RDs ongoing capacity
to play familiar musical pieces on the piano.
This case study has some limitations. First, the rating scale for the as-
sessment of musical performance used emphasized the motor aspects of
playing a musical instrument (e.g. accuracy of the notes and rhythm). The
ability to convey the mood and nuances of musical pieces was not studied.
Development of a scale to measure these aspects of musical performance will
be of interest, particularly as deficits in emotion processing are a prominent
feature of patients with frontotemporal dementia. Second, while there was an
increase in atrophy over time, it is unclear where this may have occurred.
Analysis of atrophy in the motor regions of the brain, relative to the frontal
and temporal lobar areas, would also be of interest in the future.
Finally, RD was seen relatively early in the course of FTD. It is unclear
whether there is a resilience of musical skills overall in dementia or whether
different abilities decline at different rates with disease progression. Further
longitudinal assessments of musical performance skills as well as cognition
and activities of daily living should also be examined.
Acknowledgments
This research is supported in part by NHMRC Project Grant (#570106) and ARC Fed-
eration Fellowship (FF0776229). SH is supported by the Australian Postgraduate
Award. We thank Patient RD and his family.
Sharpley Hsieh, Prince of Wales Medical Research Institute, PO Box 82, St Pauls, Syd-
ney, New South Wales 2031, Australia; Email: s.hsieh@powmri.edu.au
References
Baird A. and Samson S. (2009). Memory for music in Alzheimers disease: Unforgetta-
ble? Neuropsychology Review, 19, pp. 85-101.
Beatty W. W., Rogers C. L., Rogers R. L. et al. (1999). Piano playing in Alzheimers dis-
ease: Longitudinal study of a single case. Neurocase, 5, pp. 459-469.
Crystal H. A., Grober E., and Masur D. (1989). Preservation of musical memory in Alz-
heimers disease. Journal of Neurology, Neurosurgery and Psychiatry, 52, pp.
1415-1416.
Fisher A. G. (2003). Assessment of Motor and Process Skills (vol. 1). Fort Collins, Colo-
rado, USA: Three Stars Press.
Mioshi E., Dawson K., Mitchell J. et al. (2006). The Addenbrookes cognitive examina-
tion revised (ACE-R): A brief cognitive test battery for dementia screening. Inter-
national Journal of Geriatric Psychiatry, 21, pp. 1078-1085.
Neary D., Snowden J. S., Gustafson L. et al. (1998). Frontotemporal lobar degeneration:
A consensus on clinical diagnostic criteria. Neurology, 51, pp. 1546-1554.
Smith S. M., Zhang Y., Jenkinson M. et al. (2002). Accurate, robust and automated
longitudinal and cross-sectional brain change analysis. NeuroImage, 17, pp. 479-
489.
Smith S. M., Jenkinson M., Woolrich M. W. et al. (2004). Advances in functional and
structural MR image analysis and implementation as FSL. NeuroImage, 23, pp.
208-219.
Chin force in violin playing
Hiroshi Kinoshita and Satoshi Obata
Graduate School of Medicine, Osaka University, Japan
The aims of the present study were to develop a chinrest that could di-
rectly assess chinrest force during musical performance and to provide
baseline force data. A force transducer-mounted chinrest was designed
and built by the present authors. Data were obtained from 11 elite violin-
ists while they performed scale tasks at different dynamics (p, mf, and f),
tempi (1, 4, and 8 Hz), and hand positions (first and eighth). Data were
also obtained from playing the tasks with vibrato, and the excerpts were
from the Bruch and Dvorak concertos. A mechanical test of the chinrest
confirmed good linearity of the force up to 100 N. During the scale tasks,
chinrest force was around 14 N at p, which increased to 18 N at f. Neither
the playing tempo nor the hand position largely affected the force, but it
was significantly increased during vibrato. During playing the Bruch con-
certo, the force was elevated to 28 N (peak force=39 N; the peak-force
range in all players=26-76 N). Typical chinrest force to stabilize the violin
during ordinary musical performance is thus less than 20 N, but it can be
tripled or more when performing technically demanding musical tasks.
Keywords: violin; chinrest force; force sensor; pedagogy
Hiroshi Kinoshita, Graduate School of Medicine, Osaka University, 1-17 Machikane-

yama, Toyonaka, Osaka 560-0043, Japan; Email: hkinoshita@moted.hss.osaka-u.ac.jp
The nature of professional accompanists and

their roles: Performing with musical excellence
and enjoying communicative interaction
Yuriko Kubota
University of London, UK
The aim of this paper is to explore the traits and roles of professional ac-
companists. This study is based on an internet survey. Professional ac-
companists were asked to complete a questionnaire through an internet
website. The participants (n=96) identify themselves mainly as being
sensitive, thoughtful, logical, and optimistic. At the start of rehearsal, ac-
companists generally try to establish a collaborative tempo and timing
while seeing the big picture of the music. At the end of rehearsal, they
concentrate on mood/feeling, dynamics, and aesthetic goals much more
than at the start. Although some insist that accompanists should follow
the soloists expression, most accompanists enjoy and expect communi-
cative musical activity within music. Expert accompanists point out that
they have an important role in making music that is better than that cre-
ated by the soloist him/herself. It would seem that professional accompa-
nists should have an ability to perform with an overall structure of music
with confidence from the very start of rehearsal. Furthermore, it is likely
that particular techniques for accompaniment, including practical com-
munication skills (social, physical, and musical), with flexibility and ar-
tistic excellence, have considerable influence on their work.
Keywords: accompanist; personality; rehearsal; roles; ability
Accompaniment has been researched by only a few authors, including Moore

(1943) and Adler (1965), whereas there are more ensemble studies, for exam-
ple, by Shaffer (1984) and Davidson and King (2004). Furthermore, Kemp
explored musicians personalities (1981, 1996).
The aim of this research is to investigate the nature of professional ac-
companists and their roles. In undertaking this research, I have set my heart
on raising the status of professional accompanists, many of whom have excel-

lent artistry.
METHOD
Participants
Two hundred and twenty professional accompanists were asked to fill in a

questionnaire. The questionnaire was completed by 96 professional accompa-
nists (experience of accompaniment: mean=23 years; men: n=38 and
women: n=58) from 11 different nationalities. They are musicians (55%),
school or piano teachers (11%), teaching staff in an academic position (10%),
and others (19%). They accompany instrumentalists (34%), vocalists (32%),
choirs (12%), chamber ensembles or orchestras (3%), all styles of ensemble
(18%), and others (1%).
The criteria used in this study for professional accompanists are as fol-
lows: (1) people who have more than ten years of accompaniment experience;
(2) people who have had formal education in piano performance or the rele-
vant experience; and (3) people who identify themselves as professional ac-
companists and are paid from their accompaniment work.
Materials
This research is based on an internet questionnaire with 15 multiple choice

questions, 23 rating scales, and 8 open-ended questions. The questionnaire
covered personality, interests, rehearsal, solo/ensemble concerts, and ability.
Procedure
The questionnaire was uploaded onto a website, and the answers from the
participants were analyzed.
RESULTS
Nature of professional accompanists
Personality
The accompanists describe themselves as being sensitive (60.4%), thoughtful

(46.7%), optimistic (38.5%), logical (37.5%), inquisitive (31.3 %), calm/gentle
(17.7%), quiet (12.5%), stubborn (11.5%), argumentative (7.3%), or something
else.
Kemp (1981, 1996) explored musicians personalities. He found high lev-

els of tendency to introversion (self-sufficiency). In contrast, he found high
levels of extraversion among keyboard players. In this research, the results
also did not show the typical trends of introversion of the accompanists. Just
28.0% of respondents were classed as introverts, 39.6% as neutral, and 31.3%
as extraverts. Furthermore, there are more accompanists who prefer to be a
leader (38.9%), or are neutral (32.6%), than a follower (28.4%).
Interests
The accompanists take an interest in their own accompaniment work; 70.8%

of the accompanists prefer ensemble to solo performance, 15.6% prefer solo
performance, and 12.5% like both. Through their work, 83.3% of the accom-
panists expect musical communication within music, 57.3% expect musical
aspects (e.g. sound) in ensemble, and 32.3% expect social aspects.
Accompanists with less than 25 years of experience expect more social as-
pects than those with above 25 years (p<0.01). On the other hand, performers
who have accompanied for more than 25 years expect musical communica-
tion within the music more than less experienced players (p<0.05). Further-
more, accompanists aged over 60 expect more individual improvements than
those in their 20s and 40s (p<0.05).
Roles and ability
Gaps between performers
The biggest gaps, that is, differences and difficulties between their co-per-
former(s) and themselves, which the accompanists experience are: tempo
(35.8%), ritardando (30.5%), articulation (30.5%), and boundaries involving
places (27.4%) and treatment (27.4%). They also feel gaps of different expres-
sion caused by a different instrument (22.1%), accelerando (18.9%), dynamics
(18.9%), mood (17.9%), and tone quality (16.8%).
Pianists who have accompanied for more than 30 years, especially those
aged 41-50 (p<0.05), feel more gaps of dynamics than less experienced play-
ers (p<0.05). Furthermore, pianists who accompany instrumentalists feel
more gaps of ritardando than those who accompany singers (p<0.001).
Aims of rehearsal
Although the accompanists aims changed throughout their rehearsal, the

levels of attention toward every goal at the end increased considerably
(p<0.001). Their main aim in rehearsal is synchronization. At the start, ac-
Table1. Aims of rehearsal with level of importance (1-10, with 10=very important).
At the start Mean SD At the end Mean SD

1 Synchronization 8.5 1.9 1 Mood/feeling 9.7 1.0
2 Tempo 8.4 1.8 2 Synchronization 9.5 1.1
3 Mood/feeling 7.7 2.5 3 Aesthetic 9.3 1.3
4 Overall structure 7.5 2.4 4 Dynamics 9.3 1.2
5 Timing 7.3 2.3 5 Tempo 9.1 1.6
6 Dynamics 7.2 2.3 6 Characteristics 9.1 1.8
7 Characteristics 7.2 2.5 7 Overall structure 9.0 1.7
8 Motion 7.2 2.5 8 Timing 8.9 1.7
9 Aesthetic 7.1 2.6 9 Motion 8.9 2.0
10 Structure (local) 6.9 2.4 10 Structure (local) 8.4 2.2
Figure 1. Abilities required as a competent accompanist.
companists generally try to establish a collaborative tempo and timing, while

seeing a big picture of the music and its overall structure. At the end, they
concentrate on mood/feeling, aesthetic goals, and dynamics much more than
before (see Table 1).
Ability
As far as the accompanists ability is concerned, three themes emerged: effec-

tive technique, personality, and musical excellence (see Figure 1).
Firstly, accompanists require a particular (effective) technique for their
work. Most accompanists insisted that sight-reading, listening skills, and
ability to have an immediate grasp of the overall structure of the music are
important. Furthermore, accompanists report that communication skills have
a significant influence on their work; these involve social, physical (keys of
synchronization/leading the co-performer), and musical (give and take)
skills.
Secondly, personality affects their work considerably. Most reported that
sensitivity and tactfulness are important for their effective work and flexibil-
ity is a necessity. Without accepting and leading the others expression and
opinions, collaborative expression will not work satisfactorily.
Finally, musical excellence (artistry) was mentioned by a number of par-
ticipants. Although some accompanists insist that they should follow the solo-
ists expression, the majority of accompanists would agree with expressions
like these:
Accompanists have an important role in making music better than that

created by the soloist him/herself (participant 7).
Make your partner comfortable, while not losing your own musical integ-
rity (participant 91).
As a result, it seems that the levels of both performers musicality and

technique are the keys of successful performance.
DISCUSSION
The expression of a pianist (timing/dynamics) with a violinist (including their

rehearsal) was investigated in my previous research (Kubota 2008). Student
performers played a Mozart Sonata for Violin and Piano.
It seems that there is a difference in the process of rehearsal between stu-
dents and professional accompanists. The student performers aims of re-
hearsal at the beginning were solving technical problems, sharing general
ideas, and synchronization, and at the end the aims were developing ideas
with broad structures, more creative expression, and synchroniza-
tion/fluency. The student accompanists expression developed considerably
throughout the experiment, especially in the period before a concert. It is
likely that less experienced accompanists grasp the overall structure of the
music gradually while practicing together. On the other hand, it seems that
expert accompanists should have the ability to perform with appropriate and
artistic interpretation of overall structure from the beginning.
At the beginning of rehearsal, I prefer to seek a big picture rather than

details. If all parties are well-prepared and experienced, the music will
sync with minimum negotiation (participant 85).
Ensemble performance has the social and psychological aspects of music-

making. Accompanists might be attracted by the art of synthesis.
Accompaniment is the most rewarding and fun thing one can do as a mu-
sician! (participant 48).
Acknowledgments
I would like to thank all the participants in this survey.
Yuriko Kubota, 496-19 Ichi-cho, Kawachinagano-shi, Osaka-fu 586-0002, Japan;

Email: yk25@student.london.ac.uk
References
Adler K. (1965). The Art of Accompanying and Coaching. New York: Da Capo Press.
Davidson J. W. & King E. C. (2004). Strategies for ensemble practice. In A. Williamon
(ed.), Musical Excellence (pp.105-122). Oxford: Oxford University Press.
Kemp A. E. (1981). Personality differences between players of string, woodwind, brass
and keyboard instruments, and singers. Council for Research in Music Education
Bulletin, 66/67, pp. 33-38.
Kemp A. E. (1996). The Musical Temperament. Oxford: Oxford University Press.
Kubota Y. (2008). A pianists expression in the role of co-performer: Changes in timing
and dynamics through communicative interaction with a violinist. In K. Miyazaki et
al. (eds.), Proceedings of the Tenth International Conference on Music Perception
and Cognition (pp. 140-149), Sapporo, Japan: ICMPC.
Moore G. (1943). The Unashamed Accompanist. New York: Macmillan.
Shaffer L. H. (1984). Timing in solo and duet piano performances. Quarterly Journal of
Experimental Psychology, 36, pp. 577-595.
How memory fades:

Very-long-term recall of Bach
Tania Lisboa1, Roger Chaffin2, and Topher Logan2
1 Centre for Performance Science, Royal College of Music, UK

2 Department of Psychology, University of Connecticut, USA
A cellist memorized the Prelude from J. S. Bachs Suite No. 6 for solo
cello and identified performance cues (PCs) that she attended to in per-
formance. During the next three years, she recalled the piece twice,
playing and writing out the score from memory on both occasions, in
counterbalanced order. Played recall was better than written recall.
Written recall was better at expressive and structural PCs, suggesting
that these cues provided content accessible access to declarative memory.
Written recall was worse at PCs for basic technique but less so when
written recall followed played recall. In written recall, serial cuing was
impaired by the absence of sensorimotor cues, particularly at basic PCs.
These directed the cellists attention to her actions. Reinstating sensori-
motor memory by playing through the piece reduced the impairment
caused by the absence of these actions during written recall.
Keywords: music; performance; memory; cello; recall
The demands placed on memory during solo performance in Western art

music are extraordinary. Not surprisingly, memory and attentional lapses are
not uncommon. Thus, when preparing for a memorized performance, it is
important for musicians to develop a memory retrieval system that is flexible
and that will permit the performance to proceed whatever may go wrong
(Chaffin et al. 2002, Lehmann and Ericsson 1998). During music perform-
ance, memory for what comes next is normally activated by serial cuing as
the current passage cues motor and auditory memory (Chaffin et al. 2009a).
Serial cuing has the limitation that the chain of cues starts at the beginning of
the piece so that if the performance is disrupted the musician must start over.
For this reason, experienced performers usually prepare an alternative mem-
ory retrieval system that provides content addressable access, allowing the
musician to recall any passage in the piece by simply thinking of it (e.g. the G
section). Content addressable access is provided by performance cues (PCs)
representing landmarks in the music that the performer is able to think about
consciously during performance. PCs provide a mental map of the music that
allows the performer to monitor the performance as it unfolds and to recover
from mistakes and memory lapses.
Written recall of the score has proved an important source of evidence
that PCs provide content addressable access to memory (Chaffin and Logan
2006, Chaffin et al. 2009b). Recall is better at PCs representing musical ex-
pression and structure and declines in the bars that follow (an effect of serial
position). This pattern suggests that musicians have content addressable ac-
cess to memory at these points and then retrieve the following bars by serial
cuing. In contrast, recall is poorer at PCs representing decisions about basic
technique. One possible explanation is that musicians attend more to details
of technique at these points and so pay less attention to the notes. We tested
this explanation by comparing written and played recall of a well-prepared
piece. We have reported elsewhere that the written recall was better at
structural and expressive PCs and poorer at basic PCs (Chaffin et al. in press).
Here we ask whether the same effects occurred when the musician played the
piece at around the same time.
METHOD
Participants
Tnia Lisboa, the cellist and first author was trained in classical cello and
piano in Brazil, England, and France and currently lives in London perform-
ing as a cello soloist.
Materials
The Prelude from J. S. Bachs Suite No. 6 for solo cello explores both the
mellow quality and virtuoso aspects of the instrument. The cellist had never
learned the Suite No. 6 for performance before, although she was very famil-
iar with it and had played other works by Bach throughout her career. Writ-
ten for an instrument with five strings, Suite No. 6 presents contemporary
cellists with substantial technical challenges, as fingerings and left-hand po-
sitions must be adapted to play the notes written for the fifth string on the
four strings of a modern cello. Musically, however, the Prelude is comparable
to the other five Bach cello suites. Notated in 104 bars in 12/8 time, the piece
takes about five minutes to perform.
Percent Recall
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Written Played Played Written
Recall 1 Recall 2
Figure 1. Percentage correct on two tests of written and played recall, shown in tempo-
ral order from left to right.
Table 1. Regression coefficients for the effects of serial position of half-bars from PCs on
probability of correct recall for first/second written recalls.
Effect of serial position following Effects Interaction indicating

difference between recall tests
Expressive PCs -0.073* 0.033
Structural PCs -0.032 0.025
Interpretive PCs 0.031 -0.007
Basic PCs 0.086** -0.050*
Structural PCs x Expressive PCs -0.018*
Note. *p<0.05, **p<0.01, ***p<0.001.
Procedure
The cellist learned the Prelude for a series of eight public performances. She
then provided reports about the musical structure and PCs she attended to
during performance (expressive, interpretive, intonation, and basic tech-
nique) as part of another study (Chaffin et al. in press). Written and played
recall were each tested twice in counterbalanced order. The first test began
ten months after the last public performance with the cellist writing out the
score from memory. She then played it from memory seven weeks later, re-
cording her playing. Twenty months later, she began the second test by play-
ing the piece from memory again, and then wrote it out for a second time four
weeks later. She did not otherwise play or study the piece during this time.
Written and played recall were scored for accuracy.
First Recall--Section Bounds Second Recall--Section Bounds
100 100
75 75
% Correct Recall
% Correct Recall
50 50
25 25
0 0
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8
Half-Bars From Section Bounds Half-Bars From Section Bounds
First Recall--Expressive Cues

Second Recall--Expressive Cues
100
100
75
75
% Correct Recall
% Correct Recall
50
50
25 25
0 0
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8
Half-Bars From Expressive Cues Half-Bars From Expressive Cues
First Recall--Basic Cues

Second Recall--Basic Cues
100
100
75
75
% Correct Recall
% Correct Recall
50 50
25 25
0 0
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8
Half-Bars From Basic Cues Half-Bars From Basic Cues
Figure 2. Mean probability of correct recall [written (black) and played (grey)] as a
function of serial position of half-bars numbered sequentially from beginnings of sub-
sections (structural PCs), expressive PCs, and basic PCs for first/second recall tests.
RESULTS
Recall was almost perfect for the two played tests and substantially lower for
the two written tests (Figure 1). The difference demonstrates the large role
played by motor and auditory memory. Written recall was better (71%) in the
second round of testing than in the first (56%).
Table 1 summarizes the results of a mixed hierarchical regression analysis
testing the effects of serial order and their interaction with the first and sec-
ond set of tests. For beginnings of sub-sections and expressive PCs, recall
declined as distance increased (see Figure 2, top and middle panels, respec-
tively). For basic PCs, the effect was in the opposite directionprobability of
recall was lowest at basic PCs and increased with distanceand was larger on
the second test (see Figure 2, bottom panels).
DISCUSSION
The cellists ability to play almost without error after more than two years of
not playing or thinking about the music is notable. The cellist described the
experience in an email to the second author shortly afterward:
It is awful to play without having practiced the piece for so long because,
besides memory, the hands feel soggy and I have no technical control of
anything even when I remembered it. Iwas hesitating all the way
through but managed to get to the end. At some placesmy fingers
seemed to go by themselves. Mostly, it was thinking of bowing and fin-
gering (basic PCs) thatgot me through.
At the end of this account, she points to the importance of the sensorimotor
cues created by her playing, which provided effective serial cuing of her ac-
tions, even in the absence of any declarative memory for what came next.
Sensorimotor cues were important to both played and written recall. The
reduction in sensorimotor cues in written recall explains why (1) written was
worse than played recall, (2) written recall was worse on first test than on the
second, and (3) written recall was worse at basic PCs. First, written recall was
worse than played recall because it provided fewer sensorimotor cues for
what came next. Second, written recall was better on the second test than the
first, despite the passage of two years, because the cellist had recently re-
freshed her sensorimotor memory by playing the piece in the played recall
test four weeks earlier. Third, written recall was worse at basic PCs because
they directed the cellists attention toward her actions and away from the
music when learning the piece, so that in recalling it, she relied more heavily
on cuing by the sensorimotor context. In written recall, the absence of the
sensorimotor context provided by her playing had a bigger impact at basic
PCs because she relied more heavily on these contextual cues at these points.
Thus, basic PCs operated as part of the serial chain of associations that re-
minded the musician of what came next (Chaffin et al. in press).
Structural and expressive PCs, in contrast, provided content addressable
access to the cellists declarative memory, allowing her to recall a passage
simply by thinking of it. Direct access to these landmarks in memory pro-
duced better recall by allowing the cellist to recover and to begin writing
again after gaps in her memory where she was unable to recall anything. Once
begun, the memory of each passage cued recall of what followed until, at
some point, a link failed and the chain was broken, resulting in a poorer recall
as distance from the landmark increased (Roediger and Crowder 1976).
The results for the second written recall replicated the previously reported
finding that her recall was better at expressive PCs than at the beginnings of
subsections (Chaffin et al. in press). The difference supports the idea that ex-
pressive PCs marked the highest level in the cellists hierarchical organization
of the music into harmonic sections (marked by expressive PCs) and melodic
subsections. By contrasting played and written recall, this study has increased
our understanding of how basic PCs differ from other kinds of PCs.
Tnia Lisboa, Centre for Performance Science, Royal College of Music, Prince Consort
Road, London SW7 2BS, UK; Email: tlisboa@rcm.ac.uk
References
Chaffin R., Ginsborg J., and Dixon J. (2009b). Serial position effects in a singers long
term recall. Paper presented at Society for Music Perception and Cognition Confer-
ence (SMPC), Indianapolis, Indiana, USA.
Chaffin R, Imreh G., and Crawford M. (2002). Practicing Perfection. Mahwah, New
Jersey, USA: Lawrence Erlbaum Associates.
Chaffin R., Lisboa T., Logan T., and Begosh K.T. (in press). Preparing for memorized
cello performance: The role of performance cues. Psychology of Music.
Chaffin R. and Logan T. (2006). Practicing perfection: How concert soloists prepare for
performance. Advances in Cognitive Psychology. 2, pp. 113-130.
Chaffin R., Logan T. R., and Begosh K. T. (2009a). Performing from memory. In S.
Hallam, I. Cross, and M. Thaut (eds.), Oxford Handbook of Music Psychology (pp.
352-363). Oxford: Oxford University Press.
Lehmann A. C. and Ericsson K.A. (1998). Preparation of a public piano performance:
The relation between practice and performance. Musicae Scientiae, 2, pp. 69-94.
Roediger H. L. and Crowder R. C. (1976). A serial position effect in recall of United
States presidents. Bulletin of the Psychonomic Society, 8, pp. 275-278.
What do children think of music teachers?

Their conceptions about cello teaching and
learning
Guadalupe Lpez, J. Ignacio Pozo, and Alfredo Bautista
Department of Cognitive Psychology, Autonomous University of Madrid, Spain
We present an exploratory study about the conceptions held by basic

level conservatory students about what they think of cello teachers and
how different teaching strategies could improve their learning skills, fo-
cused on the educational-evolutionary variable. In this research, twelve
Spanish children participated, and they were evenly in four different
grades of basic level in Spanish conservatories. The main aim of this
study was to describe, from a qualitative perspective, the different con-
ceptions about cello teaching and learning. This project helped to develop
and test the goodness of the materials, tasks, and criteria of analysis
needed. Data were collected through a structured interview, which con-
tained questions and tasks on five different studios related to teaching
and learning musical instruments. It seems that children come with con-
structive ideas to the music lessons, but when they receive instruction,
these ideas disappear gradually. Theoretical and educational implications
suggested by the findings of this investigation are discussed.
Keywords: learning conceptions; music students; teaching approaches;

learning outcomes; conceptual change
As is happening in other educational areas, learning and instruction of in-

strumental music performance is being subjected to changing demands. Both
the curriculum reforms that have taken place and the development of recent
research require a more constructivist approach to learning and teaching in
this domain, which, however, is hardly seen in the music classrooms. We
could find many different causes for the resistance to this educational change
in music performance classrooms (e.g. Bautista and Prez Echeverra 2008,
Pozo et al. 2008, Torrado et al. 2005). Among them, this project focuses on
Table 1. Assumptions of the different theories about learning and teaching (adapted
from Bautista et al. in press).
Assumptions Direct theory Interpretive theory Constructive theory

Epistomological Knowledge (partial Knowledge reflects Knowledge is a con-
and/or complete) reality in a blurred struction elaborated
reflects reality in a and distorted way. by the student, who
direct way. builds own and per-
sonal models to in-
terpret reality.
Ontological Learning is con- Learning happens Knowledge is a con-
ceived in terms of over time, in a basic struction elaborated
states or static way, and it is con- by the student, who
products. ceived in terms of builds own and per-
processes. sonal models to in-
terpret reality.
Conceptual A direct and linear A direct and linear A complex and inter-
relation is estab- relation is estab- active relation is
lished between lished between established between
conditions and learning conditions, the three learning
learning outcomes. learning processes, components.
and learning
outcomes.
investigating the role of the conceptions that students have about teaching
and learning music, as an essential component of their learning practices.
Research on the conceptions that students have on how to learn is still a
nascent area, but remarkable importance has been obtained in the last two
decades thanks to the research on metacognition, theory of mind, strategies
and styles of learning, and conceptual change from the implicit theories in
various domains (Prez Echeverra et al. 2006). The studies carried out at the
Autonomous University of Madrid concerning these conceptions have shown
that achieving a constructivist approach requires a real conceptual change, a
new mentality in teachers and students. This means that the implicit direct
theory of learning (see Table 1), close to behaviorism, or interpretive, which
assumes a cognitive activity by the student but subordinate to the achieve-
ment of results or learning product previously defined, should be changed to
promote constructivist ideas, in which the students mental activity is not
only the engine but the goal of learning (Pozo et al., 2006).
METHOD
Participants
This study was carried out with 12 cello students aged 8-12 years (5 male, 7
female) who were in four different courses of basic level (BL) in Spanish con-
servatories. While we were carrying out this study, all participants developed
their activity in various official conservatories from the regions of Madrid and
Valencia.
Materials
We designed and implemented structured individual interviews to assess the

conceptions of children, through questions about processes, conditions, and
results (in line with Pozo 2008). Starting from the interviews used by Scheuer
et al. (2002, 2006) in their research with children, we designed different
tasks which were the goal of this preliminary investigation.
We prepared three different videos responding to cases of the three im-
plicit learning theories of teaching music. In these videos, a girl makes a pitch
mistake in her weekly cello lesson and her teacher reacts to the situation in
three different ways to help her, according to the epistemological, ontological,
and conceptual assumptions of the theory in question (see Table 1). After the
children watched the three videos, they were asked to choose both the worst
and the best way of teaching, as well as justify their choice.
Procedure
The four teachers of the 12 students were first asked about the students
availability, and the parents consented to record the interviews. After this, the
children were interviewed in their conservatory or at their homes, and the
interviews were recording with a video camera.
In order to describe qualitatively these students conceptions about cello
teachers, tasks of choice and rank provided data that enabled descriptive sta-
tistical analysis (taking as independent variables the developmental and edu-
cational level). The relationship between their responses and the theory
maintained was analyzed to check the consistency of these implicit theories.
RESULTS
While we should take the results with caution, because of the small sample
size, they indicate that the children in the 1st basic level (BL) chose the con-
structive teacher as the best, while children in the 2nd and 3rd BL preferred the
Table 2. Selection of the best and the worst videos of ways of teaching for students in
each basic level (BL).
Participants Direct Interpretive Constructive

1st BL (8, M) Worst Best
1st BL (8, M) Worst Best
1st BL (8, F) Worst Best
2nd BL (9, F) Best Worst
2nd BL (10, M) Best Worst
2nd BL (10, F) Best Worst
3rd BL (10, F) Worst Best
3rd BL (10, M) Worst Best
3rd BL (11, F) Best Worst
4th BL (11, F) Best Worst
4th BL (12, F) Worst Best
4th BL (12, M) Worst Best
direct or interpretive way of teaching. Finally, if the students in the 4th BL

chose the direct way of teaching as best, they also thought the constructive
was worst. Inversely, the students who chose the constructive did not like the
direct way.
As appears in Table 2, the constructive method was chosen least as the
worst method, while the direct method was chosen most as the worst. Ac-
cording to the childrens responses and choices, it seems that they think both
direct and constructive theories of teaching and learning could be good for
learning, and their reasons for this are simple. For example, in their opinion,
a good direct teacher is demanding of their students and uses techniques such
as repetition and correction as teaching methods.
Asking the students to play it, to play it, and, and, and, if they make mis-
takes, ordering them to repeat it until things work out more or
lessdemanding from them, because the mistakes should be mentioned,
and if they make mistakes again, they should be asked to repeat again,
and sometimes, if necessary, he may play with students.
When talking about why they felt the constructive theory was best, they
said that a teacher using this method should involve the student in his own
learning process in an active and reflective way.
He shouldnt mention the error if you make a mistake, or mention what

the error was, or how the error was made, but just talk with you, if a
teacher doesnt let you figure out the mistake you made and the way to
correct it.
Because the teacher let the girl think, which was a good idea, and, and
the girl will do so at homebecause she helped the girl think of what
happened and, and analyze what she can do to correct herself. In fact,
you learn a lot in this wayin order that she can realize her errors, and it
was the method the teacher used.
Finally we asked about the worst theory, most students immediately cited
the direct method. They felt this method could obtain results but that the
student would not necessarily understand the process to achieve them.
She ends up doing itbecause finally she managed to do it due to the

repetition, but its not, not the best solution she can find. She may be
gaining bad habits at home because theres no one who can tell her,
wrong, repeat 5 times again. Well I dont know, but I think that it is the
worst, at least for me, particularly for other students, but, because a
teacher can be there without doing anything, but, always, always in some
cases he does so.
DISCUSSION
We would need to interview more children in order to confirm these findings,

but the trends suggest that children arrive at the conservatory with construc-
tive ideas, and then, in advanced courses, they opt for simpler, less desirable
designs, focusing primarily on learning outcomes. Such conceptions evolve
gradually according to the educational-evolutionary variable, reaching again a
more complex theory in the more advanced grades. Therefore, teachers
should consider what to prioritize in their lessons and what is happening in
the classroom when they have a student who comes with previous construc-
tive ideas. Finally, they should observe if a student loses or modifies that idea
when he/she comes into contact with the education center.
Presently, we are working on refining the tasks and extending the sample
to compare children educated according to different types of practices.
Acknowledgments
The authors are grateful for the assistance from the project SEJ2006-15639-C02-01,
whose principal researcher is M del Puy Prez. We also express our gratitude to Jos A.
Torrado, Amalia Casas, and Cristina Marn who have collaborated in the development of
this research, giving very helpful suggestions and insightful comments.
Guadalupe Lpez iguez, Department of Cognitive Psychology, Autonomous University

of Madrid, C/ Ivan P. Pavlov 6, Campus de Cantoblanco, Madrid 28049, Spain; Email:
guadalupe.lopez@uam.es
References
Bautista A. and Prez Echeverra M. P. (2008). Qu consideran los profesores de in-

strumento que deben ensear en sus clases? Cultura y Educacin, 20, pp. 17-34.
Bautista A., Prez Echeverra M. P., and Pozo J. I. (in press). Music performance teach-
ers conceptions about learning and instruction: A descriptive study of Spanish pi-
ano teachers. Psychology of Music.
Prez Echeverra M. P., Mateos M., Scheuer S., and Martn E. (2006). Enfoques en el
estudio de las concepciones sobre el aprendizaje y la enseanza. In J. I. Pozo, N.
Scheuer, M. P. Prez Echeverra et al. (eds.), Nuevas Formas de Pensar la En-
seanza y el Aprendizaje (pp. 55-93). Barcelona, Spain: Gra.
Pozo J. I. (2008). Aprendices y Maestros. La Psicologa Cognitiva del Aprendizaje.
Segunda Edicin. Madrid: Alianza.
Pozo J. I., Bautista A., and Torrado J. A. (2008). El aprendizaje y la enseanza de la
interpretacin musical. Cultura y Educacin, 20, pp. 5-15.
Pozo J. I., Scheuer N., Mateos M., and Prez Echeverra M. P. (2006). Las teoras im-
plcitas sobre el aprendizaje y la enseanza. In J. I. Pozo, N. Scheuer, M. P. Prez
Echeverra et la. (eds.), Nuevas Formas de Pensar la Enseanza y el Aprendizaje
(pp. 95-132). Barcelona, Spain: Gra.
Scheuer N., de la Cruz M., and Pozo J. I. (2002). Children talk about learning to draw.
European Journal of Psychology of Education, 17, pp. 101-114.
Scheuer N., de la Cruz M., Pozo J. I., and Neira S. (2006). Childrens autobiographies as
learners of writing. British Journal of Educational Psychology, 76, pp. 709-725.
Torrado J. A., Casas A., and Pozo J. I. (2005). Las culturas de la educacin musical.
Estudios de Psicologa, 26, pp. 259-270.
The use of musical scores in order to perform:

An exploratory study with flute players
Cristina Marn Oller1, Puy Prez Echeverra1, and Susan Hallam2
1 Department of Cognitive Psychology, Autonomous University of Madrid, Spain

2 Institute of Education, University of London, UK
Sixteen flute students at two levels of expertise from conservatories in

Madrid participated in an individual semi-structured interview designed
to explore the activities they carry out when they learn a new piece of
music, as well as the elements of the musical score that they work on. The
learning process was divided into three stages: beginning, middle, and
end. Twelve categories of analysis were developed and chi-square was
used. Differences in responses were noted between the students at each
level of expertise for each stage of practice. Findings are discussed in re-
lation to previous research in development of musical expertise.
Keywords: musical score; expert; novice; instrumental learning; flute
In the last decades there has been an increasing research interest in the de-
velopment of expertise in musicians through the ways they learn and practice
music. The musical score, a main element in most of the Western music
learning processes taking place in conservatories, constitutes a semiotic sys-
tem such as texts, graphs, maps, and numbers. These systems are powerful
tools through which we develop and accumulate knowledge. Drawing on evi-
dence from different fields, researchers have described different levels of
comprehension relating to some representational systems (see Bautista and
Prez Echeverra 2008, Friel et al. 2001, Kintsch and van Dijk 1978). Studies
of the development of interpretation suggest that some musicians plan inter-
pretation at the outset, primarily letting the expressive ideas guide the techni-
cal work (Chaffin et al. 2003), while others develop a performance plan after
mastering most of the technical challenges (Nielsen 2001). Although practic-
ing notes and difficult passages are important aims for students at all levels,
the experts are also interested in expressing a musical idea of the piece from
the beginning of learning it, as professional musicians state they do (Chaffin
and Imreh 2001). Novice students are mainly focused on reading and per-
forming the explicit symbols presented on the score (Gruson 1988, Hallam
1994).
The aim of this research was to explore how Spanish flute students at two
different levels of expertise used the musical score during their normal
learning practices outside the lessons and the nature of the activities they
undertook.
METHOD
Participants
Sixteen flute students from five different conservatories in Madrid (four in-
termediate and one tertiary) took part voluntarily in the study. They consti-
tuted two groups of different levels, each composed by eight students (five
females and three males). The Intermediate (I) group (mean age=21.1
years), was in its tenth year in the Spanish specialized musical education sys-
tem, which is the last year before starting the music performance degree. The
Tertiary (T) group (mean age=24.4 years), was composed of flutists in the
last two years of the music performance degree.
Materials
An individual interview was carried out with each participant. They were
asked to describe the usual study process that they follow when learning a
new piece of music. The interview was designed to explore the activities that
the students do when practicing a new piece, as well as the elements of the
score in which they focus their attention throughout their practice. The proc-
ess was divided into three stages, corresponding to the beginning, the middle,
and the end of the process. We suggested to the participants that they imag-
ined they were preparing the piece to perform in a concert. In order to give
them a context to explain the process, we used the piece An Evening in the
Village by Bla Bartk (Sz. 39), in an arrangement for flute and piano.
Procedure
Participants were asked to read Bartks piece in order to get an overview of

it. This took around seven mins. After, they were interviewed individually for
approximately 10-15 mins. Data were audio recorded and coded into themes
using an iterative process described by Tesch (1990) as empirical phenome-
nology. A double category system was produced, which is presented in Table
1.
Table 1. Category systems.
Category Definition
System A: Practicing
Reading Mentioning any kind of entire reading of the score.
Passages Mentioning any group of bars that seem difficult for the students, to
practice them separately.
Recording Mentioning any kind of activity using a recording of the piece, but
not when referring to record himself/herself.
Pianist Mentioning any activity to be done with the piano accompanist.
Speed Mentioning any activity that involves modifying speed of the score.
Memory Mentioning any process of memorizing the piece.
System B: Score or instrumental elements

Notational level Mentioning any explicit mark on the score, apart from dynamics and
regulators (rhythms, notes, accents).
Dynamics Mentioning dynamics and regulators.
Syntactic level Mentioning any term that implies to relate two or more elements
from the scorefor example, melody, structure, motif, phrase, etc.
Expressivity Mentioning expressive terms, about character or musicality.
Style/composer Mentioning something about composer, style and/or historic period.
Skills Mentioning the practice of skills that cannot be included in any of
previous categories, like pitch, breathing, sound, etc.
System A refers to activities to carry out when practicing a piece. System

B refers to elements from the musical score or to the instrument. Three inde-
pendent judges categorized 20% of the total responses. Inter-rater agreement
level was calculated by Cohens Kappa coefficient, which ranged from 0.75-
1.00. We use statistic chi-square to analyze differences between groups for
each category at each stage.
RESULTS
In the beginning stages (see Figure 1), the I group reported being mainly
focused on reading activities, listening to recordings, and working on specific
passages. References to musical elements included focus on the notes,
rhythms, and indications about tempi. Students from the T group referred
to reading processes and to selecting and working on difficult passages. In
contrast to the I group, they referred to activities related to syntax such as
Reading
6
Passages
Recording
5
Pianist
Speed
4 Memory
Notational Level
Dynamics
3
Syntactic Level
Expressivity
2 Style/Composer
Skills
1
0
1 2
Figure 1. Categories reported in the beginning of the process (1=intermediate group, 2=

tertiary group).
Reading
6
Passages
Recording
5 Pianist
Speed
4 Memory
Notational Level
Dynamics
3
Syntactic Level
Expressivity
2
Style/Composer
Skills
1
0
1 2
Figure 2. Categories reported in the middle of the process.
Reading
6
Passages
Recording
5 Pianist
Speed
4 Memory
Notational Level
Dynamics
3
Syntactic Level
Expressivity
2
Style/Composer
Skills
1
0
1 2
Figure 3. Categories reported at the end of the process.
identifying the structure of the piece, main themes, etc. These differences
were statistically significant (X2=6.35, p=0.012). In the middle stages, most
students in the I group (88%) mentioned expressivity, and there was an
increasing focus on dynamics. The T group demonstrated a less marked
change from the beginning stage than the I group. Reading activities de-
creased with a greater emphasis on working at different speeds, developing
skills, and focusing on specific passages and dynamics (see Figure 2). In the
end stage, the I students continued to focus on expressivity and dynamics.
The T group referred to more activities (15 statements) than musical ele-
ments at this stage (5 statements). They seemed to return to a consideration
of the piece as a whole with a reduction in the focus on specific elements from
the score (see Figure 3).
DISCUSSION
The findings support those from earlier research demonstrating changes in

the use of practicing strategies as expertise develops (Gruson 1988, Hallam
2001). The most expert participants approximated the pattern of learning
reported by professional musicians, getting an initial overview of the work,
identifying difficult passages and focusing on them, then as the performance
neared focusing on the piece as a whole (Hallam 1995, Chaffin et al. 2003).
Lane (2006) conceptualizes this as moving from macro to micro and return-
ing to macro. Nonetheless, this pattern is not too clear and would need to be
checked in further studies using a broader sample.
It is also relevant that the T group establishes relations among elements
present on the score, which can be noticed by the presence of the syntactic
level from the first stage of practice, whereas participants from the I group
are mainly focused on notational elements at the beginning, moving from
there to expressivity in the next stages. Although neither of these two groups
can be considered novice, the high frequency of the notational level is rele-
vant, mainly in the early stages of the learning process. According to previous
research, experts are focused on expressivity from the beginning (Chaffin and
Imreh 2001, Chaffin et al. 2003). This presence of notational level can be
understood as the need to practice the elements on the score, which is a tech-
nical issue not related with the comprehension of the piece.
There are limitations in the extent to which the findings from this study
can be generalized. It was an exploratory study based on a small sample of
students, representing expertise on a single instrument, and who had under-
taken their studies in a specific learning environment. In addition, the find-
ings are based on self-report rather than direct observation of behavior itself.
However, there is no reason to suppose that the conceptual strategies that the
students reported adopting are not similar to those of other students studying
in Western conservatories. Using categories as those developed in this re-
search and studying the relationships established among them in each stage
of practice can provide further evidence about comprehension and use of the
musical score in instrumental learning processes.
Acknowledgments
This study was supported by the project SEJ2006-15639-C02-01, directed by Puy Prez
Echeverra. We thank the participants of this study and their teachers.
Cristina Marn Oller, Department of Cognitive Psychology, Autonomous University of

Madrid, Calle Ivn Pavlov 6, Madrid E-28049, Spain; Email: cristina.marin@uam.es
References
Bautista A. and Prez Echeverra M. P. (2008). Qu consideran los profesores de in-

strumento que tienen que ensear en sus clases? Cultura y Educacin, 20, pp. 17-
34.
Chaffin R. and Imreh G. (2001). A comparison of practice and self-report as sources of
information about the goals of expert practice. Psychology of Music, 29, pp. 39-69.
Chaffin R., Imreh G., Lemieux A., and Chen C. (2003). Seeing the big picture: Piano
practice as expert problem solving. Music Perception, 20, pp. 465-490.
Friel S. N., Curcio F. R., and Bright G. W. (2001). Making sense of graphs: Critical fac-
tors influencing comprehension and instructional implications. Journal for Re-
search in Mathematics Education, 32, pp. 124-158.
Gruson L. M. (1988). Rehearsal skill and musical competence: Does practice make per-
fect? In J. A. Sloboda (ed.), Generative Processes in Music (pp. 91-112). Oxford:
Hallam S. (1994). Novice musicians approaches to practice and performance: Learning
new music. Newsletter of the European Society for the Cognitive Sciences of Music,
6, pp. 2-10.
Hallam S. (1995). Professional musicians orientations to practice: Implications for
teaching. British Journal of Music Education, 12, pp. 3-19.
Hallam S. (2001). The development of expertise in young musicians: Strategy use,
knowledge acquisition and individual diversity. Music Education Research, 3, pp.
7-23.
Kintsch W. and van Dijk T. A. (1978). Toward a model of text comprehension and pro-
duction. Psychological Review, 85, pp. 363-394.
Lane J. S. (2006). Undergraduate instrumental music education majors approaches to
score study in various musical contexts. Journal of Research in Music Education,
54, pp. 215-230.
Nielsen S. G. (2001). Self-regulating learning strategies in instrumental music practice.
Music Education Research, 3, pp. 155-167.
Tesch R. (1990). Qualitative Research. Hampshire, UK: The Falmer Press.
Pirouetting with pain: Attitudes surrounding

female ballet dancers dancing with pain
Rosemary Martin
Dance Studies, National Institute of Creative Arts and Industries,

While female dancers performing classical ballet are expected to present

an image of effortless grace, the actuality is often a much more painful
experience. The current study examines questions related to this experi-
ence: what do dancers understand pain to be, why do they dance in pain,
when should they stop dancing in pain, and how do they ultimately
communicate this pain to others? This ethnographic research collects the
views of current dancers, ex-dancers, dance teachers, choreographers,
and artistic directors through in-depth individual interviews. The col-
lected interviews are analyzed using Foucauldian perspectives of disci-
pline, social hierarchies, power, knowledge, and discourse, along with
feminist theory, prevalent physiological conceptions of pain, and con-
temporary theories of dance pedagogy, ethnography, and subjectivity. It
is revealed through the interviews that pain is often a part of the dancers
life, a close acquaintance constantly watching over the dancers shoulder.
Pain is also seen to be instigated, influenced, and reinforced by the aes-
thetics, technique, institutions, and culture of ballet. The collected nar-
ratives demonstrate the highly subjective nature of pain and that the
balletic environment still embrace a no pain, no gain mentality.
Keywords: ballet; pain; injury; aesthetics; ethnography
This research is driven by the key question: within the context of professional
ballet, what are the attitudes surrounding female ballet dancers dancing in
pain? The research focuses on the attitudes toward pain from the perspectives
of dancers, ex-dancers, teachers, choreographers, and artistic directors, with
my own personal experience also informing the research.
The term pain is complex, with multiple interpretations and meanings.
Pain is defined in this research as physical pain caused by noxious stimulus
or bodily harm (Melzack and Wall 1965, Woolf 1995, Loeser and Melzack
1999). This definition focuses on the physiological pain experience, rather
than the psychological pain experience. Defining pain in this way is not to
ignore how pain is also intrinsically linked and influenced by various psycho-
logical, environmental, and cultural factors. This research aims to draw on
both a biomedical model of pain, and what could be described as a contextu-
alist approach (Loland 2006) of how pain is interpreted and shaped within
the social and cultural contexts in which people live.
The pain/ballet relationship can be seen emerging in academic scholar-
ship, scientific research, autobiographical accounts, and performances, with a
growing body of literature and research that investigates wider issues related
to dancers bodies, injuries, and health. Some notable quantitative research
has investigated dancers pain (e.g. Anderson and Hanrahan 2007, Encarna-
cion et al. 1999, Krasnow et al. 1994, Tajet-Foxell and Rose 1995). This re-
search, although predominantly through quantitative investigation, is
intended to provide an understanding and awareness that dancers are experi-
encing pain.
Significant qualitative readings of subjective scholarship has surrounded
the dancers body and aesthetics (e.g. Abra 1987-88, Foster 1996, 1997, Tho-
mas 2003, Wainwright and Turner 2003a, 2003b, 2004, 2006, Wainwright
et al. 2005). This qualitative research has illuminated attitudes surrounding
dancers dancing with pain, though none of the above authors focus specifi-
cally on dancers and their attitudes toward the pain experience, which pro-
vides this research with a justification to address such a potential gap in the
literature.
Krasnow et al. (1994) conducted research which demonstrated that most
dancers do not seek medical treatment for injuries, with fewer than 50% of all
dancers injuries being treated by medical practitioners. There are various
possible explanations for why dancers are hesitant to seek professional help
for injuries, ranging from financial constraints, pressures from teachers, ar-
tistic staff or directors, as well as themselves, to continue dancing, and feeling
that there is a lack of understanding from medical practitioners about their
situation (Krasnow et al. 1994). Consequently, injured dancers may often rely
on the advice from teachers and other dancers regarding how to treat and
deal with pain and injury (Krasnow et al. 1994). Research has also queried
whether dancers really can tell the difference between different types of pain
(Anderson and Hanrahan 2007). Anderson and Hanrahan have found that
dancers will endure performance pain or injury pain, and the type of pain
experienced does not influence the cognitive appraisal made or the pain cop-
ing strategies used to manage the pain.
MAIN CONTRIBUTION
I am a dancer, dancing makes me who I am. Without it I wouldnt know

who I was. I would feel like I would have lost my identity or at least a huge
part of it (Zoe). The use of the term identity has many and varied inter-
pretations and meanings (Parfit 1971, Castells 2004). Identity can be de-
scribed as peoples source of meaning and experience (Castells 2004, p. 6).
More specifically, ones own personal self-identity could be explained to be
what makes each person uniquely who they are (Parfit 1971, p. 17), and
further, our identity can be shaped and influenced by experiences, memories,
and society (Parfit 1971, Wainwright and Turner 2004).
The subject of identity was an issue I did not foresee emerging from the
interviews, though it was a discovery which gives context and understanding
to the particular attitudes expressed by the interviewees. The dancers inter-
viewed primarily identified themselves as dancers and expressed that the act
of dancing contributed heavily to their identity. An interesting association
emerged where dancers also often identified themselves with athletes in rela-
tion to their bodily experiences, understandings, and dealings with pain.
Emily explains, the thing is we are athletes, there is no difference.
All participants interviewed stated that they had danced in pain before.
Some explained that they danced with some level of pain every day, such as
Amy who states, I would say every day in class or performance there is some
sort of pain and something not quite right.... Others stated that they did not
experience pain every day but only when their bodies had been overworked,
they were performing new choreography, or they were managing or experi-
encing an injury.
The dancers interviewed often perceived pain as a purely physiological
event, which was something they experienced in their body only, not in their
mind. This raises the Cartesian dualist notion of mind/body separation, that
the body is an object controlled by ones mind (Grosz 1994). This concept was
carried throughout many of the dancers perceptions and understandings of
what they as dancers have to do, who they are as people and how they should
respond to various pain situations. Several of the dancers revealed that they
chose to ignore pain, stating that I just block it out (Amy) or I dont
choose to have it (Tatiana). Such mind over matter in the ballet class or
performance appeared to be entrenched in their thinking, with some of the
dancers stating that they selected to ignore their thoughts and feelings of
pain.
All interviewees voiced an understanding of the significance of the ballet
aesthetic in relation to painthat having the desired body type of long limbs,
a well proportioned body, arched feet, and being slim yet strong, made the
technique of ballet somewhat easier to physically perform and more aestheti-
cally pleasing to the eye. Lisa, a dance teacher, explains this further: of
course if helps to have the right body, it is a visual art form, and I think an
audience wants to see a certain aesthetic. [Ballet] can be so much easier
when things like turn-out are naturally there, the right balance of strength
and flexibility. Foucaults (1977a) notion of normalization of the body,
through social practices and institutional structures and regimes, could be
applied to the experience dancers have in trying to normalize their body into
the ballet aesthetic.
From speaking to the interviewees it can been seen that pain is something
that is part of their daily lives, so much so that many consider pain to be nor-
mal; yet most dancers felt uncomfortable to disclose or discuss pain openly.
The lack of disclosure around the subject of pain was expressed by Kerry who
said, we dont discuss it, talking about it ruins the illusion that we work so
hard for; talking about it puts us at risk of jeopardizing our opportunities,
or as Amy explains, it goes without saying that ballet hurts in some way at
sometime, talking about it isnt going to change that; it will just make us look
weak.
The dancers, ex-dancers, and teachers and choreographers all spoke of
the pressures from the ballet culture that contributed to dancers postponing
the decision to seek treatment for pain or have time off dancing to let the in-
jury causing the pain to heal. These included practical constraints of time,
scheduling, and the need to produce a performance. Pressures felt by the
dancers came from the company management, teachers, repertoire, and
peers. This led to the feeling that their position in the company was not se-
cure and that they were easily replaceable; moreover, time off from dancing
was often seen as essentially wasting time, jeopardizing their career pro-
gression. Foucault (1977b) describes culture as a hierarchical organization of
values, accessible to everybody, but at the same time the occasion of a
mechanism of selection and exclusion (p. 173). The ballet culture and hierar-
chy of the ballet company, with its own system of beliefs and structures, can
shape how and when dancers decide to stop dancing with pain through the
surveillance and disciplinary power it holds as an institution (Foucault
1977a).
IMPLICATIONS
From the analysis of the interviews and the narratives collected, it is apparent
that dancers are dancing with pain. They are dancing for prolonged periods of
time with pain. They often do not know when to stop dancing with pain and
are fearful of the repercussions stopping dancing may have on their career
and identity. In a dance style that is evolving, extending and creating new
physical boundaries, while also trying to maintain and adhere to the structure
and form of the classical aesthetic, do these attitudes surrounding female
dancers dancing in pain need to change? It is one thing to acknowledge that
there are particular attitudes towards pain in the ballet environment; it is
another to demand that these attitudes must change. From this research it
can be concluded that if we would like dancers to have longer and healthier
careers with less physical pain, then yes, these attitudes will have to shift. The
question of how this shift might occur within the balletic culture, which holds
power over the dancers bodies and minds, imposing constraints, prohibitions
and obligations, is something that will take more than a quick fix solution
and will require much research and investigation.
This research has not attempted to uncover one particular truth or prove
a certain hypothesis; rather it has aimed to further open up a dialogue about
ballet and pain, and through the collected ethnographical interviews and
analysis, present multiple accounts and experiences. Ballet is not always easy;
achieving the physical form it requires, and the mindset that allows this form
to take place, may not always be pain free. Yet by beginning to understand
how and why dancers feel the need to excessively conceal and silence pain, to
dance through it, ignore it, even crave it, rather than listening to their bodies,
there is potential to offer alternatives and directions to allow them to experi-
ence less pain while dancing.
Rosemary Martin, Dance Studies, National Institute of Creative Arts and Industries,
University of Auckland, Building 421, Level 5, 26 Symonds Street, Auckland 1010, New
Zealand; Email: rmar161@aucklanduni.ac.nz
References
Abra J. (1987-88). The dancer as masochist. Dance Research Journal, 19, pp. 33-39.
Anderson R. and Hanrahan S. (2007). Dancing in pain: Pain appraisal and coping in
dancers. In R. Solomon and J. Solomon (eds.), Proceedings of the Seventeenth An-
nual Meeting of the International Association for Dance Medicine and Science (pp.
245-248). Canberra, Australia: International Association for Dance Medicine and
Science.
Castells M. (2004). The Power of Identity. Oxford: Blackwell.
Encarnacion M., Meyers M., Ryan M., and Pease D. (1999). Pain coping styles of ballet
performers. Medicine, Science in Sport and Exercise, 31, pp. 217.
Foster S. (1996). The ballerinas phallic pointe. In S. Foster (ed.), Corporealities (pp. 1-
24). London: Routledge.
Foster S. (1997). Dancing bodies. In J.Desmond (ed.), Meaning in Motion (pp. 235-
259). Durham, North Carolina, USA: Duke University Press.
Foucault M. (1977a). Discipline and Punish. New York: Vintage.
Foucault M. (1977b). Nietzsche, genealogy and history. In D. Bouchard (ed.), Language,
Counter-memory, Practice (pp.139-164). Ithaca, New York, USA: Cornell Univer-
sity Press.
Grosz E. (1994). Volatile Bodies. Bloomington, Indiana, USA: Indiana University Press.
Krasnow D., Kerr G., and Mainwaring L. (1994). Psychology of dealing with the injured
dancer. Medical Problems of Performing Artists, 9, pp. 7-9.
Loeser J. D. and Melzack R. (1999). Pain: An overview. Lancet, 353, pp. 1607-1609.
Loland S. (2006). Three approaches to the study of pain in sport. In S. Loland, B.
Skirstand, and I. Waddington (eds.), Pain and Injury in Sport (pp.49-62). New
York: Routledge.
Melzack R. and Wall P. D. (1965). Pain mechanisms: A new theory. Science, 150, pp.
971-979.
Parfit D. (1971). Personal identity. The Philosophical Review, 80, pp. 3-27.
Tajet-Foxell B. and Rose F. D. (1995). Pain and pain tolerance in professional ballet
dancers. British Journal of Sports Medicine, 29, pp. 31-34.
Thomas H. (2003). The Body, Dance and Cultural Theory. New York: Palgrave Mac-
Millan.
Wainwright S. and Turner B. (2003a). Reflections on embodiment and vulnerability.
Journal of Medical Ethics: Medical Humanities, 29, pp. 4-7.
Wainwright S. and Turner B. (2003b). Corps de ballet: The case of the injured ballet
dancer. Sociology of Health and Illness, 25, pp. 269-288.
Wainwright S. and Turner B. (2004). Epiphanies of embodiment: Injury, identity and
the balletic body. Qualitative Research, 4, pp. 311-337.
Wainwright S. and Turner B. (2006). Just crumbling to bits? An exploration of the
body, ageing, injury and career in classical ballet dancers. Sociology, 40, pp. 237-
255.
Wainwright S., Williams C., and Turner B. (2005). Fractured identities: Injury and the
balletic body. Health, 9, pp. 49-66.
Woolf C. J. (1995). Somatic pain: Pathogenesis and prevention. British Journal of An-
aesthesia, 75, pp. 169-176.
Cardio-respiratory responses to expressiveness

in piano performance
Hidehiro Nakahara1,2, Shinichi Furuya2, Satoshi Obata2, Peter R. Francis3,

and Hiroshi Kinoshita2
1 Faculty of Health Sciences, Morinomiya University of Medical Sciences, Japan

2 Graduate School of Medicine, Osaka University, Japan
3 Department of Exercise and Nutritional Sciences, San Diego State University, USA
This study examined selected autonomic and cardio-respiratory re-

sponses of nine pianists during solo performances of the same single mu-
sical piece. The subjects performed the piece with and without self-
perceived emotional expression, and with and without free ancillary body
movements during expressive performance. Autonomic nervous system
and cardio-respiratory parameters were continuously monitored during
all experimental conditions. These parameters were heart rate (HR),
sweating rate, the root mean square of successive difference (RMSSD) of
heart rate variability, and respiratory measurements such as oxygen con-
sumption (VO2), minute ventilation, tidal volume, and respiratory rate.
Kinematics of the trunk and arms were recorded during all conditions.
The expressive condition had significantly higher levels of HR, sweating
rate, minute ventilation, and tidal volume, and lower levels of RMSSD
and respiratory rate than the non-expressive condition. No difference
was found for VO2 between these conditions. The expressive condition
with ancillary body movements did not significantly differentiate any of
the physiological measures except for respiratory rate from those without
such body movements. These findings suggested that expressive musical
performance could modulate the emotion-related autonomic and cardio-
respiratory responses that are independent of the effect of physiological
load due to expressive ancillary body movements in playing the selected
music on the piano.
Keywords: expressive piano performance; heart rate; heart rate variabil-

ity; autonomic nerve system; respiratory measurements
Hidehiro Nakahara, Faculty of Health Sciences, Morinomiya University of Medical

Sciences, 1-26-16 Nankokita Suminoe, Osaka, Osaka 559-8611, Japan; Email: nakahara
@morinomiya-u.ac.jp
The Listening Gallery: Integrating music with

exhibitions and gallery displays
Giulia Nuti1, Ashley Solomon2, and Aaron Williamon1

2 Department of Historical Performance, Royal College of Music, London, UK
The Listening Gallery is a collaboration between the Royal College of Mu-

sic (RCM) and the Victoria and Albert Museum (V&A), the UKs National
Museum of Art and Design. Stemming from recent research in music, art,
design, and technology, the project connects objects in the V&As collec-
tions with music that shares their rich and distinctive pasts. Specifically,
new and existing recordings of music have been integrated into the V&As
spring 2009 exhibition, Baroque 1620-1800: Style in the Age of Mag-
nificence, and into the museums newly renovated Medieval and Renais-
sance Galleries. The impact of the project has been far reaching, as a
novel approach to music in museums is achieved: the choice of pieces is
underpinned by musicological research, performance follows the prac-
tices of the time, the instruments used are originals or faithful copies,
and the provenance of the pieces is described. This article focuses on four
new recordings that were made for the project. The objects with which
the music is associated are briefly described, the connection with the mu-
sic is explained, and details of the recorded pieces are given.
Keywords: museum; performance practice; medieval and renaissance art

and music; recordings; historical instruments
Music has played a central role in popular and elite culture throughout Euro-
pean history, spanning sacred and secular spaces and representing private
and communal experiences. Yet there was, until now, no museum of art and
design that connected music and its performance with other art forms and
objects in interactive and historically meaningful ways.
The Listening Gallery is a project in which new and existing recordings of
music have been integrated into two major exhibitions at the V&A:
Baroque 1620-1800: Style in the Age of Magnificence, a temporary exhi-

bition in Spring 2009, which has subsequently travelled to other interna-
tional venues
Medieval and Renaissance Europe, a series of eleven newly refurbished
permanent galleries, which re-opened in December 2009
The exhibitions provided an opportunity to incorporate music into gal-

leries covering sacred and secular, northern and southern European art, be-
tween 1100-1800. The unparalleled variety of musical styles, genres, and
instruments covered across this period required musicological knowledge
drawn from a wide range of sources (e.g. Medieval: Testi 1969; Renaissance
Gallico 1978, Fenlon 2002; Baroque: Fortune 1987, Bianconi 1991). In addi-
tion, the input provided by performers for the recording of new works proved
essential, particularly for the earliest works where hands-on music making is
often the best means of addressing interpretative questions (Nuti 2007).
The Listening Gallery demonstrates how the past, so clearly portrayed in
the V&As galleries, is to be discovered in music as well as other arts by pro-
viding music appropriate to the objects on display. Historical events, patron-
age, and fashions all contribute to changes in style that are reflected as much
in music as they are in art and sculpture.
MAIN CONTRIBUTION
The expertise in music, history, and art that the project has brought together
has given rise to new knowledge on performance. The examples of new re-
cordings provided below are presented under separate headings, either Ba-
roque 1620-1800 or Medieval and Renaissance Europe. The rooms in which
the music is heard, as well as the objects with which the music is associated,
are briefly described in order to place the music in context.
Baroque 1620-1800
The magnificence and splendour of Baroque, one of the most opulent styles of
the seventeenth and eighteenth centuries, was the subject of the V&As 2009
spring exhibition.
A recording of Davide Perez (1711-1778), Domine Deus, from the Messa con 5
strumenti, c.1750
The Sacred Spaces section of the exhibition was dedicated to religious devo-
tion and the ritual of Mass. Displays included Portuguese artefacts made in
Rome, for consumption in Portugal, during the reign of John V. The Model of
the Chapel of St John the Baptist in the Church of So Roque, Lisbon (Rome,
1742-44), by Giuseppe Palmes, after drawings by Luigi Vanvitelli (1700-73), is
a testimony to the characteristic of artefacts built in Portugal during this pe-
riod: the influence of Italian style and artists. This became the focal point
around which the music was chosen.
In 1743-44, the Portuguese king John V ordered a team of artists based in
Rome to build, decorate, and fit out a royal chapel for the Jesuit church of So
Roque in Lisbon. In 1751, John V called the Neapolitan composer Davide
Perez to Lisbon. Under his influence, Portuguese musicboth sacred and
secularbecame entirely dominated by the Italian operatic style, as is imme-
diately apparent upon hearing this recording.
The music associated with this splendid altarpiece highlights the stylistic
influences between Rome and Lisbon, felt as strongly in music as in art and
architecture, and the influence of Italian opera in Portuguese music post-
1750, including sacred music.
The manuscript of Perezs Mass is held in the RCM Library, MS 977. The
Domine Deus was recorded in the RCM Studios in February 2009, with the
orchestra and five soloists directed by Ashley Solomon. It is the only existing
recording of this piece.
Domestic music for spinet and other instruments: A recording of RCM MS

2093
As well as the music for great religious occasions and occasions of state, the
baroque period enjoyed performance of more intimate music for the
delectation of a smaller audience; lesser households, too, could command
music as integral to that richness of lifestyle.
The Secular Spaces section of the exhibition, in which the recording of
music for spinet and other instruments could be heard, focused on the most
intimate part of a household: the bedroom and closet. Beyond the public
rooms of a house, in which guests were welcomed, the bedchamber was open
only to the select few. Into the inner sanctum of the dressing room and closet
was packed a level of furnishing and luxury unseen outside. The exclusivity of
these rooms and their relative lack of ceremonial function also made them the
setting for influential innovations in furniture and interior decoration. Among
the objects that would have been found in these most intimate rooms is the
spinet, a small keyboard instrument of the harpsichord family.
The main instrument used in this recording was built in London in the
1680s by Stephen Keene, preserved in the RCM Museum of Instruments
(RCM 179). The instruments case is made of walnut wood, and there is a spe-
cial decorative laburnum wood veneer on the panel above the keyboard.
The music in this recording is taken from a single, late-seventeenth cen-
tury English keyboard manuscript belonging to the RCM Library, MS 2093.
Although its ownership is uncertain, the manuscript is typical of the virginal
books used to tutor players, especially women, at the time. It is remarkable
for two reasons. Firstly, its contents are divided into two collections, one of
preludes and the other of fugues. Secondly, it preserves pre-civil war music by
Byrd, Bull, and Weelkes, alongside later Baroque composers such as Locke
and Blow.
The recording of RCM MS 2093 was made in the RCM Museum in Febru-
ary 2009, with RCM students of harpsichord, violin, and recorder performing
under the direction of Terence Charlston. This is the first recording of the
manuscript, an important exemplar of virginal music of the period, as well as
a testimony of the intimate type of music that would have been heard in a
seventeenth century English home.
Medieval and Renaissance Europe
Medieval and Renaissance Europe is a major re-interpretation of a series of

eleven permanent galleries in the V&A that opened in December 2009. The
galleries will have a minimum lifespan of 25 years. The museums collection is
one of the finest in the world and covers European art and architecture from
800 to 1600.
A Missal from St Denis, Paris, c.1350
A missal is a book that contains the texts and music needed by a priest and
choir to celebrate Mass. This highly decorated missal was probably made
around 1350 in Paris for use at one of the altars at the royal abbey of St Denis.
Research into the service books and the liturgical life of the abbey (Robertson
1991) led to the identification, performance, and recording of the Dionisian
sequence Salve Pater Dyonisi, which would have been performed on the
Feast Day of St Denis.
Many recordings of Gregorian chant are available commercially, yet this is
the first recording to be made of Salve Pater Dyonisi, or indeed any part of
the St Denis Missal. There are seven verses that praise St Denis and his two
co-martyrs, St Rusticus and St Eleutherius. The words are sung in Latin to a
tune that was borrowed and adapted from pre-existing pieces to create a fresh
and new work. The recording was made in June 2009 at the Parish Church of
Our Lady of Mount Carmel and Simon Stock, London. Five male voices were
used, under the direction of Jennifer Smith.
Notation knives
One of the rarest sixteenth-century objects on display in the new Medieval

and Renaissance Galleries at the V&A is both a knife and a piece of music. On
one side of the broad blade of this hybrid exhibit is a clear, perfectly notated
musical inscription of a blessing of the table, to be sung before a meal; on the
other, a prayer giving thanks, to be sung when the meal has ended. Although
we do not know where or for whom this curious object was made, it reveals a
fascinating relationship between music and the material culture of the table
during the sixteenth century, and represents the important formal role played
by music in the rituals of dining (Dennis in press).
As part of the project, the beautiful Grace and Benedictus engraved on
this unusual notation knife were transcribed and recorded. It is the only ex-
isting recording of these blessings, or indeedand surprisinglyof any of the
music that is not infrequently found on Renaissance household objects such
as cutlery and maiolica plates.
Four singers were recruited to sing a cappella for this recording; RCM
students, under the direction of Giulia Nuti, recorded the notation knives in
June 2009 in the RCM Studios.
IMPLICATIONS
The Listening Gallery has far reaching implications. The most obvious and
immediate benefit has been for visitors to the V&A. The music performed on
the four recordings discussed above originates from precisely the same years,
from the same cities, and is chosen from works commissioned by the same
patrons who commissioned artefacts on display in the exhibitions. The
recordings of Perez and RCM MS 2093 are contemporary with their
corresponding objects and have the same provenance; in the case of the
missal and the notation knives, the music is actually written on the objects.
Much of the music associated with the objects in the Galleries was re-
corded for the first time for this project. Much of the music can now be heard
via the V&As website, with annual online traffic of approximately 15 million
people. For further information on the pieces discussed above, see www.vam.
ac.uk and www.listeninggallery.rcm.ac.uk.
The project has demonstrated that music and musical performance were
central to the domestic and public life of those periods and how the
experience of music and art reunited reveals much about Europes geographi-
cal and cultural diversity. The objects on display, together with the worlds
they represent, have been experienced today as they would have been in the
past.
Acknowledgments
We wish to thank the following collaborators: Paul Banks, Clifford Bartlett, Terence
Charlston, Glyn Davies, Flora Dennis, Sebastian Durkin, Stuart Frost, Jakob Lindberg,
William Lyons, Peta Motture, Jenny Nex, Joanna Norman, Jennifer Smith, and Lance
Whitehead. The Listening Gallery is funded by the Arts and Humanities Research Coun-
cil (AH/F016840/1).
Giulia Nuti, Centre for Performance Science, Royal College of Music, Prince Consort
Road, London SW7 2BS, UK; Email: gnuti@rcm.ac.uk
References
Bianconi L. (1991). Il Seicento. Turin, Italy: EDT.

Dennis F. (in press). Scattered knives and dismembered song: Cutlery, music and the
rituals of dining. Renaissance Studies.
Fenlon I. (1995). Music and Culture in Late Renaissance Italy. Oxford: Oxford Univer-
sity Press.
Fortune N. (ed.) (1987). Music and Theatre. Cambridge: Cambridge University Press.
Gallico C. (1978). L'Et dell'Umanesimo e del Rinascimento. Turin, Italy: EDT.
Nuti G. (2007). The Performance of Italian Basso Continuo. Aldershot, UK: Ashgate.
Robertson A. (1991). The Service Books of the Royal Abbey of Saint-Denis. Oxford:
Clarendon Press.
Testi L. (1969). La Musica Italiana nel Medioevo e nel Rinascimento. Milan: Bramante
Editrice.
Difficulty of violin vibrato in novice players:

Fingerboard reaction force analysis
Satoshi Obata1, Hidehiro Nakahara1,2,Takeshi Hirano1, and

Hiroshi Kinoshita1

The aim of this study was to investigate the difference in the nature of
shaking (the longitudinal and lateral components) and press (the vertical
component) forces during the production of a vibrato tone by 10 novice
players as compared with 10 expert players of the violin. A violin in-
stalled with a 3-D force transducer was used for the measurement of the
force while performing a successive A (open) and D (vibrato and force
measurement) tone production task for 30 s at 4.5 Hz vibrato rate at mf
(75-77 dB). The index, middle, ring, and little fingers were used for the
measurement. The average, amplitude, and peak-to-peak time of shaking
and press forces were evaluated for each trial data. The results indicated
that the intra-individual variability of the amplitude of shaking force and
its peak-to-peak time for all fingers was significantly larger for the nov-
ices than the experts. The novices had smaller mean values of shaking
and press forces than the experts, and this group difference was larger for
the index and little fingers than for the other fingers. Novices have been
believed to use a grip too firm to shake the hand. The present data sug-
gest that it is the opposite.
Keywords: violin; vibrato; finger force; novice; finger difference
Satoshi Obata, Graduate School of Medicine, Osaka University, 1-17 Machikaneyama,

Toyonaka, Osaka 560-0043, Japan; Email: s-obata@moted.hss.osaka-u.ac.jp
Musical learning and cognitive performance
Carlos Santos-Luiz1, Daniela Coimbra2,3, and Carlos Fernandes da Silva4
1 College of Education, Polytechnic Institute of Coimbra, Portugal

4 Department of Educational Sciences, University of Aveiro, Portugal
According to literature, there are broad associations between music and

cognitive abilities, which apparently result from the frequency of music
lessons. The aim of the present work was to test these associations be-
tween a group of music students and a control group of students with no
music education, thereby contributing toward understanding the link
between music learning, intelligence, and academic achievement. The
Battery of Reasoning Tests (BPR/7-9), collated for the Portuguese popu-
lation, as well as the academic classifications of two groups of students,
provided the basis for this studys quantitative analysis. The sample
population comprised students who frequented Year 7 of the basic music
course in specialist music education and of the basic course in main-
stream education (no music lessons). Compared with the students in the
control group, those in the groups which included formal music educa-
tion showed increases in their general capacity of reasoning (g factor)
and in tasks of spatial and numerical reasoning. Equally, there were in-
creases in academic achievement. The results suggest that the duration of
exposure to music lessons is associated with an increased intelligence
and increased academic achievement.
Keywords: specialist music education; mainstream education; general

capacity of reasoning; academic achievement; cognitive skills
It has been suggested that music may be more than just an art form, and that
it may indeed be able to produce benefits in other domains, such as cognitive
abilities or general intelligence. According to Schellenberg (2005, 2006a),
different lines of research have shown that both passive listening to music
and formal musical training reveal such a connection. However, the differ-
ences between these two activities make it unlikely that both could have
similar effects upon the non-musical aspects of human behavior (Schellen-
berg 2006a). This study focuses upon the hypothesis that music lessons could
provide benefits in non-musical areas of cognition.
Of the musical activities that appear to improve cognitive performance,
the most significant appear to be the playing of a musical instrument (par-
ticularly keyboard instruments), vocal training, musical learning in general,
and rhythmic training. The underlying transfer effect (Barnett and Ceci 2002)
may prove to be unique in children that have received music lessons for pro-
longed periods; as such, lessons involve activities such as:
Focused attention, regularpractice, reading music, memorizing ex-

tended passages and entire pieces, learning about rules of pattern forma-
tion that define western musical structures, incremental improvement
of fine motor skills, and learning to express emotions through music
(Schellenberg 2006b, p. 466).
The combination of these activities with genetic predisposition and environ-

mental factors may play an important role, not only in music but also in other
areas of cognition (Barbro 2006), particularly during childhood when cere-
bral development is still very flexible (Huttenlocher 2002).
Some researchers believe that music lessons are associated to specific
sub-areas of intellectual skills. These include spatial skills (Bilhartz et al.
2000, Rauscher et al. 2006), reading (Lamb and Gregory 1993, Butzlaff
2000), subcomponents of literacy such as vocabulary (Orsmond and Miller
1999) and verbal memory (Ho et al. 2003), mathematics (Vaughn 2000,
Maureen 2008), and mathematical subcomponents such as arithmetic
(Rauscher et al. 2006) and geometry (La Mont 2008). According to Schellen-
berg (2004) the most parsimonious explanation of these diffuse associations
is that they stem from a common component, such as general intelligence
(Schellenberg 2004, p. 511). Schellenberg (2004, 2006a) also claims that
music improves general intelligence. Increased intelligence and enhanced
performance in subtests for academic competence were found in music stu-
dents. There is also evidence that the duration of musical training is positively
associated to intelligence, and that this association is general (i.e. not limited
to a specific subgroup of intellectual skills) (Schellenberg 2006b).
The research described in this paper examines associations between mu-
sic lessons and general capacity of reasoning (the so-called g factor), and
between the duration of the musical training and the increase in the general
capacity of reasoning. The performance of music students in each reasoning
task (abstract, numerical, verbal, mechanical, and spatial) was compared with
that of students who had not had music lessons. The general academic per-
formance of the students was also taken into account.
METHOD
Participants
The sample population (n=134) consisted of students from the seventh grade
of the basic music course (specialized musical education program), and oth-
ers undergoing conventional education. The average age was 13 years. The
students were divided into three groups, two experimental and one control:
(1) students receiving specialized musical education, who had had between 5-
7 years of musical training (n=63, SME 5-7), (2) students from specialized
musical education with 3-4 years of musical training (n=21, SME 3-4). The
control group (n=50) consisted of students undergoing conventional school-
ing (CS), which did not include music lessons. All students were from a
similar socioeconomic and cultural background.
Materials
The data used in this study was obtained using a Battery of Reasoning Tests,
BPR/7-9 (Almeida and Lemos 2006), designed to assess general capacity of
reasoning (g factor). These tests, which had been specially adapted to the
Portuguese population, included five instruments to evaluate performance in
different tasks of reasoning: abstract, numerical, verbal, mechanical, spatial.
Procedure
The participants were recruited from ten schools around Portugal by means
of an invitation addressed to their parents or guardians. The pupils were
tested individually using the BPR/7-9 at the beginning of the school year.
Statistical tests were then performed to gauge differences between groups.
RESULTS
Comparing the average g factor scores, it was found that the SME group with
5-7 years of musical education had the highest average, followed by the SME
group with 3-4 years of musical education (see Figure 1). Sheffs post hoc
tests showed that only the differences between the groups CS and SME 5-
7 were significant (mean difference=-2.12, SE=0.493, p=0.000). In short,
the students that had musical education displayed a higher general capacity
16
12
Average
8
4
0
CS without SME 3-4 yrs SME 5-7 yrs
musical training musical training musical training
Figure 1. Mean general capacity of reasoning for the CS, SME 3-4, and SME 5-7 groups.
of reasoning than the students from the control group, a difference which
became more pronounced with the duration of musical training.
Using MANOVA, the performances of the three groups was compared
with relation to each task. Scheff tests were used to assess the differences
between the various tests in the three groups. Although the students from the
SME 5-7 and SME 3-4 groups scored better than those from the CS group in
abstract and verbal reasoning and those from the SME 5-7 group scored bet-
ter than the CS group for mechanical reasoning, the differences were not sig-
nificant. However, in the spatial reasoning tasks, the difference between the
averages obtained by the CS group and the SME 5-7 and SME 3-4 groups was
significant (respectively, mean difference=-2.97, SE=0.710, p=0.000, and
mean difference=-2.44, SE=0.974, p=0.046). In the numerical reasoning
task, there was a significant difference between the CS group and the SME 5-7
group (mean difference=-3.15, SE=0.666, p=0.000).
DISCUSSION
The results of this study corroborate the hypothesis that there is a significant
relationship between musical education and cognitive growth. This associa-
tion between the general capacity of reasoning and the number of years in-
volvement with musical education is confirmed by literature dealing with the
relationship between musical education and cognitive ability (Schellenberg
2005). As in some of the studies cited above, this research showed that asso-
ciations were limited to specific tasks. However, when we considered educa-
tional performance as a whole, it was found that music lessons affected
general intellectual abilities. Students that had a musical education generally
obtained higher grades in academic achievement than their counterparts who
had not, and this difference seemed to increase with the number of years of
music lessons. We might infer that music lessons play an important role in
the results achieved, rather than extracurricular activities as this was tested
elsewhere (Santos-Luiz et al. 2009).
The music students in our sample had been tested for musical aptitude,
among other things, before embarking on their musical studies. Researchers
examining whether musical aptitude is associated to other cognitive capaci-
ties have noted that performance in this type of task tends to be positively
correlated with general intelligence (Lynn et al. 1989). Therefore, it is possi-
ble that increased musical aptitude through musical training is accompanied
by improved general cognition (Schellenberg 2006a).
The association of music with intelligence and academic performance may
result from a simple characteristic or from a cluster of abilities that are fos-
tered by musical education or musical stimulation. Musical structures are
abstractions that involve a range of transformations, and mental representa-
tions of music must be sufficiently abstract to allow recognition of similarities
between varying patterns. Formal musical training may promote the ability to
recognize regular patterns and think in a flexible way (Schellenberg 2006b).
These abilities are central to the g factor. This study therefore appears to cor-
roborate the assumption that there are associations between music lessons
and intellectual ability, although further research is required before this can
be asserted with confidence.
Acknowledgments
This work has been supported by the Fundao para a Cincia e Tecnologia (FCT) and
Programa Operacional Cincia e Inovao 2010 (POCI 2010), co-funded by the Portu-
guese Government and European Union by the FEDER Program.
Carlos Santos-Luiz, Department of Music, College of Education, Polytechnic Institute of

Coimbra, Praa Heris do Ultramar, Coimbra 3030-329, Portugal; Email: cluis@esec.pt
References
Barbro B. J. (2006). Music and brain plasticity. European Review, 14. pp. 49-64.
Barnett S. M. and Ceci S. J. (2002). When and where do we apply what we learn? A
taxonomy for transfer. Psychological Bulletin, 128, pp. 612-637.
Bilhartz T. D., Bruhn R. A., and Olson J. E. (2000). The effect of early music training on
child cognitive development. Journal of Applied Developmental Psychology, 20,
pp. 615-636.
Butzlaff R. (2000). Can music be used to teach reading? Journal of Aesthetic Education,
34, pp. 167-178.
Ho Y.-C., Cheung M.-C., and Chan A. S. (2003). Music training improves verbal but not
visual memory: Cross sectional and longitudinal explorations in children. Neuro-
psychology, 17, pp. 439-450.
Huttenlocher P. R. (2002). Neural Plasticity. Cambridge, Massachusetts, USA: Harvard
University Press.
Lamb S. J. and Gregory A. H. (1993). The relationship between music and reading in
beginning readers. Educational Psychology, 13, pp. 19-27.
La Mont K. A. (2008). Exploring the Foundations of the Relationship between Music
Training and Mathematical Cognition. Unpublished doctoral thesis, Harvard Uni-
versity.
Lynn R., Wilson R. G., and Gault A. (1989). Simple musical tests as measures of Spear-
mans g. Personality and Individual Differences, 10, pp. 25-28.
Maureen H. (2008). The effects of music instruction on learning in the Montessori
classroom. Montessori Life, 20, pp. 24-31.
Orsmond G. I. and Miller L. K. (1999). Cognitive, musical and environmental correlates
of early music instruction. Psychology of Music, 27, pp. 18-37.
Rauscher F. H., LeMieux M., and Hinton S. C. (2006). Quality piano instruction affects
at-risk elementary school childrens cognitive abilities and self-esteem. Paper pre-
sented at the Ninth International Conference on Music Perception and Cognition,
Bologna, Italy.
Santos-Luiz C., Coimbra D., and Silva C. F. (2009). Impacto do ensino especializado da
msica no desempenho acadmico. Paper presented at the IV Encuentro de Prima-
vera, University of Granada, Spain.
Schellenberg E. G. (2004). Music lessons enhance IQ. Psychological Science, 15, pp.
511-514.
Schellenberg E. G. (2005). Music and cognitive abilities. Current Directions in Psycho-
logical Science, 14, pp. 322-325.
Schellenberg E. G. (2006a). Exposure to music: The truth about the consequences. In G.
E. McPherson (ed.), The Child as Musician (pp. 111-134). Oxford: Oxford University
Press.
Schellenberg E. G. (2006b). Long-term positive associations between music lessons and
IQ. Journal of Educational Psychology, 98, pp. 457-468.
Vaughn K. (2000). Music and mathematics: Modest support for the oft-claimed rela-
tionship. Journal of Aesthetic Education, 34, pp. 149-166.
A longitudinal observation of one-to-one

singing lessons: The effects of personality and
adult attachment
Sofia Serra-Dawa1,2
1 Department of Music, University of Sheffield, UK

Instrumental and singing teaching has previously been studied with par-
ticular focus on pedagogical, cognitive, technical, and developmental as-
pects. However, the relationship and interactions that take place between
the teachers and students in that setting deserve more extensive explora-
tion. This study approaches the singing teacher-student relationship with
a particular focus on the observation of personality traits and the levels of
attachment identified between teacher and student. Eleven singing
teachers with 54 students were observed in one-to-one singing lessons
through video observations made during one academic year. Addition-
ally, teachers and students completed the questionnaires NEO FFI-R and
Adult Attachment Scale to evaluate personality and attachment. The in-
teraction between teacher and student include a wide list of variables:
personality combination, the individual background, kind of attachment
felt toward each other, and many other singing aspects, suggesting that
the relationship with the student could have major impact on other as-
pects of singing.
Keywords: singing; relationship; adult attachment; longitudinal; per-

sonality
Instrumental and singing teaching has previously been studied with particu-
lar focus on pedagogical, cognitive, technical, and developmental aspects.
However, the relationship and interactions that take place between the teach-
ers and students in that setting deserves extensive exploration.
Teacher-student interactions have powerful potential to improve and fa-
cilitate learning (Siebenaler 1997), with relationships playing a role in stu-
dents development. Burland and Davidson (2002) suggest that the teacher in
a master-student relationship plays a vital role in the development of the

students performance, as well as in their overall artistic personality, and that
students who do not find the ideal teacher may have consequences reflected
in their professionalism and artistry. The interaction of teacher and student
may constitute an important base for the students knowledge building. With
the purpose of exploring these interactions between the teacher and student,
this study aims to (1) identify behavior in a longitudinal relationship of sing-
ing teacher and student, (2) analyze the effects of personality and adult at-
tachment on singing teacher and student relationship outcomes, (3)
distinguish combinations of singing teacher and student characteristics.
METHOD
Participants
The participants in the current study were 11 singing teachers and 54 under-
graduate and postgraduate students from six colleges and universities in the
UK and Portugal. All students were studying singing as their main instru-
ment. From the same sample, 35 participants completed personality and
adult attachment scale measures.
Materials
The study had three stages where participants were video recorded during
one academic year. These recordings were complemented with existing psy-
chological questionnaires in order to access the personality types and the
levels of attachment in each dyad. For the personality identification, the NEO
FFI-R by Costa and McCrae (1985) was used, and adult attachment was
measured using the Adult Attachment Scale by Collins and Read (1990).
Procedure
Six higher education institutions (universities and colleges) from the UK and
Portugal were contacted through their heads of vocal studies. These two
countries were chosen to allow a wider range of comparison and a diverse
cultural environment. The head of vocal studies were asked to choose several
singing teachers who were seen as being different or contrasting in relational
terms, and these staff were approached for a longitudinal video collection of
their singing lessons. The teachers who agreed to participate were then asked
to choose at least two contrasting students that would follow these differ-
ences: (1) age, (2) academic year, (3) levels of development, (4) that the
teachers felt they related to in different ways, and/or (5) students they felt
approached their teaching differently. No reference to gender was stipulated.

At the second and third stages of video recordings, the teachers and students
completed the personality and attachment tests.
RESULTS
Ongoing analysis illustrates that the interaction between teacher and student
includes a wide list of variances: personality combination, the individual
background, kind of attachment felt toward each other, and many other as-
pects suggesting that the relationship with the student could have major im-
pact on other aspects of singing. The longitudinal observation indicated that
the relationships develop into becoming more personal and often include
elements from the personal lives of both teacher and student. These elements
of students lives are often used toward interpretative development and be-
come interesting strategic tools for the singing teachers. The relational inter-
actions between teacher and student are in all cases conditioned by the
academic years schedule and there is a noticeable peak of relationship ap-
proximation during the middle stage of the year in relation to the beginning
of year (where participants are getting to know each other) and the last stage
(where the pressure of examination normally drives the lesson).
The observational part of the study presented fluctuations of behavior and
adaptability between teacher and student during the course of the one aca-
demic year. Particularly interesting to note is the teachers adaptability to
each student, making the relationship unique. The personality and attach-
ment types of each participant seem to play an important role in the individu-
ality of each relationship.
DISCUSSION
This study observed the psychological involvements of the teacher-student

relationship and the effects of those in the behavior in each lesson and from a
longitudinal perspective. The personality combination between teacher and
student, and the kind of attachment that is implicit in this relationship, may
be a predisposition for successful or unsuccessful relationships.
Sofia Serra-Dawa, School of the Arts, Portuguese Catholic University, Rua Diogo Bo-
telho 1327, Porto 4169-005, Portugal; Email: sofiaserra99@yahoo.com
References
Burland K. and Davidson J. W. (2002). Training the talented. Music Education Re-
search. 4, pp. 121-140.
Collins N. and Read S. (1990). Adult attachment relationships, working models and
relationship quality in dating couples. Journal of Personality and Social Psychol-
ogy, 58, pp. 644-683.
Costa P. T. and MacCrae R. R. (1985). The NEO Personality Inventory Manual. Odessa,
Florida, USA: Psychological Assessment Resources.
Siebenaler D. (1997). Analysis of teacher-student interactions in the piano lessons of
adults and children. Journal of Research in Music Education, 45, pp. 6-20.
Emotional lingering: Facial expressions of

musical closure
William Forde Thompson1, Rachel Bennetts1, Bojan Neskovic1, and

Caroline Palmer2
1 Department of Psychology, Macquarie University, Australia

We report evidence that singers maintain emotional facial expressions

after vocalization has terminated, introducing a form of emotional lin-
gering. Emotional lingering extends and complements acoustic signals of
emotion, providing a visual signal of musical closure. We first describe
evidence from production studies that emotional facial expressions con-
tinue beyond the acoustic dimension of music. We next describe evidence
that perceivers are sensitive to facial expressions that occur beyond the
production of sound, and that such signals carry reliable emotional in-
formation. We note that audiovisual experiences of music are compatible
with current understandings of multisensory integration in the central
nervous system. As such, investigations of performance should include
consideration of facial expressions and other performance gestures.
Keywords: music cognition; emotion; singing; closure; facial expression
Music is a powerful medium for conveying emotion (Juslin and Sloboda

2001). Emotional information is carried not only in musical sound; it is also
signaled in the facial expressions of performers (Livingstone et al. 2009,
Thompson et al. 2005, Thompson et al. 2008). This paper concerns the na-
ture and significance of facial expressions during sung music performance,
focusing on expressions that occur immediately following the cessation of
vocal production.
Effective singing requires a stage in which plans for vocal and emotional
expression are created, coordinated, and implemented. Vocal planning fa-
cilitates accurate production of pitch, timing, intensity, and vocal quality
(timbre). It recruits a range of muscle groups including the vocal cords,
throat, face, and other parts of the body that support vocal production. The
nature and time course of vocal planning can be inferred by examining subtle
movements in these muscle groups that occur prior to sound production.
Such movements reflect a dynamic process in which action and decision-
making are tightly coupled (McKinstry et al. 2008).
Muscular activity associated with emotional communication overlaps with
muscular activity used for effective singing, but it is not identical. While a
performer is singing it may be difficult to disentangle movements that sup-
port accurate vocal production from those that support emotional communi-
cation. Similarly, movements that occur immediately prior to vocal
production combine plans for accurate vocal production with plans for emo-
tional communication.
Once vocal production has terminated, however, the bodily movements
and facial expressions that persist cannot be interpreted as a reflection of
vocal production mechanisms. Emotional facial expressions linger beyond the
time it takes to deactivate muscular activity, often lasting several seconds or
more. Such activity can only be interpreted as a constituent of emotional
communication and is referred to as emotional lingering.
Emotional lingering can be investigated in both production and percep-
tion studies. First, facial expressions and other movements that occur fol-
lowing the cessation of vocal production can be captured and analyzed using
technologies such as motion capture and electromyography. These strategies
can be used to confirm the existence of emotional lingering and determine its
properties and time course. Second, perceptual studies can be used to deter-
mine how perceivers are affected by emotional lingering phenomena. Do fa-
cial expressions that persist beyond the termination of sound provide reliable
emotional information, and do they influence emotional interpretations of the
music?
Livingstone et al. (2009) confirmed that singers maintain emotional fa-
cial expressions beyond the cessation of vocal production. Participants were
recorded with motion capture or electromyography (EMG) as they watched
and imitated phrases of emotional singing. All were shown audiovisual re-
cordings of sung phrases performed with happy, sad, or neutral emotional
expressions. Their task was to imitate the target stimulus, emphasizing the
emotion expressed. Facial expressions for happy and sad sung phrases were
significantly different from facial expressions for neutral sung phrases. How-
ever, these differences were observed not only during the imitation, but also
before and after vocal production. The authors concluded that facial expres-
sions not only support music performance during vocal production, but also
reflect processes of emotional planning and lingering.
In the current investigation, we examined whether the facial expressions

that occur after vocal production provide reliable emotional information to
listeners. If so, it would support the notion that such expressions can be used
to clarify, underscore, or modify emotional intentions. A full description of
the investigation will be described in a forthcoming publication.
METHOD
Participants
Seventeen first year psychology students took part in the experiment in re-
turn for course credit. All had normal or corrected to normal vision.
Materials
Stimuli were excerpts of video recordings of facial expressions used by four

singers (Livingstone et al. 2009). Stimuli were presented as full video or
point-light display (13 markers). Full videos were recorded using a Sony DV
camera, positioned roughly 45 degrees to the right of participants at a dis-
tance of 2.5 m. Motion capture data were collected from 13 markers on the
face using an 8-camera Vicon optical motion capture system (Oxford Met-
rics), calibrated to less than 1 mm error. Six stimuli were chosen from each
singer, depicting two sung phrases for each of three emotions: happy, neutral,
and sad. Clips from three time periods or epochs were excerpted from each
trial: pre-production (three beats prior to vocalization, or 2.4 s), production
(the middle three beats during vocalization), and post-production (the first
three beats after the singer stopped singing). Each epoch was depicted in two
formats: video and point-light display. Videos were edited in Final Cut Pro
(Apple) to reveal only the face and shoulders of the singer on a neutral (beige)
background. Point-light displays were produced by processing motion cap-
ture data for each trial in Matlab (Mathworks) and consisted of black dots on
a white background.
The experiment was run on a MacBook Pro laptop, using Experiment
Creator 1.4 (available from WFTs website). Stimuli were presented on a 24
Samsung monitor. Participants were seated roughly 50-100 cm from the
screen.
Procedure
Each participant completed 144 trials (4 singers, 2 phrases, 3 epochs, 3 emo-

tions, 2 display types). Display type (video or point-light display) was blocked
and counterbalanced and the order of presentation within blocks was ran-
Figure 1. Timeline of a trial involving point-light display. Each black dot represents a
marker used to track facial motion. Markers were placed on the forehead, temples,
nose-bridge, nose-tip, eyebrows (2 each), and lips (upper, lower, and corners).
domized. Participants completed six practice trials for each display type be-
fore completing each block. Illustrated in Figure 1, each trial began with 1 s of
black screen, followed by the 2.4 s stimulus and then another 1 s of black
screen. Participants then rated the emotion perceived on a scale from -3 (very
sad), to 0 (neutral), to +3 (very happy). A computer mouse was used to select
ratings displayed on the monitor. Once participants rated the stimulus they
could progress to the next trial when ready.
RESULTS
Preliminary analyses revealed significant interactions involving display type,

motivating separate analyses for video and point-light display conditions. For
each, an Analysis of Variance (ANOVA) was conducted with repeated meas-
ures on emotion (happy, sad, neutral) and epoch (pre-production, pro-
duction, post-production).
For full video trials, there was a main effect of emotion (F2,32=235.82,
p<0.001) with high ratings assigned to happy trials (M=1.59, SE=0.10), in-
termediate ratings assigned to neutral trials (M=-0.47, SE=0.06), and low
ratings assigned to sad trials (M=-1.29, SE=0.11). Although this general pat-
tern was observed in all three epochs, there was a significant interaction be-
tween emotion and epoch (F4,64=20.48, p<0.001). This interaction is
illustrated in Figure 2 (left panel), which displays mean ratings of emotion for
each of the three epochs. Analysis of data for the lingering epoch separately
revealed a main effect of emotion (F2,32=177.31, p<0.001), and planned com-
parisons verified that ratings for the three emotions differed significantly
from each other. Thus, facial expressions that lingered beyond song produc-
tion provided reliable emotional cues.
Figure 2. Mean emotion ratings for full video (left panel) and point-light display (right
panel), and standard error bars.
For point-light display, there was a main effect of emotion (F2,32=25.05,

p<0.001), with highest ratings assigned to happy trials (M=0.25, SE=0.09),
intermediate ratings assigned to neutral trials (M=-0.02, SE=0.08), and low-
est ratings assigned to sad trials (M=-0.39, SE=0.11). There was also a signifi-
cant interaction between emotion and epoch (F4,64=4.60, p<0.01), illustrated
in Figure 2 (right panel). Analysis of the post-production epoch separately
revealed a main effect of emotion (F2,32=5.38, p<0.05). However, planned
comparisons indicated that only the neutral and sad conditions differed sig-
nificantly in emotion ratings. During the post-production epoch, point-light
display did not provide participants with sufficient information to differenti-
ate happy facial expressions from other facial expressions.
DISCUSSION
Results confirm that facial expressions that linger beyond the cessation of
vocal production contain reliable emotional cues. Indeed, when full video was
available, discrimination of emotional facial expressions was observed prior,
during, and after vocalization. Such facial expressions of emotion may func-
tion to shape a listeners interpretation of the music (Thompson et al. 2008).
When only point-light display was available, discrimination was observed
prior and during vocalization, but participants could not discriminate happy
expressions from other emotional intentions in the post-production epoch,
indicating that marker movement alone carried insufficient information for
decoding happy expressions in that epoch. It is possible that isolated happy
emotional expressions are ambiguous in the lingering epoch because they
begin with raised lip corners and end with a neutral expression, implying
movement away from a positive emotional expression. Alternatively, in the

absence of recognizable facial expressions, movement that comes to a stop
may generally have negative connotations.
Facial expressions of emotion are likely to be integrated with auditory
signals of emotion, yielding audiovisual experiences of music. Such experi-
ences are consistent with current understandings of multisensory integration
in the central nervous system, and underscore the importance of considering
facial expressions and other performance gestures in studies of music and
emotion.
Acknowledgments
This research was supported by an ARC Discovery Grant DP0987182 awarded to the
first and last authors. Alex Chilvers, Catherine Greentree, and George Narroway pro-
vided research assistance.
William Forde Thompson, Department of Psychology, Macquarie University, Sydney,

New South Wales 2109, Australia; Email: bill.thompson@mq.edu.au
References
Juslin P. N. and Sloboda J. A. (2001). Music and Emotion. Oxford: Oxford University
Press.
Livingstone S. R., Thompson W. F., and Russo F. A. (2009). Facial expressions and
emotional singing: A study of perception and production with motion capture and
electromyography. Music Perception, 26, pp. 475-488.
McKinstry C., Dale R., and Spivey M. (2008). Action dynamics reveal parallel competi-
tion in decision making. Psychological Science, 19, pp. 22-24.
Thompson W. F., Russo F. A., and Quinto L. (2008). Audio-visual integration of emo-
tional cues in song. Cognition & Emotion. 22, pp. 1457-1470.
Thompson W .F., Graham P., and Russo F. A. (2005). Seeing music performance: Visual
influences on perception and experience. Semiotica, 156, pp. 203-227.
Pianists with carpal tunnel syndrome:

Conservative versus surgical treatment
Hara Trouli1 and Nikos Reissis2
1 National Music Conservatory of Athens, Greece

2 Northwood Orthopedic and Musculo-Skeletal Clinic, London, UK
We studied retrospectively 18 patients, all professional pianists or ad-

vanced piano students, who had presented with carpal tunnel symptoms
in the past and had undergone surgical or conservative treatments. We
found that delayed diagnosis was more likely to lead to surgical treat-
ment and that fear of diagnosis and fear of surgery prevented early
treatment. Although surgical release was successful in 50% of the pa-
tients, re-education was essential to maintain good results according to
long follow-up. Finally, medical practitioners are in a better position to
help pianists when they make consideration of the demands and techni-
cal peculiarities of this profession.
Keywords: carpal tunnel syndrome; pianists hand pain; pianists finger

numbness; pianists finger weakness; nerve entrapment
Carpal tunnel syndrome is the most common cause of entrapment neuropa-

thy in the upper limb. Compression of the median nerve within the carpal
tunnel may be caused by an increase in the volume of its contents either due
to local events such as tenosynovitis of the tendons within the tunnel or due
to general disease such as diabetes, myxoedema, or rheumatoid arthritis.
Compression of the median nerve gives rise to symptoms of numbness, pain,
or weakness (clumsiness) in the affected hand. When a patient visits the
medical practitioner complaining of symptoms suspicious of carpal tunnel
syndrome, the clinical examinationcomprising of specific diagnostic tests
may be able to confirm the condition and to some extent its severity. In cases
of doubt, further electrophysiologic tests (electromyography and nerve con-
duction studies) may become necessary, where the prolongation of distal
motor latency will confirm the presence of median nerve compression. It is,
however, worth mentioning that a normal result on electromyography does

not rule out the diagnosis.
The treatment of carpal tunnel syndrome has been traditionally either
conservative (rest, splints, oral or injectable anti-inflammatories), surgical
(division of flexor retinaculum with or without neurolysis or tenosynovec-
tomy), or both, depending on the severity of the symptoms. There is, how-
ever, a natural progression of the condition, with deterioration over time, and
this becomes essential in the awareness that the patient should have in order
to identify the early symptoms. Particularly for pianistswho are among the
professionals more prone to develop carpal tunnel syndrome due to the re-
petitive tasks they performearly diagnosis is crucial for the successful out-
come of their treatment. In this study, we tried to detect how the time of
diagnosis affected the treatment and outcome of carpal tunnel syndrome in
the pianists studied. We also aimed to examine the progress of the condition
in the pianists who received conservative treatment and the pianists who re-
ceived surgical treatment. The surgeon-patient relationship was studied in an
effort to establish potential correlation between the medical approach to the
condition and the awareness of the particular technical issues of the pianist
when making management decisions.
METHOD
Participants
Fifteen female and three male pianists, aged between 17-57 years, partici-
pated. They all studied or worked at the National Music Conservatory in Ath-
ens and their treatmentsurgical or conservativewas also provided in
Athens. The diagnosis had been made on the basis of clinical presentation,
and on six patients, nerve conduction and electromyographic studies were
also performed. Twelve patients had undergone surgical decompression and
six had had non-surgical treatment with a combination of advice on modify-
ing their practice, physiotherapy, splints, local anti-inflammatory injections,
and oral anti-inflammatories.
Procedure
To assess the patients follow up we distributed medical as well as practice

questionnaires to collect information on pre- and post-treatment symptoms.
In the pre-treatment period, the time of initiation of symptoms was studied in
relation to the subsequent type of treatment and the relation of patterns and
habits of practicing, as well as adverse psychology in relation to carpal tunnel
symptoms. In the post-treatment period (follow-up 4-14 years), the results of

the pianists progress were grouped into poor, good, and excellent according
to specified criteria and assessed in relation to the treatment received.
RESULTS
In the pre-treatment period, our results showed that the early diagnosis was
followed with relief of the symptoms without the need for surgery. Most de-
layed diagnoses were either due to fear of diagnosis or due to fear of potential
surgical treatment (see Table 1). We found a correlation between symptoma-
tology and technical difficulties in practice or occasions where low morale and
social or career pressures were prominent (see Table 2).
In the post-treatment period, we defined our results according to the out-
lined criteria (poor: no change/possible deterioration, good: improvement/
possible relapse, excellent: improvement/no relapse) for both conservatively
and surgically treated pianists (see Table 3) and found that although surgical
treatment had been of good initial relief in 50% of the patients, a further re-
education of practicing was essential in order to maintain the outcome.
In the evaluation of this result, it is difficult accurately to detect retro-
spectively the surgeon/physicians familiarity/awareness with the particular
technical issues that affect the pianists. In our questionnaire, we asked
whether the patient felt that the medical practitioner was able to discuss and
understand the pianistic demands and peculiarities of their technique, and
we based our results on their replies (see Table 4). It could be said that, the
less aware the surgeon, the more his/her inclination toward surgical treat-
ment.
DISCUSSION
Peripheral nerve entrapment syndromes of the upper extremities are well

documented in musicians. Lederman (2004) evaluated 226 instrumentalists
for playing-related disorders and found that 29% had a peripheral nerve dis-
order. Gohl et al. (2006) studied 19 pianists and concluded that 16% had
early evidence of median neuropathy and that by completing a thorough
medical history, good physical examination, and nerve conduction testing,
early neuropathy may be detected.
In our present study, we found that the early diagnosis was followed with
relief of the symptoms without the need for surgery. Most delayed diagnoses
were either due to fear of diagnosis or due to fear of potential surgical treat-
ment. It is essential that the pianists learn to identify early symptoms. During
the early stages one can reflect upon the actual technique and postural ad-
Table 1. Pre-treatment period: time of diagnosis and subsequent treatment.
Diagnosis <3 months Diagnosis 3-6 months Diagnosis >6 months

Conservative 1 (5.5%) 5 (27.7%) 0 (0.00%)
Surgical 2 (11.1%) 2 (11.1%) 8 (44.4%)
Fear of diagnosis 1 (5.5%) 2 (11.1%) 6 (33.3%)
Fear of surgery 2 (11.1%) 0 (0.00%) 7 (38.8%)
Table 2. Pre-treatment period: symptoms and associated circumstances.
Pain Weakness Paresthesiae Other Total

No technical difficulties 2 2 2 1 7 (38.8%)
No adverse psychology 0 0 1 1 2 (11.1%)
No career pressures 1 1 1 0 3 (16.6%)
Technical difficulties 4 2 3 1 10 (55.5%)
Low morale 4 2 4 2 12 (66.6%)
Career pressures 4 3 4 4 15 (83.3%)
Table 3. Post-treatment period: outcome of treatment.
Poor Good Excellent

Conservative 2 (33.3%) 2 (33.3%) 2 (33.3%)
Surgical 3 (25.o%) 6 (50.o%) 3 (25.o%)
Table 4. Post-treatment period: patient/surgeon relationship.
Surgeon aware Surgeon unaware

Conservative 3 (50%) 3 (50%)
Surgical 3 (35%) 9 (75%)
justments in order to reverse and further prevent compression of the median

nerve in the carpal tunnel. Early signs should not be neglected by pianists as
during these times one may not seek medical help; it relies on the musician to
consider modification in their practice. If, however, symptoms and signs de-
teriorate, medical help should be sought without delay. Conservative treat-
ment may come to benefit and treat the condition without need for surgery. It
is also recognized that from some point onwards the damaged median nerve
will not recover with conservative means and surgery will become unavoid-
able.
We also found a correlation between symptomatology and technical diffi-
culties in practice or occasions where low morale and social or career pres-
sures were prominent. It has been discussed that misuse of the tendons
and/or excessive wrist flexion are contributing factors to the development of
carpal tunnel syndrome. It may also be the case that psychological matters
that give rise to tension in the upper limbs play some role in this or other
conditions that affect piano players. Further studies should be conducted to
determine the statistical significance of the psychological effects.
We found that although surgical treatment had been of good initial relief
in 50% of the patients, a further re-education of practicing was essential in
order to maintain the outcome. This is either because the pianist had not
modified their practice during the earlier stages of the condition or because
the relief from the acute symptoms following surgery gave the freedom and
reassurance to the pianist to practice without deliberating their technique. It
is therefore essential for any re-education to happen soon after the operation
and to be maintained during the rehabilitation period and beyond in order to
prevent reoccurrence.
It could be said that, the less aware the surgeon, the more his/her incli-
nation toward surgical treatment. This we cannot prove with this study, as
due to its retrospective nature it cannot reflect accurate assumptions. How-
ever, we do take the opportunity to highlight this issue and emphasize that as
the pianist needs to be aware of the early warnings of a possible carpal tunnel
syndrome, the surgeon should be also aware of the possible technical aspects
that can contribute to this condition and be in a position to discuss them with
his patient before applying treatments.
We conclude that it is of major importance to maintain the clinicians
awareness of the particular demands that the pianists have from their hands
all the way through the diagnostic and treatment process. The debate of sur-
gical versus conservative management still remains open, as there are cases
that do equally well with either type of care. Carpal tunnel syndrome should
be broadly made known to piano players, and the early clinical symptoms
should be highlighted and possibly taught to the pianists early in their studies
and careers. Whether the symptoms are due to wrong technical habits or
whether they are due to unavoidable pathologies, it is crucial to understand
that early diagnosis and management can relieve the pianist from the anxiety
that carpal tunnel syndrome may cause.
Hara Trouli, Northwood Orthopedic and Musculo-Skeletal Clinic, 6 Jonathans, Dene

Road, Northwood, Middlesex HA6 2AD, UK; Email: haratrouli@googlemail.com
References
Gohl A. P., Clayton S. Z., Strickland K. et al. (2006). Median and ulnar neuropathies in
university pianists. Medical Problems of Performing Artists, 21, pp.17-25.
Lederman R. J. (2004). Peripheral nerve disorders in instrumentalists. Annals of Neu-
rology, 26, pp. 640-646.
Musical training facilitates brain plasticity:

Short-term training effects on sensorimotor
integration
C. Carolyn Wu, Vanessa K. Lim, Jeffrey P. Hamm, and Ian J. Kirk
Research Centre for Cognitive Neuroscience, Department of Psychology,

Efficient sensorimotor integration is essential for music performance.

Previous research has indicated that auditory-motor associations form
not only as a result of long-term training but after a very brief period of
training. After short-term training, it has been demonstrated that pre-
motor areas are recruited during passive listening of trained music, sug-
gesting that these mappings can rapidly become automatic. It has been
argued that these mappings rely on activity in mirror neuron systems
(involved generally in imitating and learning actions). Action-observa-
tion studies in this field have associated changes in EEG mu-rhythm ac-
tivity with the mirror neuron system. We utilized this technique in our
action-listening study in order to detect involuntary motor co-activation
during passive listening to melodies and rhythms. We investigate
whether motor co-activation during passive listening occurs specifically
for newly acquired sound-action mappings after training. Subjects par-
ticipated in a short-term training scheme in which they were trained to
accurately play back randomly generated basic piano melodies. Prelimi-
nary results show changes in the mu-rhythm activity in post-training
EEG recordings. These initial findings support the hypothesis that sen-
sorimotor experience is important for the mirror neuron system. This
study demonstrates that musical training research can make a valuable
contribution to brain plasticity research.
Keywords: sensorimotor training; brain plasticity; neuroimaging; EEG;

audio-motor
C. Carolyn Wu, Research Centre for Cognitive Neuroscience, Department of Psychology,

University of Auckland, 10 Symonds Street, Auckland 1142, New Zealand; Email:
carolyn.wu@auckland.ac.nz
Symposium:
Musicians dystonia: New aspects in
pathophysiology and treatment
Is musicians dystonia an inherited condition?
Alexander Schmidt1, Hans-Christian Jabusch2, Eckart Altenmller3,

Johann Hagenah1, Rachel Saunders-Pullman4,5, Susan B. Bressman4,5,
Alexander Mnchau6, and Christine Klein1
1 Department of Neurology, University of Lbeck, Germany

4 Department of Neurology, Albert Einstein College of Medicine, USA
5 Department of Neurology, Beth Israel Medical Center, USA
6 Department of Neurology, University Medical Center, Hamburg-Eppendorf, Germany
Musicians dystonia (MD) is generally considered a sporadic disorder

that presents with loss of voluntary motor control of extensively trained
movements. To test the hypothesis of a genetic etiology in at least a sub-
set of MD, we initiated a large clinical genetic study. The families of 28
index patients with MD, 14 with a reported positive family history of focal
task-specific dystonia (FTSD) and 14 with no known family history (FH-),
underwent a standardized telephone screening interview using the Beth
Israel Dystonia Screen. Videotaped neurological examinations were per-
formed on all participants who screened positive, and consensus diagno-
ses were established. All patients were investigated for the GAG deletion
in DYT1; suitable families were tested for linkage to DYT7. A diagnosis of
dystonia was established in all 28 index patients and in 19 of 97 exam-
ined relatives (MD: n=8, other FTSD: n=9, other dystonias: n=2), 5 of
whom were members of FH- families. In total, 18 families were multiplex
with 2-4 affected members. The GAG deletion was not present in any of
the tested patients. Linkage to DYT7 could be excluded in one of the 11
informative families. Our results suggest a genetic contribution to MD
with phenotypic variability including FTSD.
Keywords: musicians dystonia; focal task-specific dystonia; genetics;

movement disorders; DYT1 gene
Musicians dystonia (MD), a type of focal task-specific dystonia (FTSD), pre-

sents with painless muscular incoordination or loss of voluntary motor con-
trol of extensively trained movements when a musician is playing the
instrument (Altenmller 2003). While the pathophysiology remains largely
elusive, MD has been associated with intensive training regimes and thus
been considered a form of occupational cramp. However, the epidemiology
also suggests a possible hereditary component: 10% of MD patients report a
positive family history of dystonia (Altenmller 2003). In rare cases of focal
dystonia, a hereditary component has been demonstrated, such as the GAG
deletion in the DYT1 gene (Defazio et al. 2007). However, this mutation was
excluded in a small group of MD patients (Friedman et al. 2000). In two
families, focal dystonia has been linked to a specific gene locus on chromo-
some 18 (DYT7) (Defazio et al. 2007). More recently, the report of three
families with putative autosomal dominant inheritance of FTSD in relatives of
patients with MD has lent support to the concept of a genetic contribution to
MD (Schmidt et al. 2006). To test the hypothesis of a genetic etiology in at
least a subset of MD and to explore a possible relationship between MD and
other forms of FTSD, we initiated a large clinical genetic study of MD based
on systematic examination of 28 families (Schmidt et al. 2009).
METHOD
Recruitment of study sample and onsite examination
After approval of the study by the local ethics committee and obtaining in-
formed consent, we included 28 professional musicians diagnosed with focal
dystonia at the outpatient clinic of the Hanover Institute of Music Physiology
and Musicians Medicine (index patients). Based on history and clinical fea-
tures, all were classified as having likely primary dystonia. Fourteen of these
index patients had a reported positive family history (FH+) of FTSD and were
matched to 14 patients with no known family history (FH-) for age, sex, in-
strument group, and type of dystonia (limb versus embouchure). In a first
telephone contact, all index patients were asked to report known cases of
dystonia in their families (family history interview). All available first- and
second-degree relatives with no known dystonia underwent a standardized
telephone screening interview using a modified version of the Beth Israel
Dystonia Screen containing additional questions screening for MD (BIDS,
adapted from Saunders-Pullman et al. 2005). Videotaped neurological ex-
aminations were performed at a home visit by the same examiner (AS) in all
28 index patients, in all 15 relatives with a known or reported form of FTSD,
and in all 11 relatives who screened positive for dystonia in the BIDS.
Diagnostic criteria, video rating, and consensus diagnosis
A diagnosis of dystonia was made following previously published criteria

(Altenmller 2003, Bressman et al. 2002):
Definite: muscle contractions producing characteristic twisting, flexion,

or extension movements and postures consistently present
Probable: movements and postures of insufficient intensity or consistency
to merit classification as definite
Possible: muscle contractions not considered abnormal but remotely sug-
gestive of dystonia
No dystonia
All examined family members were diagnosed in a three-step process:

first, by onsite examination including information about pedigree structure
and medical history (AS); second, by blinded independent video review by
four movement disorders specialists (EA, JH, CK, and AM), one of whom is
an expert in MD (EA); third, by evaluation of questionable cases by two
blinded external collaborators (SB and RSP). Finally, a consensus diagnosis
was established.
Molecular analysis
Peripheral blood samples were collected from all probands, and DNA was
extracted and screened for the known three-nucleotide (GAG) deletion in the
DYT1 gene. To test for a possible involvement of the DYT7 locus, suitable
families were investigated for linkage using the following six DNA microsa-
tellite markers: D18S481, D18S54, D18S976, D18S452, D18S843, and
D18S1153.
RESULTS
Consensus diagnoses
The study procedure and main results are illustrated in Figure 1. MD was
established by onsite examination and videotape review in all 14 FH+ (defi-
nite: n=13, probable: n=1) and in all 14 FH- (definite: n=14) index patients. In
total, 97 (56 FH+, 41 FH-) first- and second-degree relatives of these index
patients were examined. Seventeen of the 56 FH+ relatives had a previously
known FTSD (MD: n=7, writers cramp (WC): n=10), two of these relatives
were deceased. Family history interview revealed another two deceased rela-
tives with MD, one FH+ and one FH-. Using the BIDS in 78 putatively unaf-
Figure 1. Flow chart of the study displaying study procedure and results. FH+=reported
positive family history of FTSD, FH=no known family history, pts.=patients,
MD=musicians dystonia, FTSD=focal task-specific dystonia, BIDS=Beth Israel Dysto-
nia Screen, def=definite, prob=probable, poss=possible, (+)=additional dystonias or
additional other movement disorders in some individuals definitely, probably, or possi-
bly present. (See full color version at www.performancescience.org.)
fected family members, an additional 11 individuals screened positive for

dystonia, 7 of whom were members of FH- families.
A type of dystonia was confirmed as definite or probable in 19 of the 26
relatives who had a previously known form of FTSD or screened positive with
the BIDS. Fourteen of these 19 probands were members of FH+ (MD: n=7,
other FTSD: n=6, other dystonias: n=1) and five of FH- (MD: n=1, other
FTSD: n=3, other dystonias: n=1) families. In 27 of the 47, FH+ and FH-
family members with definite or probable dystonia one (n=18), two (n=8), or
three (n=3) additional forms of dystonia were present at least possiblyi.e.
these patients were not only affected with one dystonia type (e.g. MD) but
also with another FTSD (e.g. WC) or additional other dystonias (e.g. cervical
dystonia). In 23 of the aforementioned 47 family members, one (n=20), two,
(n=2), or three (n=1) additional other movement disorders or movement ab-
normalities were diagnosed as definitely or probably present, including
tremor (n=6), tics (n=4), chorea (n=2), involuntary perioral movements
(n=2), athetosis (n=1), mirror movements (n=8), and parkinsonian features
(n=4). In total, 18 (13 FH+, 5 FH-) of the entire set of 28 families were multi-
plex families (based on definite, probable, and possible diagnoses of dystonia)
with two to four affected family members in one (n=6), two (n=10), or three
(n=2) generations, compatible with autosomal dominant inheritance in at

least 12 families.
Molecular findings
The DYT1 GAG deletion was not present in any of the tested patients. Due to
the small number of affected family members, linkage to the DYT7 locus
could not be excluded in 10 of the 11 informative families. In one family, af-
fected offspring did not share a common DYT7 haplotype with their affected
father. Therefore, linkage to the DYT7 locus could be excluded in this family.
DISCUSSION
The present study expands previous findings of the presence of dystonia in a

considerable number of relatives of index patients with MD with an auto-
somal dominant pattern of transmission (Schmidt et al. 2006). Affected rela-
tives were identified both of index FH+ and FH- patients. Although none of
the 14 FH- index patients reported any cases of dystonia in their families, five
of them (36%) had affected relatives with dystonia on clinical examination.
MD, however, has long served as a textbook example of a purely occupational
dystonia, even more so than other forms of FTSD such as writers cramp
(WC). Due to the large number of familial cases observed also in MD, the
concept of MD as a sporadic and solely environmentally acquired type of
dystonia needs to be reconsidered. Surprisingly, a considerable number of
identified patients with MD or WC displayed additional types of dystonia. In
addition, an unexpected number of other hypo- and hyperkinetic movement
disorders, some of them unusual, were present in a considerable number of
participants. As a broad intra- and interfamilial phenotypic spectrum is
known for many genetic movement disorders, it is tempting to speculate that
at least part of the observed additional movement disorders in our patients
are due to a shared underlying genetic cause. Not surprisingly, none of our
patients carried the GAG deletion in the DYT1 gene that has only rarely been
linked to focal dystonia (Defazio et al. 2007). Due to relatively small number
of affected family members, linkage to the DYT7 gene locus that has been
described in two families with focal dystonia (Defazio et al. 2007) could not
be definitively excluded in all but one family. Based on the results of the pre-
sent study, our main hypothesis is that at least some cases of MD, other forms
of FTSD, and possibly even other types of movement disorders may have a
shared underlying genetic cause. The genetic factors that cause or contribute
to focal dystonia still have to be identified.
Acknowledgments
The authors thank the patients and family members for their participation in the study.
This work was supported by a grant from the Dystonia Medical Research Foundation,
the Bachmann Strauss Foundation, and the Volkswagen Foundation. CK is a recipient of
a Schilling Award from the Hermann and Lilly Schilling Foundation.
Alexander Schmidt, Department of Neurology, University of Lbeck, Ratzeburger Allee

160, Lbeck 23538, Germany; Email: alexander.schmidt@neuro.uni-luebeck.de
References
Altenmller E. (2003). Focal dystonia: Advances in brain imaging and understanding of

fine motor control in musicians. Hand Clinics, 19, pp. 523-538.
Bressman S. B., Raymond D., Wendt K. et al. (2002). Diagnostic criteria for dystonia in
DYT1 families. Neurology, 59, pp. 1780-1782.
Defazio G., Berardelli A., and Hallett M. (2007). Do primary adult-onset focal dystonias
share aetiological factors? Brain, 130, pp. 1183-1193.
Friedman J. R., Klein C., Leung J. et al. (2000). The GAG deletion of the DYT1 gene is
infrequent in musicians with focal dystonia. Neurology, 55, pp. 1417-1418.
Saunders-Pullman R., Soto-Valencia J., Costan-Toth C. et al. (2005). A new screening
tool for cervical dystonia. Neurology, 64, pp. 2046-2049.
Schmidt A., Jabusch H. C., Altenmller E. et al. (2006). Dominantly transmitted focal
dystonia in families of patients with musicians cramp. Neurology, 67, pp.691-693.
Schmidt A., Jabusch H. C., Altenmuller E. et al. (2009). Etiology of musicians dystonia:
familial or environmental? Neurology, 72, pp. 1248-1254.
New aspects in action planning and execution

in musicians with dystonia
Mara Herrojo Ruiz1, Hans-Christian Jabusch1,2, and Eckart Altenmller1

Recent neurophysiological studies have associated focal-task specific

dystonia (FTSD) with impaired inhibitory function. However, it remains
unknown whether FTSD also affects the inhibition (INH) of long-term
overlearned motor programs. Consequently, the question whether or not
musicians dystonia (MD) affects the inhibition (INH) of long-term
overlearned motor programs was addressed. By means of electroe-
ncephalography (EEG), the neural correlates associated with INH of
long-term overlearned motor memory traces were investigated in MD
and healthy pianists in a Go/NoGo paradigm. The findings support the
hypothesis of a deficient phase coupling between the neuronal assemblies
required to inhibit motor memory traces in patients with MD. Further-
more, in NoGo trials, the movement related cortical potentials showed a
positive shift after the NoGo signal related to inhibition and was signifi-
cantly smaller over sensorimotor areas in musicians with MD. Lastly,
EMG recorded from the right flexor pollicis longus muscle confirmed
that patients with MD had a disrupted INH in NoGo trials.
Keywords: music; performance; dystonia; inhibition; EEG
Musicians dystonia (MD), a form of focal task-specific dystonia (FTSD), is

characterized by a degradation of these motor memory traces. MD is a
movement disorder, which occurs while a musician is playing the instrument,
and is marked by the painless loss of voluntary motor control of extensively
trained movements (Altenmller 2003). Recent neurophysiological studies
have provided evidence for the hypothesis that musicians dystonia is associ-
ated with impaired inhibitory function and abnormal movement preparation
(for a review, see Lim et al. 2001). The latter has been reported, for instance,
in data of the Bereitschaftspotential (BP) (Deuschl et al. 1995) and the con-
tingent negative variation (CNV) (Lim et al. 2001). Furthermore, in a recent
study (Hummel et al. 2002), deficient inhibition of simple motor patterns
was demonstrated in six patients with FTSD using transcranial magnetic
stimulation (TMS) and EEG-alpha oscillatory activity. However, it remains
unknown whether FTSD also affects the inhibition (INH) of long-term over-
learned motor programs. Consequently, the aim of the present study was to
investigate, using a Go/NoGo paradigm, the neural correlates associated with
the activation (ACT) and inhibition (INH) of motor memory traces in pianists
with MD during a pianistic motor task under constraint timing conditions
with multichannel EEG. Our main hypothesis was that the functional cou-
pling during INH is impaired for pianists with MD compared with healthy
pianists.
METHOD
Participants
Nine healthy pianists (8 males, age range=26-47 years, mean=36.5 years) and
nine pianists with MD (8 males, age range=27-50 years, mean=35.3 years)
participated in this study. In all patients, the right hand was affected. All par-
ticipants were professional pianists (with accumulated practice time over
10,000 hours). Eight of the nine participants in each group were right-
handed, according to the Edinburgh Inventory (Oldfield 1971).
Procedure
Participants were seated at a digital piano (Wersi Digital Piano CT2). The
keyboard and the right hand of the participant were covered with a board to
prevent participants from visually tracking hand and finger movements. In a
modified Go/NoGo study, the task was to play upward C-major scales over
two octaves. Scales were played as sixteenth notes, and the tempo was stan-
dardized at 80 beats/min for a quarter note (one key stroke every 187.5 ms)
and paced by metronome-like auditory cues. Scales were played using the
conventional C-major fingering: 1,2,3,1,2,3,4,1,2,3,1,2,3,4,5 (the fingers 1-5
refer to thumb, index, middle, ring, and little finger, respectively). The speci-
fications of the Go/NoGo study were as follows: a first visual cue (S1) indi-
cated that participants should be prepared to start playing. The metronome
was started 2,750 ms after S1. Participants were instructed to play the first
note of any scale coinciding with the third metronome beat. A second visual
cue (S2) was presented 250 ms before the third metronome beat, indicating
that the participant should either execute (Go, green ellipse) or not execute
(NoGo, red ellipse) the motor sequence. Continuous EEG was recorded from
22 electrodes placed over the scalp according to the extended 10-20 system
referenced to linked mastoids. Data were sampled at 500 Hz; the upper cutoff
was 100 Hz, and the time constant was set to DC (DC amplifiers and software
by NeuroScan, Herndon, Virginia, USA). One bipolar EMG channel was re-
corded from surface electrodes positioned over the right flexor pollicis longus
muscle, located 6 cm apart. The bandpass filters for EMG were set to 5 Hz
(highpass) and 100 Hz (lowpass).
We performed the following analyses of the EEG signals: (1) standard
time averaging technique to analyze the slow shift of movement-related corti-
cal potentials (MRCPs), (2) wavelet based time-frequency representations
(TFR) to analyze (3) the spatiotemporal dynamics of the inter-electrode phase
coupling. All statistical tests were performed by means of nonparametric uni-
variate permutation tests and multivariate synchronized permutations (Good
2005).
RESULTS
EMG analysis
The amplitude mean value in the analysis interval and the EMG peak were
selected as EMG activity parameters. In NoGo trials, the selected time win-
dow was 0-350 ms to detect whether pianists initiated a movement around
250 ms in spite of the NoGo signal. Burst of EMG activity showed that par-
ticipants initiated movements of the thumb in some NoGo. Interestingly,
however, the first key of the MIDI piano was not actually pressed. This result
confirmed that inhibition in our paradigm demanded active suppression of
the motor program. The EMG peak was found significantly higher in pianists
with MD (mean=14 V, range=5.2-50 V) than in healthy ones (mean=5.5
V, range=3.1-16 V).
Movement-related cortical potentials
Both groups showed similar pre-movement activity over the sensorimotor

cortex, characterized by the slow negative MRCP termed as CNV (Figure 1).
In our paradigm, the CNV reflected the maintenance of a motor response in
readiness (Haider et al. 1981). In NoGo trials, the pre-movement negativity
returned to baseline levels and later a positive shift post-S2 was observed.
The positive peak after S2, which could be related to the inhibition of the
Figure 1. Movement-related cortical potentials (MRCPs) analysis. Grand average of the

MRCPs at electrode FCz for NoGo trials in pianists with MD (blue line), healthy pianists
(black line), and difference (MD minus healthy, red line). The temporal interval of sta-
tistical significance (p<0.0017, permutation test) in the between-groups difference is
indicated by the pink region. (See full color version at www.performancescience.org.)
motor pattern, had a significantly larger amplitude across sensorimotor areas

in healthy pianists than in pianists with MD (p=0.0016).
Inter-electrode cortico-cortical phase synchronization
NoGo trials were associated with a robust increase in the degree of global
synchronization across the sensorimotor cortex in the theta (4-7 Hz) and
lower alpha band (7-8 Hz) with a time span of 200-400 ms, hence coinciding
with the latency when the participants were required to begin playing (Figure
2). This increase was significantly more enhanced between 230-330 ms for
healthy pianists than for pianists with MD (p=0.004, permutation test, uni-
variate test). This effect was due to a weaker phase synchronization in pian-
ists with MD between electrode Cz, representing the supplementary motor
areas (SMA), and left premotor and sensorimotor electrodes (FC3, C3, CP3).
This result confirmed our main hypothesis, that the functional coupling dur-
ing INH is impaired for pianists with MD compared with healthy pianists.
DISCUSSION
Our study focused on the execution and inhibition of long-term overlearned

motor programs, due to its relevance in real playing conditions. Our assump-
tion was that in the non-retrieval condition, motor memory traces, strongly
activated after the first metronome beat, needed to be suppressed after S2
Figure 2. Phase synchronization analysis during INH. The time-frequency plots of the
inter-electrode phase synchronization index, averaged across electrodes over sensori-
motor and prefrontal areas, are presented for pianists with MD (A), for healthy pianists
(B), and for the between-group difference (C, A-B). A pointwise paired permutation test
between groups yielded significant differences (p=0.004) between 230 and 330 ms and
at 7-8 Hz, due to higher global synchronization in healthy pianists than in pianists with
MD in this time-frequency window. This region is indicated by the green contour. (See
(Hummel et al. 2002). Here, we aimed at studying a task that is close to natu-
ralistic piano performance. Accordingly, we imposed high temporal con-
straints on the task, we used a large sample of patients suffering from MD,
and we had healthy musicians as controls.
In this setting, (1) the role of the inter-electrode functional coupling in the
sensorimotor integration of inhibitory processes turned out to be the most
relevant physiological marker. Our study further showed that in pianists with
MD, (2) the non-retrieval of the motor program was associated with a weaker
positive shift after-S2 over cortical sensorimotor areas. Finally, (3) the EMG
peak in NoGo trials was found to be significantly higher in pianists with MD
than in healthy pianists. Our findings, thus, offer evidence that patients with
MD, as compared with healthy pianists, have a significantly higher innerva-
tion input of the flexor pollicis longus during NoGo trials. This outcome sup-
ports the main hypothesis of deficient inhibition in pianists with MD.
The SMA is thought to play an important role in the functional control of
movement in that it has direct projections to the primary motor cortex and
the spinal chord (Matsuzaka et al. 1992). Recent data has proven the suppres-
sive influence of SMA on the primary motor cortex (M1) in motor imagery,
thus reflecting the inhibitory function of the forward connection between the
SMA and M1 (Kasess et al. 2008). Hence, our results could be interpreted as
a deficient higher order motor functioning in pianists with MD: the phase
coupling between the SMA and the left premotor and sensorimotor cortex,
which is required for the non-retrieval of the motor program, is weaker in
pianists with MD. Consequently, these data can be regarded as an electro-
physiological correlate of the impaired inhibition in pianists with MD.
Acknowledgments
This work was funded by the Center of Systems Neuroscience, Hanover, and a Marie
Curie Early Stage Training Contract (MEST-CT-2005-021014) to Mara Herrojo Ruiz.

References

Deuschl G., Toro C., Matsumoto J., and Hallett M. (1995). Movement-related cortical
potentials in writers cramp. Annals of Neurology, 38, pp. 862-868.
Good P. (2005). Permutation, Parametric, and Bootstrap Tests of Hypotheses (2e).
New York: Springer.
Haider M., Groll-Knapp E., and Ganglberger J. A. (1981). Event-related slow (DC) po-
tentials in the human brain. Reviews of Physiology, Biochemistry, and Pharmacol-
ogy, 88, pp.125-197.
Kasess C. H., Windischberger C., Cunnington R. et al. (2008). The suppressive influence
of SMA on M1 in motor imagery revealed by fMRI and dynamic causal modeling.
NeuroImage, 40, pp. 828-837.
Lim V. K., Altenmller E., and Bradshaw J. L. (2001). Focal dystonia: Current theories.
Human Movement Science, 20, pp.875-914.
Matsuzaka Y., Aizawa H., and Tanji J. (1992). A motor area rostral to the supplementary
motor area (presupplementary motor area) in the monkey: Neuronal activity during
a learned motor task. Journal of Neurophysiology, 68, pp. 653-662.
Oldfield R. C. (1971). The assessment and analysis of handedness: The Edinburgh in-
ventory. Neuropsychologia, 9, pp. 97-113.
Functional and morphological changes of brain

structures in patients suffering from
musicians dystonia
Eckart Altenmller1, Oliver Granert2, Martin Peller2, Hans-Christian

Jabusch3, and Hartwig Siebner2,4

2 Department of Neurology, University Hospital Schleswig Holstein,
Campus Kiel, Germany
4 Danish Research Centre for Magnetic Resonance, Hvidovre
University Hospital, Copenhagen, Denmark
Focal hand dystonia has been associated with morphometric changes and
distorted somatotopic representations in the putamen. Our objective was
(1) to test for morphometric alterations of the putamen in pianists with
musicians dystonia (MD) relative to healthy pianists without dystonia
and (2) to identify structural changes in the basal ganglia that correlate
with performance during piano playing. Eleven pianists with MD and 12
healthy pianists without dystonia underwent high-resolution T1-
weighted MRI of the whole brain. Additionally, motor performance was
investigated in a music-related task according to a protocol previously
described as a valid and reliable method to assess motor control in pian-
ists. When playing major scales on the piano, the timing of key strokes
was more variable in patients with MD than in pianists without dystonia.
Healthy musicians had a smaller grey matter volume in the right middle
putamen compared with MD patients. In dystonic and non-dystonic pi-
anists, the middle part of the left and right putamen was smaller in indi-
viduals with higher temporal accuracy during piano playing. A smaller
associative territory of the motor putamen is a structural marker for
manual skillfulness in highly trained pianists. Since this structure-func-
tion relationship is preserved in MD, we argue that the relative increase
in grey matter volume in this region reflects impaired performance in
dystonic musicians rather than a specific structural substrate of focal

hand dystonia.
Keywords: focal dystonia; pianists; basal ganglia; morphometry; per-

formance skill

Setting the stage for prevention and treatment:

New therapeutic approaches in musicians
dystonia
Hans-Christian Jabusch1, Franziska Buttkus2, Volker Baur2,3, Maricruz

Gomez-Pellin2, Laurent Boullet4, and Eckart Altenmller2

3 Psychiatric University Hospital, University of Zurich, Switzerland
4 Department of Music Physiology, International Piano Academy, Lake Como, Italy
Musicians dystonia (MD) is probably the most challenging disorder in

musicians medicine. It has a tremendous negative impact on the careers
of affected musicians and is difficult to treat. We have investigated
treatment effects of established therapies and new approaches on play-
ing-related motor control in pianists with MD.
Keywords: focal dystonia; musicians dystonia; retraining; botulinum

toxin; transcranial direct current stimulation
Focal dystonia in musicians (MD) is a task-specific movement disorder which

presents itself as a loss of voluntary motor control in extensively trained
movements while the musician is playing the instrument (Altenmller 2003).
For those who are affected, the disorder is highly disabling, and in many
cases, it terminates musical careers. According to estimations, 1% of all musi-
cians are affected. Defective inhibition on different levels of the central nerv-
ous system is thought to be involved in the pathophysiology (Lin and Hallett
2009). Treatment of patients with MD is still a challenge. Therapeutic options
include anticholinergic medication with trihexyphenidyl (Trhx), botulinum
toxin (BT) injections, and pedagogical retraining (PR) (Jabusch and Alten-
mller 2006). Transcranial direct current stimulation (tDCS) has been shown
to modulate excitability of the motor cortex: anodal tDCS facilitated motor
learning, and cathodal tDCS improved performance in overlearned tasks
(Nitsche et al. 2003, Antal et al. 2004). It may be hypothesized that tDCS
facilitates retraining effects in musicians with dystonia. In this paper, we pre-
sent outcome data in 33 pianists with focal dystonia after an average follow-
up period of 24 months. Pianists were treated using BT, Trhx, or PR, either as
monotherapies or in combination. Treatment monitoring was made using an
established protocol that allows assessment of motor control in a relevant
musical task. Additionally, short-term development after tDCS-supported
retraining in a single patient is reported.
METHOD
Participants
Study 1: 33 pianists with MD (mean age=37 years, range=21-71 yrs, 24 men, 9

women) were included in the follow-up study. Study 2: the short-term follow-
up study was performed in a 43 year-old male pianist suffering from MD. All
patients were diagnosed at the Institute of Music Physiology and Musicians
Medicine, Hanover University of Music and Drama. The diagnostic procedure
included a complete neurological examination as well as visual inspection
while patients were playing the piano.
Procedure
Assessment of motor control: motor control at the piano was assessed in

scale playing because this motor task is early affected during onset of MD.
MIDI-based scale analysis was done according to the following protocol
(Jabusch et al. 2004). Scales were performed with the affected hand on a
digital piano that was connected to a computer. Sequences of 10 to 15 C major
scales were played over two octaves in both playing directions. Scales were
played using the conventional C major fingering. The tempo was standardized
and paced by a metronome (one keystroke every 125 ms). The temporary un-
evenness of inter-onset intervals (IOI) has previously been identified as a
valid, reliable, and precise indicator of the impairment of motor control in
pianists with dystonia (Jabusch et al. 2004). For each participant, temporary
unevenness of IOI was analyzed for the affected hand and for both playing
directions by calculating the median standard deviations of IOI (mSD-IOI) of
all scales. The mSD-IOI score of the more severely affected playing direction
was used for further analyses. Motor control was assessed before and after
follow-up in Study 1 as well as before and after each treatment condition in
Study 2.
Treatment, Study 1: therapeutic approaches, as monotherapies or in si-

multaneous or successive combination, included the following options: PR
was applied in patients who preferred a non-medication treatment approach.
PR took place under the supervision of a piano instructor (LB) specialized in
dystonia retraining. PR included elements based on the following principles
reported previously (e.g. Boullet 2003): (1) movements of affected body parts
were limited to a level of tempo and force at which the dystonic movement
would not occur, (2) compensatory movements (e.g. of adjacent fingers) were
avoided, partially under the application of splints, (3) instant visual feedback
with mirrors or monitors helped patients to recognize dystonic and non-
dystonic movements. BT injections were applied in patients in whom primary
dystonic movements could be clearly distinguished from secondary compen-
satory movements. Target muscles were identified by visual inspection of the
dystonic movement patterns while patients were playing their instruments. A
lyophilized botulinum toxin A powder (Dysport, Ipsen Ltd., Berkshire, UK)
was injected using an EMG-guided technique. Trhx was applied as monother-
apy when treatment with PR or BT was not desired or possible and no con-
traindication was present. Adjustment of the dosage was made depending on
beneficial effects and side effects. Patients with little response during PR were
additionally treated with BT or Trhx when no contraindications were present.
Treatment, Study 2: in a double-blind single case study, tDCS was com-
bined with retraining on the piano that took place according to the aforemen-
tioned principles. The patient was treated with three stimulation protocols
consecutively, with a minimum of 5 weeks between treatment sessions: an-
odal tDCS, cathodal tDCS, and placebo stimulation (3x5 days). In the verum
stimulation conditions, tDCS (2 mA) was applied for 20 mins on the primary
motor cortex contralateral to the affected hand. During stimulation, the pa-
tient practiced slow, non-dystonic movement patterns on the piano. Motor
control was assessed before and 1 min, 60 mins, 120 mins, and 180 mins after
the respective treatment session.
Statistical analyses: Mann-Whitney-U tests were applied to analyze per-
formance differences. The alpha level was set at 0.05. In Study 2, alpha ad-
justment for multiple testing was made according to Bonferroni-Holm.
RESULTS
Study 1: monotherapies were applied in 23 patients: BT (n=8), Trhx (n=1),

PR (n=14). Treatment combinations were applied in five patients: PR+BT
(n=3), PR+Trhx (n=2). Five patients refused any treatment. Follow-up
monitoring after an average period of 24 months (range=3-57) revealed the
Figure 1. Study 1: each line indicates the mSD-IOI values of one patient before (left
endpoint) and after follow-up (right endpoint). A high score for mSD-IOI denotes a high
level of temporal unevenness in the scales (impaired motor control), while a lower score
for mSD-IOI denotes a lower level of unevenness (less or non-impaired motor control).
Solid black lines indicate significant improvement of motor control, solid grey lines
indicate no significant change in motor control, dashed grey lines indicate significant
deterioration in motor control. Gray window: range of motor control of healthy pianists.
following outcome. A significant improvement of motor performance was

seen in 20 patients (71% of the 28 treated patients) after BT (n=7), PR (n=8),
PR+BT (n=3), and PR+Trhx (n=2). Four of these patients (14% of all treated
patients) returned to a level of normal motor control as seen in healthy pian-
ists (PR: n=2, PR+BT: n=1, PR+Trhx: n=1). No change of motor performance
was seen in 8 patients after treatment (BT: n=1, Trhx: n=1, PR: n=6). In the
five untreated patients, motor performance remained unchanged (n=2) or

deteriorated (n=3). Detailed outcome of individual patients is displayed in
Figure 1.
Study 2: motor control at baseline did not differ between the stimulation
conditions. In the placebo condition, motor control was improved 1 min
(p<0.001) and 120 mins (p<0.05) after treatment as compared with before
treatment. In the anodal condition, motor control was improved 1 min
(p<0.001), 60 mins (p<0.001), 120 mins (p<0.01), and 180 mins (p<0.05)
after treatment. In the cathodal condition, motor control was improved 1 min,
60 mins, 120 mins, and 180 mins after treatment (each p<0.05). Inter-treat-
ment comparisons revealed a better performance outcome in the cathodal
condition compared with placebo 180 mins after treatment (p<0.001).
DISCUSSION
The results underline the potential benefit of a behavioral approach in the

treatment of pianists with MD. PR as a monotherapy resulted in improved
motor control in the majority of patients after follow-up. This result indicates
a beneficial effect of an active involvement of patients. As a limiting factor, PR
requires patience and compliance of affected pianists. BT injections were
successful in all but one of the patients treated with this option. When the
muscles involved in the dystonic movements can be clearly identified, BT
treatment may be recommended to patients, especially to those who are not
able to invest the time and patience required for PR. A treatment attempt
with Trhx was unsuccessful in the only patient treated with this option as a
monotherapy. Other publications report an improvement rate of 33% in pa-
tients with MD who tolerated the medication (Jabusch and Altenmller
2006). The applicability is, however, limited due to the frequent occurrence
of side effects. Treatment combinations such as PR+BT or PR+Trhx are
promising, especially in patients with little response to PR alone. In Study 2,
retraining resulted in prolonged effects when combined with cathodal tDCS
as compared to placebo stimulation. Further studies with large numbers of
patients and longer follow-up periods are, however, required to determine the
effects of (1) combination therapies based on PR and (2) tDCS-supported
retraining. Our results show that the currently available therapies help to
improve the situation in the majority of patients. However, they also demon-
strate that only exceptionally, affected pianists return to normal motor con-
trol.
This observation underlines the crucial importance of a successful pre-
vention. Previous findings suggest that different risk factors such as exagger-
ated perfectionism and anxiety can trigger the manifestation of MD (Jabusch

and Altenmller 2006) in predisposed musicians. These risk factors should
be addressed in the musical education from the first lesson onward. Music
teachers should strive to create a friendly, supportive atmosphere focusing on
creativity, curiosity, and playful experiences in the world of sounds. We as-
sume that the risk of MD may be reduced by including retraining principles
even in the teaching process for healthy pianists and by the application of
healthy practice strategies. Reasonable practice schedules, an economic
playing technique, prevention of overuse and pain, mental practice, a variety
of movement patterns, maintenance of motivation, avoidance of mechanical
repetitions and frustration, sufficient breaks, warm-up and cool-down exer-
cises, regular physical exercise, and sleep are the cornerstones of healthy mu-
sical practice. Future projects will have to investigate the preventive potential
of these strategies.
Hans-Christian Jabusch, Institute of Musicians Medicine, Dresden University of Music

Carl Maria von Weber, Wettiner Platz 13, Dresden 01067, Germany; Email: jabusch@
hfmdd.de
References

Antal A., Nitsche M. A., Kruse W. et al. (2004). Direct current stimulation over V5 en-
hances visuomotor coordination by improving motion perception in humans. Jour-
nal of Cognitive Neuroscience, 16, pp. 521-527.
Boullet L. (2003). Treating focal dystonia: A new retraining therapy for pianists. In R.
Kopiez, A. C. Lehmann, I. Wolther, and C. Wolf. (eds.), Abstracts of the Fifth Trien-
nial ESCOM Conference (pp. 273-274). Hanover, Germany: Hanover University of
Music and Drama.
Jabusch H. C., Vauth H., and Altenmller E. (2004). Quantification of focal dystonia in
Jabusch H. C. and Altenmller E. (2006). Focal dystonia in musicians: From phenome-
nology to therapy. Advances in Cognitive Psychology, 2, pp. 207-220.
Lin P. T. and Hallett M. (2009). The pathophysiology of focal hand dystonia. Journal of
Hand Therapy, 22, pp. 109-113.
Nitsche M. A., Schauenburg A., Lang N. et al. (2003). Facilitation of implicit motor
learning by weak transcranial direct current stimulation of the primary motor cor-
tex in the human. Journal of Cognitive Neuroscience, 15, pp. 619-626.
The effects of pianistic retraining via video

conferencing as a means of assisting recovery
from focal dystonia: A case study
Rae de Lisle1, Dale Speedy2, and John Thompson3
1 School of Music, University of Auckland, New Zealand

2 Department of General Practice and Primary Healthcare,
3 Department of Pediatrics, University of Auckland, New Zealand
Focal dystonia (FD) is a devastating neurological condition that can re-

sult in the loss of a musicians playing ability. In pianists, involuntary
muscle contractions can cause abnormal finger postures, making it im-
possible to play at concert level. It is a difficult condition to treat, and
although some improvement has been reported with Botulinum toxin
therapy, complete recovery is rare. Our study investigated whether re-
training with the aid of video conferencing could be helpful in the treat-
ment of a professional pianist resident in the UK with a six year history of
focal dystonia. Quality of scales was assessed before and after pianism
retraining and included assessment by a listener blinded as to which
hand was dystonic and whether they were assessing playing pre- or post-
retraining. Although full recovery was not seen, improvement was ob-
served at slow tempi and the hand was visibly less cramped as training
sessions progressed. We conclude that video conferencing could be an
acceptable medium for pianistic retraining in pianists with FD when lo-
cation prevents onsite retraining. However, in this study it did not seem
as effective as one-to-one retraining in the same location.
Keywords: focal dystonia; pianism; piano technique; retraining; video

conferencing
Focal dystonia is usually painless and most commonly affects only one hand,
often involving involuntary flexion of just two or three specific fingers. The
incidence may be as high as 1% of all professional musicians (Jabusch and
Altenmuller 2006). In focal dystonia, the areas in the brain responsible for
the movement of adjacent fingers have become enlarged, due to overuse, and
can overlap (Elbert 1998). Treatment options have included administration
of Trihexyphenidyl or Botulinum Toxin-A, splinting, and limb immobiliza-
tion, but only exceptionally do musicians with focal dystonia return to normal
motor control (Altenmuller and Jabusch 2007). Different methods of re-
training alone without the above medical interventions have reported anec-
dotal success, but few have applied scientific method. We have previously
reported successful treatment of FD with intensive one-to-one pianistic re-
training without the use of Botulinum Toxin. The aim of the present research
was to see whether it is possible to improve the condition of a pianist suffer-
ing from FD through pianistic retraining using the medium of video confer-
encing.
METHOD
Participants
A 65 year old male professional pianist with a five year history of FD affecting
digits 3-5 (D3-5) of his right hand participated in a specific retraining pro-
gram based on a biomechanically sound way of playing with minimal tension
through the medium of video conferencing. Subject characteristics are shown
in Table 1.
Materials
The following equipment was used: video conferencing facilities, two fixed
and two mobile cameras, and two pianos.
Procedure
The subject underwent pianistic retraining using video conferencing technol-

ogy. A collaboration was set up between the University of Auckland and the
Royal College of Music, London. In each location, two cameras were used:
one fixed and one mobile. Scales and test repertoire were recorded at the
commencement of the study, after ten sessions within three weeks, and then
after a further ten follow up sessions within the next year. The sound only of
three different scales and a diminished seventh arpeggio at tempi ranging
from 60-144 bpm (quarter note) was assessed by a professional pianist (the
blinded listener), blinded to which hand was playing and whether the play-
ing was pre-, post-ten sessions, or at the end of the retraining process. The
assessment used a Scale Quality Evaluation (SQE) and a Dystonic Hand
Identification Evaluation (DHIE) where the listener was asked to identify
Table 1. Subject characteristics.
Characteristic Data
Age 65 years
Sex Male
Level of performance Professional soloist
Years of playing 60
Duration of dystonia Six years
Hand dominance Right
Hand affected Right
Digits affected D3-D5
whether the hand playing was dystonic or not. The above evaluation scales
have been previously described by the author (de Lisle et al. 2006). Statistical
analysis was carried out using Generalised Linear Models in SAS v9 for Win-
dows. Estimates are the change in SQE from the initial assessment to the end
of the three week 10 session end point or the end of one year endpoint. Seven
months following commencement of retraining the author had two face-to-
face retraining sessions with the subject in London, but otherwise all training
sessions were carried out via video conferencing.
The subject had been used to playing with a very curved hand position
and had trained using finger independence exercises involving antagonistic
muscle movements simultaneously (e.g. Pischner exercises). The excessively
rounded hand position had caused the subject to play at the edge of the white
keys with the thumb over the key surface, and D2-5 rarely played between the
black keys. An ascending scale passage revealed radial deviation and any up-
ward movement of the wrist caused D3-5 to curl excessively. Descending scale
passages involved shoulder abduction when crossing over the thumb, causing
the elbow to elevate away from the body and the hand and wrist to pronate
while pivoting on the thumb. Therefore, the initial aim was to lower the wrist
and to use a flatter hand position and play with fingers less flexed. In scale
passages, the subject was instructed to avoid lifting the wrist when passing
over the thumb and instead to ride forward on D3, moving the arm toward
the fall of the piano.
Particularly challenging was any tremolo movement using rotation, where
D5 became stuck to the keys and the wrist would flex excessively. The subject
had a lack of freedom in forearm rotation and tended to direct the movement
from the wrist. This was corrected by reducing wrist palmar flexion and in-
Change in SQE by hand at two time points
0.8 Dystomic
.
0.7 Non-dystonic
0.6
Change in SQE
0.5
0.4
0.3
0.2
0.1
0
Initial sessions 1 year
Time of sessions
Figure 1. Change in SQE by hand at two time points.
creasing the rotation of the forearm (pronation and supination). Lifting the
fingers simultaneously from the metacarpophalangeal (MCP) joints was
problematic when pivoting on D2 because D5 tended to lag behind the other
fingers and this was given as a specific exercise. This lagging was also appar-
ent in descending scales but could be prevented if the movement of crossing
over the thumb was directed from D5 without flexing the wrist. In ascending
broken chord playing, passing from D5 to D1 caused excessive curling of D5,
but this was preventable at a slow tempo by actively extending D5 away from
the cramping movement. Passing from D2-D3 caused cramping of D5, but
this could be prevented by a downward convex movement of the wrist to align
the fingers with the keys. Chord playing using D1,3,5 was difficult, but D1,4,5
was accomplished with ease.
RESULTS
The analysis showed a significant improvement of the SQE from the pre-as-
sessment to the point in time of the three week, 10 session intensive retrain-
ing of 0.46 points [95% CI=(0.11,0.81), p=0.01], after controlling for
dystonic/ non-dystonic hand, this effect remained statistically significant. An
analysis of the effect by hand showed a small non-significant improvement in
the non-dystonic hand of 0.20 points [95% CI=(-0.13,053), p=0.23], while
the dystonic hand showed a statistically significant improvement of 0.68
points [95% CI=(0.20,1.17), p<0.01]. When the blind assessor was assessing
the playing, they correctly identified the non-dystonic hand 85% of the time
with the pre-assessment and 95% of the time at the end of the 10 sessions;
this was not statistically different (p=0.29). Whereas with the dystonic hand
the assessor was able to detect the dystonic hand 95% of the time at the pre-
assessment, this decreased significantly to 44% after training (p=0.0006).
When we compared the data from the pre-assessment to the end of year
assessment (see Figure 1), we again found an overall improvement, in this
case of 0.35 points [95% CI=(0.11,0.59), p=0.0044]. After controlling for the
effect of hand, this again remained significant. Similarly, analysis by hand
found a small but non-significant improvement in the non-dystonic hand of
0.15 points [95% CI=(-0.07,0.36), p=0.17] but a significant improvement in
the SQE for the dystonic hand of 0.53 points [95% CI=(0.24,0.82),
p=0.0004]. Again, assessment of the non-dystonic hand was successfully
assessed 81% of the time at the pre-assessment, and this improved slightly to
91% and the end of training (p=0.26). As before, the dystonic hand was accu-
rately detected 91% of the time at the pre-assessment, and this decreased
significantly to 69% at the end of training (p=0.03).
DISCUSSION
We have previously shown that it is possible to treat FD effectively in pianists

with intensive retraining (de Lisle et al. 2006). The novel finding of the pre-
sent study is that this retraining can be undertaken using video conferencing
technology. Although full recovery was not seen, improvement was observed
at slow tempi, and the hand was visibly less cramped as training sessions
progressed. We conclude that video conferencing could be an acceptable me-
dium for pianistic retraining in pianists with FD when location prevents on-
site retraining. However, in this study, it did not seem as effective as one-to-
one retraining in the same location. Many factors influenced this. First, tech-
nical problems often meant that the sessions were shorter than the planned
time of one hour. Also apparent was the fact that when retraining sessions
were too far apart, movement patterns were not easily maintained, and it was
difficult to maintain the motivation of the subject. Another distracting factor
was that this subject did not stop other concert playing during the retraining
process, and this tended to interfere with new movement patterns being
strongly established. The clearest understanding of the process occurred
when the sessions were close together (e.g. 10 sessions within three weeks). A
greater improvement in this process may have been possible had the follow-
ing sessions been at more regular intervals and if the subject had ceased all
other playing during the retraining process. However, feedback at distance to
both retrainer and subject lacks the element of touch, which is the most im-
mediate way of conveying the fine technical adjustments that are necessary
for recovery from focal dystonia. While video conferencing may be useful in
establishing new movement patterns the results are probably slower than
those possible in the same location. However, it could be an effective tool to
supplement and monitor progress after an initial period of onsite retraining.
Acknowledgments
The technical expertise of the IT departments at the University of Auckland and Royal
College of Music, London, is gratefully acknowledged.
Rae de Lisle, School of Music, University of Auckland, Private Bag 92019, Auckland,
New Zealand; Email: r.delisle@auckland.ac.nz
References
Altenmuller E. (2000). From Laetoli to Carnegie: Evolution of brain and hands as pre-
requisites of music performance in the light of music physiology and neurobiology.
Paper presented at the Tenth Symposium of the International Study Group of the
Archaeology of Music, Kloster Michaelstein, Germany.
Altenmller E. and Jabusch H. C. (2007). Focal dystonia in musicians: Recent results
and new developments. Paper presented at Medical Problems of Musicians and
Dancers, Aspen, Colorado, USA.
Altenmuller E. and Jabusch H. C. (2006). Focal dystonia in musicians: From phenome-
de Lisle R., Speedy D. B., Thompson J., and Maurice D. G. (2006). Effects of pianism
retraining on three pianists with focal dystonia. Medical Problems of Performing
Artists, 21, pp. 105-111.
Elbert T. (1998). Alteration of digital representations in somatosensory cortex in focal
hand dystonia. NeuroReport, 9, pp. 3571-3575.
Jabusch H. C., Zschucke D., Schmidt A. et al. (2005). Focal dystonia in musicians:
Treatment strategies and long-term outcome in 144 patients. Movement Disorders,
20, pp. 1623-1626.
Jabusch H. C. and Altenmller E. (2006). Focal dystonia in musicians: From phenome-
Schuele S., Jabusch H. C., Lederman R. J., and Altenmller E. (2005). Botulinum toxin
injections in the treatment of musicians dystonia. Neurology, 64, pp. 341-343.
Thematic session:
Understanding performance
Doing without thinking? Aspects of musical

decision-making
Daniel Bangert
School of English, Media, and Performing Arts, University of New South Wales,
Australia
When asked to explain their processes of musical decision-making, musi-

cians attribute their behavior to various causes. This paper explores how
musical decisions are made. A semi-structured interview was conducted
with a Baroque violinist about her interpretation of the violin line in the
aria Ich bin vergngt in meinem Leiden from J. S. Bachs Cantata BWV
58. Using interpretative phenomenological analysis (IPA), the interview
was coded to create categories describing differing influences on the
performers musical interpretation: feelings, score analysis, research,
technical considerations, and specific experiences. A second analysis
identified intuition and analysis as contrasting approaches to musical de-
cision-making. This paper focuses on the nature of intuitive methods of
making interpretative choices with reference to the interview data and
psychological literature on intuition.
Keywords: music performance; interpretation; decision-making; intui-

tion; phenomenology
Musicians are involved in complex and often rapid processes of decision-

making when preparing and performing a musical work. This paper explores
how musical decisions are made, focusing on the experiences of an individual
musician to identify influences on their artistic practice. Within this frame-
work of musical decision-making, this paper aims to distinguish between
intuitive and non-intuitive methods of making interpretative choices, focus-
ing on the nature of intuitive processes in particular.
Intuition has been conceptualized and defined in many ways (see
Sprenkle 2005). Betsch (2008) identifies three differing approaches to un-
derstanding the concept and how it may operate:
Some scholars focus on intuition as a source of knowledge. Accordingly,

intuition is what we know without knowing how we learned it. Others
suggest a process view by equating intuition with automatic or implicit
processes of thinking. Finally, one spots proponents of a system view
claiming that intuition is a distinct faculty of the human mind (p. 4).
This paper takes a process view of intuition based on dual process theo-
ries of cognition that distinguish between processes that are unconscious,
rapid, and automatic (System 1), and those that are conscious, slow, and de-
liberative (System 2) (see Hammond et al. 1987, Epstein 1991, 1994, 2003,
Hogarth 2001, Evans 2008). Many of the characteristics of System 1 can be
found in definitions of intuition:
Intuition is a process of thinking. The input to this process is mostly pro-

vided by knowledge stored in long-term memory that has been primarily
acquired via associative learning. This input is processed automatically
and without conscious awareness. The output of the process is a feeling
that can serve as a basis for judgments and decisions (Betsch 2008, p. 3).
METHOD
Procedure
A semi-structured interview was conducted with Baroque violinist Alice Ev-

ans in January 2009. Evans chose to discuss a piece by J. S. Bach with a
prominent violin line: the soprano aria Ich bin vergngt in meinem Leiden
from Cantata BWV 58. The interview questions were designed to elicit reflec-
tion on the interpretative choices involved in performing the piece. The inter-
view began with a hypothetical masterclass scenario asking Evans how she
would teach her interpretation of the piece and progressed to broader ques-
tions about musical decision-making.
Analytical framework
The interview transcript was analyzed using interpretative phenomenological

analysis (IPA) within the qualitative data analysis software program Atlas.ti
(version 6). IPA studies explore the meaning of a particular experience for the
participant by focusing on the participants views and perceptions of the topic
being investigated (see Landridge 2007, Willig 2008). IPA is a version of the
phenomenological method that does not separate description and interpreta-
tion; instead, it acknowledges that the researchers own conceptions are re-
Table 1. Influences on musical decision-making.
Category Discussion points Language examples

Feelings Musical characters I feel
Tempo choice You get the sense
Ornamentation
Baroque bow and instrument
Bass line
Score analysis Harmony I look for

Beat hierarchy Be aware
Placement of rests Noticing
Chord position
Articulation markings
Clarity of different parts
Research Text of cantata I learned
Technique Intonation Basic things
Producing good tone
Rhythm
Experiences Playing other pieces by Bach I played
Playing from different editions He said
Using first Baroque bow
Rehearsals with Sigiswald Kuijken
quired to make sense of the participants life world through a process of in-
terpretative activity (Smith et al. 1999, pp. 218-219).
RESULTS
During the interview, Evans discussed 20 different influences on her musical

decision-making in relation to interpreting the piece. Analysis of the tran-
script resulted in the creation of five thematic categories: feelings, score
analysis, research, technical considerations, and experiences (see Table 1).
The language used to discuss an aspect of the piecefor example, feel,
sense, look, or learninformed this process of categorization. Based on
the definition claiming that the output of intuition is a feeling (Betsch 2008),
5 of the 20 elements of Evans interpretation could be classed as operating via
intuition, while the categories score analysis and research demonstrate more
conscious, controlled decisions. The categories technique and experiences
Table 2. Discussion of decision-making processes.
Category Discussion points Language examples

Intuition Instinct Accumulative learning
Subconscious realization Thoughtlessness
Absorption
Innate sense
Reaction
Analysis Taste Thoughtfulness

Consciousness
seem to influence decision-making at varying levels of conscious awareness.

For example, Evans recalled some specifics from influential rehearsals and
performances but suggested that knowledge gained through experiences usu-
ally took place at a more general, less explicit level.
Although the categorization process demonstrates a mixture of influences
on her musical decision-making, Evans labeled herself an instinctive rather
than a predominantly learned player. This distinction is similar to Dunsbys
(2002) differentiation between performers who are more inclined to the
intuitive approach, and those who feel that some kind of thorough cognitive
underpinning yields better results (p. 226). Concepts related directly to in-
tuitive and non-intuitive or analytical methods of decision-making discussed
during the interview were identified and categorized through a second analy-
sis of the transcript (see Table 2).
DISCUSSION
While Evans frequently referred to instinct during the interview, instinct is

not guided by deep knowledge structures, prior learning, and expertise (Carl-
son 2004), and what Evans described could perhaps be more accurately
termed intuition:
For me there is an aspect in music which is accumulative learning and so

you do something automatically that is sort of the culmination of what
youve learnt up until that point and so you dont make a conscious deci-
sion about what youre doing. You do it completely without thinking. So
for me instinct, whether or not semantically thats correct, for me instinct
includes partially reacting without thought using the knowledge that
youve had up until that point. Therefore it can change, so the more you
learn, your instincts towards something can change. But for me, I use in-
stinct as the opposite of thoughtfulness, consciousness, so that you do
something without being conscious of why youre doing it in the moment.
Evans description of intuitive processes as being automatic, having low

conscious awareness, and being based on accumulated knowledge corre-
sponds to the characteristics of intuition as defined by Betsch (2008) and
other authors (see Hodginkson et al. 2008). These characteristics result in
decisions that make sense or feel right and cannot be readily explained by
a specific reason.
According to Weber and Lindemann (2008), the lack of conscious access
associated with expert intuition is either because previously conscious, ana-
lytic processes have become automated to a point in which conscious atten-
tion is no longer necessary or as the result of cumulative, associative learning
that has never been conscious (p. 191). The interview data suggests that for
Evans, much of her knowledge was not gained consciously, but through ex-
perience and what she termed absorption. The automation of analytic proc-
esses on the other hand is similar to Simons (1987) definition of intuition as
analyses frozen into habit (p. 63), which in musical contexts could apply to
learning technical skills or the activity of practicing in general.
This preliminary study has attempted to present a framework of decision-
making through which intuitive and non-intuitive processes involved in in-
terpreting a musical score can be identified. The study concentrates on the
subjective experience of musical intuition while acknowledging that self-re-
port data can present difficulties for the researcher. For example, musicians
may inaccurately attribute their behavior, certain processes may not be intro-
spectively available to them, or they may be anxious, defensive, or unaware
of what they do (Chaffin and Crawford 2007, p. 156). The author intends to
develop an understanding of how musical intuition is experienced by period
instrument string players through further interviews in combination with
other approaches to data collection.
Acknowledgments
I thank Alice Evans for participating in this study and am grateful for the assistance of
my joint PhD supervisors at UNSW, Dorottya Fabian and Emery Schubert.
Daniel Bangert, School of English, Media and Performing Arts, University of New South
Wales, Sydney, New South Wales 2052, Australia; Email: danielbangert@gmail.com
References
Betsch T. (2008). The nature of intuition and its neglect in research on judgement and
decision making. In H. Plessner, C. Betsch, and T. Betsch (eds.), Intuition in
Judgement and Decision Making (pp. 3-22). New York: Taylor and Francis.
Carlson N. R. (2004). Physiology of Behaviour (8e). New York: Pearson Education.
Chaffin R. and Crawford M. (2007). Unresolved dissonance? Subjectivity in music re-
search. In A. Williamon and D. Coimbra (eds.), Proceedings of ISPS 2007 (pp. 155-
160). Utrecht, The Netherlands: European Association of Conservatoires (AEC).
Dunsby J. (2002). Performers on performance. In J. Rink (ed.), Musical Performance
Epstein S. (1991). Cognitive-experiential self-theory: An integrative theory of personal-
ity. In R. Curtis (ed.), The Relational Self (pp. 111-137). New York: Guildford Press.
Epstein S. (1994). Integration of the cognitive and the psychodynamic unconscious.
American Psychologist, 49, pp. 709-724.
Epstein S. (2003). Cognitive-experiential self-theory of personality. In T. Millon and M.
J. Lerner (eds.), Handbook of Psychology (pp. 159-184). Hoboken, New Jersey,
USA: John Wiley and Sons.
Evans J. (2008). Dual-processing accounts of reasoning, judgment, and social cogni-
tion. Annual Review of Psychology, 59, pp. 255-278.
Hammond K. R., Hamm R. M., Grassia, J., and Pearson, T. (1987). Direct comparison of
the efficacy of intuitive and analytical cognition in expert judgment. IEEE Transac-
tions on Systems, Man, and Cybernetics, 17, pp. 753-770.
Hodginkson G. P., Langan-Fox J., and Sadler-Smith E. (2008). Intuition: A fundamen-
tal bridging construct in the behavioural sciences. British Journal of Psychology,
99, pp. 1-27.
Hogarth R. (2001). Educating Intuition. Chicago: University of Chicago Press.
Landridge D. (2007). Phenomenological Psychology. Harlow, UK: Pearson Education.
Simon H. A. (1987). Making management decisions: The role of intuition and emotion.
Academy of Management Executive, 1, pp. 57-63.
Smith J. A., Jarman M., and Osborn M. (1999). Doing interpretative phenomenological
analysis. In M. Murray and K. Chamberlain (eds.), Qualitative Health Psychology
(pp. 218-240). Thousand Oaks, California, USA: Sage Publications.
Sprenkle J. M. (2005). Defining Intuition: A Framework for Understanding Intuition
in Psychology. Unpublished doctoral thesis, Alliant International University.
Weber E. U. and Lindemann P. G. (2008). From intuition to analysis. In H. Plessner, C.
Betsch, and T. Betsch (eds.), Intuition in Judgement and Decision Making (pp. 191-
208). New York: Taylor and Francis.
Willig C. (2008). Introducing Qualitative Research in Psychology (2e). Maidenhead,
UK: Open University Press.
Applications of formal analysis: Musical

comprehension and memory consolidation in
performance
Elena Esteban Muoz1,2
1 Faculty of Education, Complutense University of Madrid, Spain

2 Royal Conservatoire of Music, Madrid, Spain
This paper reflects principally on the perception of musical form and its
dependence on the temporality in front of the conceptual system, related
to the nominalizations and numerical groupings of the theoretical repre-
sentation of the structure. Through the assumption of practice as the only
way to experience music, this paper calls for sensations and emotional
intelligence with regard to musical appreciation, including form as one of
the most important elements of musical comprehension. The capacity of
the performers for evoking musical form, through its knowledge syn-
chronically to its temporal experience, is defended as a potential tool for
consolidating memory due to the security it provides in performance.
Keywords: formal analysis; memory; musical comprehension; time;

performance
Independent of the consideration of music as a language and consequently a

communication system (Maconie 1990), or as an abstract expression of
something not related to the will of meaning, a structure exists that defines
it in length and articulates the presentation of its sonorous content. Consid-
ering music as a language or a discourse the form refers to its syntax (Vega-
Rodrguez and Villar-Taboada 2001) and, taking into account the exclusive-
ness of its abstract expression, avoiding a semantic quality, and noticing its
ineffable condition (Janklvitch 1983), the form constitutes the design of
those presented. In both cases, the complex organizational analysis of the
developed elements in a composition fosters the comprehension, apprehen-
sion, and assessment of its contents by the performer. The knowledge of the
form of a piece, through its analysis, allows the performer to connect the syn-
chronic relationship of execution and aural capacity with the prevision or, as
Huron (2006) explains, with a sweet anticipation that makes possible the
evocation/recovery (the last period of memory capacity after appropriation
and storage) of that which is to be played. It is absolutely feasible to play,
even by heart, ignoring the formal dimension of a piece, although it is not
possible to perform (Stravinsky 1942) a musical work without a solid as-
sumption of its form. Tovey reminds us that players should understand what
they play (Lester 1995), and formal analysis is a relevant tool for connecting
the multiple musical elements (values, frequencies, rhythm, melody, nuances,
tempo, agogic, harmony, motives, cadences, etc.) for this purpose.
Performing from memory can trigger fears or insecurity about final re-
sults in live performances. But at the same time, playing by heart awakens in
players a sensation of freedom with regard to the musical appropriation
which provides a very convenient state of introspection and individuality
(Delannoy 2008) with the support of the emotional intelligence (Goleman
1995). While in theory the formal analysis rests on the number (number of
sections, phrases, motives, measures, notes, etc.), the execution embraces
time as the perception of form. Music is a temporal artistic representation,
and this is the crucial point through which a performer may transcend what
he or she appreciates at a cognitive level to the experience of practice (Elliot
1995). Rousseau (1760) explained that only through our own movement we
acquire the idea of extension (p. 82). In a parallel way, it is possible to affirm
that only through the experience of time can we acquire the final idea of mu-
sical form. Intellect and perception (Renaud 1999) plus imagination are the
keys for consolidating musical memory (Snyder 2000, Barbacci 1987) in or-
der to attempt a rhetoric expression of structure.
MAIN CONTRIBUTION
To know the limits of the musical form does not mean to understand it. The
border that divides the knowledge and the comprehension of the musical
structure can be as wide or as narrow as the capacity of the performer for
letting himself or herself be attracted by the temporal experience. For a theo-
retical analysis of the structure of a piece it is indispensable to develop logi-
cal-mathematical and spatial intelligences (Gardner 1993). Logical-
mathematical intelligence is required for synthesizing, through the number, a
pieces parts and limits, while spatial intelligence is required as a visual help
for speeding up the logical-mathematical intelligence in the preliminary stage
of a pieces progress. Perhaps both intelligences were considered by Gardner
as constituent parts of another of his defined intelligences: the musical intel-
ligence. However, for transcending the results of the theoretical recognition
of the form to the practical perception, it is necessary to turn to the emo-

tional intelligence bonded to the physical sensations. Goleman (1995, p. 15)
insists on the experience as the basic vital attitude in front of the concept
in a proper free flowing emotional intelligence.
What is the route of experience? The route of experience is time. But the
perception of time is evidenced only through the experience. In this process
of complicities, the experience of the physical/vital movement in performance
is essential in order for the performer to be conscious synchronously to the
execution of the start and ending of the frequencies, musical cells, phrases or
sections of a composition. This is just what form represents. Because of past
assumptions this reflection focuses the attention on the comprehension of
music, including its form or structure, as an act. Elliot asserted that the
aesthetic concept of music-as-object obscures the more fundamental reality of
music as a form of deliberate doing and making (Elliot 1995, p. 49). The
musical comprehension through the perception of form via time illuminates a
very important demand in professional performance: memorization.
The duration: The formal essence of music
The form is what defines, configures, or holds up an object due to the limits
that it imposes on the matter of which they are made. In music, once it is
transcended from the score/object to the sonorous world thanks to practice,
the only way to appreciate its form is hearing it attentively and letting the
time occur, trying to perceive when the sound and the multiple associations
that the composer establishes with it start and end. This is the perception of
the sound duration. Kivy (2001) affirms:
That musical form is spatial I can only take to be a claim based on the
conflation of musical form with its mode of representation. It is an un-
fortunate choice of words to have started calling sonata, variation, rondo,
and so forth formsat all, for the obvious reason that form is a word
more closely connected in our common usage with two-dimensional or
spatial images. But musical forms are not, except metaphorically (al-
though the metaphor is ubiquitous) musical shapes. They are temporal
patterns (p. 211).
Similarly Tarchini (2004), pianist and formal analyst, affirms that:
The possibility of a conscious monitoring of the temporal progress im-

plies a great capacity for compiling, synthesizing, associating and under-
standing the obtained data from the score. This capacity allows us to
corroborate with certainty the perceptive and formative unities, their
functions, hierarchies and the different levels of tension-rest (p. 311).
What is important in Tarchinis words for this study is the acceptance she
assumes of a temporal awareness in performance and its connection with the
musical form.
The number: The organizer concept of the musical form for theory
Although for the performers and the audience the musical form is essentially
perceived in time, in theory the form is conceptually measured in the num-
ber of figures and silences, measures, cells, semi-phrases, phrases, themes,
motives, sections, or movements registered through a graphic code in a par-
ticular space: the score. The resultant quantities of all the elements of the
musical form are, in theory, as static and tangible as the papers fragments
that contain them. Musical theory has named some of the most common mu-
sical forms throughout history. Sonata, lied, or fugue are some of these stan-
dardized forms. But, obviously, they do not represent the fingerprint of any
particular piece. When theory wants to recognize and describe a specific
piece, it uses the number to elaborate the exhaustive and individual structure
of its content. As an example, the formal fingerprint at the beginning of
Chopins Scherzo Op.20 is represented in Table 1. Evidently, the names motif,
exposition, section, introduction, or transition that appear in the analysis-
table are valid for many other masterpieces, usually in the same order, but
what is almost impossibleor would be the result of chanceis that the se-
quences that conform to the different named parts are the same in other
pieces regardless of the number and the association with the measures they
have.
Strategies of musical memory: Formal evocation
Through the formal analysis of a composition (theoretical and temporal-per-

ceptive), the musical contents remain mentally delimited. The resultant com-
partments of the contents delineation depend, among other aspects, on the
meaning, inspiration, aesthetic, harmony, style, syntax, rhythm, melody,
musical prosody, expressiveness, or technical exigencies that affect them;
qualiasinherent proprieties of sensorial states (Delanoy 2008, p. 111) and
so on. The order imposed by performers to these formal compartments de-
mands of the musicians a volatile impulse of interaction and expectation of
their elements, which represents in a hypothetical memorized performance,
Table 1. Chopins Scherzo No. 1 Op. 20 in B minor for piano.
Section Measures
Introduction 1-8
Section A 9-68
Exposition Theme A-1
Motif A-1 9-16
Presentation motif A-1 9-11
Transport motif A-1 11-13
Conclusion motif A-1 13-16
Motif A-1 transported 17-24 etc.
two important resources for avoiding the so feared lapse. Even when a slip
emerges, the musical mind tries to reconstruct the material with reference
to the specific details of the general elements. Form is essentially a general
element so, if performers remember it, the last opportunity to reconstruct
the musical material is in some way guaranteed. Memory survives partially
thanks to the growth of security; when a commanded formal evocation be-
gins, the consciousness of the musical content is more effective and the per-
formance evolves synchronically from a passive verification of audition
synchronically to a simple and physical execution to a controlled mental
complex. This is thanks to the re-creative intention, which in symbiosis with
the bodily experience provides a sensation of pleasure and security in prac-
tice.
IMPLICATIONS
The benefits of involving formal analysis in performance as a memory re-

source include: (1) the potential for music comprehension, (2) the evidence of
the acquired knowledge as a tool for continuing to learn from it, (3) the cul-
tural information that it provides about structural models, (4) the mental
appropriation of musical content, (5) the effective use of practice and teach-
ing hours, (6) the prevention of lapses due to the effort of construction and
de-construction of musical material, (7) the knowledge of formal patterns
useful for improvising and composing, (8) the attention to the musical breath
and musical discourse apart from technical necessities, and (9) the mathe-
matical stimulus of mental processing. To resort to the formal analysis repre-
sents an exercise of compromise and faithfulness to the composer and his or
her creation, which forms the basis of a possible performance.
Acknowledgments
I wish to acknowledge Juan Pineda (Spain) and Pamela Huebner (Armstrong School,
Hoffman Estates, Illinois, USA).
Elena Esteban Muoz, Royal Conservatoire of Music, C/Doctor Mata 2, Madrid 28012,
Spain; Email: elenaesteban@edu.ucm.es
References
Barbacci R. (1987). Educacin de la Memoria Musical. Buenos Aires: Ricordi.

Delannoy L. (2008). El Espejo. Ensayos Sobre la Consciencia Musical Seguidos de la
Consciencia Inacabada. Mrida, Yucatn, Mexico: Centro de Investigaciones en
Neuroesttica y Neuromusicologa (CINN).
Elliot D. J. (1995). Music Matters. Oxford: Oxford University Press.
Gardner H. (1993). Inteligencias Mltiples (translated 2005). Barcelona, Spain: Paids-
Surcos.
Goleman D. (1995). Inteligencia Emocional (translated 2006). Barcelona, Spain:
Kairs.
Huron D. (2006). Sweet Anticipation. Cambridge, Massachusetts, USA: MIT Press.
Janklvich V. (1983). La Msica y lo Ineffable (translated 2005). Barcelona, Spain:
Alpha Decay.
Kivy P. (2001). New Essays on Musical Understanding. Oxford: Oxford University
Press.
Lester J. (1995). Performance and analysis: Interaction and interpretation. In J. Rink
(ed.), The Practice of Performance (pp. 197-216). Cambridge: Cambridge University
Press
Maconie R. (1990). La Msica Como Concepto. Barcelona, Spain: Acantilado.
Renaud J. (1999). Cmo Adquirir una Supermemoria. Barcelona, Spain: Iberia Edito-
rial.
Rouseau J. J. (1760). Emilio, o de la Educacin (translated 2007). Madrid: Alianza
Editorial.
Snyder B. (2000). Music and Memory. Cambridge, Massachusetts, USA: MIT Press.
Stravinsky I. (1942). Potica Musical (translated 2006). Barcelona, Spain: Acantilado.
Tarchini G. (2004). Anlisis Musical. Sintaxis, Semntica y Percepcin. Buenos Aires:
Cooperativa Chilavert.
Vega-Rodrguez M. and Villar-Taboada C. (2001). Msica, Lenguaje y Significado.
Valladolid, Spain: University of Valladolid.
That blissful feeling: Phenomenological

conceptions of music performance from one
performers perspective
Andrew Geeves and Doris McIlwain
Department of Psychology, Macquarie University, Australia
Inspired by the small amount of relevant past research available (Berger

1999, Berliner 1994, Monson 1996, Sudnow 1978), this paper focuses on
the type of performance experience an individual musician views as
worthy of striving toward (and avoiding) and the possible way(s) in
which this can be accomplished. Using Strauss and Corbins (1998) take
on Grounded Theory (GT) as a methodology, data obtained from a semi-
structured, in-depth interview with Jeremy Kelshaw (JK), a professional
musician, are examined. JKs subjective, phenomenological experience of
music performance comprised a detailed understanding of an ideal per-
formance which emerged from JKs understanding of excellence and the
uniquely uncertain nature of music performance. Also important in this
experience were a number of strategies implemented by JK and his band
Cloud Control in an attempt to establish, maintain, and regain vibe, the
mysterious key ingredient of a desirable performance experience. Along-
side the implications it holds for music education, this research also pro-
vides a unique insight into an individual musicians understanding of an
ideal performance experience and the strategies used to achieve this.
Keywords: phenomenology; experience; interview; idiographic
The bulk of past studies examining music performance within the discipline
of psychology have utilized an experimental research paradigm centered on
the music listener. These studies have particularly focused on the relationship
between music and the listeners emotion experience (Gabrielsson and Juslin
1996, Huron 2006, Minsky 1982, Sloboda 1991, Waterman 1996). In com-
parison to the music listener, the perspective of the music performer has been
relatively underrepresented in this literature. The majority of studies have
used an experimental research paradigm to examine the importance of per-
former-related factors such as emotion and memory (Dunsby 2002, Palmer

2005, Williamon 2002). The specific nature of these studies restricts the
amount to which they are able to contribute to an understanding of a musi-
cians gestalt experience of performance.
A handful of studies have examined a musicians broader experience of
music performance using self-report and in-depth interviews. One difficulty
among many when discerning the relevance of this research to the current
research project is that different genres make different use of a large number
of variables including musical form, convention, and performance setting.
These variables must have an impact at the level of phenomenology. In this
way, work by Berliner (1994), Monson (1996), and Sudnow (1978) examining
the experience of performance for jazz musicians is valuable, but naturally
cannot apply across all genres. Aligning more closely with the aims of our
research, Berger (1999) investigated the musicians experience of perform-
ance across three genres, discovering a close link between performer affect,
music structure, and aspects of the performers social life. However, notice-
able researcher bias, coupled with the strong historiographical and ethnomu-
sicological foci of Bergers work, limits the relevance of its results to our
research. We aim to allow the data to speak by being more exploratory in
our focus and choice of methodology.
Arising from this context, the main objective of our research is to ascer-
tain which sort of performance experience, if any, a musician views as worthy
of striving for (or avoiding) and the possible way(s) in which this is accom-
plished. By conducting and analyzing semi-structured, in-depth interviews
with professional musicians, this research aims to build a Grounded Theory
(Strauss and Corbin 1998) of the musicians performance experience. An in-
terview with a single musician, the first in this program of research, forms the
basis of this article.
MAIN CONTRIBUTION
Research methodology and background
Strauss and Corbins (1998) reworking of Glaser and Strausss (1967)

Grounded Theory (GT) is the ideal methodology for this research program
due to the paucity of past research available in the subject area and the
aforementioned difficulties associated with attempting to investigate phe-
nomenology and experience. Rather than demanding example verification or
perfect description, GT calls for bottom-up theory building derived from strict
adherence to data. To aid theory building, GT employs theoretical sampling, a
piecemeal approach in which sampling, rather than being predetermined
before beginning the research, evolves during the process (Strauss and Cor-
bin 1998, p. 202).
The data on which this article is based are drawn from a semi-structured
interview carried out with Jeremy Kelshaw (JK), an experienced bassist cur-
rently playing in Cloud Control, a moderately successful self-described indie,
alt-folk, pop band. It was initially thought that JKs interview would serve as a
pilot study. However, the data emerging from this interview were rich and
substantial enough to allow the interview to be included as the first in a pro-
gram of research involving interviews with professional musicians. The un-
predictability of theoretical sampling and the incomplete nature of this
research make it important to stress the preliminary nature of findings re-
ported in this article.
The interview
Excellence and the blissful moment
In describing his ideal performance experience, JK not only spells out the
specific type of performance experience that he aims to create but also reveals
his understanding of excellence:
If youre not nervous and [the] crowd is already into itand if its a song
that you know backwardsyou just go into a little bit of a zone. In that
blissful moment its the same feeling you have when you really enjoy
anything I think. Your body knows what to do, and you just go into this
trance. You just enjoy yourself, and you dont have to think. Its that
feeling of success and a feeling of accomplishment and privilege at the
same time. Its just bliss.
Phenomenologically, excellence in performance for JK comprises feelings

of bliss, success, accomplishment, and privilege. Trance and zone are used
by JK in an attempt to convey the other-worldly nature of this experience,
amplified by the high level of intensity and extreme pleasure accompanying
these feelings. Cskszentmihlyis (1990) notion of flow is helpful here. Just
like one experiencing flow, JKs experience is characterized by concentration,
absorption, and reward.
Vibe
Although the word was originally introduced into the interview by the re-
searcher (AG), JK demonstrates a predilection toward using vibe in his de-
scription of performance experience. Closely linked to JKs ideal performance

experience, vibe approximates to an emergent something that binds the
blissful moment together. Its nexus between individual band members, vibe
can expand to fill the space between band and audience and band and song,
with all three of its manifestations being interrelated:
It is the four of us, first and foremost, trying to create a connection with
each other and then that being broadcast to the audience. In the same
way that you cant force a vibe with an audience, you cant force a vibe
with a song.
A fleeting uniqueness
JKs understanding of the unique nature of music performance underpins the

performance experience that JK aims to create:
A live performance is unique in that it happens once and then its gone.
[It] is unique to you and unique to the people that youre playing with.
The inherent uncertainty in music performance resulting from its vulner-

ability to temporal and contextual specificity is, ironically, of greatest threat
and value to JK. Feelings of exclusivity, privilege, success, and accomplish-
ment stem from the informed yet inevitable gamble with uncertainty JK must
take during performance and the sense that it has, on this occasion, paid off.
Yet uncertainty also serves as the biggest obstacle to JKs desired experience,
with performance being a fleeting, nonreplicable creative experience. A cer-
tain quality of performance then, a specific type of experience, a particular
type of vibe, can all be aimed for but can never be taken for granted nor guar-
anteed.
Indefinable connection
Like vibe, there is also an element of uncertainty in the connection to audi-

ence. Although vital to a successful performance, the way in which a dynamic
and reciprocal connection to the audience emerges seems unable to ever be
completely fathomed:
So much of a successful performance is a connection with the audience

We cant pin it down but I thinkthe better performers still look for that
connection, look for that connectivity.
Although partially reliant on variables such as audience expectation,

genre, and location of performance space, the unpredictable elements of a
connection to the audience, just like those in an ideal music performance,
guarantee against its certainty. As a result, performance for JK must always
involve working hard to establish, maintain, and prevent the loss of both con-
nection and vibe.
The shaping of a successful performance
Cloud Control implement a number of tactics in an attempt both to prevent

an undesirable performance experience and to increase the likelihood of a
desirable performance experience. Among pre-performance strategies such as
ensuring familiarity with a piece of music and thinking thoroughly about a set
list, Cloud Control carry out a pre-performance routine. Prior to performance,
the band finds a physical space in which they can sit, collecting their
thoughts, singing together and listening to a motivational speech delivered by
JK. This routine ensures band members are on the same wavelength before
performance. Cloud Control also execute a number of strategies onstage. The
most important of these strategies for JK centers on his notion of intention.
For JK, performance is never a haphazardly random occurrence into which
the performer is thrown but is rather a planned experience over which the
performer can exercise control:
I think youve gotta be intentional. You can actually choose to have a

good time. You just have to give it your best, and people will respect
that. If its out of your control then thats fine. How you respond to
that is in your control. That is all you can do really.
Although some variables may be out of a performers control, for JK, the
amount of enjoyment derived from a performance and the amount of effort
put into a performance are not. Even if other factors are less than ideal, in-
tention seems conducive to establishing vibe.
IMPLICATIONS
Analysis of an in-depth interview with a single professional musician provides

a unique insight into the particular meaning that excellence holds for JK in
the context of performance. Using conceptual variables as units of
comparison, the iterative nature of theoretical sampling in GT will allow this
picture to expand continually over the course of the research project. It is
hoped that dissemination of the results of this project might also renew
appreciation for the amount of work and talent the intricacies of successful
music performance demand. This work also holds potential implications for
the ways in which music is learned, taught, and performed.
Acknowledgments
We wish to thank the following people, without whom this paper would not have been
possible: Jeremy Kelshaw, John Sutton, Wayne Christensen, and Anthony Arcuri.
Andrew Geeves, Department of Psychology, Macquarie University, Balaclava Road,

North Ryde, New South Wales 2109, Australia; Email: andrew.geeves@psy.mq.edu.au
References
Berger H. M. (1999). Metal, Rock and Jazz. London: University Press of New England.
Berliner P. F. (1994). Thinking in Jazz. Chicago: University of Chicago Press.
Cskszentmihlyi M. (1990). Flow. New York: Harper & Row.
Dunsby J. (2002). Performers on performance. In J. Rink (ed.) Musical Performance
Gabrielsson A. and Juslin P. N. (1996). Emotional expression in music performance:
Between the performers intention and the listeners experience. Psychology of Mu-
sic, 24, pp. 68-91.
Glaser B. G. and Strauss A. L. (1967). The Discovery of Grounded Theory. Hawthorne,
New York, USA: Aldine Publishing.
Huron D. (2006). Sweet Anticipation. Cambridge, Massachusetts, USA: MIT Press.
Minsky M. (1982). Music, mind, and meaning. In M. Clynes (ed.) Music, Mind, and
Brain (pp. 1-20). New York: Plenum Press.
Monson I. (1996) Saying Something. Chicago: University of Chicago Press.
Palmer C. (2005) Sequencing memory in music performance. Current Directions in
Psychological Science, 14, pp. 247-250.
Sloboda J. A. (1991). Music structure and emotional response: Some empirical findings.
Strauss A. L. and Corbin J. (1998). Basics of Qualitative Research. London: Sage.
Sudnow D. (1978). Ways of the Hand. London: Routledge.
Waterman M. (1996) Emotional responses to music: Implicit and explicit effects in
listeners and performers. Psychology of Music, 24, pp. 53-67.
Williamon A. (2002) Memorising music. In J. Rink (ed.). Musical Performance (pp.
113-126). Cambridge: Cambridge University Press.
Thematic session:
The perception of technique
Left-hand expression in cello playing:

Exploring approaches to shifting
Tnia Lisboa1,2 and Fernando Gualda3

2 Orpheus Institute, Belgium
3 Sonic Arts Research Centre, Queen's University Belfast, UK
This investigation explores aspects of musical interpretation in relation

to left-hand cello technique. It focuses on how listeners, performers, and
teachers approach shifting, both in terms of its functionality and its ex-
pressive use in portamenti. The method involved appraisal of five se-
lected interpretations through listening experiments with cellists and
other instrumentalists, analyses of portamenti, and interviews. The re-
sults revealed that listeners generally preferred contemporary recordings
with less obvious use of portamenti and that cellists tended to rate the
use of portamenti higher than other instrumentalists.
Keywords: performance; expression; string playing; shifting; preferences
During Pablo Casalss 97 years (1876-1973), the cello emerged from reclusion
to take its place alongside other popular solo instruments. Major additions to
the repertoire presented cellists with new challenges, and advances in re-
cording possibilities ensured that technical proficiency became the norm.
Although much has been written about string technique, there is no detailed
study of the expressive characteristics of shifting. Pleeth (1982) suggests that
the feel for distance in movement and the variety of tension in the left-hand
are essential in creating good shifts. He also links portamenti to singing and
states that it is the notes that give birth to the slide, not the other way
around (p. 37). However the study of such a technical, yet expressive, tool
can be problematic. Mantel (1975) suggests that technique cannot be sepa-
rated from the experience of making music. For practical purposes [however]
we can separate technique from interpretationa point of departure for the
analyses of this technique (p. XV). Although such separation may be possible
in order to refine technique, there seems to be little point in practicing tech-
nical movements without relating them to a musical objective. Nonetheless,

in the mechanical sense, musicians have been the object of study for quite
some time, as examples of practitioners with highly skilled sensorimotor co-
ordination in terms of precision, speed, and dexterity (e.g. Kay 2003). The
relationship between mental representations and motor actions in pianists
has also been studied (e.g. Drost et al. 2005). However, studies on the rela-
tionship between musical understandings and ways of shifting are nonexis-
tent. Music performance research often considers expressive intonation,
vibrato, dynamics, and timing, but with the exception of a few studies on the
violin, expressive shifting is rarely mentioned.
This investigation examines expressive shifting from both technical and
interpretive angles, exploring approaches to portamenti from the perspectives
of the listener, the performer, and the teacher. This article reports on the first
two perspectives: (1) listeners evaluations of portamenti and impact on in-
terpretation and (2) cellists individual approaches to shifting and portamenti
in performance.
METHOD
Participants
Thirty-eight listeners were recruited: cellists (n=27) and other instrumental-

ists (n=11). The 27 cellists comprised undergraduate (n=14) and postgraduate
(n=8) students of the Royal College of Music and Trinity College of Music,
amateurs (n=2), and professional performers (n=3).
Materials
Five recordings by distinguished cellists (see Table 1) of the Finale from Don
Quixote by R. Strauss were selected. This excerpt was chosen for its extensive
potential for expressive portamenti.
Procedure
All recordings were analyzed and compared focusing on portamenti. For the
purposes of homogeneity, recordings were normalized and noise reduction
filters were applied. The participants were asked to evaluate an extract of the
Finale (see Figure 1), with the five recordings played in a random order, using
a scale of 1 to 7 and to provide the reasons for their ratings. Subsequently, the
participants were required to listen to the same extract a second time, rating
the use of portamenti for the following categories: (1) quality (QP), (2) appro-
priateness for the interpretation (HA), and (3) overall importance (OI). Semi-
Figure 1. Cello part of the extract played from the Finale of Don Quixote by R. Strauss.
Table 1. Cellists, recording labels, and dates of the five recordings, as well as the esti-
mated onsets, offsets, and durations of the selected portamenti.
Recording Cellist Label Onset-offset (s) Duration (ms)

R1 (1976) M. Rostropovich EMI 4-76903 19.07-19.18 ~120
R2 (1969) P. Fournier CBS 61110 16.62-16.79 ~160
R3 (1938) E. Feuermann IGI-372 17.08-17.31 ~240
R4 (1958) P. Tortelier HMV ASD-326 18.96-19.11 ~150
R5 (1953) G. Piatigorsky HMV ALP-1211 18.87-19.18 ~200
structured interviews were carried out with the cellists, probing their ap-
proaches to shifting, including practice strategies, decisions about choices of
portamenti, relationship with stage fright, and training received.
RESULTS
In order to determine whether there were differences in the listeners evalua-

tions of interpretations and portamenti, repeated measures Analyses of Vari-
ance (ANOVA) were conducted separately for each type of rating. The within-
group variable was the recording, and the between-group variable was cel-
list/non-cellist. Mauchlys tests of sphericity were significant for each test
and, therefore, F values are reported with the Greenhouse-Geisser correction.
Thee main points arose.
Firstly, significant between-group differences were found. Cellists ratings
of interpretations and portamenti were generally higher than non-cellists.
The differences between the two groups were significant in QP (F1,35=8.55,

p<0.05) and HA (F1,35=6.57, p<0.015). Since all recordings were by eminent
cellists, it is perhaps expected that cellists, most of them students aiming to
acquire the technical and musical skills present in the recordings, would in-
deed rate QP and HA higher than non-cellists. However, evaluations of OI
seemed to reflect simply the amount of portamenti employed (i.e. the more
portamenti, the higher the rating).
Secondly, the analysis revealed significant within-group effects across all
performances for all four outcome measures: PE (F2.86,99.95=4.32, p<0.05), QP
(F3.06,106.92=2.98, p<0.05), HA (F2.68,93.87=7.20, p<0.05), OI (F3.16,110.8=9.45,
p<0.05). Although the exact pattern of differences varied between the four
outcome measures, inspection of the marginal means (see Figure 2) sug-
gested that R3 was generally least preferred in terms of PE, QP, and HA.
However, the profile for OI was reversed, indicating that the importance of
portamenti was high in R3. This seems to suggest that participants found the
portamenti particularly striking in R3, but they did not generally like it. The
preferred interpretations (PE) were R4 and R5, and the highest ratings for QP
and HA were for R1.
Thirdly, the only significant interaction between recording and group was
for PE (F2.86,99.95=3.48, p<0.05). As can be seen in Figure 3, while the general
between group effect (cellists rating higher than non-cellists) was maintained
for recordings 1, 2, and 4, the difference disappeared for R3 and R5. This may
be due to the fact that most cellist participants are students used to contem-
porary styles and, hence, more biased in their appreciation of interpretations
set in a different era than other instrumentalists (i.e. the lower ratings may
reflect their lack of familiarity with older styles of interpretation).
A preliminary analysis of the recordings was also carried out, and as ex-
pected, a variety of approaches to portamenti was found. For example, the
number of portamenti heard varied considerably for each recording (R1
n=22, R2 n=18, R3 n=28, R4 n=22, R5 n=16), and there were only two places
where all cellists commonly produced portamento (see Figure 4): third to
fourth beat of bar 4, and last beat of bar 15 to first beat of bar 16.
The analysis of shifts in bar 4 also highlights the individual approaches
adopted. Although spectrograms cannot offer a clear representation of por-
tamenti due to the presence of orchestral accompaniment, it is still possible
nevertheless to identify some of the partials of the most salient pitches (F#
and D) as well as changes in frequency. Spectrograms were, therefore, created
using 512 points Fast Fourier Transforms with 87.5% overlap. The approxi
mated durations (see Table 1) are based on the steady state of pitches since
the presence of vibrato and interference of orchestral sounds hindered the
Figure 2. Listeners mean ratings (cellists and non-cellists) for PE (top left), HA (top
right), QP (bottom left), and OI (bottom right).
Figure 3. Graph of interactions (mean ratings for cellists and non-cellists) for PE.
Figure 4. Places of portamenti in each recording. Pronounced portamenti are shown by

the segmented lines and less pronounced portamenti shown by the diamond shapes.
Arrows indicate common places of portamenti for all cellists.
precision of the estimation of onsets and offsets. The spectrograms showed

that portamenti differed not only in their duration but also in the vibrato em-
ployed at the beginning of the shift; their speed from the departure to the
arrival pitch, and the changes in loudness.
DISCUSSION
This exploratory study shows that (1) approaches to portamenti are highly
individual and (2) this plays an important role in listeners preferences for
certain interpretations. The results of the evaluations highlight the prefer-
ences for contemporary styles with fewer expressive shifts as well as the
higher ratings by cellists when asked to focus upon specific technical and
musical aspects of shifting.
This preliminary study will be further developed by focusing on teaching.
Observations, interviews, and intervention studies will provide insights into
contemporary and artistic approaches involved in shifting that may have an
impact on how cellists prepare for performance. In these ways, the continua-
tion of the research will have implications for performing, teaching, and
learning, stimulating performers toward more creative and artistic ap-
proaches to the role of the left-hand in cello playing.
Acknowledgments
We wish to acknowledge the help and assistance of Sam Thompson, Aaron Williamon,
and Alexander Demos.
Tnia Lisboa, Centre for Performance Science, Royal College of Music, Prince Consort
Road, London SW7 2BS, UK; Email: tlisboa@rcm.ac.uk
References
Drost U., Rieger M., Brass M. et al. (2005). Action-effect coupling in pianists. Psycho-
logical Research, 69, pp. 233-241.
Kay B. A. (2003). An early Oscillator model: Studies on the biodynamics of the piano
strike. Motor Control, 7, pp. 1-45.
Mantel G. (1975). Cello Technique. Bloomington, Indiana, USA: Indiana University
Press.
Pleeth W. (1982). Yehudi Menuhin Music Guides: Cello. New York: Schirmer Books.
Fingering force in violin vibrato
Satoshi Obata1, Hidehiro Nakahara1,2,Takeshi Hirano1, and

Hiroshi Kinoshita1

This study investigated the spatio-temporal features of the longitudinal

(shaking) and normal (press) string clamping forces by the left hand
during vibrato sound production. A violin installed with a 3-D force
transducer was used for the measurement of the force at the D5 tone po-
sition. Twelve trained violinists produced an A (open) tone for 2 s fol-
lowed by a D (vibrato and force measurement) tone for 30 s at different
vibrato rates (4.5 and 6 Hz, and no-vibrato), dynamics (p, mf, and f), and
with the use of different fingers (index, middle, ring, and little fingers).
The average, amplitude, and peak-to-peak time of shaking, and press
forces, and the longitudinal-lateral shear force relationship were evalu-
ated. During vibrato, an oscillated pattern was observed in each of the
three forces, while the longitudinal component demonstrated the largest
periodic oscillation. The average press force and the amplitude of shaking
force significantly increased with the rate of vibrato as well as dynamics
of the sound generated. These force variables did not differ among the
four fingers. The shaking force showed considerable inter-player differ-
ence (1.0-4.2 N). The feedback training was found to help in some reduc-
tion of the force.
Keywords: violin; vibrato; fingerboard reaction force; acoustic parame-

ters; finger difference
Satoshi Obata, Graduate School of Medicine, Osaka University, 1-17 Machikaneyama,

Toyonaka, Osaka 560-0043, Japan; Email: s-obata@moted.hss.osaka-u.ac.jp
Fluctuation strength of tremolo played on the

mandolin: How is tremolo evaluated as good?
Nozomiko Yasui1, Masafumi Kinou2, and Masanobu Miura2

The tremolo played on the mandolin is a continuous sound produced by

the repetition of attenuating sounds. The amount of acoustic amplitude
of tremolo is usually fluctuated in terms of time. Therefore, the tremolo is
assumed to give a listener the feeling of fluctuation, which is thought to
be concerned with subjective evaluation for performance proficiency. In-
troduced here is an evaluation method that strongly depends on the
fluctuation strength (FS), which has been suggested as an index for
evaluating the feeling of fluctuation for fluctuating sounds at low fre-
quencies. Past studies have investigated the index for modulated pure
tones and broadband noise. In this study, tremolo fluctuations are de-
scribed by FS, and the relation between FS and tremolo proficiency was
investigated through evaluation experiments. Specifically, we conducted
four studies using performance sounds and two kinds of synthesized
sounds. As a result, highly rated tremolos had a relatively low physical FS
and were correlated with highly subjective evaluations. Therefore, we
confirmed that skilled tremolos satisfied both playing restrictions and
psychoacoustical criteria.
Keywords: mandolin; tremolo; feeling of fluctuation; fluctuation

strength; subjective evaluation
The tremolo is one of the popular ways of playing the mandolin (Fujita 2005).
Though there is almost no agreement concerning the physical characteristics
of tremolo played by experts, the feeling of fluctuation on tremolo is thought
to be concerned with subjective evaluation for performance proficiency. The
fluctuation strength (FS) has been suggested as an index for evaluating
hearing sensation in fluctuating sounds at low frequencies, and past studies
have investigated that of modulated pure tones and broadband noise (Ter-
hardt 1968, Fastl 1982, Fastl and Zwicker 1990). However, few have reported
on the FS of fluctuating musical sounds produced by musical instruments and
proficiency. In addition, there are no suggestions for an index corresponding
to their proficiency.
Therefore, tremolo fluctuations are described by FS, and the relation be-
tween FS and tremolo proficiency was investigated through evaluation ex-
periments. The plucking rate, onset deviation, and amplitude deviation are
set on the basis of our assumption that they affect the hearing sensation and
subjective evaluation for tremolo more strongly than the other factors. Here,
both onset and amplitude deviation are called irregularity on tremolo. Aes-
thetic performances of tremolo are said to be smooth or not fluctuating by
trained players, so the amount of FS for tremolo is originally calculated by
extracting a fluctuation component of about 4-8 Hz from acoustic data. The
procedure for objectively calculating the amount of FS is designed based on
facts reported by past studies (Terhardt 1968, Fastl 1982, Fastl and Zwicker
1990) and an assumption that the characteristics of the tremolo are some-
what similar to the AM SIN at slow speeds (Roads 2001).
Therefore, we hypothesized that the fluctuation components extracted
originally from a tremolo influences the feeling of fluctuation (or an aesthetic
evaluation). Here, the calculated FS is called physical FS, while the FS pro-
posed by Fastl et al. is called psychological FS. We conducted four studies
using performance sounds and two kinds of synthesized sounds. In Study 1,
the region of plucking rates used in actual tremolo performance was investi-
gated. In Study 2, in order to describe tremolo fluctuation by FS, the relation
between feeling of fluctuation, and physical FS was investigated using synthe-
sized sounds that have just a feeling of fluctuation without tremolo profi-
ciency, played at several plucking rates (2-16 Hz). In Study 3, the relation
between physical FS and tremolo proficiency was investigated using perform-
ance sounds played at three plucking rates (6, 8, and 9 Hz). However, factors
that affect the feeling of fluctuation and subjective evaluation for tremolo
proficiency are not only plucking rate but also onset deviation and amplitude
deviation. The unified irregularity on tremolo is rarely found by simply ob-
serving actual performance. So, in Study 4, the relation between physical FS
and tremolo proficiency was investigated using other synthesized sounds by
controlling three parameters that have both tremolo proficiency and feeling of
fluctuation. Here, synthesized sound used in Study 2 is called imitated
sound, while the sound used in Study 4 is called simulated sound.
METHOD
Participants
In Study 1, four mandolin players with over two years of experience were used
as listeners. In Study 2, two of the listeners were mandolin players, and three
of the listeners had no experience of playing the mandolin. In Study 3, seven
mandolin players (P1-P7) with over three years experience were used as man-
dolin players. Six other mandolin players with over a half year of experience
were used as listeners. In Study 4, five other mandolin players with over a
year of experience were used as listeners.
Materials
In Study 1, simulated tremolo sounds with onset deviations and amplitude

deviations on each plucking were used, 60 in total (3 deviation patterns and
20 plucking rate patterns). In Study 2, imitated tremolo sounds with onset
deviations and amplitude deviations on each plucking were used, 45 in total
(3 deviation patterns and 15 plucking rate patterns). Those sounds have just a
feeling of fluctuation without tremolo proficiency. It Study 3, 21 tremolo
sounds played by 7 players were used (i.e. 3 plucking patterns times 7 play-
ers). In Study 4, simulated tremolo sounds with onset deviations and ampli-
tude deviations on each plucking were used, 27 in total (3 deviation patterns
and 9 plucking rate patterns).
Procedure
In Study 1, participants were asked to evaluate whether or not stimuli were

significantly perceived as tremolo sounds. In Study 2, an experiment using
the method of magnitude estimation was conducted to subjectively evaluate
the magnitude of fluctuation for imitated sounds synthesized at several
plucking rates (2-16 Hz) under three deviation patterns by five listeners. In
Studies 3 and 4, the two tremolo sounds were presented one immediately
after the other, and listeners were asked to subjectively select the better one.
RESULTS
The result of Study 1 is shown in Figure 1; it was found that simulated sounds
synthesized at plucking rate of 6-13 Hz were evaluated as tremolo sounds.
The result that compared FS with amplitude-modulated pure tone (Terhardt
1968, Fastl 1982, Fastl and Zwicker 1990), as well as comparing to physical
FS, is shown Figure 2. The magnitude of fluctuation for imitated sounds in
100
90
ess
caf 80
o ool 70
re m
b etr 60
m
un sa
50
of de
tioa ivce 40
r re 30
gea p
erv 20
A
10
0
1 3 5 7 9 11 13 15 17 19
Plucking rate [Hz]
Figure 1. Results of investigation of the region of plucking rates.
Magnitude of fluctuation for imitated sounds

100
Physical FS for imitated sounds
Psychological FS for amplitude pure tone
]% 80 (Fastl, 1982; Fastl & Zwicker, 1990)
[
SF 60
eiv
tla 40
eR
20
0
2 4 6 8 10 12 14 16
Modulation frequency [Hz]
Figure 2. Results of investigation of feeling of fluctuation.
Figure 2 is the average of all results, represented as a percentage. Percentages

were obtained by assigning the number 100 to the maximum value of each
listeners answer. As can be seen in Figure 2, it was confirmed that magnitude
of fluctuation for imitated sounds is high at plucking rate of 4-8 Hz and low at
plucking rate of 10-16 Hz. It revealed that the feeling of fluctuation for simu-
lated sounds has a strong positive correlation with a high FS of amplitude-
modulated pure tone and physical FS (0.81r0.98, n=15).
As a result of Study 3, it is confirmed that tremolos performed at plucking
rates of 8 or 9 Hz were evaluated as good. Also, it is confirmed that faster
rates correspond to smaller amounts of physical FS.
Figure 3. Relation between our results and psychological FS. (See full color version at
As a result of Study 4, it was found that the tremolos synthesized at

plucking rates of 8-11 Hz were evaluated as good. Also, it was found that a
plucking rate of 7 Hz corresponds to larger amounts of physical FS, and one
after 7 Hz corresponds exponentially to smaller amounts of physical FS.
DISCUSSION
As seen from Study 4, highly rated tremolos had a relatively low physical FS.
Furthermore, the physical FS has a strong negative correlation with a high
subjective evaluation (r=-0.60, n=27). Therefore, this shows that physical FS
can be used as an evaluation index of the tremolo performance proficiency
played on the mandolin.
The relation between results obtained in this study and psychological FS
is shown in Figure.3. The realizable region is obtained from the results of
Study 1. As can be seen in Figure 3, two strong relations were confirmed
among the calculated physical FS: the subjective evaluation results and the
psychological FS.
Acknowledgments
We are grateful to Prof. Hugo Fastl for discussing with us the application of fluctuation
strength to actual musical contents, and Profs. Yanagida and Yamada for their help in
giving suggestions and ideas for this study. We also express our sincere thanks to
OB/OG, as well as the players belonging to the Ryukoku University Mandolin Orchestra
and the listeners for their help. This study is partly supported by High-Tech Research
Center project for private universities: matching fund subsidy from the Japanese Min-
istry of Education, Culture, Sports, Science, and Technology (MEXT), 2006-11.
Nozomiko Yasui, Graduate School of Science and Technology, Ryukoku University, 1-5
Yokotani, Oe-cho, Seta, Otsu, Shiga 520-2194, Japan; Email: n-y@mail.ryukoku.ac.jp
References
Bksy G. (1935). ber akustische Rauhigkeit. Zeitschrift fr technische Physik, 16, pp.
276-282.
Fastl H. (1982). Fluctuation strength and temporal masking patterns of amplitude-
modulated broadband noise. Hearing Research, 8, pp. 56-69.
Fastl H. and Zwicker E. (1990). Psychoacoustics Facts and Models. Berlin: Springer-
Verlag.
Fujisawa N., Yamada M., and Nakayama I. (2000). Fluctuation in frequency of Japa-
nese pressed voice, Dami-Goe. Journal of the Acoustical Society of Japan, 21, pp.
365-367.
Fujita O. (2005). The Correct Way to Play Mandolin: Illustrated for Beginners to In-
structors with Concert Repertories. Tokyo: Zen-On Music Company Ltd.
Kurakata K., Kuwano S., and Namba S. (1993). Factors determining the impression of
the equality of intensity in piano performance. Journal of the Acoustical Society of
Japan, 14, pp. 441-447.
Rauber A., Pampalk E., and Merkl D. (2002). Using psycho-acoustic models and self-
organizing maps to create a hierarchical structuring of music by sound similarity.
Paper presented at the Third International Conference onMusic Information Re-
trieval, Paris.
Roads C. (2001). The Computer Music Tutorial. Cambridge, Massachusetts, USA: MIT
Press.
Terhardt E. (1968). ber akustische rauhigkeit und schwankungsstrke. Acustica, 20,
pp. 215-224.
Thematic session:
Emotion in performance
The effects of musical syntax on perception of

music performance
Katty Kochman, Marc Leman, and Jana DeClerk
Institute for Psychoacoustics and Electronic Music, Ghent University, Belgium
The focus of this study is to examine the effects of social interaction and
musical syntax in a music performance with emphasis on emotion per-
ceptions. The syntactical elements that were analyzed during the course
of these experiments included pitch, tuning, and timing. Two experi-
ments were conducted to examine the relationship of syntax and per-
formance. In the first experiment, subjects were asked to state preference
between altered stimuli with higher levels of expressivity and unaltered
stimuli with lower levels of performer expressivity. In the second experi-
ment, subjects were asked to sing familiar or traditional songs in social
and individual settings. Results confirm that participants demonstrated a
definitive preference toward social interaction when actively engaged in
embodied musical performance. However, the outcome was less defini-
tive in passive listening environment of the first experiment.
Keywords: music perception; syntax; musical embodiment; music per-

formance; social interaction
Musical performance situations provide opportunities for social interaction

and joint attention (Baldwin 1995). Musical enjoyment in a performance set-
ting often relies on the communicative interaction of participants and the
transmission of the musical messages may impact the enjoyment of the par-
ticipants. According to the shared affective motion experience (SAME)
model developed by Overy and Molnar-Szakacs (2009), music perception is a
process of interactive social communication involving neural processing of
pitch and time dimensions that result in complex action sequences involving
the perception and/or agency of another person, whose actions are inter-
preted, imitated, and predicted. While accuracy of musical syntax is generally
considered a critical feature of emotional perception of music, in many per-
formance situations the quality of syntax may be secondary to social interac-
tion and group attuning (Leman 2008). De Bruyn et al. (2009) found that
social interaction between music listeners affected music perception and mu-
sical meaning formation processes. Participants in a social setting were better
able to synchronize with music and responded with increased intensity of
corporeal articulations in social conditions.
The purpose of the present study was to explore the settings in which the
importance of syntax is undermined by the context in which the music is
performed. The first experiment measured the effect of changes in syntax of
the music on listeners preferences and perceptions. The second experiment
measured the effect of changes in syntax of the music on subjects preferences
in musical performance environments. In both experimental settings, we
examined methods to measure the elements involved in embodied music
cognition when syntactical elements are altered.
METHOD
Participants
Thirty-two subjects, aged 20-26 years, participated in each experiment. Par-

ticipants were from different sample populations. Background data regarding
the participants musical experience were collected through the use of pre-
surveys and groups were divided to include musically experienced partici-
pants and less musically experienced participants.
Materials
In the first experiment, video/audio segments of approximately one minute in

length were recorded with singer and piano. Syntax alteration was introduced
through audio software and included tonal changes varying the sharpness or
flatness of pitch of approximately 0.67 semitones and timing changes to the
piano accompaniment gradually increased through the duration of the musi-
cal piece. Survey data was used to collect participant responses after each two
excerpts.
In the second experiment, four songs were recorded approximately two
minutes in length (two Dutch, two English). Audiovisual excerpts were then
altered to include syntactical alteration in timing and pitch. Pitch was gradu-
ally altered to -6 semitones. Sliders, which consisted of a scale of 1-10, re-
corded participant responses while stimulus was played and were
interconnected to collect data for both group and individual settings through
an Arduino interface. A Max Patch allowed for data to be recorded along a
timeline for each song. Participants were asked to reflect their musical en-
joyment using the slider as part of the experimental task, allowing for the
collection of data that reflected changing musical preference during active
participation in the performance. Data were analyzed according to a timeline
corresponding with video data.
Procedure
In Experiment 1, all participants listened to stimuli with and without syntac-

tical alteration, together in a group setting. Subjects were then divided into
two groups based on musical experience. Each group was presented a set of
four musical excerpts performed in two modes of expressivity. Group A was
presented stimuli in the original syntax. Group B was presented all stimuli
with an altered syntax. Subjects were asked to state their musical preferences
through the use of a survey. Responses were compared in all listening condi-
tions.
In Experiment 2, all participants performed stimuli together in a group
setting and individually. Lyrics of the songs were presented simultaneously
with visual stimuli and participants were asked to sing along. Stimuli con-
sisted of the same pieces in both performance situations. Subjects were asked
to state their musical enjoyment simultaneously while performing experi-
mental tasks. This data was collected through the use of multiple sliders
(scaled 1-10) connected through an Arduino interface and through the use of
a post-survey.
RESULTS
Experiment 1
In Group A, stimuli were presented with unaltered syntax. According to data

based in participant responses, the majority of subjects preferred the more
expressive performance (see Figure 1). Of the musically experienced partici-
pants, 66.7% on average preferred the more expressive performance through-
out the musical excerpts. Of the less musically experienced participants,
85.7% preferred the more expressive performance on average throughout the
excerpts. This data confirmed that a more expressive performance is pre-
ferred when syntax is unaltered.
Group B was presented stimuli with altered syntax. The majority of par-
ticipants preferred the less expressive performance (see Figure 2). Of the mu-
sically experienced participants, 70.8% preferred the less expressive
performances throughout the excerpts on average. Of the less musically ex-
perienced participants, 72.5%preferred the less expressive performance on
Figure 1. Statistical analysis of syntax perception Group A.
Figure 2. Statistical analysis of syntax perception Group B.
average throughout the musical excerpts. These data demonstrated that a less
expressive performance is preferred when syntax is altered in a passive lis-
tening situation.
Experiment 2
Subject responses demonstrate a significant preference toward group per-

formance over individual performance. Sixty-six percent of participants re-
sponded that the group performance situation appealed to them more than
the individual performance situation. Fifty-six percent of respondents men-
tioned they were most comfortable when singing together in the independent
responses. When describing the group experimental setting, 47% of partici-
pants responses could be organized into two categories specifying the ap-
pealing aspects of the group situation: the social interaction (including not
singing alone, listening to each other, feeling supported, and
group/social feeling) and that the atmosphere was more enjoyable or fun.
Of our subjects, 70% said that singing in the group setting was easier because
of appreciable differences between their individual and group performance.
In the group setting, subjects responded that it was easier to keep the right
tone, there was more fun of singing together and the combination of the
voices was more aesthetically pleasing. The altered syntax was not mentioned
as a factor in responses for the group singing evaluations. Seventy-five per-
cent felt uncomfortable in their individual performance. In addition, sharing
difficulties when performing music increased the enjoyment of some 19% of
participants, who stated that sharing of the experience made them happy
and comfortable.
Slider data confirmed the overall preference for the group setting when
syntax was altered in a musical performance setting. Fifty-nine percent of
participants had a greater number of increases in enjoyment along the scale
when performing stimuli in the group setting. Twenty-two percent had more
increases of enjoyment in the individual setting. Instances where the degree
with which increases in enjoyment were more substantial than decreases in
enjoyment occurred for 46% of participants in the group setting and 15% of
participants in the individual setting. Instances where the degree with which
decreases in enjoyment were more substantial than increases in enjoyment
occurred for 12% of participants in the group setting and overall 38% of par-
ticipants in the individual setting. Fifty percent of participants spent more
time at higher levels of enjoyment in the group, as opposed to 9% who spent
more time at higher levels of enjoyment in the individual setting.
DISCUSSION
Analysis of data facilitated the observation of the cultural and social dynamics
influencing music performance. In Experiment I, participants preferred the
more expressive performance overall when syntax was unaltered. However,
fewer participants with musical experience preferred more expressive per-
formances than those without musical experience. In situations where the
syntax was altered, the majority of participants selected the less expressive
performance. The introduction of an additional level of expressivity may in-
fluence experimental results. In Experiment 2, participants preferred social
interaction in the group setting, as it allowed them to form a common under-
standing regarding music perception. In future research, it may be possible to
integrate syntax alteration through a more ecological approach. Ideally syntax
change would be introduced through the performance of participants or per-
formers themselves. A more nuanced distinction between corporeal and cere-
bral understanding of music and an implementation of a taxonomy of
structural cues to analyze movement data in relation to syntactical change
may assist researchers to further understand the impact of behavioral reso-

nance with corporeal articulations.
The study effectively quantified subjective musical understanding in so-
cial and individual environments. The majority of participants also indicated
that a social interaction condition was preferred. Participants responses
demonstrated that they actively attempt to construct an understanding of
each others intentional actions when performing musical stimuli in a group
setting. Results may indicate the effect of corporeal articulations in influenc-
ing musical and syntactical perception. In support of the findings of Overy
and Molnar-Szakacs (2009), music functioned as social activity where per-
ceptual activities of participants were integrated and interdependent.
Acknowledgments
Thanks to Ivan Schepers and Pieter Coussement for assistance with technological devel-
opment and design and to Michiel Demey for his assistance with MaxMSP. The authors
also wish to acknowledge Frank Desmet for his assistance with the statistical analysis.
Special thanks also to Liesbeth Bodyn and Anne Claessens for the practical assistance in
experimental administration.
Katty Kochman, Institute for Psychoacoustics and Electronic Music, Ghent University,
Blandijnberg 2, Ghent 9000, Belgium; Email: kjkochman@yahoo.com
References
Baldwin D. (1995). Understanding the link between joint attention and language. In C.
Moore and P. J. Dunham (eds), Joint Attention (pp. 131-158) Hillsdale, New Jersey,
USA: Erlbaum.
De Bruyn L., Leman M., Moelants D., and Demey M (2009) Does social interaction
activate music listeners? In S. Ystad and Kronland-Marinet R. (eds), Computer Mu-
sic Modeling and Retrieval (pp. 93-106). Heidelberg, Germany: Springer Verlag.
Koelsch S. E., Kasper D., Sammler K. et al. (2004). Music, language and meaning: Brain
signatures of semantic processing. Nature Neuroscience, 7, pp. 302-307.
Leman, M. (2008) Embodied Music Cognition and Mediation Technology. Cambridge,
Overy K and Molnar-Szakacs I. (2009). Being together in time: Musical experience and
the mirror neuron system. Music Perception, 26, pp. 489-504.
Emotional arousal and the automatic detection

of musical phrase boundaries
Steven R. Livingstone1, Emery Schubert2, Janeen D. Loehr1, and

Caroline Palmer1

2 School of English, Media, and Performing Arts,
University of New South Wales, Australia
This study investigated correspondences between listeners emotional re-

sponses to an orchestral work and the underlying phrase structure. An
algorithm for the automatic detection of musical phrase boundaries was
developed, based on listeners continuous ratings of perceived emotion
with a two-dimensional tool (valence and arousal). Rates of change in
arousal (velocity) and change in change in arousal (acceleration) for each
musical phrase were combined into a single metric. Similarity in listen-
ers responses for musically related phrases was higher than those for
musically contrasting phrases. Based on listeners responses to identical
sections, a recursive algorithm identified all phrase boundaries. These
findings indicate novel measures of change in listeners emotional
arousal that correspond with musical phrase structure.
Keywords: emotional response; arousal; phrase structure; automatic

boundary detection; similarity metric
Emotional response to music is influenced by contributions of both the com-

position and performance (Kendall and Carterette 1990, Livingstone and
Thompson 2009). Although significant research has investigated the role of
musical cues such as tempo and articulation on emotional response (Gabri-
elsson and Lindstrom 2001), fewer studies have investigated the role of
phrase structure. Recent work (Krumhansl 2002, Vines et al. 2005) found
that listeners tension ratings coincided with the phrase structure, suggesting
that listeners emotional responses could be used for the computational
(automatic) detection of phrase structure (Cambouropoulos 2001).
We investigated listeners emotional responses to an orchestral work to

identify response patterns across phrases that varied in melodic and har-
monic context. Two measures of emotional response were analysed: velocity
(first derivative) and acceleration (second derivative) of arousal (amount of
emotional activity). These instantaneous measures reduce effects of the pre-
ceding musical context. The velocity-acceleration arousal responses were
compared with the theoretical phrase structure. Based on the recurrent na-
ture of listeners responses, a detection algorithm was developed to identify
segment boundaries, which confirmed that listeners emotional arousal re-
flects musical phrase structure.
METHOD
Participants and materials
Sixty-seven adult listeners with varied amounts of musical training (9 had no

training, 19 had 1-10 years, 39 had 10+ years), most of whom were under-
graduate music education students participated (further details in Schubert
1999). An orchestral recording of the Pizzicato Polka, Op.234 (J. and J.
Strauss), length 2:37 mins, was used, which contains the following phrase
structure: introduction (4 bars), phrase A (8 bars), B (12 bars), A' (8 bars),
followed by a middle section (32 bars), and a repeat of the introduction,
phrases A, B, and A', followed by a coda. The A, B, A' section that occurred
twice within the composition was the focus of the analyses; the two occur-
rences are referred to as section 1 and section 2. Phrases A and A' had related
melodic and harmonic content, whereas Phrase B was not related to A or A'.
Sections 1 and 2 were further subdivided into 4-bar sub-phrases: a1, a2, b1,
b2, a'1, a'2, as shown in Figure 1.
Procedure and analysis
Listeners continuously reported their emotional response using Emotion-

Space Lab, a two-dimensional tool that records valence from positive (100) to
negative (-100) and arousal (amount of emotional activity) from strong (100)
to weak (-100), sampled at 1 Hz (Schubert 1999). Listeners valence responses
did not correspond to the phrase structure, and therefore we focus on arousal
responses. Responses were smoothed with functional data analysis (Ramsay
and Silverman 2005). Listeners mean arousal responses and their velocity
(first derivative) and acceleration (second derivative) values are shown in
Figure 1. Phase-plane plots that combined velocity and acceleration were gen-
erated for each sub-phase (for an example, see inset of Figure 1). Similarities
Figure 1. Phrase structure of sections 1 and 2 of the Pizzicato Polka, with listeners mean
arousal, velocity, and acceleration values. Theoretical phrase boundaries indicated as
dots. Phase-plane plot for sub-phrase a1 in section 1 (inset) shows velocity and accel-
eration values and begins at the filled square.
between phase-plane plots, which reflect the 2-dimensional responses to

arousal, were evaluated with procrustes analyses (2-dimensional measure of
similarity).
Listeners responses to different sub-phrases were divided into three
groups: repetition, within-category, and between-category. The repetition
group contained pairs of responses to the same sub-phrase from section 1 and
section 2 (n=6). The within-category group compared all musically related
sub-phrases (same letter, e.g. a1 to a2) in and across both sections, excluding
repetition pairs (n=28). The between-category comparisons contained pairs
of responses to musically unrelated sub-phrases (different letters, e.g., a1 to
b1) in and across both sections (n=38).
RESULTS
Listeners mean emotional arousal values across the entire section 1 were
highly similar to those for section 2, which occurred later in the musical con-
text (mean arousal: r=0.84, p<0.01, mean velocity: r=0.70, p<0.01, mean
acceleration: r=.69, p<0.01). Next we measured response similarity among
the 2D phase-plane plots, shown in Figure 2, at the level of individual sub-
phrases. The procrustes similarity values for repetition pairs (Table 1, first
column) indicated that listeners emotional responses to individual sub-
Table 1. Procrustes similarity values for repetition, within-category, and between-cate-

gory sub-phrase comparisons for original and new phrase structures. All p<0.01.
Original phrase structure New phrase structure (with B3)

Repetition 0.75 0.81
Within-category 0.69 0.75
Between-category 0.63 0.77
Figure 2. Phase plane plots for sub-phrases sections 1 (top row) and 2 (bottom) of the
Pizzicato Polka. Sub-phrases begin at the filled squares and possess a cyclical structure.
Figure 3. Automatic phrase detection outcomes aligned with listeners mean arousal
responses (velocity) for section 1. Filled circles indicate theoretical phrase boundaries.
Bars indicate points in time at which a phrase boundary was algorithmically detected
(taller bars=smaller p-values).
phrases in section 1 reheard in section 2 were highly similar. Furthermore,

emotional responses for within-category sub-phrases were perceived as next
most similar and responses to between-category sub-phrases as least similar.
The phase-plane plots in Figure 2 show a cyclical structure to listeners

emotional responses; each sub-phrase generated semi-elliptical responses
that shifted clockwise around the midpoint. However, b2 sub-phrase re-
sponses possessed an unusual double swirl. These phrases were also consid-
erably longer in duration than other sub-phrases (see Figure 1). Therefore, b2
sub-phrases were subdivided into 2 sub-phrases (b2 and b3) and procrustes
analyses were recalculated for the new phrase structure (Table 1, second col-
umn). The similarity values based on the new phrase structure were higher
than those based on the original phrase structure, suggesting that the new
phrase structure is a more accurate representation of listeners emotional
responses to the Pizzicato Polka.
The high similarity in listeners emotional responses to repeated phrases
suggests that this information may benefit the automatic detection of phrase
boundaries in this piece, which has a simple harmonic structure. A recursive
algorithm was developed which takes as input the listeners velocity responses
for sections 1 and 2. Beginning with a window size of 2 s, responses for sec-
tions 1 and 2 were correlated and the window length grew iteratively (+1 s) up
to the entire 41 s response. The resulting correlation function was used to
identify peaks of most similarity, for comparison with locations of sub-phrase
boundaries. Peaks in the correlation function were defined by two successive
increasing values followed by two successive decreases. The algorithm was
then reiterated once at each value, with the new window onset time equal to
the sequential location of the peak. To evenly distribute the effect of sequence
beginnings and endings on window lengths, the algorithm was run in both
sequence directions (forward/backward) and the outcomes averaged. Figure
3 shows the p-values associated with each detected boundary. All six bounda-
ries were correctly identified automatically within two seconds of the hy-
pothesized phrase boundaries, and no false positives.
DISCUSSION
Listeners emotional arousal responses to repeated musical sections were

highly consistent. Similarity analyses revealed that the first and second de-
rivatives of emotional arousal were more similar for repeated musical sub-
phrases than for musically related or less related sub-phrases. These findings
indicate that listeners had distinct arousal responses at the 4-bar sub-phrase
level in the Pizzicato Polka and these patterns recurred during musical repe-
titions.
The phrase boundary detection algorithm used the similarity in listeners
responses to repeated musical sections to identify all phrase boundaries suc-
cessfully. The algorithm operates under the premise that listeners responses
to identical musical sections are most similar at phrase boundaries. Phrase
boundaries are points of perceptual salience, and listeners emotional re-
sponses may accelerate at these locations. Use of the first and second deriva-
tives of emotional arousal helped to provide a more precise temporal marker
of change in listeners responses. This research demonstrates novel analytic
techniques for measuring listeners emotional responses and their correspon-
dence with the underlying phrase structure of music.
Acknowledgments
This work was supported by the Canada Research Chairs program and NSERC Grant
298173 to the fourth author. We thank Erik Koopmans and Frances Spidle.
Steven Livingstone or Caroline Palmer, Department of Psychology, McGill University,

1205 Dr Penfield Ave, Montreal H3A 1B1, Canada; Email: steven.livingstone@mcgill.ca
or caroline.palmer@mcgill.ca
References
Cambouropoulos E. (2001). The local boundary detection model (LBDM) and its appli-
cation in the study of expressive timing. Proceedings of the International Computer
Music Conference (ICMC01) (pp. 232-235). Havana, Cuba.
Gabrielsson A. and Lindstrom E. (2001). The influence of musical structure on emo-
tional expression. In P.N. Juslin and J. S. Sloboda (eds.), Music and Emotion. Ox-
ford: Oxford University Press.
Kendall R. A. and Carterette. E. C (1990). The communication of musical expression.
Krumhansl C. L. (2002). Music: A link between cognition and emotion. Current Direc-
tions in Psychological Science, 11, pp. 45-50.
Livingstone S. R. and Thompson W. F. (2009). The emergence of music from the theory
of mind. Musicae Scientiae, 10, pp. 83-115.
Ramsay J. O. and Silverman B. W. (2005). Functional Data Analysis. New York:
Springer.
Schubert E. (1999). Measuring emotion continuously: Validity and reliability of the two-
dimensional emotion-space. Australian Journal of Psychology, 51, pp. 154-165.
Vines B.W., Nuzzo R. L., and Levitin D. J. (2005). Analyzing temporal dynamics in
music: Differential calculus, physics and functional data analysis techniques. Music
Perception, 23, pp. 137-152.
Communicating emotion in piano performance
Cristina C. Gerling, Regina A. T. dos Santos, and Catarina Domenici
Graduate Music Program, Federal University of Rio Grande do Sul, Brazil
Music expression involves tacit knowledge, an elusive and frequently

lacking aspect of music education. This paper considers activities related
to learning new repertoire. In a supervised performance lab at the Fed-
eral University of Rio Grande do Sul, students gather weekly to play and
discuss the results of their practice activities and performances. Building
on two previous studies (conducted in 2007 and 2008) wherein students
were introduced to Russells circumplex model in order to establish a
common vocabulary, they were encouraged to label both intended and
perceived emotions. In the present study, we present data on the prepa-
ration of a lesser-known piece by Schumann without guidance from their
piano teachers. After nine weeks, participants were encouraged to dis-
cuss their level of achievement by listening to the results of three re-
cording sessions. Parameters such as contour, articulation, tempo,
timing, dynamics, movement/gesture, and global coherence were evalu-
ated in order to assess their performances. Two independent referees
also graded the third recording session, and these results were quantita-
tively and qualitatively compared with those of the students. The results
suggest a lack of correlation among the parameters. After two more
weeks of practice, there was no significant increase in the level of ability
and there was also a low degree of correlation between intended and per-
ceived emotions among participants.
Keywords: expression; piano performance; communication; circumplex

model
A relevant issue to the study of music performance and perception concerns

the ability of players to express emotional intentions and the ability of listen-
ers to recognize them. This subject has already been investigated from several
different points of view. For instance, Timmers (2007) investigated the rela-
tionship between vocal expression, musical structure, and emotion in re-
corded performances by famous singers of three Schubert songs. Timmers

observed that variations in tempo, dynamics, and pitch were systematically
related to music structure, emotional activity, and valence. Several experi-
ments have dealt with the encoding and perception of emotional intentions in
music performance with different musical instruments (e.g. Gabrielsson and
Juslin 1996, Juslin 2003, Juslin and Person 2002). From a strategic point of
view, Hultberg (2008) recently observed in two case studies that the process
of music interpretation is characterized by complex strategies, based on indi-
vidual familiarity with the conventions of musical expression.
From the music education point of view, Juslin and Persson (2002)
claimed that instrument teachers lack a guiding theory of expression in per-
formance. Karlsson and Juslin (2008) have recently investigated instrumen-
tal teaching with a focus on expression and emotion, suggesting that the focus
of teaching was mainly on technique and on the written score. Although there
were differences among teachers, common features pointed to lack of clear
goals, specific tasks, and systematic teaching patterns.
As part of the present study, the piano laboratory at the Federal Univer-
sity of Rio Grande do Sul (UFRGS) promotes discussions of multifaceted as-
pects of performance. We provide an environment in which students are free
to explore their attitudes and beliefs about music expression, to reflect upon
their own performances, and to gauge the achievement of their peers. In a
previous study, we investigated the emotion that undergraduate and graduate
students intended to communicate through the piano and the emotions per-
ceived by the musically educated audience (Gerling and Santos 2007). Un-
known to the participants, the chosen piece was based on a ratio of shorter
and longer note values devoid of time or dynamic signatures. In order to fol-
low up on our exploratory research, another study considered the case in
which Russells circumplex model (1980) was employed as a guide for de-
scribing intended and perceived emotions in the performance of a well known
Baroque work (Gerling et al. 2008). In the present study, we discuss the
preparation by undergraduate and graduate students of a lesser-known
Schumann piece without guidance by their music teachers.
METHOD
Participants
A semester-long course with one two-hour weekly meetings and a laboratory

was set up for undergraduate and graduate piano students at UFRGS. For this
activity, four males and two females participated as performers and listeners.
Materials
Students received a romantic score (Schumanns Anhang discarded from

Op.12) without the composers name. All but one recognized the author and
was asked not to divulge the information.
Procedure
The students preparation of the music was monitored in three phases. The
first phase comprised nine weeks of practice without guidance. The perform-
ances and interviews were recorded in weeks 2, 5, and 9 of the students work.
Complementarily, the students at the laboratory (1) watched videos of profes-
sional pianists playing Brahms, Liszt, Schumann, and Chopin works (week 3),
(2) discussed the relationship between bodily expression and performance
(e.g. Maria Joo Piress masterclass) (week 6), (3) received tips from a profes-
sional pianist concerning practical ways to achieve sound effects as indicated
on the score (week 7), and (4) discussed the structural characteristics of the
score among themselves (week 8).
In the second phase, a sample from the third recording session was evalu-
ated by two referees who focused on contour, articulation, tempo, timing,
dynamics, movements/gesture, and global coherence. A stimulated recall
interview was conducted after each student watched and graded his or her
individual recorded sample. Before entering the third phase, the German
terms present in the score (feurigst and rascher) were again discussed with
the student without reference to actual playing.
During phase three, students practiced for a further two weeks, at the end
of which there was another recording session. Students were then requested
to reveal the emotion they had intended in their performance, making use of
terms defined in Russells circumplex model.
RESULTS
Grades were assigned by two independent piano teachers, who acted as refe-
rees (R1 and R2), and by each student after analyzing the third recorded per-
formance on video in terms of phrase contour, articulation, tempo, timing,
dynamics, movements/gestures, and global coherence on a scale from 1-10,
with 10=highest.
The accuracy for the mean grade and standard deviation (SD) attributed
by the two referees, as well as those given by the student himself or herself,
was calculated for each parameter. Results (see Figure 1) express the degree
attributed by the student himself/herself with the corresponding SD. The
Figure 1. Mean score assigned in each category by each student and respective SD in
comparison to the mean grades assigned by the two referees. Students A, B, and C are
freshmen, D and E are juniors, and F is a graduate student From top left, the graphs
indicate phrase contour, articulation, tempo, timing, dynamics, movements/gestures,
and global coherence.
students were then requested to express their intended emotion in a recorded

performance. All participants including the performer evaluated audio and
audio/video samples using the terms from Russells circumplex model. The
results contradict previous observations both in the case of Bachs Prelude
(Gerling et al. 2008) and in similar discussions, such as that in Vines et al.
(2006). In this study, the limits imposed by the aural channel improved the
correlation between intended and perceived emotion.
DISCUSSION
Juslin and Madison (1999) used a combination of analysis and synthesis af-
forded by the ability to resynthesize human performance by means of com-
puter technology. This approach was employed in the investigation of the
relative importance of timing patterns in the communication of emotions
through musical performance. The authors gradually removed different ex-
pressive cues (tempo, dynamics, timing, articulation) from piano perform-
ance played with various intended expressions (anger, sadness, happiness,
fear) to see how such manipulations would affect a listeners ability to decode
the intended emotional expression. The results indicated that (1) removing
the timing pattern yielded a decrease in the listeners decoding accuracy, (2)
timing patterns alone were capable of communicating at least some emotions
with high accuracy, and (3) timing patterns were less effective in communi-
cating emotions than were tempo and dynamics.
According to these authors, listeners can use timing patterns to decode
the emotional expression of a performance. In our present study, the low
degree of correlation between intended and perceived emotion may be asso-
ciated with a high SD in scores of timing, dynamics, and movement/gesture.
The results suggest that in their practice our participants neglected some if
not many of the most fundamental means for achieving global coherence in
terms of the demands of the score and its expression in their musical per-
formances.
Our results suggest that the students have to be made aware of the im-
portance of choosing a tempo (speed, pacing) in the planning and enunciation
of all other parameters evaluated in this study in order to convey their in-
tended expressive qualities in their performances. Furthermore, timing and
movement/gesture seem to develop along and to increase with expertise and
maturity (see Figure 1). For this particular study group, contour was shown to
be well assimilated and less dependent on students level of expertise. We
conclude that expertise in the establishment and manipulations of tempo has
been an aspect largely overlooked in a musicians instruction, and this will be
the focus of our future research. It seems that the appropriate preparation of
a score depends fundamentally on the articulation of all parameters. It is our
hypothesis that the deliberate choice of tempo sets a precedent in which other
parameters such as contour, articulation, timing, and dynamics will be pro-
portioned, thus fostering global coherence and the expression of the intended
expressive qualities during performance.
Cristina C. Gerling, Graduate Music Program, Federal University of Rio Grande do Sul,
Rua Prof. Annes Dias 112, Porto Alegre, Rio Grande do Sul 90020-090, Brazil; Email:
cgerling@ufrgs.br
References
Gerling C. C. and dos Santos R. A. T. (2007). Intended versus perceived emotion. In A.

Williamon and D. Coimbra (eds.), Proceedings of ISPS 2007 (pp. 233-238).
Gerling C. C, Domenici C., and dos Santos R. A. T. (2008). Exploring intended and per-
ceived emotion in piano performance supported by the circumplex model. Paper
presented at the Twenty-eighth International Society for Music Education World
Conference, Bologna, Italy.
Hultberg C. (2008). Instrumental students strategies for finding interpretations: Com-
plexity and individual variety. Psychology of Music, 36, pp. 7-23.
Juslin P. N. (2003). Five facets of musical expression: A psychologists perspective on
music performance. Psychology of Music, 31, pp. 273-302.
Juslin P. N. and Madison G. (1999). The role of timing patterns in the decoding of emo-
tional expressions in music performance. Music Perception, 17, pp. 197-221.
Juslin P. N. and Persson R. S. (2002). Emotional communication. In R. Parncutt and G.
E. McPherson (eds.), The Science and Psychology of Music Performance (pp. 219-
236). Oxford: Oxford University Press.
Karlsson J. and Juslin P. N. (2008). Music expression: An observational study of in-
strumental teaching. Psychology of Music, 36, pp. 309-334.
Russell J. A. (1980). A circumplex model of affect. Journal of Personality and Social
Timmers R. (2007) Vocal expression in recorded performance of Schubert songs. Musi-
cae Scientiae, 11, pp. 237-268.
Vines B. W., Krumhansl C. L., Wanderley M. M., and Levitin D. J. (2006) Cross-modal
interactions in the perception of musical performance. Cognition, 101, pp. 80-113.
Workshops
Hearing and the noise of performance:

Solutions for sound monitoring
Aaron Williamon1, Bradford C. Backus2, Stephen Dance3, and

Georgia Zepidou3

2 Ear Institute, University College London, UK
3 Department of Engineering Systems, London South Bank University, UK
Performing artists must be able to practice, rehearse, and perform safely.

With respect to hearing and the noise of performance, however, the
delicate nature of their work and the dedication of performers themselves
may mean that theyas well as their employers, educators, and manag-
ersare placed in a difficult position when complying with new
international noise at work regulations. This workshop showcases recent
initiatives that have brought together artists and scientists to generate
practical solutions and new technology for monitoring noise and pro-
tecting hearing, while having little or no impact on the quality of
performers work. The workshop consists of the following demonstra-
tions: Improving noise exposure during practice: The sound absorbing
mirror (Dance and Zepidou), The use of the iPhone as a sound meter: A
cheaper noise badge (Dance and Zepidou), and The E-meter: A new
noise badge designed for the entertainment industry (Backus and
Williamon). In each, new devices and technology are demonstrated
alongside in-depth discussion of the research underpinning their devel-
opment, as well as their cost and application across a wide range of
performing arts contexts. The workshop also considers avenues for future
research and the artistic, cultural, and scientific requisites that must be
met in pursuing them.
Keywords: hearing; noise; sound meter; acoustics; regulations
Aaron Williamon, Centre for Performance Science, Royal College of Music, Prince Con-
sort Road, London SW7 2BS, UK; Email: awilliamon@rcm.ac.uk
Taking microtonal composition and

performance from the periphery
into the mainstream
Graham Hair1, Nicholas J. Bailey1, and Lisa Swayne2
1 Centre for Music Technology, University of Glasgow, UK

2 Freelance soprano, London, UK
This workshop seeks to bring some technical and expressive aspects of

microtonal singing from experimental approaches (e.g. those represented
by the music of Harry Partch) into mainstream vocal performance. Sa-
cred Songbook is a collection of lieder by the first author, set to texts of
Abrahamic (Christian, Jewish, Islamic) provenance, using the microtonal
scale with 19 equal steps to the octave. The third author will demonstrate
(live) the preparation for performance of selections, using machine-as-
sisted rehearsal, performance, measurement, and analysis techniques
developed by the second author. The workshop will showcase: (1) ma-
chine-assisted practice techniques, (2) score-based versus aural-based
rehearsing techniques and feedback, (3) post facto empirical (machine-
based) measurement, analysis, and interpretation, (4) use of techniques
of microtonal melodic ornamentation and gesture to blend expressive
and technical means, and (5) the relationship between the meaning,
character, and aesthetic qualities of the texts and the musical expression
and structure of the performance. The workshop draws on a number of
sources from experimental composition and performance traditions, and
harnesses, develops, and extends them to invigorate mainstream ap-
proaches to vocal performance and to achieve performances that reflect
something of the ecstatic mysticism of texts from the different traditions
represented in the Sacred Songbook.
Keywords: microtonal; practice; feedback; machine-assisted; voice

Graham Hair, Centre for Music Technology, University of Glasgow, 45 St Vincent

Crescent, Glasgow G3 8NG, UK; Email: graham@n-ism.org
The art and science of historical performance
Colin Lawson1 and James Tibbles2
1 Royal College of Music, London, UK

2 School of Music, National Institute of Creative Arts and Industries,
Daniel Trks Klavierschule (1789) makes the important observation that

some musical effects cannot be described; they must be heard. Evi-
dence for the art (as opposed to the science) of historical performance
remains elusive, deliciously inexact, but potentially inspirational. Clearly,
C. P. E. Bachs remarks about the importance of moving an audience are
of special value; and no clarinettist can afford to ignore the reviewer who
described Anton Stadlers clarinet as having so soft and lovely a tone
that no-one with a heart could resist it. Modern period performers
each occupy a position on the spectrum from historical accuracy to prac-
tical expediency, not least in their choice of original instruments, copies,
or replicas. Authentic performers have sometimes denied any form of
glorifying self expression, but acted in the service of the composer by
following textbook rules, with a strictly empirical program to verify
historical practices. With no reference to personality, this was somehow
magically transformed into a composers intentions, with the performer
dangerously close to infantile dependency. This workshop deconstructs
period performances of music for clarinet and fortepiano by Mozart,
Stadler, and Vanhal, distinguishing historical elements from those as-
pects of art and science that are arguably rooted in the twenty-first cen-
tury.
Keywords: historical performance; original instruments; personality;

clarinet; fortepiano
Colin Lawson, Royal College of Music, Prince Consort Road, London SW7 2BS, UK;
Email: clawson@rcm.ac.uk
Friday
18 December 2009
Symposium:
Student musicians motivation,
learning, and performance
Multiple motives: Profiles of young

Australians reasons for musical engagement
James M. Renwick1 and Gary E. McPherson2
1 Sydney Conservatorium of Music, University of Sydney, Australia

2 School of Music, University of Melbourne, Australia
Self-determination theory conceives of motivation as lying along an in-

trinsic-extrinsic continuum and has been investigated extensively in
school learning, but less in performance domains. This paper investigates
the ways that students manifest multiple motives for striving in the de-
manding field of musical performance. Performance examination candi-
dates (n=677) aged 8-19 years completed questionnaires. Factor analysis
produced five motives lying along the internal-external continuum.
Cluster analysis then generated four groups of participants: (1) a group
reporting high levels of internal motives and low levels of external mo-
tives, (2) a group reporting high levels of all five motives, (3) a group re-
porting low levels of internal motives and high levels of external motives,
and (4) a group reporting low levels of all five motives. Results showed
that both groups with high levels of internal motives achieved higher
performance results than the two groups with low levels of internal mo-
tives. The high-external/low-internal group did not differ in performance
achievement from the group reporting low levels of all motives. The study
confirms prior research in suggesting that high levels of internalized mo-
tives can work together with high levels of external motives to produce
excellent performance, but external motives alone are insufficient.
Keywords: music education; motivation; self-determination theory; self-

efficacy; self-regulated learning
Large-scale studies in the 1990s (Ericsson et al. 1993, Sloboda et al. 1996)
demonstrated the crucial role of deliberate practice in the acquisition of ex-
pertise in musical performance, and more recently, researchers (e.g. Austin et
al. 2006) have discussed the need for powerful motivational resources to
sustain such work, which by definition is not inherently enjoyable. Intrinsic
motivation is one construct that has been invoked to account for the devel-
opmental shift toward sustained and more self-controlled engagement in the
learning of musical performance skills (Renwick and McPherson 2009,
Sloboda and Davidson 1996). According to self-determination theory, intrin-
sic motivation is the inherent tendency to seek out novelty and challenges, to
extend and exercise ones capacities, to explore, and to learn (Ryan and Deci
2000, p. 70), whereas people are extrinsically motivated when they engage in
an activity such as musical practice in order to obtain a separate outcome,
such as a reward, or to avoid punishment or a sense of shame.
The present study builds on our earlier work examining young musicians
motivational beliefs in light of self-determination theory (Renwick 2008,
Renwick et al. 2002, 2009) by investigating the possible existence of motiva-
tional profiles consisting of different combinations of intrinsic and extrinsic
motives. Such multiple goal profiles have been documented in terms of
achievement goal theory in academic learning contexts (Meece and Holt 1993,
Seifert 1995, Valle et al. 2003), and also in terms of self-determination theory
in physical education (Boich et al. 2008) and athletics (Gillet et al. 2009).
Such findings show that high levels of intrinsic motivation (or mastery goals)
can coexist with high levels of extrinsic motivation (or ego goals) to predict
high levels of achievement. To our knowledge, this possibility has not been
investigated in the domain of music performance training; hence, this paper
aims to extend our prior variable-centered approach to a person-centered
one, by using cluster analysis to identify groups of individuals sharing similar
motivation beliefs.
METHOD
Participants
We recruited 753 candidates enrolled in a performance examination admin-

istered by the Australian Music Examinations Board [AMEB (NSW)]; 677 of
these fit our sampling frame of students enrolled in Grades 1 to 8 of the per-
formance syllabus, aged between 8.0 and 19.1 years (mean=13.25), and com-
pleted sufficient items for the present analysis. Females comprised 65% of the
sample of young people, who undertook the examination on keyboard (55%),
woodwind (21%), brass (12%), string (9%) instruments, or voice (3%).
Materials
The Academic Self-Regulation Questionnaire (Ryan and Connell 1989) was

adapted to the music-learning context. This scale assesses students motiva-
Figure 1. Mean standardized factor scores for each cluster on each motivational factor.
tion to strive in achievement settings in various ways along the intrinsic-ex-

trinsic continuum proposed by self-determination theory. The questionnaire
was subjected to exploratory factor analysis and congeneric modeling
(Jreskog 1971), producing five measures of motivation along the following
continuum: internal motivation, exam-related motivation, shame-related
motivation, social motivation, and external motivation (see Renwick 2008).
Procedure
Participants completed the questionnaire the night before their examination

and delivered it to the reception desk on the day of their examination. Per-
formance grades (ranging from D=1 to A+=7) were obtained from the AMEB
(NSW) in the days following the examinations. Following the computation of
individual factor scores for the five motivational factors described above,
these factor scores were subjected to K-means cluster analysis and differences
between cluster groups were computed with analysis of variance (ANOVA).
RESULTS
Cluster analysis identified four groups of students with distinct motivational

profiles (see Figure 1). There were 240 participants reporting high levels of
the more internal forms of motivation and low levels of the more external
forms (the dotted line on Figure 1); 169 participants reported the opposite
pattern: high levels of the more external forms of motivation and low levels of
the more internal forms (the dashed line); 174 respondents reported above-
average levels of all five forms of motivation (double line), while the remain-
ing 94 showed below-average levels of all five motives (unbroken line).
A one-way ANOVA was used to test for differences between the cluster
groups in terms of their results in the AMEB examination. Achievement dif-
fered significantly between groups (F3,668=5.13, p<0.01, 2=0.02). Post-hoc
comparisons using the Bonferroni adjustment revealed that the all-high
group (mean=4.87, SD=1.25) achieved significantly (p<0.05) higher results
than both the all-low group (mean=4.37, SD=1.44) and the high-exter-
nal/low-internal group (mean=4.43, SD=1.46). In addition, the high-inter-
nal/low-external cluster (mean=4.79, SD=1.38) achieved significantly
(p<0.05) higher results than the high-external/low-external group. Other
comparisons were non-significant.
DISCUSSION
The results of the cluster analysis showed a remarkably clear pattern (al-
though given the exploratory nature of cluster analysis, this finding should be
treated with some caution). The four groups were strongly differentiated into
(1) a large group reporting mostly internal motivation, (2) a group reporting
mostly external motivation, (3) a group reporting high levels of all five types
of motivation, and (4) a group reporting low levels of all five types. This find-
ing is consistent with prior research in self-determination theory (Boich et
al. 2008, Gillet et al. 2009): because researchers have typically found that
internal motivation and external motivation are only weakly related (r=-0.20
in the present study), a pattern of motivation whereby students can sustain
high levels of a range of motives along the continuum is to be anticipated.
Thus, even though research has shown that the provision of extrinsic rewards
can undermine intrinsic motivation (Deci et al. 1999), it would appear that as
part of long-term motivational profiles, extrinsic and intrinsic motives can
coexist in a complex interplay.
The study found that high levels of internalized motives (sometimes
called mastery goals) can work together with high levels of external motives
(sometimes conceptualized as ego goals) to produce excellent performance,
but external motives alone are insufficient.
Acknowledgments
This study was made possible by a Large Australian Research Council grant (number
A00102974) to Gary McPherson and John McCormick in 2001. We thank the partici-
pants and acknowledge the invaluable assistance of the AMEB (NSW) with data collec-
tion.
James Renwick, Sydney Conservatorium of Music, Building C41, University of Sydney,

Sydney, New South Wales 2006, Australia; Email: jrenwick@usyd.edu.au
References
Austin J. R., Renwick J., and McPherson G. E. (2006). Developing motivation. In G. E.

McPherson (ed.), The Child as Musician (pp. 213-238). Oxford: Oxford University
Press.
Boich J. C. S., Sarrazin P. G., Grouzet F. M. E. et al. (2008). Students motivational
profiles and achievement outcomes in physical education: A self-determination per-
spective. Journal of Educational Psychology, 100, pp. 688-701.
Deci E. L., Koestner R., and Ryan R. M. (1999). A meta-analytic review of experiments
examining the effects of extrinsic rewards on intrinsic motivation. Psychological
Bulletin, 125, pp. 627-668.
Ericsson K. A., Krampe R. T., and Tesch-Rmer C. (1993). The role of deliberate practice
in the acquisition of expert performance. Psychological Review, 100, pp. 363-406.
Gillet N., Vallerand R., and Rosnet E. (2009). Motivational clusters and performance in
a real-life setting. Motivation and Emotion, 33, pp. 49-62.
Jreskog K. G. (1971). Statistical analysis of sets of congeneric tests. Psychometrika, 36,
pp. 109-133.
Meece J. L. and Holt K. (1993). A pattern analysis of students achievement goals. Jour-
nal of Educational Psychology, 85, pp. 582-590.
Renwick J. M. (2008). Because I Love Playing My Instrument: Young Musicians In-
ternalised Motivation and Self-regulated Practising Behaviour. Unpublished doc-
toral thesis, University of New South Wales (available from http://handle.unsw.
edu.au/1959.4/36701).
Renwick J. M., McCormick J., and McPherson G. E. (2009). Defining relationships
between motivational beliefs and self-regulated practising behaviours using a
structural equation model. Paper presented at the Seventh Triennial Conference of
the European Society for the Cognitive Sciences of Music, University of Jyvskyl,
Jyvskyl, Finland.
Renwick J. M. and McPherson G. E. (2009). Age-related changes in the young musi-
cians beliefs about their autonomy, competence and values. Paper presented at the
Australian Society for Music Education National Conference, Launceston, Tasma-
nia, Australia.
Renwick J. M., McPherson G. E., and McCormick J. (2002). Motivational influences on

childrens self-regulated learning and musical performance achievement. In C. Ste-
vens, D. Burnham, G. McPherson et al. (eds.), Proceedings of the Seventh Interna-
tional Conference on Music Perception and Cognition [CD-ROM] (pp. 377-380).
Adelaide, Australia: Causal Productions.
Ryan R. M. and Connell J. P. (1989). Perceived locus of causality and internalization:
Examining reasons for acting in two domains. Journal of Personality and Social
Ryan R. M. and Deci E. L. (2000). Self-determination theory and the facilitation of
intrinsic motivation, social development, and well-being. American Psychologist,
55, pp. 68-78.
Seifert T. L. (1995). Characteristics of ego- and task-oriented students: A comparison of
two methodologies. British Journal of Educational Psychology, 65, pp. 125-138.
Sloboda J. A. and Davidson J. W. (1996). The young performing musician. In I. Delige
and J. A. Sloboda (eds.), Musical Beginnings (pp. 171-190). Oxford: Oxford Univer-
sity Press.
Sloboda J. A., Davidson J. W., Howe M. J. A., and Moore D. G. (1996). The role of prac-
tice in the development of performing musicians. British Journal of Psychology,
87, pp. 287-309.
Valle A., Cabanach R. G., Nez J. C., Gonzlez-Pienda J. et al. (2003). Multiple goals,
motivation and academic learning. British Journal of Educational Psychology, 73,
pp. 71-87.
Playing together in ways that cater for and

fulfill student musicians psychological needs
Gary E. McPherson
School of Music, University of Melbourne, Australia
This study applied Self-Determination Theory to study the psychological

needs of a sample of 34 highly involved year 11 and 12 student musicians
in a large Midwestern state of the United States who possessed the ability
to become professional musicians. Content analysis of survey responses
and face-to-face interviews sought to clarify how the psychological needs
of relatedness, autonomy, and competence were being met by the musi-
cians musical engagement in and outside of school. Results demonstrate
complex motivational forces in action during this key stage of develop-
ment. Being around like-minded peers, supportive teachers, and an envi-
ronment that was conducive challenged them to achieve at a constantly
higher level. Many were agonizing with personal and external influences
however, which left them with self-doubts about whether they should go
on to become a professional musician, whether they would still feel the
same way about music in the years to come, and whether they felt capa-
ble of obtaining a position in music that would provide sufficient satis-
faction for them personally and professionally.
Keywords: self-determination; psychological needs; self-beliefs; motiva-

tion; musical development
In the psychological literature on motivation, studies concerned with self-

beliefs are so prevalent that they dominate the field. Yet the powerful expla-
nations that arise from close examination of this literature have rarely been
explored as these relate to learning music. Because music is a subject that
involves a high level of personal choice, students are influenced by their own
and others beliefs about and valuing of music perhaps more than in other
areas of learning.
Among the most prominent motivation theories is Deci and Ryans
(2002) Self-Determination Theory (SDT). This organismic, dialectical per-
spective is built around the central premise that all individuals have natural,
innate and constructive tendencies to develop an ever more elaborated and
unified sense of self (Deci and Ryan, p. 5). SDT is concerned with the choices
people make on their own and how they choose to behave in a manner that
reflects their autonomy and their behavior, free of the need to receive an ex-
ternal reward. In this way, SDT is used to examine the degree to which an
individuals behavior is self-endorsed and self-determined. A key component
of SDT is the three innate and universal psychological needs which Deci and
Ryan state are essential to an individuals psychological health and wellbeing.
The need to feel competent underpins our desire to believe that we are capa-
ble and competent and in control of our environment. When we feel compe-
tent, we are better positioned to seek further challenges and to engage in an
activity at a more concentrated level. According to Deci and Ryan, when
autonomous, individuals experience their behavior as an expression of the
self, such that, even when actions are influenced by outside sources, the ac-
tors concur with those influences, feeling both initiative and value with regard
to them (p. 8). Finally, people also have a need for relatedness, to be con-
nected with and cared for by others. The sense of belongingness that can
come from participation and involvement in a musical ensemble, for example,
is one way this psychological need can be met, especially given that this di-
mension concerns the psychological sense of being with others in secure
communion or unity (p. 7). In summary, SDT proposes that individuals are
active in their pursuit to satisfy these three psychological needs and is being
applied in various disciplines to explain why people become sufficiently in-
trinsically motivated that they will engage in an activity for its own sake and
at a level that becomes fully internalized and autonomous.
The purpose of this study was to clarify the degree to which 34 highly in-
volved young school-aged musicians believed that their personal learning
agenda was being fulfilled, and the ways in which the teaching and learning
process to which they were exposed satisfied their psychological needs as
defined by Deci and Ryans (2002) SDT.
METHOD
Participants
The study involved individual interviews with 34 musicians during their last
two years of high school (years 11 and 12). All were identified by their teachers
as among the most dedicated and highly involved in a large Midwest state in
the United States. The musicians ranged from those who were section leaders
in one of the most prominent youth symphony orchestras in the country, to
students in school music programs that have received national and interna-
tional recognition for the quality of their program, through to others who
were performing and composing for highly regarded folk ensembles and rock
groups.
Materials
All students completed a background questionnaire to gather preliminary

evidence concerning the level of their involvement and commitment to music,
and then face-to-face interviews that lasted between 60 and 90 minutes in
order to understand more precisely how they felt about music, the support
systems which facilitated their progress, and the beliefs they held about their
capacity as musicians. Emergent themes were identified after a content analy-
sis of full transcripts of the interviews for each participating musician.
Procedure
As shown in Figure 1, key issues covered in the interviews focused on how the
three psychological needs were being met, the quality and quantity of in-
volvement and engagement with music, the types of networks (such as peers,
social groupings, parental support) which facilitated the musicians interest in
music, their sense of purpose as a musician, and the degree to which their
involvement in music both socially and personally was a valued part of their
lives. Additional issues concerned what these students hoped to accomplish
in music both now and in the future, and the extent to which their personal
musical learning agenda was being met through their participation in formal
and informal learning processes, plus whether any of these might comple-
ment or conflict with other aspects of their learning. To clarify this last issue,
questions were asked about the musicians short and long-term goals in mu-
sic, and the influences that were impacting on their decision whether to
choose music as a career beyond high school.
RESULTS
As confirmed by their teachers, all of the musicians had the potential to be-
come professional musicians. Most felt comfortable with their school or
community music experiences, where like-minded peers, supportive teachers,
and an environment that was conducive to their needs challenged them to
achieve at a constantly higher level. Others had ceased participating in certain
musical activities that they felt were not at the level of their own musical
skills, while others were agonizing with personal and external influences
Figure 1. Flow of influences for current and future musical identity.
which left them doubting whether they should go on to become a professional

musician, whether they would still feel the same way about music in the years
to come, or whether they felt capable of obtaining a position within the pro-
fession that would provide ongoing satisfaction for them personally and
emotionally.
Comments of a 17 year old flutist, who reported doing at least 24 hours of
practice each week and spending every weekend rehearsing in ensembles
(one of which is among the most prestigious youth orchestras in the nation)
provide evidence of the complexity of these young musicians personal beliefs.
In her case, feelings of relatedness and competence impacted on how she
viewed music both now and into the future:
Music motivates me. It consoles me and excites me. Its an intangible

means of improving my life; it gives my life more glory than it often de-
serves. Its important to me because I enjoy it so much. I love the chal-
lenge. My entire weekends are eaten up by music. The kids I meet in the
youth orchestra are amazing people. I love that companionship. To ex-
perience music with others is divine! Its the most special thing I have.
As the only musician in her family, and recognizing also the commitment
her parents had made to supporting her musically, this young musician wor-
ried that she takes the limelight away from her other two sisters. As one of
the best flutists in the country for her age, there can be no doubt about the
intensity and level of her current musical engagement. Digging more deeply
to compare her current and future self however, revealed many niggling wor-
ries and self-doubts about whether she wanted a career in music, whether she
would still feel the same way about music in the years ahead, and whether a
career in music was worth the effort, given the possibility that she may be
separated from her family and the city she loved. Many of these worries and
self-doubts were perceptions about music and musicians that she had picked
up from others, with many of the more important coming from her mother:
I remember my sophomore year of high school. I started thinking about

applying to music schools. My mom said You know youd be just as
happy being an English major or a history major, and youd be much
more secure. I was like, OK, thats fine. I still dont know what to do.
Im still applying to music schools, but its hard to make a choice and I
dont want to make a choice. My mum is supportive but worried. I think
she just wants me to be happy. She likes security. Shes worried that I
wont have a secure position, or my life will be sort of all over the place.
She hates that. Id love to play in a great orchestra. But there is an
opening once every seven years, and thousands of people are trying for
that one shot. You have to move somewhere. My mom made me promise
that I wont leave [name of city]. She left her parents, and she doesnt
want me to do the same to her. I can always break my promise, but I
would like to have options. And I think with music, you are very much
bound to who wants you and if they will, whether they will have you.
For another student, a vocalist, guitarist, and composer who performs in

three rock groups, strong competence and autonomy perceptions impacted
on how he was preparing for his musical future: it was so nice when we won
the Battle of the Bands. The next day we played again, and I just remember
looking out there. The entire front two rows were singing along to a song I
wrote. I must have smiled for like a week and a half after that. These and
other comments give a clear sense of what this young man wanted to do with
the rest of his life, possibly in part because he came from a wealthy family
that made it easier than might otherwise be the case to take more chances and
to try out a lifestyle that may well end in disappointment. He reported that
when he first told his parents that he wanted to be a musician they were hor-
rified. They thought that I was losing valuable time. Despite this, he re-
garded music as the truest expression of my soul. The expression of my own
feelings and stuff. Even if I dont make it and Im stone-broke somewhere, it
will be OK. If I dont give it my all, it will haunt me for the rest of my life.
Thats what I need to do. Its not like I want to be a musician. I need to be a
musician.
DISCUSSION
Email correspondence 18 months later revealed that the flutist decided to

pursue a double degree in music and English but was considering dropping
English because I have since discovered that it is unrewarding and unsatis-
fying. She was also still not entirely sure that I will become a professional
musician, but for now it makes me happy. The guitarist/composer on the
other hand, was pursuing a path to becoming a professional musician and
songwriter. Results such as these demonstrate the effectiveness of SDT for
contextualizing the key transitional stages of development, such as in this
study of young musicians who were devoting huge personal and physical re-
sources to their musical development while learning how to cope with a de-
manding, competitive learning environment, overcome periods of self-doubt
and performance slumps, and develop a tool kit of psychological and behav-
ioral skills that would enable them to manage the many obstacles they will
encounter along the journey to fulfilling their own personal dreams.
Acknowledgments
This study was supported by a NAMM Sounds of Learning research grant (RFP 2.7).
Gary McPherson, School of Music, University of Melbourne, 234 St. Kilda Road,
Southbank, Victoria 3006, Australia; Email: g.mcpherson@unimelb.edu.au
References
Deci E. and Ryan R. (eds.) (2002). Handbook of Self-determination Research. Roches-

ter, New York, USA: University of Rochester Press.
Focus, effort, and enjoyment in chamber

music: Rehearsal strategies of successful and
failed student ensembles
Jane Ginsborg
Centre for Music Performance Research, Royal Northern College of Music, UK
Different chamber ensembles take different approaches to individual

practice and ensemble rehearsal; in this study, we investigated the ap-
proaches of three student groups who could be described as successful
and failed. The members of two string quartets, one newly-formed and
one established, and a newly-formed wind quintet kept practice and re-
hearsal diaries for six months. There were significant differences between
what and how they practiced and rehearsed and between their ratings of
their own and their colleagues focus, effort, and enjoyment, although
these three dimensions were correlated in several ways. The members of
the successful established ensemble enjoyed practicing more than re-
hearsing, while the reverse was true for the members of the successful
new ensemble. It is also clear from the ratings that participants were re-
alistic about the efficacy of their practice and rehearsal strategies. These
findings have implications for theories of motivation and practical appli-
cations for the teaching of chamber music.
Keywords: effective; group; rehearsal; practice; students
Strategies for ensemble rehearsal have been explored via studies of duos (e.g.
Williamon and Davidson 2002, Ginsborg and King 2007), string quartets
(e.g. Davidson and Good 2002), and wind quintets (Ford and Davidson 2003,
Burt-Perkins and Mills 2008), suggesting the importance of clarity as to en-
semble members roles in order to reduce potential conflicts and effective
verbal and non-verbal communication in relation to social as well as musical
issues. Ericsson et al. (1993) and Lehmann (2002) found that positive corre-
lations exist between concentration and effort, and concentration and enjoy-
ment, but negative correlations between effort and enjoyment. The present
study investigated the extent to which successful ensemble rehearsal (defined
by the relative success of the ensemble over time) is influenced by performers

ratings of their own and their colleagues focus, effort, and enjoyment.
METHOD
Participants
Three student ensembles were recruited: one newly-formed (SQ1) and one
established string quartet (SQ3) and a newly-formed first-year wind quintet
(WQ1). SQ1 comprised two male and two female first year students matched
on the basis of their performance in college entrance auditions. SQ3 was
formed by its membersthree male third year students and one female stu-
denttwo years before the present study began. WQ1 comprised a mixed
group of four first year students and one second year female student, selected
on the basis of the instruments they played.
Printed diary templates were devised for ensemble rehearsals and practice
sessions in which music for ensemble rehearsals was prepared. The ensemble
rehearsal diaries requested information including date, place, time, overall
duration, and content/duration of each activity within the rehearsal; with
reference to up to five activities, what participants thought their ensembles
goal was, how they went about achieving it, to what extent their strategy was
effective, using a rating scale (1=useless to 10=achieved goal 100%) and
what their goal would be next time they engaged in this activity; ratings of
their own and their peers focus (1=very distracted to 10=completely fo-
cused), effort (1=just went through the motions to 10=worked really
hard), and enjoyment (1=not at all to 10=had a great time). Finally, par-
ticipants were asked to describe any performance they had given that day and
nominate one thing [they] were really happy about and one thing [they]
would like to do better in [their] next performance. The practice diaries used
similar wording, adapted for individuals, and rating scales but did not ask
about performance. During the first two terms of the academic year partici-
pants were asked to complete diaries whenever they had undertaken individ-
ual practice on material for ensemble rehearsals and, as individuals, after
each rehearsal.
RESULTS
At the end of the academic year, WQ1 won a major prize. At the time of writ-
ing, its members are still performing together: it can be considered a suc-
cessful ensemble. SQ1 agreed to disband after their end-of-year assessment

and can therefore be considered to have failed, in comparison with WQ1.
SQ3 had been extremely successful for three years although ultimately it split
up, and its members formed new ensembles.
Each of the participants reports, in the form of diary entries relating to
practice session or rehearsal content, goals/strategies/plans for each activity
within the practice session or rehearsal, and comments on performance, was
numbered. As shown in Table 1, SQ3 provided more than twice as many re-
ports as SQ1 and WQ1. They reported preparing five quartets, while SQ1 spent
nearly two terms on one quartet and learned a new work. In comparison,
WQ1 reported preparing and performing three substantial quintets. SQ3 re-
ported more than twice as much individual practice on quartet repertoire as
WQ1, and half as much again as SQ1. SQ3 reported 22 quartet rehearsals,
including one coaching session, and four performances. They rehearsed three
times as long as SQ1, who reported ten rehearsals, including two coaching
sessions, and two performances; by contrast, WQ1 reported 17 rehearsals of
which more than half were coaching sessions, and two performances.
Examples of practice strategies rated high by members of each ensemble
include: Slow practice of solo passage for intonation (keeping as many fin-
gers down for accuracy); focusing and right arm/wrist/hand for string cross-
ing and building up tempo to play up to speed (SQ3); played through music
putting fingerings and bowings in (SQ1); thought about sound and where to
breathe (WQ1). Examples of highly-rated rehearsal strategies include:
Repetition of phrases (maybe only a couple of bars) gradually elongating the
passage until we could confidently play through the section (SQ3); stricter,
stronger feel for pulse, simpler approach, more effect (SQ1); We worked out
who we play the scales with. We realised the timing was out because of a
semiquaver rest at the beginning of each scale. So we played a note on the rest
and then did the scales until they were perfectly in time. Then we took out the
extra note (WQ1).
Numbers of and mean ratings for practice and rehearsal strategies are
shown in Table 2. ANOVA revealed a main effect of ensemble on practice
strategy rating (F2,79=4.78, p=0.011); post-hoc tests showed significant differ-
ences between SQ1 and WQ3. There was also a main effect of ensemble on
rehearsal strategy rating (F2,237=31.42, p<0.0001); post-hoc tests showed sig-
nificant differences between all ensembles.
Mean ratings for focus, effort, and enjoyment are shown in Table 3. The
correlation between participants ratings of focus and effort in individual
practice (r42=0.62, p<0.0001) was stronger than that between focus and
Table 1. Practice and rehearsal reports.
WQ1 SQ1 SQ3

Repertoire Nielsen, Mozart, Dance Schumann, Mozart,
Reicha, Verdi Project Matthews, Schubert, Brahms
Reports 278 274 587
Individual practice (hrs) 6.5 9.25 14+
Rehearsal hrs (sessions) 23.3 (17) 11.3 (10) 38 (22)
Table 2.Practice and rehearsal strategies
WQ1 SQ1 SQ3

Practice strategies rated (reported) 26 (26) 26 (26) 30 (30)
Practice strategy mean ratings (SD) 7.12 (2) 7.54 (1.2) 8.25 (.81)
Rehearsal strategies rated (reported) 44 (50) 49 (50) 147 (150)
Rehearsal strategy mean ratings (SD) 8.57 (1.28) 6.6 (1.52) 7.98 (1.17)
Table 3. Mean ratings (and SD) of focus, effort, and enjoyment.
WQ1 SQ1 SQ3

Practice Self Focus 7.07 (1.86) 7.86 (1.10) 8.78 (0.70)
Effort 6.64 (2.21) 7.64 (1.22) 8.43 (0.85)
Enjoyment 5.64 (1.95) 5.86 (2.14) 7.64 (0.93)
Ensemble Self Focus 8.77 (1.33) 6.92 (1.52) 8.12 (0.90)
Effort 8.71 (1.38) 7.05 (1.76) 8.43 (0.96)
Enjoyment 9.31 (0.90) 6.26 (2.37) 7.50 (1.04)
Ensemble Other Focus 8.54 (1.72) 6.76 (1.76) 8.02 (1.00)
Effort 8.80 (1.32) 6.96 (1.56) 8.24 (0.95)
Enjoyment 9.17 (0.92) 6.76 (2.26) 7.79 (1.03)
Table 4. Ratings of own focus, effort and Table 5. Ratings of others focus, effort,
enjoyment in ensemble rehearsals. and enjoyment in ensemble rehearsals.
N=135 Focus Effort N=135 Focus Effort

Effort 0.83 Effort 0.83
Enjoyment 0.57 0.55 Enjoyment 0.41 0.46
enjoyment (r42=0.39, p=0.01). There was no correlation between effort and

enjoyment.
The correlations between ratings of participants own and others focus,
effort, and enjoyment in ensemble rehearsal are shown in Tables 4 and 5.
There were also significant correlations between participants ratings of their
own and their colleagues focus (r134=0.67), effort (r134=0.70), and enjoyment
(r133=0.74), all significant at p<0.0001.
There were main effects of ensemble on focus (F2,39=6.00, p=0.005), ef-
fort (F2,39=4.76, p=0.014), and enjoyment in individual practice (F2,39=5.48,
p=0.008) such that the ratings of SQ3 were the highest; post-hoc tests
showed significant differences between SQ3 and WQ1 on all three dimen-
sions, and between all three ensembles on enjoyment. The main effects of
ensemble on self-ratings were all significant at p<0.0001focus (F2,131=22.1),
effort (F2,131= 17.04), and enjoyment (F2,131=37.44)as were their ratings of
their colleagues focus (F2,132=15.02), effort (F2,132=21.61), and enjoyment
(F2,132=24.76). There were significant differences between all ensembles self-
ratings on all three dimensions, except SQ3 and WQ1 on effort, and on ratings
of others between WQ1 and SQ1 on focus, SQ3 and SQ1 on effort, and all
three ensembles on enjoyment.
DISCUSSION
Lehmanns (2002) findings were confirmed for individual practice by positive

correlations between focus and effort, and focus and enjoyment. No negative
correlation was found, however, between effort and enjoyment, and all three
dimensions were correlated for rehearsals; there were also positive correla-
tions between participants rating of their own and their colleagues focus,
effort, and enjoyment. The members of the successful string quartet (SQ3)
both practiced and rehearsed considerably more than their younger col-
leagues, although this is not surprising given that they had much more rep-
ertoire to perform. On the other hand, they enjoyed their individual practice
more than did the other groups and more than their ensemble rehearsals,
although they attributed higher levels of enjoyment to their colleagues. They
also identified more sophisticated practice and rehearsal strategies, rating
their own practice strategies higher than did the less advanced students. Al-
though the members of WQ1 carried out the fewest individual practice
hoursand indeed both WQ1 and SQ1 reported least enjoyment from prac-
ticingthey had a great deal of coaching from a range of tutors, and this, to-
gether with their ratings of their rehearsal strategies as even higher than SQ3
and the failed SQ1, may in part account for their success; SQ1 took part in
just two rather unsuccessful coaching sessions early in the first term. Fur-
thermore, WQ1 reported the highest levels of focus, effort, and enjoyment in
rehearsal both for themselves and their colleagues. It may thus be possible for
student musicians to do less individual practice if they have plenty of, and
enjoy, ensemble rehearsal, although this is, of course, dependent on their role
within the group. A final contrast is between SQ1, who rated focus, effort, and
enjoyment for individual practice higher than WQ1 but provided the lowest
ratings on all three dimensions for ensemble rehearsalit was perhaps in-
evitable that they would fail as a group.
Further analyses of the participants goals, strategies, and plans remain to
be carried out; nevertheless, the findings of the present study have implica-
tions for theories of motivation in relation to practice versus rehearsal, as well
as practical applications for those engaged in the teaching and coaching of
chamber musicians.
Jane Ginsborg, Centre for Music Performance Research, Royal Northern College of
Music, 124 Oxford Road, Manchester M13 9RD, UK; Email: jane.ginsborg@rncm.ac.uk
References
Burt-Perkins R. and Mills J. (2008). The role of chamber music in learning to perform:
A case study. Music Performance Research, 2, pp. 26-35.
Davidson J. W. and Good J. J. (2002). Social and musical communication between
members of a string quartet: An exploratory study. Psychology of Music, 30, pp.
186-201.
Ericsson K. A., Krampe R. T., and Tesch-Rmer, C. (1993). The role of deliberate prac-
tice in the acquisition of expert performance. Psychological Review, 100, pp. 363-
406.
Ford L. and Davidson J. W. (2003). An investigation of members roles in wind quintets.
Ginsborg J. and King E. C. (2007). The roles of expertise and partnership in duo per-
formance. In A. Williamon and D. Coimbra (eds.), Proceedings of ISPS 2007 (pp.
61-66). Utrecht, The Netherlands: European Association of Conservatoires (AEC).
Lehmann A. (2002). Effort and enjoyment in deliberate practice: A research note. In I.
M. Hanken, S. G. Nielsen, and M. Nerland (eds.), Research In and For Music Edu-
cation (pp. 153-166). Oslo: Norwegian Academy of Music.
Williamon A. and Davidson J. (2002). Exploring co-performer communication. Music
Scienti, 6, pp. 53-72.
What predicts performance excellence in

tertiary level music students?
Dianna T. Kenny1,2, James Fortune1, and Bronwen Ackermann3

3 School of Biomedical Sciences, Faculty of Medicine, University of Sydney, Australia
Performance quality should be the dependent variable in studies of music

performance anxiety (MPA), as improvement in performance quality, as
opposed to greater subjective wellbeing or altered physiological states, is
the desired outcome for performing musicians of any intervention to re-
duce their arousal/anxiety during skilled performance, although the for-
mer factors may be necessary to achieve the latter. A predictive model of
performance quality was developed for empirical assessment using terti-
ary level flute students. Hierarchical regression analyses tested the hy-
pothesis that the specified variables (level of accomplishment on
instrument, performance preparation, psychological and physiological
measures) would predict expert ratings. The model accounted for 72.8%
of the variance in expert ratings. Three variables contributed most
highest achievement as a soloist (0.91), minutes spent practicing the test
piece (0.39), and music performance anxiety (0.38). These results indi-
cate that some arousal during skilled performance in skilled performers
is necessary to achieve a standard of performance that receives a high
rating from an expert listener. Fear of negative evaluation (FNE) had
negative beta weights in the predictive equation, showing that FNE is not
helpful for skilled performance, as it is likely to distract the performer
from the task and introduce irrelevant cognitions that would be likely to
interfere with performance. Musical aptitude and achievement are clearly
critical factors that are often neglected in models of both performance
anxiety and performance enhancement.
Keywords: music performance anxiety; skilled performance; practice;

expert ratings; prediction
Improved performance quality is the desired outcome of any intervention to

reduce cognitive or somatic arousal/anxiety during skilled performance, al-
though the reduction in the latter factors may be necessary to achieve the
former. In this paper, we assessed factors (technical mastery, specific task
mastery, quality and amount of practice, musical achievement, and psycho-
logical and physiological characteristics) associated with performance quality
in a unified model to determine their relative influences and hence to inform
pedagogy, self-management, and treatment.
METHOD
Participants
Participants were tertiary students of flute currently studying in the Bachelor

of Music (with a major in flute performance) in any of the four years of that
undergraduate degree, or Master of Music (Performance), or who had
graduated from either of these courses in the previous year. A recruitment
circular was distributed to 30 eligible flute students by post, of whom 24
expressed interest and were subsequently provided with the participant
information statement. In addition, signs were located at the Conservatorium,
Flute Society meetings, and orchestral rehearsals, which generated additional
enquiries. There were 20 flute players in the final group of participants.
Materials
Participants provided details about their highest level performing experience

as a soloist. A rating scale ranking solo performance achievement was de-
veloped and participants were awarded points on a ten-point scale from mu-
sical achievements below grade 8 (1-2) to played as soloist with a
professional symphony orchestra (e.g. ABC Young Performers Awards, 10).
The State-Trait Anxiety Inventory (STAI, Spielberger et al. 1983) consists
of two 20-item scales that measure state and trait anxiety. Internal consis-
tency of the state (Cronbachs =0.90) and trait scales (=0.82) in the pre-
sent sample was strong.
The Kenny Music Performance Anxiety (K-MPAI, Kenny 2004) was de-
veloped to assess the emotion-based theory of anxiety proposed by Barlow
(2000) as it applies to anxiety in the context of music performance; it as-
sesses uncontrollability, unpredictability, negative affect, situational cues,
attentional shift (e.g. task or self-evaluative focus, fear of negative evalua-
tion), physiological arousal, and memory. Questions are answered on a 7-
point Likert scale ranging from 0 (strongly disagree) to 6 (strongly agree).
Higher scores indicate greater anxiety and psychological distress. This scale
demonstrated excellent internal reliability (=0.94, Kenny et al. 2004).
Fear of Negative Evaluation (FNE, Watson and Friend 1969) assesses
expectations of being evaluated negatively and consists of 30 items that are
answered either true or false. The FNE has high internal consistency
(=0.94-0.96), strong one-month test-retest reliability (r=0.78-0.94), and
criterion validity (Watson and Friend 1969).
Surface electromyography (EMG) is a safe and non-invasive method for
measuring the nerve activation that occurs in a muscle as a result of voluntary
or reflex activation (Turker 1993). Stress levels and occupational loads are
two factors that influence surface EMG output readings, with particularly
responsive sites for these measures in the upper trapezius and frontalis mus-
cles (Westgaard 2000, Attebrant et al. 1996). Ten-second window measure-
ments were taken from the EMG recording of the upper left trapezius muscle
of the flute players in this study. A ratio of the maximal contraction from the
EMG (in Hz) to a reference voluntary contraction was recorded.
A silver chloride electrode was used to measure heart rate (ECG).
Maxima from the heart rate trace were averaged over blocks of time immedi-
ately before performance to achieve a resting heart rate and during perform-
ance of the musical item.
Till Eulenspiegels Lustige Streiche (R. Strauss, Op. 28), was chosen as the
set work because it is a well-known excerpt from the orchestral flute reper-
toire. A scoring protocol designed specifically for the study was developed by
a flute player from a premier national orchestra. Players were scored on a
criterion-referenced scale from 1 to 10 (each mark required the standard to be
met for all rankings below the ranking given).
Procedure
Ten days before the scheduled participation, participants were sent the FNE
and STAI-T questionnaires, the orchestral extracts, an instruction letter, a
map of how to get to the venue, and clothing instructions. These practices are
employed in most orchestral auditions, thus giving participants exactly the
same time-frame in which to prepare the music. Participants had an electro-
myographic trace of the left trapezius muscle taken and their HR recorded
during their performance in the acoustic laboratory at the Australian Centre
for Applied Research in Music Performance (ACARMP), Sydney Conservato-
rium of Music, University of Sydney. An expert flute player with extensive
adjudicator experience blind rated all the (recorded) performances in terms
of musical and technical proficiency from the industry standard recording.
Prior to commencement, participants completed the State Anxiety Inventory

(STAI-S). Powerlab recording was commenced and the neutral maximal vol-
untary contraction (MVC) was recorded with the levels adjusted appropri-
ately. Once measured successfully, the MVC was repeated for flute playing
posture. Participants were then instructed to begin playing after the re-
searcher tapped on their shoulder. At the end of the performance component,
participants completed the player physical profile questionnaire and the K-
MPAI.
RESULTS
The sample comprised 20 flute players (3 males, 17 females) who were all
students of flute at the Sydney Conservatorium of Music studying in the
Bachelor of Music (n=17) or Master of Music (Performance) (n=3) degrees.
Participants were aged between 19 and 32 years of age. They had been study-
ing flute for a minimum of 10 years and a maximum of 22 years (Table 1).
Fifteen of the participants (75%) wanted to become professional musicians,
three were undecided, and two did not wish to become professional musi-
cians.
A number of univariate analyses were undertaken to assess the relation-
ship of demographic and other factors with performance quality. None were
significantly associated (see Discussion). Hierarchical regression analyses
tested the hypothesis that the specified variableslevel of accomplishment on
the flute (ranking given for the highest level of achievement as a soloist),
performance preparation (number of minutes the test pieces were practiced),
psychological measures (K-MPAI, FNE, STAI), and physiological measures
(HR, EMG)would account for variation in expert ratings of the Strauss ex-
cerpt. Because the variables in the predictive model were on different scales,
they were converted to z-scores with a mean of 0 and a standard deviation of
1 before being entered into the regression analysis. ANOVA indicated a sig-
nificant F-statistic (F7,12=4.59, p=0.01), showing that using the model in-
creased the probability of correctly predicting expert ratings over chance. The
overall R2 for the model was 0.73 (i.e. the model accounted for 73% of the
variance in expert ratings in the Strauss excerpt, see Table 2).
DISCUSSION
In the univariate tests, the amount of time participants had been playing their
instrument was not associated with how much MPA, fear of negative evalua-
tion, or state or trait anxiety they experienced. Length of time one has been
playing ones instrument is a poor proxy for both the quality of practice that
Table 1. Means and standard deviations (SD) for player characteristics and expert rat-
ings of musical performance.
Characteristics Mean SD
Age 23.05 3.76
Years playing flute 13.98 3.15
Practice episodes per week 6.90 3.09
Average daily practice in minutes 131.25 58.71
Highest level of achievement as a soloist 4.50 2.12
Highest level of achievement in ensemble 5.05 2.26
Minutes practiced test piece 41.25 30.30
Expert rating Strauss 5.60 1.96
Physiological measures
EMG percent Strauss excerpt 24.42 9.22
Heart rate playing Strauss excerpt 99.15 13.31
Table 2. Beta coefficients, t-tests, and significance for the regression model predicting
expert ratings for the Strauss excerpt.
Unstand. Stand.
Strauss coeff. coeff. t Sig.
Z scores of predictors B SE Beta
(Constant) 1.78 2.75 0.65 0.53
Highest achievement as soloist 0.85 0.16 0.91 5.34 0.001
Minutes practised test piece 0.03 0.01 0.39 2.19 0.05
MPA total 0.04 0.02 0.38 1.83 0.09
FNE total -0.03 0.03 -0.15 -0.80 0.44
EMG percent Strauss excerpt 0.04 0.04 0.20 1.07 0.31
Heart rate Strauss excerpt -0.04 0.03 -0.26 -1.42 0.18
the performer undertakes and the amount of musical ability the performer
has. A better measure to assess these elements is the highest level of achieve-
ment attained in solo and/or ensemble performance, which was associated
with significantly lower reported MPA. Those with the most ensemble experi-
ence and achievement also had significantly less fear of negative evaluation. A
very robust yet simple model was developed for the prediction of expert rat-
ing of skilled performance. Three factorshighest level of achievement on
flute, minutes practiced the test piece, and MPAaccounted for 73% of the
variance in expert ratings of performance quality, although the Beta coeffi-

cient for MPA did not reach statistical significance. It is no surprise that the
first two factors predicted expert rating of performance quality. Any model of
performance quality must take into account the level of expertise attained and
the amount of practice undertaken. It is of concern that some models and
some treatment programs for performance anxiety do not take into account
these essential elements in both performance quality and gene-
sis/maintenance of MPA. Replication of the model on larger samples is
needed, as is further analysis of the role of MPA in performance quality.
References
Attebrant M., Mathiassen S. E., and Winkel J. (1996). Normalizing upper trapezius
amplitude: Comparison of ramp and constant force production. Journal of Elec-
tromyography and Kinesiology, 5, pp. 245-250.
Barlow H. (2000). Unravelling the mysteries of anxiety and its disorders from the per-
spective of emotion theory. American Psychologist, November, pp. 1247-1263.
Kenny D. T. (2004). Music performance anxiety: Is it the music, the performance, or the
anxiety? Music Forum, 10, pp. 38-43.
Kenny D. T., Davis P., and Oates J. (2004). Music performance anxiety and occupa-
tional stress amongst opera chorus artists and their relationship with state and trait
anxiety and perfectionism. Journal of Anxiety Disorders, 18, pp. 757-777.
Spielberger C. (1983) Manual for the State-Trait Anxiety Inventory. Palo Alto, Califor-
nia, USA: Consulting Psychologists Press.
Turker K. S. (1993). Electromyography: Some methodological problems and issues.
Physical Therapy, 3, pp. 698-710.
Watson D. and Friend R. (1969). Measurement of social-evaluative anxiety. Journal of
Consultative Clinical Psychology, 33, pp. 448-457.
Westgaard R. H. (2000). Work-related musculoskeletal complaints: Some ergonomics
challenges upon the start of a new century. Applied Ergonomics, 31, pp. 569-580.
Thematic session:
Performers health
Influence of musculoskeletal dysfunction and

pain on performance excellence
Anke Steinmetz
Institute for Musicians Medicine Berlin-Brandenburg,

Hospital for Manual and Pain Medicine, Germany
Playing-related musculoskeletal disorders (PRMD) and pain are a com-

mon phenomenon in professional musicians, with a prevalence of up to
80%. A majority of musicians are not aware that pain influences their
performance excellence. Recent research data demonstrate that pain has
the impact to change motor control strategies. Musicians with pain ex-
perience coordination impairments, muscle inhibition, and changes of
motor control, influencing their movement patterns while playing their
instruments. Vice versa, motor control impairments in musicians with
poor body awareness or body control can have an immense impact on the
development of musculoskeletal dysfunctions and pain. There is evidence
that musicians suffer double the amount of musculoskeletal dysfunctions
than non-musicians. Additionally, evidence is growing that the failure of
stabilization systems as the feedforward activation of transversus ab-
dominus muscle for the lumbar spine contributes to the development of a
chronic pain syndrome. A study at our clinic has provided clinical data
demonstrating the dysfunctions in various stabilization systems. Pre-
liminary results of ultrasound investigations of the activation of the
transversus abdominus muscle in musicians with musculoskeletal dys-
functions support these findings. This research data suggest an associa-
tion of the impairment of the lumbopelvic stabilization system with
PRMD.
Keywords: playing-related musculoskeletal disorders; pain; motor con-

trol strategies; muscle function; stabilization systems
Although the last 20 years have provided an enormous amount of research

data regarding playing-related musculoskeletal disorders (PRMD), preva-
lence rates persist with up to 80% of musicians affected. PRMD and pain
compromise the capability of musicians to perform and limit their perform-

ance excellence. Affected musicians frequently report on coordination im-
pairments associated with PRMD, notwithstanding measuring motor skills or
performance excellence remaining challenging. The MIDI-based technique
(Musical Instrument Digital Interface) for piano provides an opportunity for
analyzing specific movement parameters in piano playing, such as irregulari-
ties in key pressing (timing, loudness). Therefore, MIDI-based scale analysis
was developed for objective quantification of focal dystonia in pianists
(Jabusch et al. 2003). Apart from focal dystonia, data on incoordination or
aberrant movement patterns in musician is scarce. Fry et al. (1998) investi-
gated incoordination in pianists with overuse syndrome, also by analyzing
characters of key presses (duration, velocity, interval between key presses,
time off the metronome beat). Interestingly, musicians with overuse showed
distinct incoordination already in basic tasks such as playing C major scales at
different velocities. Development of further analyzing methods for string and
wind instruments are absolutely essential and should encompass a standard-
ized questionnaire screening for incoordination signs. Moreover, clinicians
managing musicians with PRMD should be aware of the influence of pain on
motor control strategies in order to provide tailored therapy strategies.
MAIN CONTRIBUTION
Muscle activation patterns and recruitment are altered in the presence of

pain. Several studies have demonstrated delayed activation of the deep ab-
dominal muscles, especially the transversus abdominus muscle (Hodges and
Richardson 1996, Hodges 2001), and increased activity of superficial muscles
(Leinonen et al. 2001) in low back pain (LBP). In order to stabilize the spine
during limb movements the tranversus abdominus muscle precedes in a
feedforward activation the contraction of the muscles producing the limb
movement (Hodges and Richardson 1996, Hodges and Richardson 1999).
Moreover, Hodges and Richardson demonstrated that these changes persist
even after the resolution of symptoms.
In LBP, these postural control deficits have been argued to contribute to
the recurrence of episodes (Hodges and Moseley 2003, Cholewicki et al.
2005). Likewise, there is evidence that in chronic neck pain the feedforward
activation of the deep cervical flexors is delayed in voluntary arm movements
(Falla et al. 2004a, 2004b) and that neuromuscular efficiency in the superfi-
cial neck flexors as the scalene and the sternocleidomastoid muscle is im-
paired (Falla et al. 2004c). Additionally, signs of myoelectric muscle fatigue
develop in scalene and sternocleidomastoid muscles (Falla et al. 2003). Ex-
perimental pain studies provide data demonstrating changes in activation

patterns of trapezius muscle subdivisions during repetitive shoulder flexion
and alterations in task-dependent changes in cervical agonist/antagonist ac-
tivity (Falla et al. 2007a, 2007b). Furthermore, in the arm, injection of hy-
pertonic saline into biceps and triceps brachii muscles resulted in
modification of the elbow flexor and extensor muscle activation during a re-
petitive elbow movement task (Ervilha et al. 2005) and muscle pain in the
biceps brachii muscle increased EMG activity in the trapezius muscle (Ervilha
et al. 2004, Schulte et al. 2004, Ervilha et al. 2005). Also, accumulating evi-
dence points to an impairment in the lower division of the trapezius muscle in
people with painful shoulder conditions (Ludewig and Cook 2000, Cools et
al. 2003, Lin et al. 2005).
Based on these data, in an attempt to transfer these findings on the behalf
of musicians, one can presume that:
Musicians with neck pain show a delay of deep cervical flexor activation
during upper limb movement (e.g. playing their instrument).
In musicians with neck pain, muscle function of scalene and sternocleido-
mastoid muscles are impaired.
Pain in the arm and the shoulder alters muscle activation and recruitment
patterns of trapezius muscle.
Musicians with low back pain show signs of delayed deep abdominal mus-
cle activation.
Accordingly, musicians have to be aware that pain has a profound potential to

alter their muscle function, contributing to performance impairments.
Musculoskeletal dysfunction in musicians
Pain is frequently preceded by musculoskeletal dysfunction. Motor control

changes contribute to inhibition or weakness in some muscle groups and in-
creased activity and hypertonicity of others. In consequence, trigger points
(local contraction bands or areas causing referred pain) occur and muscle
activation patterns, so called stereotypes, change. There is evidence that mu-
sicians experience twice the rates of functional disturbances as compared
with non-musicians (Steinmetz 2007). A study at our clinic, examining 84
musicians consulting our outpatient clinic for PRMD, has provided clinical
data demonstrating impairments in various stabilization systems (Steinmetz
and Seidel submitted). Dysfunctions of the postural stabilization systems
were present in 77 (92%) of the patients. Most frequently, in 85% impairment
of scapular stabilization system was found within the examination and in 71%
the lumbopelvic stabilization system (deep abdominal and back muscles).
Preliminary results of ultrasound investigations of the activation of the trans-
versus abdominus muscle in musicians with musculoskeletal dysfunctions
support these findings. This research data suggest an association of the im-
pairment of the lumbopelvic stabilization system with PRMD. Muscle dys-
functions and postural stabilization impairments can influence performance
excellence in various ways. An example demonstrating this phenomenon is
the case of a cellist with impaired interossei muscles, contributing to a lack of
stabilization abilities of the fourth finger of the left hand. A second example is
a violinist with neck and shoulder pain experiencing an alteration of shoulder
muscle activation patterns, resulting in a changed up-bow pattern showing
increased trapezius activation.
IMPLICATIONS
There is a need of further research to evaluate the influence of the lumpopel-

vic as well as scapular and cervical stabilization systems on the development
of PRMD. The identification of typical motor control changes associated with
the impairment of stabilization systems is of highest importance for musi-
cians because of their potential impact on performance excellence.
Treatment regimes in chronic back pain patients have shown that training
of these stabilizing muscles can prevent relapses of back pain. Adaptations of
these treatment strategies in the therapy of musicians may contribute to a
higher success rate in treating musicians with PRMD. Additionally, the ther-
apy of musicians has to implement an approach integrating the work on mo-
tor control and muscle function during playing a musical instrument.
Anke Steinmetz, Institute for Musicians Medicine Berlin-Brandenburg, Hospital for

Manual and Pain Medicine, Sana Kliniken Sommerfeld, Waldhausstr. 1, Kremmen/OT
Sommerfeld 16766, Germany; Email: a.steinmetz@sana-hu.de
References
Cholewicki J., Silfies S. P., Shah R. A. et al. (2005). Delayed trunk muscle reflex re-
sponses increase the risk of low back injuries. Spine, 30, pp. 2614-2620.
Cools A. M., Witvrouw E. E., Declercq G. A. et al. (2003). Scapular muscle recruitment
patterns: Trapezius muscle latency with and without impingement syndrome.
Ervilha U. F., Farina D., Arendt-Nielsen L., and Graven-Nielsen T. (2005). Experimen-
tal muscle pain changes motor control strategies during dynamic fatiguing contrac-
tions. Experimental Brain Research, 164, pp. 215-224.
Ervilha U. F., Arendt-Nielsen L., Duarte M., and Graven-Nielsen T. (2004). The effect of
muscle pain on elbow flexion and coactivation tasks. Experimental Brain Research,
156, pp. 174-182.
Falla D., Rainoldi A., Merletti R., and Jull G. (2003). Myoelectric manifestations of
sternocleidomastoid and anterior scalene muscle fatigue in chronic neck pain pa-
tients. Clinical Neurophysiology, 114, pp. 488-495.
Falla D., Bilenkij G., and Jull G. (2004a). Patients with chronic neck pain demonstrate
altered patterns of muscle activation during performance of a functional upper limb
task. Spine, 29, pp. 1436-1440.
Falla D., Jull G., and Hodges P. W. (2004b). Feedforward activity of the cervical flexor
muscles during voluntary arm movements is delayed in chronic neck pain. Experi-
mental Brain Research, 157, pp. 43-48.
Falla D., Jull G., Edwards S. et al. (2004c). Neuromuscular efficiency of the sternoclei-
domastoid and anterior scalene muscles in patients with neck pain. Disability and
Rehabilitation, 26, pp. 712-717.
Falla D., Farina D., and Graven-Nielsen T. (2007a). Experimental muscle pain results in
reorganization of coordination among trapezius muscle subdivisions during repeti-
tive shoulder flexion. Experimental Brain Research, 178, pp. 385-393.
Falla D., Farina D., Kanstrup Dahl M., and Graven-Nielsen T. (2007b). Muscle pain
induces task-dependent changes in cervical agonist/antagonist activity. Journal of
Applied Physiology, 102, pp. 601-609.
Fry H. J., Hallett M., Mastroianni T. et al. (1998). Incoordination in pianists with over-
use syndrome. Neurology, 51, pp. 512-519.
Hodges P. W. and Richardson C. A. (1996). Inefficient muscular stabilization of the
lumbar spine associated with low back pain. A motor control evaluation of transver-
sus abdominis. Spine, 22, pp. 2640-2650.
Hodges P. W. and Richardson C. A. (1999) Altered trunk muscles recruitment in people
with low back pain with upper limb movement at different speeds. Archives of
Physical Medicine and Rehabilitation, 80, pp. 1005-1012.
Hodges P. W. and Moseley G. L. (2003). Pain and motor control of the lumbopelvic
region: Effect and possible mechanisms. Journal of Electromyography and Kinesi-
ology, 13, pp. 361-370.
Hodges P. W. (2001). Changes in motor planning of feedforward postural responses of
the trunk muscles in low back pain. Experimental Brain Research, 141, pp. 261-
266.
Jabusch H.-C., Vauth H., and Altenmller E. (2003). Quantification of focal dystonia in
Leinonen V., Kankaanpaa M., Luukkonen M. et al. (2001). Disc-herniated back pain
impairs feed-forward control of paraspinal muscles. Spine, 26, pp. E367-372.
Lin j.-J., Wu Y.-T., Wang S.-F., and Chen S.-Y. (2005). Trapezius muscle imbalance in
individuals suffering from frozen shoulder syndrome. Clinical Rheumatology, 24,
pp. 569-575.
Ludewig P. M. and Cook T. M. (2000). Alterations in shoulder kinematics and associ-
ated muscle activity in people with symptoms of shoulder impingement. Physical
Therapy, 80, pp. 276-291.
Schulte E., Ciubotariu A., Arendt-Nielsen L. et al. (2004). Experimental muscle pain
increases trapezius muscle activity during sustained isometric contractions of arm
muscles. Clinical Neurophysiology, 115, pp. 1767-1778.
Steinmetz A. (2007). Differences in musculoskeletal dysfunctions between music and
nonmusic students. Medical Problems of Performing Artists, 22, pp. 169-178.
Steinmetz A. and Seidel W. (submitted). Impairment of postural stabilization systems in
musicians with playing-related musculoskeltal disorders.
Do pianists play with their teeth?
Sofia Loureno1,2, Miguel Pais Clemente3, Daniela Coimbra1,2,

lvaro Barbosa2, and Joo Carlos Pinho3

3 Faculty of Dentistry, University of Porto, Portugal
The aim of this study was to find out whether the masticatory and pos-
tural muscles are used by pianists during their music performances. The
study also aimed to ascertain whether the complex neuromuscular activ-
ity involved in the act of playing the piano also encourages hyperactivity
in terms of the masticatory muscles. The bio-electric potentials of the
masticatory and postural muscles of 20 pianists were recorded by surface
electromyography (EMG). The EMG recordings obtained regarding the
temporal and masticatory muscles are much higher than those recorded
when in the resting position. These recordings, which are not the same as
those obtained, for example, when the individuals are chewing hard food
such as a carrot, are nonetheless indicative of daily parafunctional activ-
ity in musicians who often study for as much as seven hours per day.
Keywords: pianists; electromyography; muscular hyperactivity; masti-

catory muscles; postural muscles
Music performance is based on the knowledge musicians acquire through a

great deal of deliberate practice to obtain high levels of skills (Ericsson et al.
1993, Hallam 1997). Consequently, the incidence of focal dystonia, for exam-
ple, may be as high as one in 200 professional musicians (Altenmller 2000).
Research has shown the musculoskeletal system to be the most frequently
involved area of impairment (Morse et al. 2000). Orofacial problems may
also be included as result of the musicians professional activity. Orthodontic
problems, focal dystonia, herpes labialis, dry mouth, and temporomandibular
joint disorders (TMD) have been identified especially among wind and string
players, as well as vocalists. These problems may result from the impact of
the selected instrument on the orofacial structures of the musician (Raney
2006) or from an inadequate dental treatment and increase with stress. In all
cases, it may be detrimental to the musicians careers.
Pianists are among the highest in number of music college students in
Portugal, both in classical and jazz degree courses. Research on piano related
injuries is mainly focused on the upper limbs, namely the hand adaptation to
the keyboard, hand pain, focal hand dystonia, finger joints or tendon, and
arm overuse related problems (Shields and Dockrell 2000, Sakai 2008). Yet,
little has been written about their orofacial activity. Given the duration and
intensity of the daily practice these pianists might undertake (as much as
seven hours per day), they may well develop a parafunctional activity, espe-
cially in the masticatory and postural muscles. It is extremely important to
monitor such activity so that muscular hyperactivity does not interfere in
maintaining a functional equilibrium of the cranio-cervic0-mandibular com-
plex. Hence, the aim of the present study was to determine whether the mas-
ticatory muscles, whose main functions are chewing, swallowing, and speak-
ing, are used by pianists during their music performances and so to ascertain
whether the complex neuromuscular activity involved in the act of playing the
piano also encourages hyperactivity in terms of the masticatory and postural
muscles. Since we had both classical and jazz pianists in our sample, an addi-
tional aim was to ascertain whether different styles have a different effect on
the masticatory and postural muscles.
METHOD
Participants
The experiment was conducted with 20 participants, ranging from 18 to 27

years old with classical and jazz piano training. In this article, we present the
most significant cases.
Materials
The electromyographic (EMG) activity was recorded using the Bio EMG 2
electromyograph with eight channels (Bioresearch Associates Inc., Milwau-
kee, Wisconsin, USA).
Procedure
An alcohol pad was wiped in the area where the sensors were to be placed and
a conductive gel was placed on a 9 mm diameter disposable silver/silver-
chloride bipolar surface electrodes (Duo-Trode, Myotronics Inc., Seattle,
Washington, USA), before their attachment to the skin. These bipolar surface
Figure 1. Pianists performing a jazz (left) and a classical piece (right).
electrodes were located on the muscular bellies of the anterior temporalis,

masseter, digastric, and trapezious muscles of both sides (right and left), and
parallel to the muscle fibers.
The electromyographic activity was recorded during the following proce-
dures: (1) at rest, (2) maximum voluntary clenching in the intercuspal posi-
tion, (3) maximum voluntary opening, (4) playing piano, and (5) eating a
cake. Subjects played a piece of their choice in a classical and jazz style (see
Figure 1). A resting period of one minute between recordings was allowed to
avoid muscular fatigue.
RESULTS
The head posture adopted by piano players during their performance gave
different patterns of EMG activity in the masticatory and postural muscles.
The visual system used in this study played an important role in the percep-
tion that there are variations in seated positions between classical piano and
jazz players, with direct implications on the cranio-cervico-mandibular com-
plex.
Likewise, it is possible to verify an adoption of an anterior head position
on jazz players, with a tendency of most of these subjects to maintain a jaw
position by using the anterior temporal more than the masseter muscles. In
classical players, there were higher values on the EMG activity of the masse-
ter muscles compared with the jazz players. A contributing factor can be that
jazz players are used to improvising during their performance while classical
players have a prolonged and strained stress placed on the masticatory mus-
cles, specifically the masseter muscle. For the particular case of classical pi-
anists, we were able to compare the activity of a mandible elevator muscle
Figure 2. Pianist with high electromyographic activity of the masseter muscle.
(masseter) during one of its main functions: eating. When eating a cake, the
bioelectrical potential of the masseter muscles reached 15-18 V, while play-
ing for example the C minor Rachmaninoff concerto, the masseter muscle can
reach values of 49.6 V.
The findings of this study illustrate that pianists may feel persistent neck
and orofacial pain due to the large number of hours of piano exposure, high-
lighting the physical demands as well as the psychosocial factors involved in
such a demanding profession. Understanding what kind of muscles are being
used in the cranio-cervico-mandibular complex during the performance of a
pianist is of vital importance as it can help in the correct diagnosis of a con-
crete problem originated by repetitive movements.
The aim of the study was to understand how the orofacial muscles behave
during voluntary exercise such as piano playing, where the activity of some
muscleslike masseter, digastrics, or the anterior temporalshould have
minimum activity given that their key activity should occur mainly during
mastication, speech, and swallowing. In this particular case it was not a ques-
tion of being able to evaluate if the fatigue of these muscles could restrict the
pianist to the point of no longer being able to performing the task: piano
playing. An overuse of these muscles may have a direct impact on the pianists
quality of life when performing their maximal voluntary performance of these
orofacial muscles during normal tasks like eating.
Figure 3. Pianist with high electromyographic activity of the trapezious muscle.
In time, these kinds of restrictions can induce pain that can be related to
certain movements or occur at any time of the day. This special attention
toward performers health issues and specific needs will provide a working
diagnosis, allowing health care professionals to focus their examination
rather than conducting a series of tests that are usually time consuming.
Having piano players, teachers, and performing arts medicine profession-
als conscious of what is actually happening to the orofacial region while
playing piano will encourage the daily supervision of any parafunctional habit
like clenching their teeth during performance.
DISCUSSION
Do pianists really play with their teeth? The essential point is yes, some
pianists do in fact play with their teeth, since they have an activity of their
masseter and anterior temporal muscles that acts as an elevator of the man-
dible, forcing the mandible teeth to contact the maxillary teeth. So here piano
players, during their daily performance, have parafunctional habits inducing
hyperactivity of some of the orofacial muscles that very often is associated
when an individual is concentrated on a particular task or when anxiety levels
rise.
Further studies would benefit from a multimodal approach in which this

method would run in parallel with high resolution sound recording and mo-
tion capture.
Acknowledgments
This work has been supported by the Fundao para a Cincia e Tecnologia (FCT) and
Programa Operacional Cincia e Inovao 2010 (POCI 2010), co-funded by the Portu-
guese government and European Union by the FEDER program.
Sofia Loureno, Research Centre for Science and Technology of the Arts (CITAR),
School of the Arts, Portuguese Catholic University, Rua Diogo Botelho 1327, 4169-005
Porto, Portugal; Email: slourenco@porto.ucp.pt
References
Altenmller E. (2000). From Laetoli to Carnegie: Evolution of brain and hand as pre-
requisites of music performance in the light of music physiology and neurobiology.
Paper presented at the Tenth Symposium of the International Study Group of the
Archeology of Music, Kloster Michaelstein, Germany.
Ericsson K., Krampe R., and Tesch Romer C. (1993). The role of deliberate practice in
the acquisition of expert performance. Psychological Review, 100, pp. 363-406.
Hallam S. (1997). Approaches to instrumental practices of experts and novices: Impli-
cations for education. In H. Jorgensen and A. Lehmann (eds.), Does Practice Make
Perfect? (pp. 89-107). Oslo: Norwegian Academy of Music.
Morse T., Ro J., Cherniack S., and Pelletier S. (2000). A pilot population study of mus-
culoskeletal disorders in musicians. Medical Problems of Performing Artists, 15,
pp. 81-87.
Raney N. (2006). The effects of orthodontic appliances on wind-instrument players.
Journal of Clinical Orthodontics, 40, pp. 384-387.
Sakai N. (2008). Keyboard span in old musical instruments: Concerning hand span and
overuse problems in pianists. Medical Problems of Performing Artists, 23, pp. 169-
171.
ISSTIP performing arts clinics at the London

College of Music 1990-2005: The first ideas,
the realization, and the lessons for the future
Andy Evans1 and Hara Trouli2
1 London College of Music, Thames Valley University, UK

2 International Society for the Study of Tension in Performance, UK
We take this opportunity of presenting data from the International Soci-

ety for the Study of Tension in Performance (ISSTIP) clinics in London
between 1990-2005. The ISSTIP clinics were multi-disciplinary and were
based at the London College of Music. They were founded by Carola
Grindea (1914-2009), a piano pedagogue specializing in posture and re-
laxation techniques. ISSTIP clinics were among the very first performing
arts clinics in the UK, and they aimed to address the pressing issues of
performing artistsgeneral psychological and physical tension and pain
and posture problems in practice and performance. The aim of this retro-
spective study is to examine the structure and success of the ISSTIP clin-
ics and to look at what kind of models these early clinics represented
compared with contemporary theory and practice.
Keywords: performing arts health; performing arts psychology; per-

forming arts clinics; posture; relaxation techniques
The International Society for the Study of Tension in Performance (ISSTIP)

clinics offered free consultations to musicians and other performers, and in
the first years, took place twice a month. Carola Grindea acted as the focal
point and coordinated the clinics in collaboration with medical specialists.
Additional practitioners included psychologists specializing in the arts, Alex-
ander and Feldenkrais technique specialists, and therapists from the fields of
complementary medicine. This multi-practitioner approach was typical of
ISSTIPs holistic and inclusive outlook. During the period 1990-94, we also
studied two other clinics run by leading performing arts practices in London:
the British Performing Arts Medicine Trust (today known as BAPAM) and the
Arts Psychology Consultants (APC). Together, such comparative data forms
an interesting and valuable profile of how performing artists were treated in

London during the early 1990s.
In terms of the development of organizations dealing with the health of
performing artists, the 1980s was the pioneering decade where new ideas and
new people came into the field. The 1990s was the decade of conferences and
clinics. Practitioners interested in the performing arts flocked to conferences,
and a large number of performers came to the clinics to benefit from the new
initiatives. This new decade since 2000 has seen the emphasis shifting to the
education of new practitioners, with courses in performance health such as
the masters program in performance health and personal development at
Thames Valley University and a number of short courses, meetings, and
seminars organized by ISSTIP, the Centre for Performance Science at the
Royal College of Music, the Institute of Musical Research, BAPAM, and indi-
vidual music colleges in the UK. The UK clinics active in the 1990s and for a
big part of the 2000s were largely staffed by competent professional musi-
cians like pianist Carola Grindea and health professionals, most of whom
were also amateur musicians. One reason for the effectiveness of the new raft
of practitioners was their own intimate knowledge of performing and per-
formers, enabling them to speak the language of musicians and to identify
with many of their more technical, social, and psychological issues.
METHOD
Participants
We have looked at data from three clinics active in London in the period
1990-94: ISSTIP, BAPAM, and APC. For the period 1995-2005, we only ex-
amined the data from the ISSTIP clinics. The ISSTIP clinic had a break in
operation in 2005 but is due to restart in 2009.
Procedure
We analyzed the collected data from those attending the clinics in two peri-
ods: 1990-94 and 1995-2005. During the first period of our study, we exam-
ined comparative data from ISSTIP, BAPAM, and APC clinics. In the second
period of the study we looked exclusively at the ISSTIP clinic, and we com-
pared these results with those of the earlier period of the same clinic.
RESULTS
ISSTIP and BAPAM offered three areas of help: medical, postural, and psy-
chological. APC offered just psychological help and referred cases needing
Table 1. Overall profile of the three clinics studied for the period 1990-94.
ISSTIP BAPAM APC

Number 363 478 199
Male/female (%) 37.0/63.0 35.5/64.5 50.0/50.0
Average age (years) 26 31 n/a
Years of data 5 5 5
Table 2. Professions established by the Arts Council of Great Britain (1990-94).
ISSTIP BAPAM APC

Music (%) 96.4 80.0 79.0
Dance (%) 1.7 7.0 5.0
Drama (%) 0.3 9.0 7.5
Media, literature, art (%) 1.6 4.0 8.5
Table 3. Type of musicians and type of instrument (1990-94).
ISSTIP BAPAM APC

Classical/pop, jazz (%) 98.0/2.0 n/a 79.0/21.0
Strings (%) 29.0 15.0 22.1
Wind (%) 7.9 7.0 11.0
Brass (%) 2.9 6.0 6.4
Percussion (%) 0.9 7.0 7.1
Keyboard (%) 36.0 16.0 12.5
Plucked instruments (%) 7.1 21.0 9.9
Voice (%) 15.0 11.1 26.1
Conductors (%) 0.9 n/a 3.5
Composers (%) 0.3 n/a 1.4
Table 4. Employment status (1990-94).
ISSTIP BAPAM APC

Professional (%) 36.0 79.0 66.8
Semi-professional (%) 0.6 6.0 3.0
Amateur (%) 3.8 2.0 4.5
Students (%) 51.2 13.0 25.7
Teachers (%) 8.4 n/a n/a
medical or postural treatment to the other two bodies. As such its number of
cases is proportionately less. The preponderance of females in the first two
clinics is interesting since numbers were equal in those seeking psychological
help (see Table 1). The breakdown by profession (see Table 2) shows an
ISSTIP sample composed almost entirely of musicians. The choice of a music
college setting with its attendant students would bias the numbers toward
musicians in the case of ISSTIP, though it is clear that the clientele most in-
terested in using such clinics was musicians in general. Musicians seem more
willing to seek direct help from performing arts organizations, although
dancers and actors do also have their own health networks. Table 3 shows
very considerable bias to classical musicians which is not mirrored in the
general pool of active musicians. With a music college setting we might expect
this bias in the ISSTIP clinics, but it also occurs in the APC clinics.
ISSTIP clinics saw a proportionately larger number of students and
teachers than BAPAM or APC (see Table 4), and the assumption is that again
this is because clinics took place in a music college where teachers also sat in
and took an interest. Even so, the fact that 36% of ISSTIP clients are profes-
sional musicians shows that the clinic was known and used by a much wider
clientele. Also interesting is that music students, who would typically have a
generalist student counselor available to them, preferred to seek specialized
help with their performance problems. One quarter of APC clients were from
music schools that had their own student counselors, indicating that special-
ist knowledge of dealing with performance problems responds to a need in
music students. When Andy Evans was student counselor at the Royal Acad-
emy of Music, three quarters of the cases he saw were related to performance
issues, rather than general counseling matters (Evans 2003).
ISSTIP clinics had an unexpectedly high percentage of keyboard players;
other instruments being broadly comparable (see Table 3). This may be a bias
due to the fact that Carola Grindea was a well-known piano pedagogue.
Clearly, the profile of the organization affects the profile of those using it.
Given the already established weighting toward classical musicians we would
also expect larger numbers of string players to reflect their distribution in
orchestras. Among the wind players, flute and clarinet predominated. Gui-
tarists tended to go more to BAPAM where they would get medical help, while
singers tended to make more use of psychologists. The nature of each organi-
zation determined the primary service offered to those who used it, and the
primary focus in the ISSTIP clinics was on posture control, breathing, and
music ergonomics (player-instrument interaction), with a secondary empha-
sis on medical interventions and a tertiary emphasis on psychological meth-
ods (see Table 5). This was the other way around in APC consultants who
offered specialized psychological help, while BAPAM as a medical organiza-

tion offered medical assessment and intervention with posture and psycho-
logical methods available for further referral.
Later data has been collected for the ISSTIP clinics only between 1995 and
2005 (see Table 6). After 1995, the psychologists were working privately and
medical participation in the clinics had also declined since the doctors were
mostly working through BAPAM. The ISSTIP clinics had therefore reduced
internal medical and psychological input, and any cases needing such atten-
tion were referred externally. Observations include (1) the total number at-
tending ISSTIP clinics per annum declined considerably, (2) treatment
methods reverted almost entirely to tension control through postural adjust-
ment, (3) there was a small increase in popular/jazz musicians, (4) the divi-
sion between male and female had reverted to almost 50:50, (5) the
percentage of keyboard players had increased dramatically, and (6) the num-
ber of students stayed at around 50%, while teachers attended more.
DISCUSSION
Clinics all have a distinct profile depending on the founders and their influ-
ence, the name of the organization and its implications, and the location of
the clinics. This profile acts as a triage system for not only those who attend
but the focus of the treatment they are thought to need and the treatment
they actually receive. Clinics have been largely built around musiciansnot
only as clients but also music as the secondary skill of the practitioner. The
early 1990s were a particularly active time for clinics as performers aired their
problems in many cases for the first time. Ideally a performer who attends a
clinic will get an impartial assessment entirely congruent with his or her
needs. This sample of data shows that this was not happening because of in-
herent bias within the organizations. This may be an insoluble problem, and
it is left ultimately with the client to make their own choices. In an ideal world
referrals would take place routinely to the appropriate specialist, and it is to
be hoped that teamwork will always be important in the field of performance
health. Another issue is that medical practitioners are trained to pass clients
to specialists, but this is less the case with therapists and posture teachers
who believe they deal with the whole person, not the problem. The field of
performance health however relies on specialists, and it is important that
specialization is respected by all who practice in this field, passing clients on
where appropriate for the client and the problem.
From these conclusions, we can ask wider questions as to whether such
clinics will always represent a symbiosis between the individual nature of
Table 5. Practitioners used by each clinic (1990-94).
ISSTIP BAPAM APC

Medical specialists (%) 42.7 79.0 n/a
Psychologists (%) 5.8 13.0 100.0
Posture, physio (%) 51.5 8.0 n/a
Table 6. ISSTIP clinics data comparisons 1990-94 and 1995-2005.
1990-94 1995-2005
Overall number seen 363 376
Male/female (%) 37.0/63.0 48.0/52.0
Musicians/dancers, actors, etc. (%) 96.4/3.6 98.0/2.0
Classical/pop, jazz (%) 98.0/2.0 94.0/6.0
Piano, keyboard (%) 36.0 64.0
Professional (%) 36.0 25.0
Students (%) 51.2 57.0
Teachers (%) 8.4 11.0
Medical treatment (%) 42.7 6.5
Psychology (%) 5.8 0.0
Posture (%) 51.5 93.5
their setup and how they function, or whether we should seek to move toward
a more all-embracing provision of care that could be replicated in a more
generic way for future clinics. Some such questions include (1) whether there
are a set of protocols that would be helpful, (2) how staff should be chosen
and trained, and (3) what should be the relationship between clinics, support
organizations, educational establishments, and the wider world of the arts
with a goal of providing students, teachers, and professional performers with
a better and more effective service. When ISSTIP clinics resume later in
2009, these questions will be in the forefront of discussions.
Hara Trouli, ISSTIP Head Office, 6 Jonathans, Dene Road, Northwood, Middlesex HA6
2AD, UK; Email: isstip@gmail.com
References
Evans A. (2003). Secrets of Performing Confidence. London: A&C Black.

Thematic session:
Performance and life factors
Diversity of dancer experience in a dance

program
Shona Erskine1 and Mary Ainley2
1 Western Australian Academy of Performing Arts, Edith Cowan University, Australia

2 School of Medicine, Dentistry, and Behavioural Science,
University of Melbourne, Australia
This research describes the relationship between dance program struc-

ture and the types of experiences had by participating dancers. The re-
search followed 48 participants through a 6-month dance project that
culminated in a weeklong performance season. Recognizing that reported
positive developments occur within the structure of programs, the as-
pects of a dance program that were significant in forming participants
different experiences were examined using a technique known as Q-
methodology. The process of factor extraction and the content of factors
are summarized to identify the different dancers perspectives on their
experience of the program. A discriminant function procedure was then
used to determine whether demographic variables significantly discrimi-
nated between the groups. Divergent Q-sorts were analyzed to highlight
that some dancers experiences were different to the main types of ex-
perience available in the program. As dance participation has long been
assumed to increase self-esteem, self-concept was measured using the
Self-Description Questionnaire (SDQ). The paper concludes by outlining
a multi dimensional view of dancers experience.
Keywords: dance; education; Q-sort; dancer experience; SDQ-III
There is little research into how young students experience the educational
dance activities designed by adults (Bamford 2006). Phenomenological re-
search suggests that dancers share common experiences across age, culture,
and gender (Alter 1997, Bond and Stinson 2000, Bracey 2004, Lindqvist
2001, Stinson 1997, Stinson et al. 1990). It has also been suggested that envi-
ronmental factors may influence this experience (Fensham and Gardner
2005, Green 1999, Hefferon and Ollis 2006, Oliver 2000, Thomas 1993). In
addition, within such experiences enhanced self-esteem appears to be one of

the strongest measures of arts participation (Burton et al. 2000, Heath and
Roach 1999, National Advisory Committee on Creative and Cultural Educa-
tion 1999, Winner and Hetland 2000). In light of this, the present study ex-
amines the aspects of a dance program that are significant in forming
dancers different perspectives on dance experience while testing whether
self-concept changes over the course of a dance program. A technique known
as Q-methodology was utilized to investigate the structure of dancers per-
ceptions and evaluations of their experience as it provides an empirical
method for the study of subjectivity (Brown 1996, McKeown 1988). Self-con-
cept was measured using the Self-Description Questionnaire (SDQ) (Marsh
1992). The SDQ instruments have provided particularly strong tests of the
Shavelson et al. (1976) theory of self-concept that posits a multifaceted, hier-
archical model of self-concept.
METHOD
Participants
The participants comprised 48 dancers that had accepted a place in the pro-
gram. This group contained equal numbers of female and male dancers
(mean age=16.25 years, SD=2.41). The second group of participants con-
tained 9 former dancers, aged between 17 and 23 years, who participated in
focus groups designed to generate material for the Q-sort.
Materials
Basic demographics were taken for each participant including sex, age, and
how many years they had been involved in (1) dance and (2) the program. The
content of the focus group discussions served to generate Q-sort materials: 19
experience statements, which represented overarching descriptions of
grouped quotes, and 8 categories, which grouped similar experience state-
ments. The future in dance category contained only a future in dance experi-
ence statement. The program elements category contained performing,
production week, rehearsals, and dancing experience statements. The cho-
reographic process category contained the idea/concept, team-
work/groupwork, and making up movement experience statements. The
peers category held only one experience statement: other dancers. The role
models category included choreographers and the artistic director. External
people contained outside friends and parents. The personal attributes cate-
gory included growth/maturity, self-disciple, and confidence. Finally, the
program values category contained equality, supportive, and competitive

experience statements. The 136-item self-report inventory SDQ-III was used
to assess three areas of academic self-concept (mathematics, verbal, aca-
demic), nine areas of non-academic self-concept (problem solving, physical
abilities, physical appearance, same sex relations, opposite sex relations, pa-
rental relations, spiritual values/religion, honesty/trustworthiness, emotional
stability) and a global (general) self.
Procedure
The focus groups were conducted leading up to and beginning the program.
The Q-sort was conducted between 2-3 weeks after the conclusion of the
performances. The instrumental basis of the Q-methodology is the Q-sort
technique (McKeown 1988). This required the participants to rank-order
systematically the 19 experience statements in a forced choice procedure ac-
cording to their own personal perspective of the program. The Self Descrip-
tion Questionnaire (SDQ-III) was administered as a pre and post measure.
RESULTS
The 48 Q-sorts were subjected to a factor analysis using principal compo-

nents extraction. The four factor solution had three dancers with either nega-
tive factor loadings or low factor loadings. It was decided to treat these three
participants as separate cases and use deviant case analysis to highlight that
their experiences were different to the other dancers. With these three cases
removed, the remaining 45 Q-sorts were subjected to the same principal
component analysis as previously described. Evidence suggested that four
factors accounting for 67.66% of the variance would be the most interpretable
solution. The members of the four groups were labeled performers, investi-
gators, personal growthers, and apprentices. The performers three most
important experience statements were performing, production week, and
dancing. The three most important experience statements for the investiga-
tors were the idea/concept, teamwork, and equality. The three experience
statements with the highest factor scores for the personal growthers were
growth/maturity, self-discipline, and confidence. The apprentices three ex-
perience statements with the highest factor scores were future in dance, the
artistic director, and choreographers. The three dancers who were not mem-
bers of any factor group are labeled peers, competitive performer, and future
growth. Peers rated outside friends, self-discipline, and other dancers as im-
portant. Performing, dancing, and competition were important to competitive
performers. Future growth thought other dancers, future in dance, and

growth/maturity were important.
In the second part of the Q-sort analysis a discriminant function proce-
dure was used to determine whether demographic variables significantly dis-
criminated between the four groups. The performers were more likely to be
female with above average years dance experience. The investigators in-
cluded an equal ratio of female and males but had lower than average years
dance experience. The personal growthers were male with lower than aver-
age dance experience, and the apprentices were male with above average
years dance experience. All three case studies were males aged 17-20 years,
had been with the program for between 1-2 years, and had between 1-3 years
dance experience. Clearly these were participants who were least experienced
with dance.
Paired t-tests were run to analyze the differences in SDQ-III scale scores
across time. Results revealed the largest differences across time on the emo-
tional stability subscale (mean=-0.36, SD=0.70, d=0.37; t45=-3.47, p=0.001)
and general self-concept subscale (mean=-0.24, SD=0.68, d=1.49; t45=-2.41,
p<0.05). To determine if there were any significant differences between the
four groups of dancers on emotional stability and general self-concept sub-
scales, a repeated measures ANOVA was run. For emotional stability, the
within subject test indicated that there was a significant change in emotional
stability over time for all four groups. The between subjects test indicated that
there was no significant difference between the groups (F3,39=1196.91,
p>0.05). However, there was a significant interaction effect by time
(F3,39=1196.92, p<0.001, partial 2=0.97). There was an increase in emotional
stability across time with the investigators and personal growthers increas-
ing more than the performers and apprentices. The within subject test for
general self-concept indicated that there was a significant change in general
self-concept across time (F3,39=8.55, p<0.05, partial 2=0.18). The between
groups test indicated that there was no significant difference between groups
(F3,39=0.29, p<0.05). Again, there was a significant interaction effect
(F3,39=1255.93, p<0.001, partial 2=0.97). The apprentices had the largest
increase in general self-concept across time.
DISCUSSION
The program provided different types of dancers with different ways of con-
necting into dance and demonstrated how dance programs can promote ef-
fective differentiated learning and opportunities for both male and female
dancers. The dance as performance experience produced educational out-
comes in relation to the intelligent and considered presentation of dance to

audiences. Dance as investigation taught dancers about the creation of dance
as a contemporary art. The educational outcomes inherent in the dance as
personal growth experience taught the value of application and self-reflection
to achievement. The dance as apprenticeship experience created educational
outcomes in relation to learning from role models and accessing information
in others. Analysis of the divergent cases provided valuable information on
the recognition of important friendships both inside and outside of the studio.
Whether they offered support through mentorships or friendships, an op-
portunity to improve through competition, or a pathway to a future in dance,
the other dancers were integral to these young mens perspective on dance
experience.
Conclusions can be drawn on where and why self esteem may be in-
creased through dance participation. With their emotional stability increasing
over the program, the investigators appear to have been experiencing posi-
tive psychological benefits from the psychologically challenging experience of
composition and creation in dance. The personal growthers also demon-
strated increased emotional stability across the program indicating that their
ability to grow personally may have been a result of a supportive environment
and positive peer relationships. The apprentices had the largest increase in
general self-concept. They reported feelings of support from those around
them, and it may be through strong role modeling that they were developing
as young people. One of the features that differentiated them from those with
different perspectives was their ranking of other dancers as neither important
nor unimportant. Each of the other perspectives identified recognized their
peers as important to their perspective on dance.
Shona Erskine, Dance Department, Edith Cowan University, PO Box 432, Bassendean,
Western Australia 6054, Australia; Email: shonaerskine@gmail.com
References
Alter J. B. (1997). Why dance students pursue dance: Studies of dance students from
1953 to 1993. Dance Research Journal, 29, pp. 70-89.
Bamford A. (2006). The Wow Factor. Berlin: Waxmann Mnster.
Bond K. E. and Stinson S. W. (2000). I feel like Im going to take off! Young peoples
experiences of the superordinary in dance. Dance Research Journal, 32, pp. 52-87.
Bracey L. (2004). Voicing connections: An interpretive study of university dancers
experiences. Research in Dance Education, 5, pp. 7-24.
Brown S. R. (1996). Q methodology and qualitative research. Qualitative Health Re-

search, 6, pp. 561-567.
Burton J. M., Horowitz R., and Abeles H. (2000). Learning in and through the arts: The
question of transfer. Studies in Art Education, 41, pp. 228-257.
Fensham R. and Gardner S. (2005). Dance classes, youth cultures and public health.
Youth Studies Australia, 24, pp. 14-20.
Green J. (1999). Somatic authority and the myth of the ideal body in dance education.
Dance Research Journal, 31, pp. 80-100.
Heath S. B. and Roach A. (1999). Imaginative actuality: Learning in the arts during the
nonschool hours. In E. B. Fiske (ed.), Champions of Change (pp. 19-34). Washing-
ton, DC, USA: The Arts Education Partnership.
Hefferon K. and Ollis S. (2006). Just clicks: An interpretive phenomenological analy-
sis of professional dancers experience of flow. Research in Dance Education, 7, pp.
141-159.
Lindqvist G. (2001). The relationship between play and dance. Research in Dance Edu-
cation, 2, pp. 41-52.
Marsh H. W. (1992). Self-Description Questionnaire (SDQ) III. Macarthur, New South
Wales, Australia: The University of Western Sydney.
McKeown B. F. and Thomas D. B. (1988). Q Methodology. Newbury Park, California,
USA: Sage.
National Advisory Committee on Creative and Cultural Education (1999). All Our Fu-
tures. Sudbury, Suffolk, UK: DFEE Publication.
Oliver S. (2000). Aesthetic understanding in dance in community and schools. Research
in Dance Education, 1, pp. 216-220.
Shavelson R., Hubner J., and Stanton G. (1976). Validation of construct interpretations.
Review of Educational Research, 46, pp. 407-441.
Stinson S. W. (1997). A question of fun: Adolescent engagement in dance education.
Dance Research Journal, 29, pp. 49-69.
Stinson S. W., Blumenfield-Jones D., and Van Dyke J. (1990). Voices of young women
dance students: An interpretive study of meaning in dance. Dance Research Jour-
nal, 22, pp. 13-22.
Thomas H. (1993). Dance, Gender and Culture. London: Macmillan Press.
Winner E. and Hetland L. (2000). The arts in education: Evaluating the evidence for a
causal link. Journal of Aesthetic Education, 34, pp. 3-10.
A circle of life: The Caroline Plummer

Fellowship in Community Dance project
Barbara Snook
Dance Studies, National Institute of Creative Arts and Industries,

As the recipient of the Caroline Plummer Fellowship in Community

Dance at the University of Otago in 2008, I coordinated a series of pro-
jects over a six-month tenure. The fellowshipwhich is in honor of Caro-
line Plummer, an extraordinary young dance student at Otago who died
of cancer at the age of 24offers the opportunity to research, examine,
and explore the relationship between community and dance. My project
was broken into five distinct sections. The first was a dance created by
adult members of the cancer community; the second was art work cre-
ated by visual artists responding to the adult dance sessions; the third
was work facilitated with Canteen, an organization that supports young
people living with cancer. I then wrote a childrens book, Come Dance
with Me, that deals with death and the healing power of dance. The fifth
section involved working with senior high school dance students, who
prepared solo choreographies around the theme of cancer.
Keywords: cancer; dance; community; therapy; performance
Barbara Snook, Dance Studies, National Institute of Creative Arts and Industries, Uni-
versity of Auckland, Building 421, Level 5, 26 Symonds Street, Auckland 1010, New
Zealand; Email: b.snook@auckland.ac.nz
Choosing the unstable: Dancing through

the mid-career
Jo Pollitt and Dawn Bennett
Faculty of Humanities, Curtin University, Australia
While professional instability is a lifelong challenge for dance artists, it is

in mid-career that instability becomes particularly problematic and the
issue of sustainability comes to the fore. Mid-career artists build in-
creasingly diverse protean careers in a bid to generate stability while
dealing with unpredictable work patterns, aging bodies, and increased fi-
nancial, family, and future responsibilities. In addition to this is the sense
of physical and emotional loss experienced when the fundamental need
to dance is not being fully met. In this study, we underscore these big
picture issues with a more personal dialogue with three Australian mid-
career dance artists. Through these case studies, we examine the constant
negotiation, reassessment, and instability inherent in dance careers. Our
research reveals that putting the body down often entails putting down
work, status, and identity.
Keywords: dance; unstable; protean; mid-career; sustainability
Since the early 1990s, work within contemporary dance in Australia has been
undertaken almost entirely by independent dance artists. Existing company-
based employment usually takes the form of contract-based employment
interspersed with other work. As we have argued previously (Pollitt 2001,
Bennett 2009), what was once a political choice in response to previous tra-
ditions or loyalties is now the mainstream in contemporary dance. As a result,
the term independent artist today encapsulates the majority of contempo-
rary dance artists, who typically complete tertiary dance training before en-
tering a competitive market where they either secure one of few contract
positions or, most commonly, join the industry as an independent artist. As
such, independent status often occurs immediately upon graduation and is a
default position: less a choice than a necessity.
The trend toward independent practice is reflected in an increasing num-

ber of Western settings and has resulted in career development centers such
as the Australian SCOPE for Artists program and the Canadian Dancer Tran-
sition Resource Centre, both of which recognize that transitions between
physical and other dance practice occur throughout dancers careers. Being
an independent dance artist in Australia encapsulates a portfolio of roles,
including the responsibility for career development, which is managed solely
by the artist.
Such relentless career navigation and management is physically and
emotionally exhausting. In perpetual transition, the dance artist is further
challenged in mid-career by increased social and familial responsibilities and
shifting priorities. It appears to be at this point that artists begin to question
the long-term sustainability of this career model. Indeed, the very existence of
a career model, with its inference of some kind of logical progression, is
brought into question. The constancy of instability is an ongoing concern for
dance artists at all levels and ages, as expressed by Joysanne Sidimus of the
Dancer Transition Resource Centre in Canada: it is perfectly possible to be
at the top of your profession and not be earning a living wage (Levine 2005,
p. 33). With dance now a formal university entrance course within several
Australian states (most recently in Western Australia), and a corresponding
increase in the numbers of dance majors, it is timely to examine the realities
of sustaining a dance career through the mid-career period.
This paper acknowledges and locates some of the ongoing pressures, po-
tential pitfalls, and personal costs of professional instability in a protean
dance career. The paper aims to provide an intimate snapshot of independent
dance artists mid-career lives. It seeks to bring attention to the particular
shift felt by independent dance artists in their thirties and to make a case for
continued support throughout this period of heightened instability.
METHOD
Participants
The study reported here focused on mid-career dance artists, all with between
10 and 16 years experience in the sector, and asked the questions: what are
the impact and direct affects of ongoing instability on the professional and
personal identity status of mid-career dance artists? What are the available
support mechanisms and personal strategies of artists in navigating this cru-
cial career phase? What does a real-life mid-career in dance look like? All of
the survey and interview participants were Australian-based dance artists
with an active dance practice. The first phase of research involved a survey of
71 dance artists from which the responses of nine mid-career dance artists (7
F, 2 M) were extracted. This was followed by in-depth case studies of three
dance artists in their mid-thirties: Maya and Steven, who are both identified
with pseudonyms, and Jo, the third participant and principal investigator. It
is the results of these case studies that form the basis of this paper. Jos inclu-
sion positioned her as both investigator and participant, giving her the op-
portunity to contribute in terms of both her lived experience as a dance
artist and from an academic standpoint. This methodology proved to be
highly effective in eliciting deep responses.
Procedure
Respondents were identified from within professional networks, and pur-

poseful sampling was employed to locate successive informants likely to give
a wealth of information (Patton 1990). The survey collected information re-
lating to demographics, work, and career aspirations and was presented in
print and electronic format. Each of the three case study interviews began by
building a short profile of the participant including qualifications, profes-
sional background, and experience. The interview encompassed instability,
professional and personal identity, status, support mechanisms, working life,
economic circumstances, and career lifespan. The interviews were recorded
and transcribed, then independently coded using Glasers constant compara-
tive method of analysis (Flick 2002, p. 231). Seeking new responses to the
topic, aspects of grounded theory were adopted to develop exploratory inter-
pretations of data that would focus future data collection.
RESULTS
Results of the study can be summarized in three key areas. First, the study
highlighted the need for professional and economic support to be continued
through mid-career. This is contrary to the traditional model of supporting
emerging artists on the basis that initial support will lead to a level of stabil-
ity. Second, the three case studies brought to the fore the complexities of
family responsibilities and between male and female parenting and artist
roles. Finally, the study shed some light on the quest for long-stability and the
impact of this on creative dance practice.
DISCUSSION
Dance is inherently unstable, and the instability comes with the dance career
rather than being chosen: its not necessarily that I would like to be in a high
risk business, it just happens to be [that] what I have chosen is most high
riskIm an artist (Maya). All three case studies are dance artists who juggle
a multitude of roles, often back-to-back or simultaneously. All accepted that
the protean nature of working as a dance artist is a reality from the outset.
The impacts of instability were seen as both positive and negative. Nega-
tive impacts relate to the difficult economic and ecological landscape of
working in dance. Card (2006, p. 13) offers that dance artists in the small to
medium sector are akin to a bunch of responsive scavengers who function in
a state of perpetual crisis though a life of irregular support and erratic em-
ployment histories. However, positive impacts of instability include the de-
velopment of a strong sense of self preservation and adaptability, alongside
the freedom to be creative, unorthodox, or innovative.
The instability of dance is a given, but in many ways it is manageable. The
difficulty seems to lie more in the disconnect between the instability of dance
and the stability of everyday life; most participants reported to being person-
ally and emotionally stable, in contrast to the unknown components under-
pinning the instability of their professional lives: working in arts in Australia
youre often able to work if other people allow you to work, rather than just
continuing with your work. So thats unstable, because its not up to you
(Maya). Multiple participants identified a need to justify their dance practice
to friends and family who have difficulty understanding, accepting, and
valuing irregular, often erratic work patterns. Such irregular project-based
work was noted to be problematic in terms of accessing community resources
such as childcare, which demands commitment to a regular schedule well in
advance.
Acceptance of and rebellion against instability appears to be a gradual
shift in attitude and priorities rather than a sudden decision point or age. For
two of the case studies, parenthood was also a factor in the reprioritizing. Yet
dance is not something that can be put down and picked up, either physically
or emotionally. Jo reported that project-based practice results in a deep sense
of loss during down times when she is financially unable to justify continu-
ing work without funds. The reality that physicality needs to be available on
demand, regardless of other priorities and activities, was for Jo becoming
both difficult and impractical. For both Jo and Steven the pursuit of stability
was coupled with an overt grief at the potential loss of the physical dance
practice. For Steven, this occurred when he made the decision to focus on
business interests to work smarter, not harder. Stevens failed attempt to
create a stable income in support of an integrated work/art/life philosophy,
when theory and practice didnt add up, resulted in a loss of physicality that
denied him opportunities to pursue his physical dance practice. His desire to
teach also diminished as it was his artistic practice that gave him the juice
to teach: a duel loss of both practice and other work potential related to prac-
tice. For Jo, the grief was a recurring state of pre-heartbreak during every
project, knowing an end is imminent.
Reflecting on his life in dance, Steven considered himself to have ca-
reered rather than had a career, implying a lack of control in his careering
through both dance and jobs outside of the arts. In identifying his motivation
for working, he was clear that I work for my family now, not for myself.
Because of time constraints and family responsibilities, Jo stressed the need
to generate income directly from artistic projects. For Maya, too, it was finan-
cially necessary to maintain a reliable income source. She separated her paid
work from her artistic practice, though she still expected to generate her
future income from her arts practice. This is in line with our previous re-
search, which found that the income source of artists may not align with their
professional identity.
In summary, this paper has provided insights in three key areas. First, the
real life snapshot of a dance artist in mid-career is complicated and unique to
each individual: Maya remained hopeful and reconciled, the same as al-
ways; Jo was building stable frames around her to house the continuation of
an ongoing practice in increasingly unstable circumstances, yet she did not
envisage a departure from dance; and Steven continued to research job op-
tions and was prepared to let go of dance in order to be the provider,
though he was still deeply engaged in an ongoing inquiry of his artistic prac-
tice. All but one of the participants expressed a deeply felt need to stay con-
nected to their own physical/emotional constitution, and grief when this need
was not met. Secondly, the case studies, with their different stages and rela-
tionships with dance, all acted as reflexive balancers of the unstable. Possibly
in direct response to the inherent instability in the dance environment, they
actively sought to cultivate sustainability in their work, family, and lives, and
all desired a home base in which to locate and situate the chaos (Jo). The
role of parenthood here arose as a major factor in slowing the artistic no-
madism. Finally, the results reiterate the vital need for support once dance
artists have moved beyond the emerging phase. Ironically, though mid-ca-
reer is contemporaneous with the period at which many dance artists begin to
enjoy an established practice and senior artist status, the factors of mid-
career deeply challenge the equilibrium. The need for support is evidenced by
the 107% increase in applications for the Scope for Artists program, with over
half of participants currently in their thirties. Further funding schemes, such
as the new mid-career dance fellowship in Western Australia will allow artists
to benefit from the available career development assistance and continue to
develop their practice. However, much more needs to be done in order to

cultivate an environment in which dance careers can be sustained in the
longer term, and where experienced Australian dance artists can be empow-
ered to challenge and grow the art form.
Acknowledgments
We would like to acknowledge the time so generously given to this project by the dance
artists who shared their experiences and expertise.
Jo Pollitt, 28 Aughton Street, Bayswater, Western Australia 6053, Australia; Email:

joannapollitt@hotmail.com
References
Bennett D. (2008). Understanding the Classical Music Profession. Aldershot, UK: Ash-
gate.
Bennett D. (2009). Dance careers: Beyond performance to the real world of work. Jour-
nal of Dance Education, 9, pp. 27-34.
Card. A. (2006). Body for Hire? The State of Dance in Australia (Platform papers No.
8). New South Wales, Australia: Currency House Inc.
Flick E. (2002). An Introduction to Qualitative Research. London: Sage Publications.
Levine M. (2005). Beyond Performance. The Hague, The Netherlands: International
Organization for the Transition of Professional Dancers (available from www.iotpd.
org/beyond_performance.pdf).
Patton M. Q. (1990). Qualitative Evaluation and Research Methods. Newbury Park,
California, USA: Sage Publications.
Pollitt J. (2001). Accumulated Experience in Live Improvised Dance Performance.
Unpublished masters thesis, Edith Cowan University.
Thematic session:
Physicality of performance
Physical movement and imagery in

professional and undergraduate student solo
marimba practice
Mary Broughton and Catherine Stevens
MARCS Auditory Laboratories, University of Western Sydney, Australia
Cognitive strategies and imagery are important features of expert musical

practice. Movement imagery is important for developing note-accurate
marimba performance. We report findings of a study where four profes-
sional marimba players and four undergraduate student marimba play-
ers completed questionnaires regarding their practice strategies and use
of imagery in performance preparation. Since pedagogical marimba
practice involves developing movement imagery, each participant also
completed the Movement Imagery Questionnaire (MIQ-R) to measure
subjective kinesthetic and visual imagery ability. As expected, profes-
sional musicians qualitative reports of practice were driven by cognitive
strategies. The student musicians reported more regular practice habits
focusing primarily on physical practice. Professional musicians reported
use of imagery in practice more often than student musicians. As pre-
dicted professional musicians kinesthetic imagery was significantly
greater than student musicians. Contrary to prediction no significant
difference was observed between professional and student musicians
visual imagery. Results offer some support for a pedagogical approach to
marimba playing specifically aimed at developing note-accurate per-
formance through kinesthetic awareness and imagery.
Keywords: practice; music imagery; movement imagery; marimba per-

formance; musical expertise
Cognitive strategies in musical practice involve mental rehearsal of skills,

music analysis, and metacognition to direct efficient practice (Barry and
Hallam 2002). Whereas novices tend to focus on technical physicality, such
as playing the correct pitches in the right rhythm (Hallam 2001), expert mu-
sicians demonstrate well-integrated cognitive and motoric strategies in pre-
paring for performance (Chaffin and Logan 2006). An important early facet
of an experts approach to preparing a musical work for performance is to
form an artistic image of the piece (Chaffin et al. 2003). This includes imag-
ined musical character and emotions to be realized as sound through move-
ment. Recent empirical evidence suggests notional audiationimagining or
hearing sounds purely from reading musical notation aloneto be a process
involving kinesthetic vocal and motor processes in expert musicians (Brodsky
et al. 2008). Hence, musical imagery and movement appear to be inextricably
linked in expert music practice. While cognitive strategies and musical im-
agery would be expected features in professional marimba players practice,
movement imagery is also vital in the Western classical art form.
Movement imagery is the mental rehearsal of visual and kinesthetic
properties of movements to enhance learning and performance of motor
skills (Gregg et al. 2007, p. 1). According to Guillot et al. (2004), visual im-
agery is self-visualization of movement whereas kinesthetic imagery requires
one to feel the movement, i.e. to perceive muscle contractions mentally (p.
190). Arvinen-Barrow et al. (2007) found elite athletes cognitive-specific
(sports skills) imagery ability was significantly higher than novices. Solo ma-
rimba performance involves overt and spatial movement where the performer
is not in tactile, or often direct visual, contact with their instrument; move-
ment imagery is anecdotally reported to be important in developing expertise.
An entire pedagogical approach to marimba playing, Ideo Kinetics (Stout
2001), aims to develop accurate movement patterns through kinesthetic
awareness and imagery. Therefore, professional marimba players might be
expected to possess greater movement imagery than student marimba play-
ers.
The aims of this study are twofold. Firstly, we offer preliminary docu-
mentation of professional and student solo marimba players practice strate-
gies, both physical and imagined. Qualitative data are expected to
demonstrate that where professional marimba players focus on cognitive
strategies and imagery, undergraduate student marimba players practice
focuses on physical or technical aspects. Secondly, since marimba playing is a
complex motor skill that develops over time, we investigate kinesthetic and
visual imagery ability of participants. It is predicted that professional ma-
rimba players self-report greater kinesthetic and visual imagery than under-
graduate student marimba players.
METHOD
Participants
Participants included staff and students of the Zeltsman Marimba Festival

2004: two female and two male internationally-renowned solo marimbists
(mean age=47 years, SD=6.18; mean years playing marimba=38.0, SD=7.66);
two female and two male undergraduate-student marimba players (mean
age=20 years, SD=1.71; mean years playing marimba=8.5, SD=1.73).
A questionnaire (Q1) gathered demographic and musical background infor-

mation; open-ended questions requested explanation of practice strategies
and use of kinesthetic, visual, auditory, and emotional/musical character
imagery. The Movement Imagery Questionnaire-Revised (MIQ-R) (Hall and
Martin 1997) assessed participants subjective movement imagery ability on
two scales: (1) kinesthetic, and (2) visual. An eight-item questionnaire, the
MIQ-R requires participants to perform one of four movement tasks and then
rate the ease with which they can kinesthetically or visually imagine per-
forming the task. The scales range from 1 (very hard to see/feel) to 7 (very
easy to see/feel). Scores for items on each scale are summed separately, giv-
ing participants a kinesthetic and a visual imagery score. A score below 16
(out of a possible 28) on either scale indicates low imagery ability on that
scale, 16-20 moderate imagery ability, and above 20 high imagery ability.
Hall and Martin (1997) reported satisfactory internal consistency for the kin-
esthetic (=0.88) and visual (=0.89) subscales of the MIQ-R. Participants
were provided with an information sheet and, following written consent being
gained, were given Q1 to complete and return. At a convenient time, partici-
pants completed the MIQ-R individually in a quiet room.
RESULTS
As expected, professional musicians practice focused on cognitive strategies:

studying the music away from the instrument (G), practice see(ing) the
notes (JG). An initial goal for a new work was to develop musical interpreta-
tion: I try to figure out what I could say through a piece from the first mo-
ment of studying it (N), Try to engage your imagination from the
beginning (M). Time constraint and demands of the repertoire affected
practice: Practicing schedule is often dependent upon mywork schedule;
often when I must do it (because of an upcoming performance, etc.)I try to
learn a piece in as little time as possible (G), To save time of playing the
whole piece through, I keep a log of the most challenging sections and focus
on those as separate exercises (N).
The student musicians reported regular practice habits (mean hours per
day=3.25, SD=0.96; mean days per week=6.00, SD=0.96). As expected, their
practice focused on physical and technical elements: Warm-ups, technique,
repertoire, mess around (D), Learn the notes, work with a metronome,
practice fast changes (J). Strategies for learning a new piece involved break-
ing it into manageable chunks: Separate the piece into chunks, and learn
each chunk until I finish the piece (S), Divide the number of measures by
the number of days (e.g. until performance); work the sections each day
until the piece is done (D). Attention was also directed toward locating and
working on the most difficult sections: I read through the entire piece at first
to get a feel for itfind the difficult spots and try to work on them first (S),
Practice black sections slowly (JS). Musical interpretation was briefly
mentioned: After I have learned all the notes and listened to a recording(s)
of the piece, I put character to it (S).
Professional musicians self-reported imagery as part of their usual prac-
tice: Kinesthetic (100%), e.g. You have to remember the movementsto play
smoothly and accurately (JG), Slow, accurate practicehelps to develop
kinesthetic memory of the notes (G); Visual (100%), e.g. You cant hit what
you dont see (JG), Develop a visual image of the printed page (G); Audi-
tory (100%), e.g. You cant shape musically what you cant hear (JG); Emo-
tional/character (100%), e.g. Projectmy relationship with the piece (JG).
The student musicians self-reported imagery less often: Kinesthetic
(50%), e.g. To train muscles for different phrasing or difficult pas-
sagesdifferent intervals using kinesthetic imagery (J); Visual (75%), e.g.
The instrument, the audience, the performance settingvisual imagery of
the keyboard (J); Auditory (50%), e.g. sometimes I try to sound like a singer
or violin; recently, I started to just sit somewhere and play my tune in my
head (J), Decide on the auditory response I want the instrument to have
(JS); Emotional/character (75%), e.g. It helps to make a story or a scene out
of the music; depending on what emotions I have while I practice, I use
them and turn them into emotions that will hopefully bring out the musicality
of the piece (S), Imagery of particular experiences Ive had which inform my
approach (J).
Mann-Whitney U tests were conducted on the kinesthetic and visual im-
agery scales separately, with musical expertise as the grouping variable. As
predicted, the professional marimba players kinesthetic imagery (Md=26.6,
n=4) was significantly greater than the undergraduate student (Md=20.5,
30
25
20
Imagery score
15 Pr ofession a l
Stu den t
10
0
Kin a est h et ic im a g er y V isu a l im a g er y
Figure 1. Median imagery scores for the professional and student marimba players on
the kinesthetic and visual imagery subscales of the MIQ-R.
n=4) marimba players (U=0.00, z=-2.34, p<0.05). Contrary to prediction,

there was no significant difference between professional (Md=25.5, n=4) and
student (Md=24.0, n=4) musicians visual imagery (U =5.0, z=-0.89,
p=0.37). Figure 1 shows the median scores for the professional and student
musicians on both the kinesthetic and visual imagery subscales of the MIQ-R.
DISCUSSION
As expected, professional marimba players practice strategies focused on

cognitive issues and imagery (Brodsky et al. 2008, Barry and Hallam 2002,
Chaffin et al. 2003, Chaffin and Logan 2006). While these habits were some-
what evident in these advanced student musicians practice, their focus was
primarily on the physical aspect (Barry and Hallam 2002, Hallam 2001). As
predicted, professional marimba players reported significantly greater kin-
esthetic imagery than the undergraduate student marimba players (Arvinen-
Barrow et al. 2007). These results conservatively support the Ideo Kinetic,
kinesthetic-imagery-based, pedagogical approach to marimba playing (Stout
2001). No significant difference was observed between professional and stu-
dent marimba players visual imagery, contradicting the prediction. This may
be because the spatial nature of movement required to play the marimba and
hit specific targets explicitly incorporates the visual sense early in training. A
more expansive study is necessary to validate these results. Future work will
gather additional data to verify and broaden the scope of these findings, and
identify links between performance practice and audience perceptual re-
sponses to recorded performances.
Acknowledgments
A University of Western Sydney postgraduate award to the first author supported this
research. Thanks are extended to the faculty and students of the Zeltsman Marimba
Festival 2004 who participated in this study and, in particular, to Nancy Zeltsman for
facilitating the research.
Mary Broughton, MARCS Auditory Laboratories, University of Western Sydney, Locked

Bag 1797, South Penrith DC, New South Wales 1797, Australia; Email: m.broughton@
uws.edu.au
References
Arvinen-Barrow M., Weigand D. A., Thomas S. et al. (2007). Elite and novice athletes'
imagery use in open and closed sports. Journal of Applied Sport Psychology, 19,
pp. 93-104.
Barry N. H. and Hallam S. (2002). Practice. In R. Parncutt and G. E. McPherson (eds.),
The Science and Psychology of Music Performance. Oxford: Oxford University
Press.
Brodsky W., Kessler Y., Rubinstein B. et al. (2008). The mental representation of music
notation: Notational audiation. Journal of Experimental Psychology: Human Per-
ception and Performance, 34, pp. 427-445.
Chaffin R., Imreh G., Lemieux A. F., and Chen C. (2003). Seeing the big picture: Piano
practice as expert problem solving. Music Perception, 20, pp. 465-490.
Chaffin R. and Logan T. (2006). Practicing perfection: How concert soloists prepare for
performance. Advances in Cognitive Psychology, 2, pp. 113-130.
Gregg M., Hall C., and Butler A. (2007). The MIQ-RS: A suitable option for examining
movement imagery ability. Evidence-based Complementary and Alternative Medi-
cine, 10.1093/ecam/nem170, pp. 1-9.
Guillot A., Collet C., and Dittmar A. (2004). Relationship between visual and kinesthetic
imagery, field dependence-independence, and complex motor skills. Journal of
Psychophysiology, 18, pp. 190-198.
Hall C. R. and Martin K. A. (1997). Measuring movement imagery abilities: A revision of
the Movement Imagery Questionnaire. Journal of Mental Imagery, 21, pp. 143-154.
Hallam S. (2001). The development of expertise in young musicians: Strategy use,
knowledge acquisition and individual diversity. Music Education Research, 3, pp.
7-23.
Stout G. (2001). Ideo Kinetics. Ithaca, New York, USA: G and C Music.
Monitoring work and rest during performance

and life-style: A study case with a principal
ballet dancer
Lus Xarez, Fernando Pereira, and Gonalo Mendona
Faculty of Human Kinetics, Technical University of Lisbon, Portugal
Traditionally, dance training is based on day-to-day technical routines:

the class. Studies in dance physiology started in the 1980s have shown
that there are significant differences between the effort made during
performance and that made in classes. The challenge for researchers and
for those in the field (dancers, teachers, choreographers) is to find out
training methodologies that allow improved performance, reducing the
high number of injuries and prolonging careers. In this paper, we present
a research methodology that can help to improve dance training by
changing the starting point: past or future. Planning based on the past is
to continue to repeat the same routines as always has been. Planning
based on the future requires an analysis of the future task (dancing a
particular role)technical, energy, informational, psychological, and ar-
tisticso that the preparation takes into account these different de-
mands; what is known today as nonlinear periodization. This training
must be specific and individualized.
Keywords: energy expenditure; work-to-rest ratio; dance training; elite

ballet dancer; performance science
The prevalence of injuries in dance, mainly for professional dancers, is ex-

tremely high. Literature suggests that 80% of dancers get injured during their
carrier (Hamilton 1989). The most frequent type of injury is normally named
as overuse, a minor injury that gets worst by overload; micro traumatisms
caused by the exhaustive repetition of the same technical exercises in order to
achieve artistic perfection. This kind of injury, according to Garricks (2008)
studies, varies between 60-90% of total injuries, and the remaining are con-
sidered acute injuries that result from macro traumatisms (sprain, colli-
sion/impact, overstretch, etc.).
There are many factors that contribute to the occurrence of injuries: fa-
tigue, excessive exercise/overtraining, inappropriate warm-up and cool-
down, continuing to work when injured, deficient physical condition, hypo- or
hypermobility, postural problems, alignment deficits, inadequate en-dehors,
poor nutrition, disordered eating, menstrual irregularity, osteoporosis, psy-
chological factors (sleeping problems, anxiety, depression), and factors re-
lated to equipment use (point shoes, floor type, wardrobe).
It seems to be consensual that injury happens when some of these factors
occur together, with it being impossible to consider the injury occurrence the
result of just one factor. However the training factors have a great contribu-
tion: 90% of the injuries happen when the dancer is fatigued (Liederbach
1997). Used to training over and over, it is common for dancers to hide the
injuries and to keep working. Hamilton (1998) was responsible for an inquiry
that collected the following results: 66% of dancers had continued to work
with a chronic injury, 52% with tendinitis, and 27% with stress fracture and
with arthritis.
One of the central problems of training consists of the specific choice of
the balance between work and rest. To find the adequate amount of load and
the adequate duration of recovery is important to improve performance and
to prevent injuries. An insufficient load does not improve performance. On
the other hand, loads too high or with a lack of time to recover are negative
for health and performance maintains or decreases.
The aim of the study is to quantify the energy expenditure of high level
performance in dance in order to characterize the dance task requirements
and energetic costs in different time periods, and also to understand the bal-
ance between load and recovery.
METHOD
Participants
A 31 year old principal ballet male dancer was studied for eight days. For this
study, the subject wore a multiple physiological sensor (SenseWear Pro Arm-
band) during a week of performing Copplia.
Materials
The arm band provided the following measures: total energy expenditure and
active energy expenditure (at 3.0 METS or higher), physical activity duration,
intensity (sedentary, moderate, vigorous, and very vigorous), and sleep dura-
tion. All the stage performances, individual and company rehearsals, and
classes were filmed and submitted to a specific observation system. The

movement analysis is centered on the kind of motor actions, and other vari-
ables, to calculate the movement density and the work-to-rest ratio.
Procedure
The subject wore the SenseWear Pro Armband on the right arm, and energy
expenditure was estimated using proprietary equations developed by the
manufacturer. The measurement periods were continuous for several days
(until 14 days). Time ranges were selected according to the desired analyses
work periods: week and day, including class, rehearsal, and performance. The
collected data were analyzed in research software. The classes, rehearsals,
and performances were filmed and recorded in digital base for posterior
analysis with the observational system.
Estimated energy expenditure from the SenseWear Pro Armband is
broadly validated in the literature, namely with standard indirect calorimetry.
The sensewear data should be accompanied with information obtained
through the completion of a simple diary. The different data were integrated
over time. The equipment proved to be reliable and easy to use for all levels of
dancers, and even during performance. A brief instruction period to the sub-
jects is enough to achieve reliable results.
For the observation of the motor actions, the system of observation of
motor behavior in dance created and developed by Rodrigues (1992, 1999)
was utilized in this study. Through the systematic observation of motor be-
havior in dance we have identified nine motor actions: steps, jumps, turns,
gestures, falls, displacements, postures, balances, and contacts. The list of
motor actions proposed is limited and their definition based on two criteria:
the support (link to the ground) and the group of specific features of each
action. The observational system was validated in 1999.
RESULTS
Energy expenditure
On average the total energy expenditure was of 2933 Kcal, oscillating between
2300 on days-off and 3500 on performance days. During the eight days ob-
served, the dancer had three performances of 2 hours and 20 mins (including
intervals) and lost on average 600 Kcal in each one. For 50 mins he was in
sedentary activity and about 90 mins in physical activity. The physical activity
was distributed in 74 mins of moderate (3.0-6.0 METS), 14 mins of vigorous
(6.0-9.0 METS), and 3 mins in very vigorous (9.0 METS and higher). The
sleep period was on average 8 hours.
The analysis of movement allows us to identify the relation between en-
ergy expenditure and density of movement: there is an increase of energy
expenditure when the number of actions per min. also increases and mainly
when those actions are moveable (jumps, displacements, and steps). How-
ever, some differences in the content are found between the classes (individ-
ual or collective) and the specificities of performance because classes are
more like a routine without a direct relationship with the different require-
ments of choreographies.
Another characteristic that is important is the difference between what
the subject does in his individual trainings before the performance and what
he does during the company class. The expenditure (about 400 Kcal) is higher
during his individual trainings of 1 hour and 15 mins (on average 4.0 METS)
than during group classes (270 Kcal, 3.0 METS) with the same time. This last
activity presents a sedentary and moderate activity, while the first situation
has periods of vigorous activity like in the performance.
Work-to-rest ratio
One of the most interesting observations in this study concerns the compari-
son of the collective training with individual training. Typically, on perform-
ance day the company (with about 70 dancers) usually does a class similar to
the other days. But on performance days our subject (principal dancer) holds
his individual preparation/training just immediately before performance,
instead of the collective class that occurs two hours before performance. What
are the differences between these two classes?
In terms of content the differences are not much (at the end of the indi-
vidual training there is some specific work/preparation with passages of the
choreography, but only in the last 5 mins), but differences arise at the level of
work-to-rest ratio: for a similar duration of 73 and 71 mins, in the collective
class there was a total of 28-45 mins from work to rest, while in the individual
training 32-39 mins of work to rest. The average time of the barre exercises is,
respectively, 90 and 70 s, with similar values (33 s, 32 s) of rest.
In the centre the differences increase, with values of work of 28 s and 30
s, but with periods of rest significantly different from 180 s and 45 s, respec-
tively. The organization of the exercises in small groups can explain this dif-
ference; in a class where there are on average 40 dancers, the rest time of
each dancer will at least triple. The problem is that the physiological effects
are quite differente.g. if one waits 3 mins to work about 30 s, or if one just
waits 45 s to do the same 30 s. The more experienced dancers are aware of

this, and when they work individually they work more (39 versus 28 mins)
and better.
DISCUSSION
The results shows that physical activity is mainly moderate (91%): on average,
the eight days observed presented 3 daily hours of moderate physical activity,
about 15 mins of vigorous, and 2 mins of very vigorous activity. The complex-
ity of the tasks, which demands high coordination, may justify these results.
Another aspect that we must enhance is the importance of individual
training for success. In the individual training periods, we found a work-to-
rest-ratio more similar to performance than in the group class company.
The underlying question that arises in this study (in our line of research)
is how to assist the performers to reach the peak level of performance at the
right moment? Planning and monitoring are key words to achieve these ob-
jectives. If we continue to train only based on tradition, with a routine daily
lesson for all the elements of the company, probably the same high rate of
injuries will continue and the individuals potential is not developed: at the
highest professional level, a lesson for all is a lesson for anyone. We need to
individualize the training, planning well to find a balance between work peri-
ods and rest periods and to avoid overuse injuries.
Lus Xarez, Faculty of Human Kinetics, Technical University of Lisbon, Estrada da

Costa, Lisbon, 1495-688 Cruz Quebrada, Portugal; Email: lxarez@yahoo.com
References
Carvalheiro S. (1999). Observao em Dana: Validao de um Sistema de Observao

do Comportamento Motor. Unpublished masters thesis, Technical University of
Lisbon.
Cohen J. L., Segal K. R., Witriol I., and MacArdle W. D. (1982). Cardiorespiratory re-
sponses to ballet exercise and the VO2max of elite ballet dancers. Medicine and Sci-
ence in Sports and Exercise, 14, pp. 212-217.
Fruin M. L. and Rankin J. W. (2004). Validity of a multi-sensor armband in estimating
rest and exercise energy expenditure. Medicine and Science in Sports and Exercise,
36, pp. 1063-1069.
Garrick J. (2008). Injury prevention in dance. Paper presented at the Twenty-sixth

Annual Symposium on Medical Problems of Musicians and Dancers, Aspen, Colo-
rado, USA.
Hamilton L. (2008). New York City Ballet. New York: St. Martinns Griffin.
Hamilton L. (1998). Advice for Dancers. San Francisco, California, USA: Jossey-Bass
Publishers.
Jakicic J., Marcus M., Gallagher K. et al. (2004). Evaluation of the Sensewear Pro Arm-
band to assess energy expenditure during exercise. Medicine and Science in Sports
and Exercise, 36, pp. 897-904.
Liederbach M. (1997). Screening for functional capacity in dancers: Designing stan-
dardized, dance-specific injury prevention screening tools. Journal of Dance Medi-
cine and Science, 1, pp. 93-106.
Pereira F. and Rodrigues L. X. (2007). Monitoring energy expenditure in dance, motor
actions, and lifestyle. Journal of Dance Medicine and Science, 11, pp. 1-15.
Rodrigues L. X. (1992). A Classificao dos Movimentos Observveis nas Diferentes
Formas de Dana. Lisbon: Technical University of Lisbon.
Rodrigues L. X. (1999). Morfologia do Movimento Danado: Gneros Coreogrficos e
Comportamento Motor na Dana Teatral Ocidental. Unpublished report, Techni-
cal University of Lisbon.
Schantz P. G. and Astrand P. O (1984). Physiological characteristics of classical ballet.
Medicine and Science in Sports and Exercise, 16, pp. 472-476.
Wyon M. (2004). Challenging habit: Planning and preparation, the art of periodisation
and optimizing performance. In M. Van Der Linden (ed.), Not Just Any Body and
Soul (pp. 66-71). Amsterdam: Uitgeverij International Theatre and Film Books.
Will supplemental fitness training have an

affect on the aesthetic components of
contemporary and classical ballet dance
performance?
Matthew Wyon1,2, Emily A. Twitchett1, Manuela Angioi1, George Metsios1,3,

and Yiannis Koutedakis1,3
1 Research Centre for Sport, Exercise, and Performance,

University of Wolverhampton, UK
2 Jerwood Centre for the Treatment and Prevention of Dance Injuries, UK
3 Department of Exercise Sciences, University of Thessaly, Trikala, Greece
Exercise scientific research in dance to date has mainly been observa-

tional in nature; there is a paucity of literature investigating the effect of
supplemental fitness training on dance performance. Many dancers have
cited fatigue as one of the most prominent causes of injury and a de-
crease in performance components. The aim of the current study was to
examine the effects of specifically tailored fitness training program on the
incidence of injury and the quality of performance of both classical ballet
and contemporary dancers, compared with relevant control groups. Par-
ticipants underwent a series of fitness tests and a proficiency in perform-
ance test at the outset and end of the study. The intervention groups
partook weekly training session that included aerobic interval training
and circuit training. Data regarding injury incidence were collected with
the assistance of the dancers physiotherapist. Results indicated no sig-
nificant difference in injury occurrence between the respective groups.
Both classical ballet and contemporary intervention groups significantly
increased their performance scores (p<0.05) compared with their
equivalent control group. It was concluded that supplementary fitness
training has a positive effect on the aesthetic components of dance per-
formance as studied herein; further research is recommended on a larger
and more varied sample.
Keywords: ballet; contemporary; performance; aesthetics; fitness

Matthew Wyon, School of Sport, Performing Arts, and Leisure, University of Wolver-
hampton, Gorway Road, Walsall, West Midlands WS1 3BD, UK; Email: m.wyon@
wlv.ac.uk
Thematic session:
Defining performance
Is composition a mode of performing?

Questioning musical meaning
Jorge Salgado Correia
Department of Communication and Art, Aveiro University, Portugal
The formalist approaches, focusing mainly on scoreswhich are just sets

of instructionshave missed a significant part of musics processes of
meaning production and its communicational dimension. This research
is an attempt both to acknowledge the role that personal meanings play
in the creation of a composition and of its performance and to verify if
composers and performers work from the same mode of knowledge when
making their respective meaning constructions. Based on gesture inter-
modality, a composer and performer created, separately, two video-clips
that express the personal meanings that have guided their respective
compositional or interpretative choices concerning two sections of a
piece for solo flute. The two pairs of video clips were analyzed, compared,
and conclusions were drawn. The results showed that, in spite of the dif-
ferent imagery used by composer and performer, the relations of tension
and relaxation, movement, and rest were analogous: it appears that the
personal meanings of the composer guided his work procedures to create
those relations and that the performer has created a personal emotional
narrative to give meaning back to those same relations. This study could
support future research on the application of gesture intermodality in
teaching strategies for musical performance.
Keywords: music; meaning; intermodality; composition; performance
The fact that performers cultivate a narrative attitude toward performance

and performance preparation is a crucial notion for the present research, as it
illuminates a range of questions on the difficult issues of musical meaning
production. Although referring to listeners, Swanwick (1999) argued that it is
our deep personal involvement with music that accounts for the high sense of
value that we accord to it:
Patterns or schemata of old experiences are activated but not as sepa-

rate entities. They are fused into new relationships. We can thus make an
imaginative leap from many old and disparate experiences into a single,
coherent new experience. It is this potentially revelatory nature of music
that accounts for the high sense of value frequently accorded to it (p. 21).
But these processes take place by the performers as well, who, consciously
or not, more or less trained and more or less (historically) informed, end up
projecting a personal interpretative style; and, we suspect, they take place
also by composers, who, again, consciously or not, more or less trained and
with more or less erudition, end up producing a score that is inevitably a pale
reflection of their creationthat is, of the performance that, intimately, they
were imagining when composing. We believe that personal meanings play a
central role in the creative process, unifying all the dimensions of the musical
work and determining decisively the actions and events of a musical perform-
ance, be it imagined or actually performed. These meanings guide the per-
formers intentionality in the unifying process of the musical creation and
embodiment.
However, it is the embodied nature of music that is perhaps most relevant
to musical understanding as it will be explored in this article. Based on the
theory of enactive cognition (cf. Gibbs 2006, Thompson 2001), Martinez
(2008) argued that structural metaphors might, to a certain extent, model
music experiences and demonstrated how listeners use the metaphorical
component of music cognition to experience the underlying structure of tonal
music. For professional performers, music is inseparable from a communica-
tive dramatic attitude, their account of the music they are about to play is
always (or, at least, hopefully) an aesthetically oriented narration of dra-
matic action (Maus 1997, p. 129). After having collected the available infor-
mationbe it anecdotal, historical, or analyticperformers create a kind of
synthesis, a personal amalgam that they apply to the music material. This
synthesis is, thus, much more an open, personal, subjective, symbolic amal-
gam, emotionally motivated to a large extent, than a reductive, closed, rigor-
ous, conceptual definition, deductively constructed. This synthesis has indeed
to be concentrated, like an embryo that develops adapting itself to the envi-
ronment, in order to pragmatically fulfill its function of (e)motionally ex-
ploring the musical material. Putting aside the issues concerning the specific
instrumental techniques of each musical genre or style, which are in fact just
means of expression, performers hermeneutic approach to scores, in other
words, their subjective, symbolically charged, embryonic amalgam has a
clear purpose: to create a communicative, convincing narrative of musical
gestures: performers tell a story and the concert begins with Once upon a
time (interview with Patrick Gallois, in Correia 2003). When applying
their subjective, symbolically charged, embryonic amalgams to explore
(e)motionally the musical pieces, performers are exploring figures, which are
constellations of embodied symbolic meanings, and invoking inevitably
analogies, metaphors, or, simply, memories. This implies that it is the realm
of tacit knowledge that is aimed in this (e)motional exploration, as if the fig-
ures would invoke the actual presence of the (e)motional memories and the
performers would re-enact them, recuperating their unconscious dimensions
to a large extent and making in this way their interpretations more authentic
and convincing. But, if composition is as a mode of performing, composers,
when composing, should be applying a subjective, symbolically charged,
embryonic amalgam as well, actively imagining a performance which would
guide their compositional decisions.
To find out more about this, and having in mind that one may express
through different sense modalities (aurally or visually) the same gestural dy-
namics and shaping, we designed the experiment we report here. It is based,
thus, on the application of the concept of intermodality (cf. Johnson 1987,
Hatten 1999), which will enable us to compare the performers and the com-
posers personal (embodied) meanings for the same piece. The simple verbal
explanation of these meanings would utterly miss the point: their embodi-
ment. Thus, we had to create the conditions under which these personal
meanings could be manifested while keeping their embodiment. In order to
do that, we asked both performer and composer to express in visual terms
(directing videos for two excerpts of the piece) the same subjective, symboli-
cally charged, embryonic amalgam they had created for the musical work.
They were asked to translate their personal (embodied) meaningswhat-
ever they intimately and introspectively have imagined for each musical ex-
cerptinto visible scenes, actions, and physical gestures; our main objective
being to understand if the role that embodied meanings play in composition
is similar to the role they play in performance. If they play a similar role, then
composition should be considered a mode of performing.
METHOD
Participants
Composer was Sara Carvalho, who teaches composition at the University of

Aveiro. The performer was Jorge Salgado Correia, who teaches flute at the
University of Aveiro. The actors were Joana Carvalho, Graciana Romeo, Juan
Capriotti, Miguel Correia, and the author.
Materials
Two excerpts with about 50 seconds duration taken from the work Solos III
for flute by Sara Carvalho (CD Nova Msica para Flauta 1999, ed. Numrica,
No. 1095) were chosen as the musical material upon which the video clips
were created. The first excerpt (M1) corresponds to the score measures from 1
to 16; the second excerpt (M2) from measure 123 to 137. For the video re-
cordings, an amateur digital camera was used (Samsung S760-7.2 mega pix-
els).
Procedure
Sara Carvalho wrote the script and directed and edited two of the video clips
expressing the view of the composer (Cv1, Cv2). She was asked to express in
this way the meanings that each of the excerpts had for her when composing
the piece Solos III. The performer was asked to do the same for the same two
excerpts, reporting himself on the time when he was preparing to perform
this piece (Pv1, Pv2). In all videos, the actors were asked to do exactly what
the directors required, neutralizing as much as possible any kind of interfer-
ence. The four videos are available on the internet: www.youtube.com, search
term jorgesalgadocorreia.
RESULTS
The videos of the composer (Cv1, Cv2) tell the story of a man who sees and
meets two women in two different scenarios; in Cv1, the woman is lying in the
sun on a beach, and in Cv2, the meeting happens in a coffee shop. In her
script the composer explained that, in her narrative for this piece, each sec-
tion corresponds to a meeting of the same man with different women. The
video of the performer for the first (Pv1) excerpt portrays the anxiety of a
woman looking for something but trying not to be seen by the boy who hap-
pened to be in the same room. In the second video of the performer (Pv2), the
camera catches a domestic discussion between a man and a woman.
DISCUSSION
Beginning with what the two pairs of videos have in common, the main thing
to highlight seems to be the profound sense of narrative that all of the videos
clearly display. Both composer and performer thought and worked in terms of
constructing a narrative, of telling a story. The main structural points of the
musicbeginning with the presentation of a conflict or an opposition (in M1
the opposition between the pizzicatos and the normal notes, in M2 the oppo-
sition or dialog between low and high registers); growing tension through
repetition (in M1 repetition and some degree of stretto, in M2 repetition and
ascendant melodic movement); and conclusion (in M1 resolution of the oppo-
sition, confluence, in M2 definitive separation of the opposites, divergence)
are clearly reproduced in the videos of both composer and performer. Of
course, they tell different stories, but it is remarkable how the important
structural features are present in both versions. For example, Cv1 and Pv1
both express in the end a sense of success, while Cv2 and Pv2 both express
the same feeling of irreparable separation; also the points of highest tension
in the videos are the same, corresponding to measure 8 in M1 and measure
135 in M2. It seems, thus, that embodied imagination assisted composer and
performer in mapping the same expressive process between different experi-
ential domains, and more importantly, this intermodal or transmodal opera-
tion was only possible because the structural features emerged from a system
of spatial-temporal organization elaborated upon patterns of image schemata
(cf. Johnson 1987).
It is this fundamental connection to a bodily ground that makes personal
meanings so decisive in musical meaning production and communication.
But there was a remarkable difference as well. The two videos of the com-
poser have both the characteristic of suggesting much more psychological
action than physical: the viewer is invited to imagine what the characters are
feeling and the musical gestures seem to portray much more these emotional
experiences than the physical actions of the actors. There are a few exceptions
where there is obvious coordination between the musical events and the visi-
ble actions in the videos of the composer: at the beginning and the end of Cv1,
measure 8 in Cv1 where some agitation in the music is expressed through the
movements of the actress, in Cv2 measure 134, and at the end where again
and respectively the actor and the actress seem to physically express, say,
more literally, the correspondent musical gesture. The physical actions in the
two videos of the performer are much more connected with the musical
events: the characters move with the music, so to speak, starting and stopping
in synchrony with the musical phrases, expressing with movement and physi-
cal gestures the relations of tension and relaxation of the music. Only in Pv2
there are two occasions (curiously, both have sustained multiphonics) where
the viewer has no physical action to interpret and is, thus, implicitly, invited
to imagine what emotions the characters are experiencing; the first is in
measure 128, where the man sits impotent and leans with his elbow on the
arm of the sofa, and the second is at the end, measure 137, where the man
walks devastated away from the woman and toward the window. To our un-
derstanding, these two different approaches result directly from the specific
conditioning of their respective professional activities: the composer is nor-

mally giving instructions and suggesting, while the performer is actually do-
ing or performing and, as we have argued, for performers, music is
inseparable from a communicative dramatic attitude. So, the videos of the
performer reflect this additional preoccupation of communicating as inten-
sively and as clearly as possible, exteriorizing in motion all the emotions. The
exploration of this attitude is very useful for performance students as it is the
application of gesture intermodality to develop intentionality and, conse-
quently, expressiveness and musicality.
Acknowledgments
Many thanks to Joana, Graciana, Juan, and Miguel.
Jorge Salgado Correia, Department of Communication and Art, Aveiro University,

Campus de Santiago, Aveiro 3810-193, Portugal; Email: jcorreia@ua.pt
References
Correia J. S. (2003). Investigating Musical Performance as Embodied Socio-Emotional

Meaning Construction: Finding an Effective Methodology for Interpretation. Un-
published doctoral thesis, University of Sheffield.
Gibbs R. (2006). Embodiment and Cognitive Science. Cambridge: Cambridge Univer-
sity Press.
Hatten R. (1999). Musical Gesture. Online lectures, Cyber Semiotic Institute, University
of Toronto (available from www.chass.utoronto.ca/epc/srb/cyber/hatout.html).
Johnson M. (1987). The Body in the Mind. Chicago: University of Chicago Press.
Martinez I-C. (2008). Enactive cognition and embodied mind: The imaginative and
metaphorical component of music listening. Estudios de Psicologa, 29, pp. 31-48.
Maus E. F. (1997). Music as drama. In J. Robinson (ed.), Music and Meaning (pp. 105-
130). Ithaca, New York, USA: Cornell University Press.
Swanwick K. (1999). Teaching Music Musically. London: Routledge.
Thompson E. (2001). Empathy and consciousness. Journal of Consciousness Studies, 8,
pp. 1-32.
Seeking excellence in
danced postgraduate degrees
Maggi Phillips
Western Australian Academy of Performing Arts, Edith Cowan University, Australia
Excellence as an evaluative category is implicit in the attainment of

doctorateness, so how can excellence be evaluated? The practices of
dance, now recognized as legitimate components of doctoral studies,
complicate evaluation processes while suggesting possible revisions as to
how these processes may be refined. The combination of tacit knowledge
and engagement, though contrary in logical senses, may provide a dy-
namic solution for evaluating studies that exceed a concept of excellence
constrained measurement.
Keywords: postgraduate dance studies; assessment; engagement; crite-

rion of excellence; conceptualization
Excellence as an evaluative category is implicit in the attainment of doctor-

ateness: as the highest level of degree conferred by most universities, the
PhD is commonly understood as indicating the height of formal educational
attainment, and has, until relatively recently, been above reproach and al-
teration (Brien 2009, p. 3). At the same time, excellence is neither a simple
attribute to define nor a concept of measurement without contestation. In-
deed, a crucial reason for embarking on the project at the core of this discus-
sion was a desire expressed by candidates, supervisors, and examiners of
doctoral degrees to gain guidance in what achievement in danced research at
the top of the academic scale might be.
Sparshott views the sole function of excellence is to signal the sum of
those qualities for which anything is praised and prized, concluding summa-
rily that excellence represents standardization and quality control and,
ultimately, compliance (Sparshott 1986, p. 137). Through clever semantic
arguments, Sparshott demolishes the validity of the term except when it
might apply to rival manifestations and persons who can be said to be doing
the same thing more or less well (Sparshott 1986, p. 138). Excellence meas-
urements may be able to be applied to virtuosity but not, he argues, to imagi-

nativeness or expressiveness, originality or beauty. His final comment,
however, is telling, introducing an ambivalence that is less characteristic of
his overall scepticism: excellence is what gives art a public identity and a
public place, without which art could not achieve whatever deeper signifi-
cance it might have (Sparshott 1986, p. 140). The idea of excellence, though
ultimately incalculable across the complexities of human behavior, thus has
its function.
Similar tensions pervade the evaluation of degrees involving artistic prac-
tices. However, whereas Sparshott and later aesthetics advocates like Stastny
(2002) cling to traditional contours of thought, the new generation of practi-
tioner-scholars abide in uncertainties that lie between excellence determined
by academic perspectives and that advocated within the professional arts
community, often emphasizing a confusion between what Melrose (2003,
2006) distinguishes as expert spectatorship and expert practitioner mastery.
Melrose highlights the divergent motivations of those who look back to re-
trieve knowledge and those who use their practical mastery (albeit skills in
movement, sound, and image developed in the past) to project forward into
the concept of discovery itself. Curiously, the two distinctive forms of mastery
resonate with similar past determinates on which excellence in the field must
be gauged.
Thus, evaluation of research acts springs from what has been learned
through time by individuals/examiners devoted to pursuits articulated in
words as much as in movement. In both, the sedimentation of thought ac-
quired through exposure to the form, an individuals tacit knowledge, plays a
vital role in measuring whether that same thing is delivered more or less
well. Judgment is, by nature, thereby enmeshed in the past even if, and this
point is significant, judgment is not confined solely to its dictates but should
also be pervious to novelty, surprise, and confrontation. Acts of doctoral ex-
amination might begin with Sparshotts observation about relative merits of
technical factors reasonably based on tacit knowledge, but they also have to
move into the less definable territory of imagination and originality, where
concepts other than excellence invariably pervade.
METHOD
Procedure and materials
This perspective on performing excellence is prompted by a recent collabo-

rative research project, Dancing between diversity and consistency: Refining
assessment in postgraduate studies in dance, conducted with funding by the
Australian Learning and Teaching Council in 2006-08. Researchers, Cheryl

Stock (Queensland University of Technology), Kim Vincs (Deakin University),
and myself, questioned why dance artists and/or scholars pursue academic
degrees alongside explorations into assessment, particularly in those pro-
grams involving artistic practices.
Extensive literature reviews into higher degree dance studies, postgradu-
ate examination/assessment issues, and peculiarities pertinent to artistic
practice components helped to structure the interview and focus group ques-
tions within the two principal disciplinary clusters of participants; the broad
professional dance community represented by Ausdance (the Australian
Dance Council) and the group of Australian universities who offered dance or
related postgraduate degrees.
Participants
Perceptions were drawn from 74 open-ended interviews with supervi-

sors/examiners, research deans and administrators, and candi-
dates/graduates as well as from forums organized through Ausdance
networks tapping into the professions views on postgraduate research. Eight-
een months into the project, responses to a draft code of practice document,
disseminated at the 2008 World Dance Alliance Global Summit in Brisbane,
also played a vital role in constructing the culminating guidelines and discus-
sion papers, as did interaction with Susan Melrose, the projects international
evaluator.
RESULTS
Legitimacy and/or engagement?
Responses indicated that the primary motivation for doctoral degrees is the
validation of dance knowledge, whether performative or choreographic,
pedagogical or therapeutic. Given the context, this emphasis on intellectuali-
zation was unsurprising and, yet, arguably slightly at odds with an aesthetic
form often propelled by the rebellious forces of its contemporary manifesta-
tions. Ironically, when faced with delineating attributes that characterize
doctorateness, the range of participants expressed a slightly divergent, though
not incompatible, view emphasizing the role of engagement or the wow
factor which ostensibly appears more consistent with traditional perceptions
of dancing than a reliance on its intellectualism.
Indeed, engagement emerges as a key attribute of evaluation when deal-
ing with ideas of excellence where the modewhether dance, writing, or
symbolic notationmatters less than a discernible brilliance or clarity of ar-

ticulation in the medium concerned. One candidates observation makes this
issue of articulation and its enmeshment in skill and form clear:
William Forsythes CD-ROMhis solo in thatits about eight minutes

long. I would call [the solo] a doctorate level of performance. Its amaz-
ingly sophisticatedtight, clear, it compresses years of dance into eight
minutes.
Encapsulated in this observation is an ability to recognize excellence al-

most without dispute. Similar resonances are perceptible in the following
selection of responses from supervisors/examiners on the characteristics of
doctorateness across the modalities of practice and word: reading involves
not a step [but] a runa dasha dance, candidates embrace the unknown
and use the unknown, they are agents and leaders of change within the dis-
cipline, their work exhibits a transformative imagination, is courageous,
thoughtful, embraces a depth of practical investigation/embodied sub-
stance, displays a mastery of craft plus inventiveness, and evokes more
than a personal capsule view; a sustained processual perspective.
Responses thus suggest that engagement is a strong indicator of excel-
lence within doctoral evaluations related, perhaps, to Csikszentmihalyis
(2001) proposition of flow: the condition that is necessary for an individual
entering the state of flow is concentration, an intense focus on what he or she
is doing (p. 8). Csikszentmihalyis analysis, honed to creative individuals and
their learning aptitudes and processes, may be equally applicable to examin-
ers and their willingness to learn within the evaluation process. Or rather, I
believe that the rigorous examiner actually embodies this notion of flow and
actively participates in the thesis journey through alignment of his/her tacit
knowledge and involvement in a deep, intrinsically motivated experi-
ence[wherein] they forget themselves (p. 7). Incorporating the two seem-
ingly oppositional intentions thus invites a dynamic concentration on, or
ideational dialogue with, the candidates research. The examiners tacit
knowledge constrains an anything goes or egocentric artistic foray but, at
the same time, should not impede concerted visions into what if questions.
Generating dynamism
Thus, there is a metaphorical give-and-take in the evaluation exercise which

involves an examiners tacit knowledge, the candidates capacity to acknowl-
edge that grounding, and the courage, on the part of both parties, to venture
into new or possibly refined territory. That said, in actuality, an examiners

immersion is more likely to be impeded by disjunctures in some aspects of
presentation and/or conceptualization within the doctoral study, since excel-
lence as noted by Sparshott is both ideal and relative. Nevertheless, aspira-
tions of excellence on the part of candidates and examiners do sustain the
disciplinary community, and it would appear that this is an important mes-
sage to convey, even if a state of flow or dynamic embodiment is not always
achieved.
How then to encourage examiners to engage in the excellence seeking
process of evaluation? Ultimately, guidelines have to be porous so as not to
delimit incursions into new knowledge at any point along the process to
product continuum. At the same time, examiners and candidates do need
guidance into expectations.
Two factors appeared crucial in the provisional recommendations that we
put forward; each candidates ability to articulate their research questions
and pivotal variables involved and each examiners willingness to engage
dynamically with the articulated situation given his/her tacit knowledge and
the unfixed environment of the unknown. The former indicates the impor-
tance of clarity, while the latter offers credibility of change. As such, our re-
port (Phillips et al. 2009) offers dynamical systems for examinations that
attempt to suggest interactions, rather than checklists, for variations that may
occur in theses. These diagrams intend to dialogically mesh academic and
professional perspectives on excellence that, though not easily achieved with
their distinct value sets, is desirable. Dynamical systems, moving and forever
challenging, should be able to embrace difference and singularity. Wherever
experimentation occurs, intellectualization and creative play should be evi-
dent. The combination may not always work, but that too is an issue of en-
gagement and adventure.
DISCUSSION
Excellence then is not a necessary measure of doctorateness, but the idea of

excellence operates like an aspiration, challenging candidate mettle and in-
viting examiner engagement. Our project points to the notion of dynamic
interactions between the examined and the examiner that have to be travelled
without certainty and without actual face-to-face communication. But the
dash and dance into knowledge does involve engagement of both parties, for
when engagement withers on either side, excellence is an idea without
meaning, even in aspiration. The project is ongoing.
Acknowledgments
I wish to acknowledge my colleagues, Cheryl Stock and Kim Vincs, the Australian
Learning and Teaching Council, Ausdance, and the participants, who willingly contrib-
uted their ideas about the difficult issues of examination and excellence.
Maggi Phillips, Western Australian Academy of Performing Arts, Edith Cowan Univer-
sity, 2 Bradford Street, Mt Lawley, Western Australia 6050, Australia; Email: maggi.
phillips@ecu.edu.au
References
Brien D. L. (2009) Unplanned educational obsolescence: Is the traditional PhD be-

coming obsolete? MC Journal, 12, pp. 1-14.
Csikszentmihaly M. (2001). Mihaly Csikszentmihaly. In R. Alston, K. Kain, D. Jowitt et
al. (eds.), Not Just Any Body (pp. 6-9). Ontario, Canada: The Ginger Press.
Melrose S. (2003). The curiosity of writing (or, who cares about performance mastery?).
Paper presented at Practice as Research in Performance, University of Bristol, UK
(available from www.sfmelrose.u-net.com/curiosityofwriting).
Melrose S. (2006). Not yet, and already no longer: Loitering with intent between the
expert practitioner at work, and the archive. Paper presented at Performance as
Knowledge, London, UK (available from www.rescen.net/archive/PaK_may06/
PaK06_transcripts4_1.html).
Phillips M., Stock C., and Vincs K. (2009) Dancing Between Diversity and Consistency.
Research report, Australian Learning and Teaching Council (available from www.
altc.edu.au/resource-dancing-between-diversity-consistency-ecu-2009).
Sparshott F. (1986) Excellence in the arts. Journal of Aesthetic Education, 20, pp. 137-
140.
Stastny K. (2002) In pursuit of excellence: A view from the field. Journal of Aesthetic
Education, 36, pp. 154-161.
Intention revisited: From composition to

performance
Helena Marinho and Sara Carvalho
Department of Communication and Art, University of Aveiro, Portugal
Within the context of Western art music performance, intention is a fun-

damental key concept, generally viewed as a representation of the com-
posers creative and interpretative intents. Even though intention is
considered a decisive factor in performance, the focus is placed on the
composer, and the performers own range of interpretative choices has
been neglected. This issue is especially pertinent in collaborative work
between composer and performer(s) in the interpretation of contempo-
rary repertoire, often informed by the intentional contribution of both
parties. This research focused on a case study, documenting the collabo-
ration between a composer and a performer, which involved the premire
of a new composition for pianoforte, and intended to demonstrate that:
(1) the concept of authorial intention, which is traditionally viewed by
performers as a mandatory interpretative framework, can also be ad-
dressed from the performers perspective and (2) a multi-level appraisal
of intention can be applied to collaborative procedures in order to pro-
mote a successful performance. The concepts of authorship and intention
provided a theoretical framework for the analysis of the collaboration
process, establishing a parallel between the dichotomies author/reader
and composer/performer, and the concept of intentional level was ap-
plied in the description of the collaboration stages.
Keywords: author; intention; composition; collaboration; performance
Within the context of Western art music performance, intention is a funda-

mental key concept, generally viewed as a representation of the composers
creative and interpretative intents. This concept has been the subject of de-
bate especially in the context of historically-informed performance, namely
by authors such as Dipert (1980), Kivy (1995), and Butt (2002), and has been
influenced and marked by similar issues raised in the field of literary studies.
Wimsatt and Beardsleys essay The Intentional Fallacy (1946), which criti-
cized the relevance of authorial intention as a criterion for critical evaluation,
defended the text, and not the author, as the focal point for interpretation.
This approach led to the announcement of the death of the author by
Barthes (1994), a position that was subsequently reformulated by Foucault
(1992).
In music, the debate has mostly been focused on the relevance of com-
posers intentions for interpretation. Intention, as discussed by Dipert or
Butt, is presented as a multi-level/multi-sided concept, following a similar
approach by some literary-studies theorists (namely Hancher 1972). The fo-
cus is mostly placed on the composer or the musical text/score, and its sig-
nificance for the performers own range of interpretative choices has been
neglected.
This neglect derives from post-romantic views of the work concept, views
that emphasize the primacy of author and text over the contingencies and
variability normally assigned to performance. Burke points out that the cru-
cial historical change in conceptions of authorship did not occur in the theo-
retical upheaval of the last thirty years but with the romantic revolution and
the eighteenth-century philosophical and aesthetic discourses upon which it
grew (Burke 1995, p. xix). This statement could also be applied to changes in
the notion of musical work during the same period, which led to differenti-
ated views of the relative relevance of composing/performing/listening as
creative actions. The intentional fallacy criticized by Wimsatt and Beardsley
located the merits of a literary work in the correct interpretation of the au-
thors intentions. The tension generated by the pressure of playing under the
ideal of Werktreue (implying fidelity to the works essence), mentioned by
Goehr (1992), is a parallel situation, placing constraints and restrictions upon
performers on the basis of authorial intention.
This article intends to demonstrate: (1) that the concept of authors inten-
tion, which is traditionally viewed by performers as a mandatory interpreta-
tive framework, can also be addressed from the performers perspective and
(2) that a multi-level appraisal of intention can be applied to collaborative
procedures in order to promote a successful performance.
MAIN CONTRIBUTION
This research was developed in the context of a case study documenting the
collaboration between a composer and a performer, which involved the pre-
mire of a new composition for the pianoforte. The concepts of authorship
and intention provided a theoretical framework for the analysis of the col-
laboration process, which was grounded on two main challenges: applying a

contemporary style of composition involving a less-known historical instru-
ment, and building on the sense of historical ambiguity without compromis-
ing the effectiveness of the performance.
Composer/performer/work
The rise of the concept of the autonomous musical work and the associated
concept of Werktreue led to the redefinition of the relation composer/per-
former and the gradual dissociation of these roles, most noticeably from the
beginning of the nineteenth century. Goehr mentions that Liszt spent a con-
siderable part of his musical life developing two distinct forms of perform-
ance, first, performances committed to faithful renditions of his works,
second, virtuoso performance devoted to the art of extemporization and the
show of impressive performance technique (Goehr 2002, pp. 232-233).
Liszts option for two different types of performance is representative of the
establishment of a viewpoint that distinguished composers and performers
activity. The text became paramount in this new outlook.
Intention revisited: From literature to music
The relevance of intention has led to an ongoing debate in the realm of liter-
ary studies, linked to the notion of authorship and the establishment of criti-
cal parameters. This research extends this theoretical stance to music,
drawing a parallel between the dichotomies author/reader and com-
poser/performer. This parallel is not fortuitous: as pointed out by Kivy, there
is an almost universal tendency since at least the eighteenth century to try to
understand absolute music on a linguistic model of one kind or another
(Kivy 1995, p. 284).
The debate surrounding the notion of authorship in the late 1960s chal-
lenged an author-centered appraisal of literary texts. Barthes announced the
death of the author in his much-debated essay, highlighting the multiplicity
of voices implied in a text: there is one place where this multiplicity reas-
sembles, and this place is not the author, as hitherto said, its the reader
(Barthes 1994, p. 595). Foucault, on the other hand, does not dismiss the in-
tellectual implications of the author figure: a discourse associated to an au-
thors nameis not a fleeting and momentary discourse, immediately
consumed, but rather a discourse that is perceived in a certain manner and
given, within a given culture, a certain status (Foucault 1992, p. 45). Fou-
cault, however, dissociates author and writer, characterizing the author as a
function that can acknowledge the existence of a multiple I.
The fact that intention as a conceptual idea remains a recurring issue in

literaryand artstudies theory reflects the enduring dichotomy between
the programmatic intention (what the author set out to say) and the opera-
tive intention (what his text ends up saying) (Burke 1992, p. 142). Some
critics approach this dichotomy through a hierarchyor categorybased
outlook, namely Hancher (1972, p. 835), who writes that the authors inten-
tion:
Can be analyzed into three kinds: his intention to make a literary work of
a certain sort (programmatic intention); his intention to be (under-
stood as) acting in a certain way (active intention); and his intention to
cause something or other to happen beyond the mere understanding of
his meaning (final intention).
The hierarchical/categorical approach has also been the preferred model re-
garding theories on musical intention.
The concept of intention in musical interpretation has been criticized for
its prescriptive stance, often based on textual and analytical constraints that
limit the performers freedom of expression. As Taruskin ironically contends,
composers do not usually have intentions such as we would like to ascertain,
andthe need obliquely to gain the composers approval for what we do be-
speaks a failure of nerve (Taruskin 1995, p. 98). Historically-informed per-
formance, nevertheless, is undeniably connected to the maintenance of
intention as a working concept, and represents an innovative trend in musical
performance that cannot be dismissed solely on the grounds of poor herme-
neutical grounding.
Part of the problem regarding the adaptation of intentional theory to mu-
sic may lie on the exclusive focus on the composers perspective, which con-
trasts with the above-mentioned trends in literary theory that question the
conventional notion of authorship and emphasize the importance of multiple
perspectives. Diperts description of intentional levels, and its re-evaluation
by Butt (2002), are, as acknowledged by Butt, unsatisfactory as theoretical
tools, as the distinction between levels of intention, despite the fact that they
are established with the composer in mind, is in fact partly or mainly depend-
ent on performers choices. Dipert, while mentioning the limited relevance of
composers intentions for performance, proposes a 3-levelled hierarchy. Butt
rejects the hierarchical outlook, and proposes a division between active in-
tentiona composers specific decisions concerning such matters as instru-
mentation, tempo, dynamic, and passive intentionthose factors over
which he had little control, but which he consciously or unconsciously as-

sumed (pp.89-90).
Intention revisited: Prelude in Fugue
From the early stages of collaboration (Prelude in Fugue was composed by

Sara Carvalho for Helena Marinho), it was decided that composer and per-
former would not assume the conventional separation of roles and that the
effectiveness of the work would require a close interaction. It also became
clear that intention was the focal concept within the collaboration process
and that former intentional models should be subject to reformulation.
The initial challenge involved mechanisms of adaptation derived from the
composers unfamiliarity with the pianoforte, the performers dilemma of
mediating between historically-informed performing practices, which are
particularly suited to the instrument, and the need to convey an original and
effective interpretative approach. The issue of intention was addressed ini-
tially as being especially pertinent in collaborative work between composer
and performer(s) in the interpretation of contemporary solo repertoire, where
this type of cooperation often leads to performances that are informed by the
intentional contribution of both parties. Thus, theoretical frameworks focus-
ing solely on the composers perspective are unsuitable to describe this type
of collaboration.
Even though the intentional model proposed by Hancher (1972) ad-
dressed the authors perspective alone, its acknowledged reliance on Austins
division of speech acts into locutionary, illocutionary, and perlocutionary
kinds, allows us to establish a parallel with musical interpretation as a com-
municative action.
Programmatic intention, in this case study, was developed through col-

laboration prior to the composition process: the performer introduced the
instrument to the composer, suggested effective techniques, and de-
scribed instrumental characteristics, namely: the type of dynamic con-
trasts, the effects obtained through specific pedaling effects, arpeggiation,
and flexible tempo choices; the composer departed from a situation of
near absence of contemporary repertoire for the instrument and had to
find a suitable way to express her intention.
Active intention was developed in alternating joint and individual work
sessions. In joint sessions matters of articulation, tempo, pedaling were
discussed, and some changes were gradually introduced; in individual
work sessions, both composer and performer worked on their own vision
of the piece for further discussion.
Final intention was shared by composer and performer, who centered
their attention in details that differed in some instances: the composer
mostly focused on structural coherence and impact, whereas the per-
former was more focused on aspects such as effective dynamic contrasts,
expressivity, and even articulation.
IMPLICATIONS
The concepts of authorship and intentionality can be successfully applied to

procedures of collaboration between composers and performers that promote
exchange and mediation. Intention emerges thus not in its now outdated per-
spective of mandatory fidelity to the composers directions, but as the core of
a collaborative and creative process in which intention is envisaged from the
standpoint of the performer as well. Furthermore, the description of the
phases and level of exchange between performer and composer demonstrate
that intention can be addressed and described as a multi-level concept shared
by composer and performer alike.
Helena Marinho, Department of Communication and Art, University of Aveiro, Campus

de Santiago, Aveiro 3810-193, Portugal; Email: helena.marinho@ua.pt
References
Barthes R. (1994). La mort de lauteur. In R. Barthes (ed.), uvres Compltes (pp. 491-
495). Paris: Seuil. (Originally published in 1968.)
Burke S. (1992). The Death and Return of the Author. Edinburgh, UK: Edinburgh Uni-
versity Press.
Burke S. (1995). Authorship: From Plato to the postmodern. Edinburgh, UK: Edin-
burgh University Press.
Butt J. (2002). Playing with History. Cambridge: Cambridge University Press.
Dipert R. R. (1980). The composers intentions: An examination of their relevance for
performance. The Musical Quarterly, 66, pp. 205-218.
Foucault M. (1992). O Que um Autor? Lisbon: Vega. (Originally published in 1969.)
Goehr L. (2002). The Imaginary Museum of Musical Works. Oxford: Clarendon Press.
Hancher M. (1972). Three kinds of intention. Modern Language Notes, 87, pp. 827-851.
Kivy P. (1995). Authenticities. Ithaca, New York, USA: Cornell University Press.
Taruskin R. (1995). Text and Act. Oxford: Oxford University Press.
Thematic session:
Psychology of performance
When repetition isnt the best practice

strategy: Examining differing levels of
contextual interference during practice
Laura A. Stambaugh
Department of Music, Western Washington University, USA
Two experiments examined the effects of blocked and random practice

schedules on the performance accuracy, speed, and temporal evenness of
performance by wind players. Blocked schedules used repetitive practice
orders, while random schedules constantly changed the order of tasks
practiced. Beginning clarinet students completed three days of practice
on three short technical tasks, in either a blocked or random order.
Twenty-four hours after practice, beginning students who had practiced
in the random order were able to play significantly faster than students
who had practiced in the blocked order (F1,38=24.95, p<0.001, 2=0.92).
Students in the blocked group performed significantly slower at 24-hour
delayed retention than immediately after practice (p<0.001). Contrary to
non-musical motor learning investigations, there was no speed-accuracy
trade-off: students maintained high accuracy scores while speed gradu-
ally improved. In Experiment II, university wind students practiced three
short technical tasks in either a blocked or random order for two days.
Retention testing occurred 24-hours and one week following practice.
Preliminary results were presented in the conference session.
Keywords: practice; contextual interference; blocked; speed-accuracy

trade-off; speed
Musicians and music educators are always seeking to better understand fac-
tors contributing to efficient and effective practice. Motor learning research,
which studies how people acquire and refine movements over time, offers a
potentially valuable framework for exploring the challenges of instrumental
performance. To the extent that playing a musical instrument is a motor task,
motor learning theories and methods may be relevant to the study of music
practice.
A series of two studies applied the motor learning principle of contextual

interference, the amount of cognitive disruption present when practicing
multiple tasks within a practice session, to instrumental practice. The pur-
pose of these studies was to examine the effects of differing levels of contex-
tual interference on performance speed, accuracy, and temporal evenness by
beginning and university-level instrumentalists. Low levels of contextual in-
terference were defined as blocked practice orders (AAA BBB CCC), while
high levels of contextual interference were defined as random practice orders
(ABC BAC ACB). The contextual interference hypothesis (Shea and Morgan
1979, Battig 1966) predicted the blocked condition would exhibit superior
performance immediately following practice (acquisition) but the random
condition would perform better at delayed retention testing. This hypothesis
is generally consistent in laboratory motor learning studies (e.g. Lee and
Magill 1983, Brady 2004), but less consistent in applied studies of sports
skills (e.g. Landin and Hebert 1997, Hall et al. 1994) and fine-motor skills
(Ollis et al. 2005, Ste-Marie et al. 2004).
In a music context, university non-percussion majors practiced right hand
lead snare drum sticking for one practice session in either a blocked or varied
(random) order or an unstructured free condition (Rose 2006). After reten-
tion delays of 30 minutes to 24 hours, no significant differences were found
between blocked and random groups for accuracy. Likewise, Stambaugh and
Demorest (under review) found second and third year clarinet and saxophone
students learned 8-measure method book songs equally well in either 6-min-
ute blocked, 1-minute serial (ABC ABC), or hybrid (2-minute serial) orders.
The current investigations expanded on previous research by using de-
signs more similar to traditional motor learning experiments and musicians
with different levels of expertise.
METHOD: EXPERIMENT I
Participants
Participants (n=41) were beginning clarinet players recruited from 16 ele-

mentary schools in the northwest USA. Students were enrolled in the first
year of their school band program, which started in either fifth or sixth grade
(age 11 or 12 years).
Materials
The stimuli (e.g. see Figure 1) were composed by the researcher to meet the
following constraints: limited performance range, task brevity, transposable
Figure 1. Practice task for Experiment I.
to two other keys to control for any aural confounds, inclusion of a novel pitch
to prevent ceiling effects, and majority tonality. There were two screening
tasks, three practice/retention tasks, and two transfer tasks.
Procedure
Students began by playing the screening tasks, and this accuracy score was
used as a covariate. The practice sessions occurred in group settings led by
the researcher. Microphones were attached to students clarinets during all
playing. Cubase multi-track software enabled individual recordings to be
made of each students playing while in the group setting.
Students within schools were randomly assigned to either the blocked or
random practice condition, attending three practice days and one delayed
testing day. Students were instructed to play as accurately and quickly as
possible. In the blocked group, students played 18 trials of one task on each
day. Random group students played 6 trials of each of the 3 tasks in a random
order. At retention testing, students within groups were further divided for
testing in either a blocked or random order (blocked practice-blocked reten-
tion, blocked practice-random retention). After performing 3 retention tri-
als of each task and 3 trials of the transfer tasks, students completed a brief
questionnaire about their attitude toward their practice condition.
Accuracy and speed scores were determined for each of the 3,554 per-
formance trials. Accuracy was scored through repeated listening by the re-
searcher using a point-deduction system, and speed was measured in Cubase
as the time lapsed from the onset of the first pitch to the onset of the last
pitch in each trial. Temporal evenness was scored for the final three practice
trials (acquisition), retention, and transfer trials (1,286) using an equation
determining the average interonset interval in relation to trial duration:
|IOIx-IOIm|=IOI, IOI/IOI.
RESULTS: EXPERIMENT I
Figure 2 presents means and standard deviations of speed and accuracy aver-
aged for every 3 practice trials. After 18 practice trials, there were no signifi-
cant differences between groups for accuracy, speed, or temporal evenness
(p>0.02). A repeated measures ANCOVA for speed found a main effect for
trial (F5,34=5.05, p=0.001, 2=0.11), indicating both groups did improve speed
Figure 2. Accuracy and speed mean scores averaged across every three trials.
during the practice trials. At 24-hour delayed retention, the random group
performed significantly faster than the blocked group (F1,38=24.95, p<0.001,
2=0.92), and pairwise comparisons indicated the blocked group played sig-
nificantly slower at retention than at acquisition (p<0.001). There were no
significant differences for accuracy or evenness at retention, and no signifi-
cant differences between groups on transfer tasks. When students within
practice groups were tested in blocked or random orders at retention, there
were no significant within-practice group differences for accuracy, speed, or
evenness. Regarding attitude toward practice, both groups exhibited a similar
(t40=-1.10, p>0.28) positive response.
METHOD: EXPERIMENT II
Participants
Participants (n=40) were members of university bands, playing woodwind or

valved brass instruments.
Materials
The researcher composed the stimuli (e.g. see Figure 3) to be a major tonality,
short task, and difficult enough to prevent a ceiling effect.
Procedure
Procedures were similar to Experiment I, with three notable differences.

First, participants completed practice sessions individually, without the re
searcher present. Second, the 18 practice trials occurred over two, rather than
three, days and a one-week delayed retention test was added. Third, the
Figure 3. Example task for Experiment II.
blocked practice order was 9 trials of each task on both days in a blocked or-
der, to control for consolidation effects.
RESULTS: EXPERIMENT II
Preliminary results were presented in the conference session.
DISCUSSION
Like many applied studies, results of Experiment I were partially consistent

with the contextual interference hypothesis: although group scores were
similar at acquisition, the random group performed faster at retention. A
possible explanation for the similarity between group performances at acqui-
sition is differences between the single-day practice design and the three day
practice design used here. In single-day designs and in the blocked group in
Experiment I, block 6/acquisition occurred on the same day as all practice
trials: there was no sleep-based consolidation interval (Duke and Davis 2006,
Duke et al. 2006). However, the random group in Experiment I distributed
their practice across three days, allowing some consolidation to occur before
playing block 6/acquisition. This potential confound was addressed in Ex-
periment II.
Notably, these results are also inconsistent with the snare drum task
practiced in Rose (2006), perhaps due to the nature of the musical task. The
clarinet stimuli were specifically designed to begin and end on tonic pitch.
During the 18 trials, students cultivated an aural imprint of the task, and as
automaticity progressed, it is possible this melodic component facilitated a
pitch-to-finger mapping that was not possible in the non-melodic task.
The learning advantage exhibited by the random group suggests this
practice strategy is beneficial for beginning clarinet students practicing short
technical tasks, contrary to the drill-type strategies promoted in many begin-
ning method books, practice software, and lessons.
Acknowledgments
Experiment I is based on the authors doctoral thesis, Effects of Blocked and Random
Practice Schedules on Performance by Beginning Wind Players, University of Wash-
ington, adviser Steven M. Demorest.
Laura A. Stambaugh, Department of Music, Western Washington University, 516 High

Street, Bellingham, Washington 98115, USA; Email: laurastambaugh@gmail.com
References
Battig W. F. (1966). Facilitation and interference. In E. A. Bilodeau (ed.), Acquisition of

Skill. New York: Academic Press.
Brady F. (2004). Contextual interference: A meta-analytic study. Perceptual and Motor
Skills, 99, pp. 116-126.
Duke R. A. and Davis C. M. (2006). Procedural memory consolidation in the perform-
ance of brief keyboard sequences. Journal of Research in Music Education, 54, pp.
111-123.
Duke R. A., Davis C. M., Allen S. E., and Goins K. R. (2006). Focus of attention affects
performance of motor skills in music. Paper presented at the MENC National Bien-
nial In-Service Conference, Salt Lake City, Utah, USA.
Hall K. G., Domingues D. A., and Cavazos R. (1994). Contextual interference effects with
skilled baseball players. Perceptual and Motor Skills, 78, pp. 835-841.
Landin D. and Hebert E. P. (1997). A comparison of three practice schedules along the
contextual interference continuum. Research Quarterly for Exercise and Sport, 68,
pp. 357-361.
Lee T. D. and Magill R. A. (1983). The locus of contextual interference in motor skill
acquisition. Journal of Experimental Psychology: Learning, Memory, and Cogni-
tion, 9, pp. 730-746.
Ollis S., Button C., and Fairweather M. (2005). The influence of professional expertise
and task complexity upon the potency of the contextual interference effect. Acta
Psychologica, 118, pp. 229-244.
Rose L. P. (2006). The Effects of Contextual Interference on the Acquisition, Retention,
and Transfer of a Music Motor Skill Among University Musicians. Unpublished
doctoral thesis, Louisiana State University.
Shea J. B. and Morgan R. L. (1979). Contextual interference effects on the acquisition,
retention, and transfer of a motor skill. Journal of Experimental Psychology: Hu-
man Learning and Memory, 5, pp. 179-187.
Stambaugh L. A. and Demorest S. M. (under review). Effects of practice schedule on
wind instrument performance: A preliminary application of a motor learning prin-
ciple.
Ste-Marie D. M., Clark S. E., Findlay L. C., and Latimer A. E. (2004). High levels of
contextual interference enhance handwriting skill acquisition. Journal of Motor
Behavior, 36, pp. 115-126.
Imaging the music: A context-specific method

for assessing imagery ability
Terry Clark and Aaron Williamon
Centre for Performance Science, Royal College of Music, London, UK
This study compared timing profiles of live and mental performances to

determine if such an approach could provide a context-specific and ob-
jective indication of musicians imagery abilities. Of further interest was
how performance on this type of task may relate to scores obtained on
standard imagery use and vividness measures. Thirty-two music students
were recruited from the Royal College of Music (RCM). They first com-
pleted two imagery use and vividness measures, followed by two live and
two mental performances of a two-minute extract of their choice. Fol-
lowing the generation of timing profiles for each performance, correla-
tions were calculated within and between the two performance
conditions. These were normalized by conversion to Z-scores and then
compared with results obtained from the imagery use and vividness
measures. While all participants achieved a significant correlation be-
tween the timing profiles from the live performance condition, only 17
did so within the mental performance condition. When comparing the
timing profiles between the two conditions, 22 obtained significant cor-
relations. Significant correlations emerged between the imagery vivid-
ness measures and Z-scores from the live performance condition, while
amount of time spent employing imagery significantly correlated with Z-
scores from the mental performance condition.
Keywords: imagery; mental skills; mental chronometry; performance

science; music education
Prior to the development and implementation of mental skills training pro-

grams for musicians, it is necessary to generate an accurate assessment of
their skills and abilities. However, assessment of musicians mental skills,
imagery included, is not straightforward. A number of self-report question-
naires for assessing imagery vividness currently exist, yet they have been
criticized for lacking objectivity. In addition, research would suggest that im-
agery ability is comprised of three distinct features: vividness, controllability,
and accuracy of reference (Denis 1991).
One potential alternative to self-report questionnaires employs temporal
comparisons between live and mental performances, also termed mental
chronometry, as employed by Repp (1999a, 1999b) and Wllner and Willia-
mon (2007). These researchers proposed that by removing sensory feedback
during performance, insight could be gained into the strength and content of
musicians mental images. The use of chronometric comparisons between live
and mental performances is also based upon work from human movement
sciences which suggests that imagery processes are mediated by neuronal
mechanisms similar to those used in perception (Mellet et al. 1998).
While previous research would indicate that musicians are capable of
performing works with similar timing profiles across repeated performances,
results derived from studies employing mental chronometry caution that
there is variability among musicians ability to perform such a task, particu-
larly in the absence of kinaesthetic feedback (c.f. Repp 1999a, 1999b, Wllner
and Williamon 2007). Due to these varying results, musicians ability to per-
form this type of chronometric task successfully is still somewhat unclear, as
is the range of factors that may help or hinder their ability to do so.
Given the current state of understanding surrounding musicians imagery
use and abilities, this study sought to investigate the efficacy of a context-
specific and objective method for assessing musicians imagery abilities. In
particular, timing profiles of live and mental performances were compared to
determine if such an approach could be a valid indicator of imagery ability
and vividness. Of interest was how performance on this type of task may re-
late to scores obtained on standard imagery use and vividness measures.
METHOD
Participants
Thirty-two undergraduate and postgraduate music students were recruited

from the RCM. The sample comprised 11 men and 21 women, ranging in age
from 20-28 years (mean=22.29, median=22.0, SD=2.20). In terms of the
participants year of study, 4 were year 1 undergraduates, 6 were in year 2, 14
were in year 3, 4 were in year 4, and 7 were postgraduates. For their instru-
ment groupings, 7 were pianists, 9 were vocalists, 12 were string players, and
4 were woodwind and brass players.
Materials
In order to gain an understanding of the participants previous experience

with imagery and mental rehearsal, as well as their mental approaches to
practicing and performing in general, a survey was developed to ascertain the
types of mental activities in which the participants engage in relation to their
musical activities. Participants were also asked to provide an approximate
indication of the amount of time they engaged in each type of mental strategy
or activity per week, how long they had been using the particular activities,
and how skilled or effective they were in using each of them by rating them-
selves on a scale from 1-7, with 7=very skilled.
The randomized short version of Betts Questionnaire upon Mental Im-
agery was employed to assess imagery vividness (Betts QMI, Sheehan 1967).
This is a 35-item self-report questionnaire in which participants are asked to
rate on a 7-point Likert-type scale the strength or vividness of suggested sen-
sory experiences, with 1=perfectly clear and vivid and 7=no image present
at all. Each of the senses is addressed, including sight, sound, taste, smell,
movement, and interoceptive and exteroceptive sensations. As per the stan-
dardized instructions, participants were asked to imagine a particular sensory
experience and then to rate the vividness of the image created. Responses are
summed for a total questionnaire score as well as scores for each of the sub-
components. A lower score indicates a greater level of imagery vividness.
Finally, a mental chronometry task was developed with the aim of pro-
viding a contextually relevant, empirical measure of imagery ability that ex-
tends beyond self-report measures. The full procedure for this task is
described below.
Procedure
Following completion of the survey and questionnaire, the participants were

requested to perform a 2-minute extract of their choice that was at a public
performance standard. The participants first gave two live performances of
their chosen extract followed by two mental performances of the same ex-
tract, tapping a light metal object upon a desk to indicate the beats of the
piece as they imagined themselves performing it. All performances, live and
mental, were recorded using a Tascam portable stereo audio recorder.
Initial data preparation involved the generation of mean scores and stan-
dard deviations for the survey and questionnaire, as well as relevant individ-
ual items or subscales within them. For the mental chronometry task, the
inter-onset-intervals (IOIs) for each of the four performances were extracted
to produce timing profiles. Following the creation of timing profiles, similar-
ity between the IOIs of the two performances within each of the two condi-
tions was assessed using one-tailed Pearson correlations. One-tailed Pearson
correlations were next used to assess the similarity of the IOIs between the
two conditions. All correlations were normalized by conversion to Z-scores
for use in further analyses with the imagery use and vividness measures.
RESULTS
When asked to identify the types of mental strategies or activities that the
participants employed as part of their regular practice activities, imagery was
identified most often (n=20 of 32). Of those who cited regularly employing
imagery, they reported using it for an average of 25.95 minutes per day
(SD=28.54), that they had been using it for 6.48 years (SD=5.87), and rated
their effectiveness with imagery at 4.80 out of 7 (SD=1.28). Within this, the
participants reported singing or hearing their music in their minds, memo-
rizing music away from their instruments, and playing on a surface other than
their instrument (or finger practice).
For the Betts QMI, the participants obtained a total mean score of 89.66
(SD=26.09). The potential range for this questionnaire is 35-245, with a
lower score indicating greater imagery vividness. The participants ability to
imagine sounds (auditory sub-component) and movements (kinaesthetic sub-
component) were the senses achieving the greatest vividness on this measure,
while the participants ability to imagine smells (olfactory sub-component)
achieved the least amount of vividness.
When considering the participants performance on the mental chrono-
metry task, all 32 participants achieved a significant correlation between the
IOIs of the two live performances. When examining the mental performance
condition, however, only 17 participants achieved a significant correlation.
Comparing the IOIs between the two conditions, 22 of the 32 participants
yielded a significant correlation between the two conditions.
Comparing the results from the different assessment measures (see Table
1), a significant positive correlation emerged between the amount of time per
day that the participants reported employing imagery and their ability to
perform consistently within the mental performance condition. A significant
positive correlation also emerged between the participants self-rated ability
to engage in imagery and their ability to give multiple temporally similar
performances within the live performance condition. A link between imagery
vividness and ability within the live performance condition was also found
with the Betts QMI. The normalised Z-scores from the live performance con-
dition achieved significant negative correlations with total scores from the
Table 1. Significant Pearson correlations between the Z-scores derived from the mental
chronometry task with the imagery use and vividness measures.
Live performance Mental performance Between

condition condition conditions
Imagery: Mins. per day used 0.31 0.46* 0.07
Imagery: Self-rated ability 0.42* 0.36* 0.25
Betts QMI Total score -0.35* -0.01 -0.13
Betts QMI Auditory -0.54** -0.06 -0.14
Betts QMI Visual -0.40* -0.10 -0.02
Betts QMI Gustatory -0.36* -0.13 -0.23
* Correlation is significant at the 0.05 level (2-tailed).
** Correlation is significant at the 0.01 level (2-tailed).
Betts QMI as well as the subcomponents of visual, auditory, and gustatory

imagery. As lower scores on the Betts QMI indicate greater levels of imagery
vividness, negative correlations indicate a promising connection.
DISCUSSION
For 63% of the participants, imagery formed a significant part of the practice
activities. Within this, auditory and kinesthetic imagery were the senses for
which the participants reported the greatest amount of vividness. Within the
mental chronometry task, while all participants were able to give two per-
formances of comparable timing profiles within the live performance condi-
tion, some musicians were less able to do so with the removal of sensory
feedback in the mental performance condition, supporting results from pre-
vious investigations of this type. Based upon examination of the results from
the three assessment measures, the three scores derived from the mental
chronometry task appear to provide an indication of either imagery vividness,
controllability, or accuracy of representation, supporting the proposal that
imagery ability is comprised of three distinct features (Denis 1991).
Possessing high levels of vividness and control over their imagery should
be of great importance for musicians, and while previous research would sug-
gest that greater imagery ability facilitates enhanced performance quality
(McIntyre et al. 2002), this link within music needs to be explored more fully.
Having reliable methods for accurate assessment of imagery ability is central
within this. Lastly, these findings support recommendations for the use of a
mixed-methods design when assessing musicians imagery use and abilities,
with a mental chronometry task such as that employed here adding an ele-
ment of objectivity to these investigations.
Acknowledgments
We wish to thank David Burnand and Andrew Earis for their advice on methods for
generating timing profiles arising from the mental chronometry task.
Terry Clark, Centre for Performance Science, Royal College of Music, Prince Consort
Road, London SW7 2BS, UK; Email: tclark@rcm.ac.uk
References
Denis M. (1991). Image and Cognition. New York: Harvester Wheatsheaf.

MacIntyre T., Moran A., and Jennings D. (2002). Is controllability of imagery related to
canoe-slalom performance? Perceptual and Motor Skills, 94, pp. 1145-1250.
Mellet E., Petit L., Mazoyer B., et al. (1998). Reopening the mental imagery debate:
Lessons from functional anatomy. NeuroImage, 8, pp. 129-139.
Repp B. H. (1999a). Effects of auditory feedback deprivation on expressive piano per-
formance. Music Perception, 16, pp. 409-438.
Repp B. H. (1999b). Control of expressive and metronomic timing in pianists. Journal
of Motor Behavior, 31, pp. 145-164.
Sheehan P. (1967). A shortened form of Betts questionnaire upon mental imagery.
Journal of Clinical Psychology, 23, pp. 386-389.
Wllner C. and Williamon A. (2007). An exploratory study of the role of performance
feedback and musical imagery in piano playing. Research Studies in Music Educa-
tion, 29, pp. 39-54.
Learning to play double bass by stimulating

mental imagery
Emmanuel Bigand1, Thierry Pozzo2, and Christophe Beraut3
1 LEAD CNRS, Institut Universitaire de France, University of Burgundy, France

2 ERT Motricit-Plasticit, INSERM, University of Burgundy, France
3 Conservatory of Music, Dijon, France
It is well establish that mental simulation of movements has beneficial

effects on learning and rehabilitation of a large range of gestures. Dot
pattern videos were found to be of great efficiency to stimulate mirrors
neurons because they stimulate dorsal pathways more than standard
video recordings. Until now, these effects were mostly reported in the
domain of sports or medicine. Only a few studies attempted to generalize
them to music. The purpose of this study was to optimize the learning of
double bass in beginner children (mean age=8 years) with a new method
that encourages mental simulation of basic movements of the right (bow)
and left hands of double bass players. The improvement in performances
in children was also documented with a light motion capture device that
tracked several parameters of the bow speed and right arm movement.
Eight children who had recently begun studying double bass at the con-
servatory in Dijon used a traditional method consisting of 24 lessons
(one per week) accompanied with a DVD. The DVD contained a set of 24
imagery training sessions, each being associated with one lesson. Each
week, the children completed approximately ten mental imagery exer-
cises displayed on the DVD. These exercises were designed to boost asso-
ciations between movement-audition, movement-sight reading, and
movement-musical expression. All videos were made with dot patterns
methods. The improvements of the child were followed by registering the
number of errors made for the mental imagery exercise and by capturing
the evolution of their motor performance on the double bass with a light
motion capture device, easily transportable to the conservatory. The de-
vice allows the measurement of velocity, acceleration-deceleration of
movements, as well as the online changes of the angles formed by differ-
ent parts of the arm.
Keywords: mental imagery; mirror neurons; music learning ergonomics;

developmental psychology
Emmanuel Bigand, Institut Universitaire de France , LEAD UMR 5022 CNRS, Ple
AAFEEsplanade Erasme, University of Burgundy, BP 26513 21065, Dijon Cedex,
France; Email: bigand@u-bourgogne.fr
Thematic session:
Perceiving performance
Computer assisted analysis and display of

musical and performance data
Stuart Pullinger1, Nicholas J. Bailey1, Jennifer MacRitchie1,

and Margaret McAllister2
1 Centre for Music Technology, University of Glasgow, UK

2 Department of Composition, Berklee College of Music, USA
The coordinated storage of performance data in such a way that it can be

used across multiple projects is problematic: general purpose systems
that can store gestural, score, and other performance data are not gener-
ally available. Using data from current projects, we aim to provide a uni-
fied database that can store and present a musical score alongside
associated performance data and musical analysis. Using a general pur-
pose representation language, Performance Markup Language (PML),
aspects of performance are recorded and analyzed. Data thus acquired
from one project is made available to others. Presentation involves high-
quality scores suitably annotated with the requested information. Such
output is easily and directly accessible to musicians, performance scien-
tists, and analysts. We define a set of data structures and operators that
can operate on musical pitch and musical time, and use them to form the
basis of a query language for a musical database. The database can store
musical information (score, gestural data, etc.) and audio/video artifacts.
Querying the database results in annotations of the musical score, po-
tentially augmented with audio/video selected from stored performances.
Two demonstrations are provided: an analytically-based query and a
performance-gesture-based one. In the former, dissonant notes/intervals
are identified in a performance of a Bach two-part invention. The score is
then graphically annotated to indicate the performers mean inter-onset
intervals in the neighborhood of these features. In the latter, a score of a
19-ET microtonal song is displayed, annotated with the deviation in the
sopranos pitch from that notated. The database is capable of storing mu-
sical score information and multimedia recordings and cross-referencing
them. It is equipped with the necessary primitives to execute music-ana-
lytical queries, and highlight notes identified from the score.
Keywords: music information retrieval; visualization; performance

analysis; score analysis; database
Nicholas J. Bailey, Centre for Music Technology, Department of Electronics and Electri-
cal Engineering, University of Glasgow, Rankine Building, Oakfield Avenue, Glasgow
G12 8LT, UK; Email: n.j.bailey@elec.gla.ac.uk
Gestural communication: Linking multimodal

analysis of performance to perception of
musical structure
Jennifer MacRitchie, Bryony Buck, and Nicholas J. Bailey
Centre for Music Technology, University of Glasgow, UK
A two-tiered experiment was conducted to assess the communication of

phrasing structure from performance nuance to audience perception.
Nine solo piano performances of two selected Chopin preludes, compa-
rable in musical structure and complexity, were recorded multi-modally
through audio, MIDI, and the Vicon motion capture system. Analyses of
performance parameters such as tempo, dynamics, and movement were
then conducted with reference to the notated score. Videos of each per-
formance were presented to observers with musical knowledge who used
a slider to determine the shape of each musical phrase. Having previ-
ously been presented performances in visual only mode, participants
were now presented the performances in three modalities: visual, audio,
and audiovisual. Further to findings that occurrence of performance
gestures correlated with notated and perceived phrase boundaries, multi-
modal analysis of performance parameters confirmed that performers
conveyed musical structure as intended in auditory as well as visual ele-
ments of performance.
Keywords: performance analysis; music perception; multimodal inter-

action; musical structure
In a western classical music performance, the performer uses certain expres-

sive functions to communicate features of the notated score and highlight
interesting points throughout the course of the music. Research in music
performance has acknowledged that this musical experience is multi-modal
and that musical information is contained not only in the aural but also the
visual stream (Davidson 1993, Wanderley 2001). Cross-modal interaction in
the perception of structural features within a piece of music have shown, for
clarinetists performance, that the visual stream can help to better decode the
information presented in audio-visual instances (Vines et al. 2006). Previous

research by the authors has focused solely on this visual information in piano
performance, particularly as performers do not require the physical act of
breathing between phrases. A relationship between overall performer ges-
tures and phrasing structure of the written score was demonstrated
(MacRitchie et al. 2009). This research aims to investigate the multi-modal
information contained within piano performances of two Chopin preludes
and to discover the relative contributions in terms of musical perception.
METHOD
Participants
Nine highly trained pianists from the Royal Scottish Academy of Music and
Drama and the Universities of Glasgow and Edinburgh were recorded playing
the selected preludes. They were paid a one off sum of 25. Fifteen musically
knowledgeable audience judges were recruited by email from the University
of Glasgow and local orchestras. They were paid 6 per hour for participation.
Materials
The two test pieces performed were Chopin Preludes Op.28 No.7 in A major
and No.6 in B minor. Prelude 7 was chosen as a control piece across perform-
ers as it comprises rhythmically repeating 2-bar phrases. Prelude 6 is slightly
more complicated, as it contains expanded phrases and changes in rhythm;
the opening 2-bar phrase of each prelude is shown in Figure 1. Eighteen au-
dience-perspective recordings were used to generate 56 perceptual stimuli
comprising two practice performances by a tenth performer and presenta-
tions of each prelude by the nine performers in three modalities: visual only
(VO), audio only (AO), and audiovisual (AV). Presentation stimuli were ~40 s
each in duration for Prelude 7 and ~90 s for Prelude 6.
Procedure
Nine highly trained pianists performed the two preludes on a Roland RD-150
88-key weighted keyboard and were recorded through audio, MIDI, and mo-
tion capture. Passive markers were attached to the upper body and head via a
cap and jacket worn by all performers. Performers were required to perform
the preludes as they would in a normal concert setting. Performances were
also filmed from an audience perspective. The 56 audience perspective
performance stimuli were presented to 15 audience participants in AO, VO,
and AV modes across two sessions. Participants were initially presented with
Figure 1. Opening 2-bar phrase of Chopins Prelude in A major Op.28 No.7. (left) and
Chopins Prelude in B minor Op.28 No.6 (right).
the VO stimuli and asked to shape the phrasing structure of each performance
using a slider bar moving from left-right-left as the phrase was shaped. They
were then presented with the stimuli in all three modalities, and the task was
repeated. Stimuli were presented in randomized order to avoid order effects.
RESULTS
Performance Analysis
Estimations of tempo, dynamics, and motion were taken from the MIDI, au-
dio and motion streams, respectively. Performed notes were matched to the
score notes using performance markup language and matcher (described in
McGilvray 2009). Queries on note information were returned by the Pullinger
database (Pullinger 2009). The inter-onset intervals (IOIs), normalized to a
quarter note beat, were divided by 1/60 to give an estimation of beats per
minute. Dynamics were estimated by the RMS amplitude of the signal. 2D
motion curves were extracted by performing principal component analysis
(PCA) of all markers and adding the weighted components. Figure 2 shows
audio, MIDI, and video data plotted against phrase boundaries as they occur
in the audio. To allow direct comparison across performers, all data was time-
warped with respect to the occurrence of phrase boundaries in the audio.
Distances to the nearest troughs in dynamics and tempo curves were ex-
tracted for each phrase boundary. Distances from either troughs or peaks
were extracted for the motion curves depending on each performers move-
ment patterns. Two-way Analyses of Variance (ANOVAs) showed significant
effects of both performer and phrase number on motion norm (F=3.56,
p<0.001 and F=6.63, p<0.01) and a significant effect of performer on dy-
namics (F=5.78, p<0.001) and of phrase on tempo (F=37.24, p<0.001). Other
effects were not significant. This suggests performers have a distinct motion
style varying between phrase boundaries and that they differ in their use of
diminuendo. While performers may not vary how they use ritardandi, this is
varied between phrases. Two-way ANOVAs for the first three phrases of Prel-
ude 6 also demonstrate a significant effect of performer on motion (F=12.85,
Figure 2. RMS amplitude (top), tempo (middle), and motion norm (bottom). Variance
weighted average of principal component scores of motion markers against phrase
boundaries (blue vertical arrows) for performances of Prelude 7 by performers 2 and 7.
Figure 3. Comparison of responses by all participants for all three presentation modes.
Troughs indicate perceived phrase boundaries, vertical lines indicate phrase transitions.
(See full color versions at www.performancescience.org.)
p<0.001). The significant effect of phrasing on dynamics (F=17.98, p<0.001)

may be due to changes in phrasing structure in phrase 3. Parameter extrac-
tion at phrase boundaries gives an indication that the data streams are vary-
ing similarly to Prelude 7. However, further investigation into the continuous
data streams and their correlations is required.
Perceptual analysis
Participants slider responses were time-warped with respect to the occur-

rence of phrase boundaries in the audio stream. Time locations of perceived
phrase endings were averaged across participants. Participants were surpris-
ingly skilled when judging phrasing structure of Prelude 7 in the VO condi-
tion, with eight clear phrases appearing for all performers. Perceived phrases
occurred after the auditory phrasing boundaries and closely follow the per-
Table 1. Three-way ANOVA for factors presentation mode, phrase, and performer.
Source SS df MS F p>F
Presentation mode 0.3374 2 0.1687 239.48 p<0.01
Phrase no. 10.6658 7 1.52369 2162.94 p<0.01
Performer 0.0426 8 0.00533 7.56 p<0.01
Mode x phrase 0.0464 14 0.00332 4.71 p<0.05
Mode x performer 0.0760 16 0.00475 6.74 p<0.01
Phrase x performer 0.0279 56 0.0005 0.71 0.9238
Error 0.0789 112 0.0007
Total 11.2751 215
formed motion trajectories (Buck et al. 2009). Comparison of presentation

modes reveals large differences in the second section of Prelude 7 between
AO and VO modes (Figure 3). Phrasing judgments in the AO and AV condi-
tions precede phrase boundaries, most likely with judgments being made as a
cadence is recognized and the final phrase note is played, rather than when
the sound diminishes and the true phrase ending occurs (Figure 1.). Spear-
mans rank correlations of perceived phrasing reveal extremely high correla-
tions between each presentation mode, the largest being between AV and VO
conditions (0.985, p<0.01). A 3-way ANOVA of performer, phrase number,
and presentation condition revealed significant effects of all three factors and
interactions between presentation mode and phrase number, and presenta-
tion mode and performer (Table 1). Individual analyses of presentation
modes indicate these differences to lie between the VO condition and the AV
and AO conditions. The significant effect of presentation mode (F=239.5,
p<0.01) and interaction between mode by phrase (F=4.71, p<0.05), and mode
by performer (F=6.74, p<0.01) corresponds to the difference in playing style,
particularly where performers vary differently in tempo and dynamics in
phrases 6 and 7. Initial analyses of Prelude 6 responses support findings for
Prelude 7, with participants demonstrating clear phrasing structure for the
initial three phrases in all modalities. The timing differences across modali-
ties are consistent with those found in Prelude 7.
DISCUSSION
Performance analysis of movement, tempo, and dynamics reveals differences

between performance styles that are relative to a performers individual in-
terpretation of the music. These performance differences are still conveyed to
and understood by an audience, assisting with clear recognition of phrasing
structure. Audio information within a performance appears to communicate a

deeper level of structural information. This is seen in the similarity between
tempo and dynamics analysis with the AO audience phrasing judgments,
where phrase boundaries are perceived to occur slightly before the actual
phrase ends. The visual aspect of performance provides clear phrasing struc-
ture through movements that extend beyond the actual sound, exaggerating
the shape of a phrase and so assisting the audience in their interpretation and
understanding of the musical structure of the piece in hand. When combined,
these visual and auditory elements produce a global effect whereby the audi-
ence is able to understand better the musical structure as well as interpreting
the weighted importance of the various parts.
Acknowledgments
We wish to thank Frank Pollick and Lukasz Piwek of the University of Glasgow Psychol-
ogy Department, and Stuart Pullinger and Douglas McGilvray at the Centre for Music
Technology for their support and access to equipment. This research is EPSRC funded.
Jennifer MacRitchie, Centre for Music Technology, University of Glasgow, Rankine

Building, Oakfield Avenue, Glasgow G12 8LT, UK; Email: j.macritchie@elec.gla.ac.uk
References
Buck B., MacRitchie J., and Bailey N. (2009) Perceptual recognition of embodied musi-
cal structure. Paper presented at The Musical Body, University of London.
Davidson J. (1993) Visual perception of performance manner in movements of solo
musicians. Psychology of Music, 21, pp. 103-113.
MacRitchie J., Buck B., and Bailey N. (2009) Visualising musical structure through
performance gesture. Paper presented at the Tenth Conference of the International
Society of Music Information Retrieval, Kobe, Japan.
McGilvray D. (2009) On the Analysis of Music by Computer. Unpublished doctoral
thesis, University of Glasgow.
Pullinger S. (2009) A System for the Analysis of Musical Data. Unpublished doctoral
thesis (forthcoming), University of Glasgow.
Vines B., Krumhansl, C. L., Wanderley, M. M., and Levitin, D. J. (2006) Cross modal
interactions in the perception of musical performance. Cognition, 101, pp. 80-113.
Wanderley M. M (2001) Quantitative analysis of non-obvious performer gestures. In I.
Wachsmuth and T. Sowa (eds.), Gesture and Sign Language in Human Computer
Interaction (pp. 241-153). Heidelberg, Germany: Springer.
Music performance venues: Keeping them in

tune with modern requirements
George Dodd
School of Architecture and Planning, University of Auckland, New Zealand
Throughout the history of music composition and performance we can

perceive a strong link between compositional style, the qualities of places
where composers imagined their music would be played, and the instru-
ment technology of the age. We must examine whether modern listening
preferences are now being shaped by the capabilities of modern re-
cording technology to the point where it is no longer possible to satisfy
these in a live performance venue using a passive design (i.e. non-elec-
troacoustical). The results of a survey of music listening habits and pref-
erences imply that live-performance venues should respect an
originality criterion and not impose features on a performance which
are inconsistent with the architecture and technology of the era (or in-
tention) of the composer.
Keywords: acoustics; electro-acoustics; concert hall; listener preference
Over the last 15 years or so we have seen a maturingand hence an increasing

acceptanceof electro-acoustical techniques for varying the acoustics of mu-
sic auditoria (e.g. Poletti 1994). However, it must not be assumed that pro-
viding a sound field via an active (i.e. electro-acoustical) technique will be as
acceptable to listeners as the same sound field provided by a passive (i.e.
purely mechanical) technique.
The first major use of an active system was in Londons Royal Festival
Hall (RFH). Peter Parkins assisted resonance (Parkin et al. 1964) was con-
ceived as a solution to a specific shortcoming in the hall which is a dedicated
concert hall, and there was considerable disquiet among musicians and con-
certgoers at the thought of loudspeakers being used in auditoria. However, it
was obvious that the active hall concept offeredin principlethe perfect
answer to the needs of a multi-purpose hall. Such multiple-use halls could be
built to havepassivelythe low reverberation time (RT) suitable for speech
(and small-group music performance) and be equipped with an electro-

acoustic system to be switched in to provide the longer RTs required for cho-
ral and symphonic works. This approach avoids acoustical conditions that are
a compromise and, compared with attempting to provide a comparable
change in acoustics by passive mechanisms, offers the advantages of econ-
omy, ease, and speed of operation. However, in these situations the active
contribution to the sound field becomes very audible and significant (com-
pared with the subtle change produced by the relatively small corrective ac-
tion of the original Assisted Resonance system in the RFH). Hence listener
acceptance of the subjective sound becomes critical.
One would hope that active technology, which epitomizes the technology
of our era, will be seen by musicians and composers as a proper element to be
incorporated in their creative works as a new artistic dimension, just as in the
past new musical instruments, made possible by technological development,
have been adopted and become the basis for new composition.
Modern recording and mixing practices are introducing features into
music production that confound the idea of a traditional live performance. By
adding differing amounts of acoustical qualities like distance, echo, and re-
verberation to separate performers in a recording, the recording engineers
are putting the performers in what our instincts (shaped by passive acoustic
experience) would interpret as different auditory scenes. Yet the modern lis-
tener accepts the result as a unified performance. Thus, there must be
changes to our perception of the nature of performance. We must relinquish
our concept of music ensemble as something necessarily mediated and uni-
fied by the singular acoustics of a performance venue.
All this puts the acoustical designer in a difficult situation. What should
modern venues provide? Can active systems (those that are available) offer
the variability and flexibility that might be needed in a modern performance
venue and do so with a sound quality that will satisfy? Since modern trends in
recorded music and reproduction systems are providing sound fields in the
home that cannot be matched in passive halls, is it still valid for us to assume
that todays listeners still wish to go to public performances in traditional
concert hall acoustics rather than listening at home to the engineered sound
from a mixing desk? Further, if we were to consider carrying out experiments
with listeners to try to establish criteria for active sound fields this would
assume that there are preferences regarding active sound fields already in-
herent in our listeners. However, if listeners are truly being presented with
extensions of their listening experience, then a learning and acclimatization
period will surely be required, or we risk these being rejected as merely dis-
tortions or perversions of familiar passive sound fields. This period will need
to be long enough for our perceptions and appreciation (and consequently

acceptance) of these new sound fields to evolve.
So, in thinking about the design of new spaces for public performance
which are likely to include active features in their realization, we should
question how new technology is influencing the habits and preferences of
listeners and, even, whether public performances are what the music con-
sumer really wants.
METHOD
Procedure
To attempt to find answers to some of these questions, a survey of people and

their listening behavior was initiated (Dodd 2001). Since one of the questions
being investigated was whether listener habits and preferences are evolving
with time, it was clear that the survey needed to be longitudinal. As it was not
possible to predict the timescale required to track significant changes, a dec-
ade was chosen as the minimum period over which to monitor listeners.
Participants
Respondents were selected to: include a representative cross section of the

populations surveyed (young, old, concertgoers, non-concertgoers etc), have
some matched groups that were nominally similar each year, and include
cohorts from France and New Zealand.
Materials
The survey used a questionnaire about respondents recall of their patterns

and preferences when listening to music and included thought experiments
where they were asked to make certain choices and then give reasons to ex-
plain their decisions.
The questionnaire was constructed to answer the following questions: (1)
is listening to music important in peoples lives, (2) are listening habits
evolving/changing, (3) is there evidence of changing attitudes toward electro-
acoustical systems or instruments, (4) do people prefer listening to live per-
formances rather than recordings and, if so, why, and (5) how many discerni-
bly different sub-groups/types of listener can we identify.
Three final questions (questions 9, 10, and 11) aimed to collect informa-
tion that could be used in the development of variable acoustics systems for
auditoria. This included finding out about attitudes to live performance and
the possible reactions to the introduction of electro-acoustical elements in

performance venues for live classical music.
RESULTS
The importance of listening to music
Approximately 30% of respondents described music as being essential for

their lives. A notable difference emerged between New Zealand and France in
the percentage that described music as a major factor of life (60% in New
Zealand and 75% in France)
The choice between a live performance, recorded performance,

and simultaneous broadcast
Respondents were asked to indicate a choice between identical performances

in either a live, recorded, or simultaneous broadcast performance where the
experience of both the live and the recorded performance would happen in
exactly the same venue. The intention was to focus respondents thoughts on
those features that differentiate the live performance from the recorded per-
formance. The responses showed unequivocally that, other things being
equal, a large majorityaround 90%of people would choose a live perform-
ance over a recording. The most frequent reason given for preferring a live
performance is atmosphere or ambience. This was cited by 45% of re-
spondents who chose live over recorded music. In second place is communi-
cation with the performers, a reason given by 15% of the group.
The use of loudspeakers in live performance
Question 10 was formulated to focus on the issue of the acceptability of loud-

speakers in live performances of music where the original instruments for
that music are ones that radiate purely mechanically produced sound. Re-
spondents were asked to choose between an organ recital on a pipe organ
compared with the same recital on an electronic organ (assuming it to be of
the highest in sound quality). The responses showed a large majority (75%)
choosing a traditional organ over an electronic organ, a strong indication
that we might expect a continuing dislike for the introduction of loudspeaker-
based systems for fashioning the acoustics of auditoria.
The most frequent reason given (accounting for 25% of respondents) for
preferring a traditional organ was truer sound. The other main reasons
(accounting for 44% of respondents)aesthetics, tradition, and authen-
ticitymake the case clearer. These cannot be linked to measurable features
of the objective sound and, hence, cannot reflect on the performance of the
electro-acoustical components. We thus conclude that consciousness of the
substitution of electro-acoustical elements, in what originally would have
been a totally mechanical chain of sound production, transmission, and re-
ception, is a determining factor for audiences. This could explain the opposi-
tion that is encountered toward the concept of active auditoria.
The demand for live performance
A comparison of the result from question 9 (which showed a large 90% pref-
erence for live performance) with the data that respondents gave on their
listening patterns indicates that there is a large discrepancy between how
people would prefer to listen to their choice of music and how they actually
do. However, the answers to question 11where the option of listening at
home was deliberately included in the experimentconfirm that there is an
unsatisfied demand for live performance.
DISCUSSION
The reasons given in the responses to the experiments of questions 9 and 11

provide some clues for designers of live performance venues. We should be
conscious of optimizing for listeners those features that will attract them to
live performances. Obviously, non-acoustical features will contribute also
(e.g. sight lines, seat comfort, visual aesthetics, and ease of access to the bar
during the interval), but our focus here is on acoustical issues.
Foremost in importance is the promotion of atmosphere. However, the
second most common reason given for the superiority of a live performance
communication with the performerscan immediately be linked to acousti-
cal performance. What is implicated here is whatever characterizes the differ-
ence between the communication happening when listeners hear instruments
directly and the communication happening when they hear them via a re-
cording. Clearly there is communication from performers to listeners taking
place in both cases, but respondents seem to sense that at a live performance
they also communicate back to the performersi.e. a two-way communica-
tion is inherent in the live performance.
This two-way communication is at its most obvious during periods of ap-
plause but one suspects there are other, less overt (as well as non-acoustical)
ways in which the audience transmit their vibes back to performers. Thus
an optimized venue is one that maximizes this sense/feeling of two-way
communication for the audience. For passive venues this isacoustically at
leastautomatic. Transmission is identical in both directions because of the
principle of acoustical reciprocity. However, in an active auditorium, the use

of amplification channels willin principleremove reciprocity.
Thus, when including active acoustics in venues and wishing to respect
the preferences of our audience we must choose those systems that try to
retain reciprocity for the communication path between performers and lis-
teners. The types of system that use microphones to pick up the direct sound
from performers, and loudspeakers arranged to radiate their processed sig-
nals primarily as direct sound to the listeners (hence described as in-line
systems), have nothing to recommend them in this respect. Other systems
(so-called non-in-line) place their transducers remotely from performers and
are effective in reverberating the sound from both performers and listeners.
This suggests that non-in-line systems shouldwherever possiblebe se-
lected in preference to in-line systems.
However, if new music, electro-acoustical musical instruments, and re-
cording practices change listener expectations and preferences we may be
unable to meet these expectations with traditional design approaches using
non-in-line active assistance. Ultimately, it may require that venues for live
performance be constructed as anechoic spaces so that passive acoustics do
not compromise the engineered sound from the performers. Transmission to
the listeners will then be achieved by the ultimate in-line system, with the
sound engineer becoming as much a player in live performances as the mu-
sicians.
George Dodd, Acoustics Research Centre, University of Auckland, Private Bag 92019,
Auckland, New Zealand; Email: g.dodd@auckland.ac.nz
References
Dodd G. (2001). Listener habits and choicesand their implications for public perform-
ance venues. Journal of Sound and Vibration, 239, pp. 589-606
Parkin P. H., Allen W. A., and Bean T. E. (1964). A special report on the experimental
Assisted Resonance system in the Royal Festival Hall. Journal of Sound and Vibra-
tion, 1, pp. 335-342.
Poletti M. (1994). The performance of a new assisted reverberation system. Acta Acus-
tica, 2, pp.511-524.
Keynote paper
The dominant artistic discourse

as a health determinant
Sylvie Fortin
Department of Dance, University of Qubec at Montral, Canada
This paper examines how elements of different discourses are appropri-

ated or resisted by artists and the effects on health and wellbeing. More
specifically, the results of a series of ethnographic studiesconducted
with professional dancers, choreographers, rehearsal directors, and stu-
dents in different contemporary dance settings in Montrealare pre-
sented. Although aesthetic and ethical issues based on grounded
empirical material in dance are discussed, many ideas can be transferred
to other performing art forms. Vocation or passion for the arts is often
presented as an explanation for why artists jeopardize their health and
accept difficult working conditions. However, these studies demonstrate
that the dominant discourse in the performing arts, which values the su-
premacy of artistic work and the surpassing of ones limits, is at the top
of a long chain of decisions that negatively impact artists health. Unless
all participants in the dance milieu, individually and collectively, criti-
cally address the different discourses and their embodied truth, posi-
tive changes in dancers health and wellbeing will remain limited.
Keywords: health; dance; passion; discourse; aesthetic
This article presents some conclusions of my latest research on the socio-

cultural dimensions of health as it relates to the artistic practice of dance.
During the last six years, my research assistants and I have conducted a series
of studies with more than 100 professional dancers, choreographers, re-
hearsal directors, and students in different contemporary dance settings in
Montreal (Fortin 2008). We conducted individual interviews and engaged in
participant observations of dance training and rehearsals, which provided us
with rich, grounded material that we analyzed qualitatively. We also obtained
data from self-report questionnaires, which we analyzed quantitatively.
Moreover, we initiated an action research to deepen our understanding of
how dancers conceptualize their health and the impact on their performing
practices and vice versa. Although my research pertains to dance, I believe
that many ideas can be transferred to other performing art forms.
Adopting the epistemological claim that individuals construct reality,
my research is part of a growing body of literature on how diverse cultural
groups conceptualize their health differently. More specifically, I look at how
artists construct themselves and are constructed in many ways by various,
and sometimes competing, discourses. According to Foucault (1963), dis-
courses are composed of ideas, attitudes, beliefs, courses of action, and prac-
tices that enable, as much as they constrain, what can be said or done in
particular times and places. Discourses construct current truths and what
power relations they carry with them.
For example, the dominant Western health discourse posits individuals as
being primarily responsible for their own health. This discourse implies con-
stant surveillance and auto-surveillance, encouraging individuals to avoid risk
and to adopt safe behaviors such as not smoking or eating junk food. Critics
of this discourse say that personal body maintenance is promoted to reduce
the economic burden on public services, and ultimately, it undermines social
responsibility to improve health and wellbeing. Assuming that individuals
will act in their own self-interest is also problematic since individual health is
often subject to external forces (Crawford 2006).
MAIN CONTRIBUTION
A delicate balance between dominant discourses
This idea is corroborated by dance research that shows how dancers privilege
the integrity of the art form over their own safety. Whereas ballet and con-
temporary dance can differ in their underlying assumption about the body,
the contemporary dancers in our sampling tended to align themselves with
the dominant discourse of Western theatrical dance that promotes an ideal
body where aesthetic criteria of beauty, slimness, virtuosity, devotion, and
asceticism prevail, and where fatigue, pain, and injury are accepted silently
(Laws 2005, Sorignet 2004, Wainwright et al. 2005).
When questioned about the relationship between health and dance, per-
forming artists in our research relegated the former to a backstage role,
valuing pushing the limits of the art form and expanding individual skills
sometimes beyond reasonable limits. The artists we interviewed regarded
dance as a risky endeavor, since dance-making involves seeking innova-
tion, and this invariably damages the body. Thus, we concluded that danc-
ers appropriated certain elements of both the contemporary dance discourse,
which assumes a certain risk associated with innovation, and the dominant
health discourse, which preaches physical self-discipline and the avoidance of
risks.
Individual responsibility and spread-out guilt
Without exception, all the artists considered themselves as healthy but dam-
aged. They did not feel that their current and past injuries conflicted with
their sense of being healthy, since being healthy means being functional
that is, being able to respond to the choreographic demands. To this end,
dancers employed various strategies to prepare, protect, and repair their
body. They attributed failure to themselves and consequently felt guilty, tell-
ing themselves they could have had better concentration, warmed-up or
trained more, stopped dancing earlier, eaten better, slept more, etc.
Confusion and uncertainty regarding the effectiveness of different options
sometimes engendered what seems to be an endless search for ways to heal
injuries and to relieve pain and psychological distress.
Individual guilt has been related to the dominant health discourse, which
is driven by a focus on the individuals responsibility and which minimizes
socio-economic and political factors (Lupton 1995). Although some dancers
spoke out against low income, unsecure employment, and difficult work con-
ditions (e.g. additional rehearsals before shows, freelancers fatigue from
holding down two jobs, inadequate studios, tight schedules aimed at mini-
mizing touring costs, etc.), they rarely acted to counter these issues, which is
not surprising in a competitive and small work environment. In this context,
the likely result of complying with the dominant health discourse is that col-
lective mobilization of dancers around health issues is restricted.
Divided about unionization
In Qubec, most dancers are freelancers. Bringing them into a worker union
was initiated ten years ago. Some individuals regarded union contracts as a
way to improve working conditions and better health for dancers. For others,
the legislative framework constituted a threat because, for them, it had a sti-
fling effect on the creative processes which need to come out of disorder,
disequilibrium, and instability. These artists revealed a profound attachment
to this dominant artistic discourse that they considered to be incompatible
with their perception of regular disciplined and ordered workplace environ-
ments. As such, they were reluctant to take on the employer-employee eti-
quette, and they did not like the term cultural industry because their
motivation to dance is aesthetic rather than economic. Choreographers did
not like to think of themselves as employers, although all of them are leading
their own dance companies. Many mentioned preferring to be called creators
rather than choreographers.
The process of unionizing dancers indirectly summons the dance com-
munity to take a critical attitude to the concept of individual responsibility for
health, and calls into question such firmly-grounded dance milieu values as
the normalizing of pain and the acceptance of precarious working conditions
as being commonplace. Dance artists narratives reflected a low propensity to
push collectively for change, which might encourage governments and agen-
cies to keep investing so little in the arts, and this eventually has conse-
quences on artists health. In our opinion, to improve the state of health of
dance artists there must be a critical analysis of the cultural, social, and eco-
nomic contexts in which dance is situated.
Money and health: Myth and reality
In our interviews, many addressed the issue of the financial health of the
dance milieu, which in Canada (as in many other countries probably) suffers
from chronic under-financing. But can it really be said that increasing fund-
ing to dance companies would directly improve dancers health? On one
hand, yes, since the equation money=health is verifiably true. For example,
all the artists raised the negative impact on their health of substandard re-
hearsal studios and theatres. On the other hand, however, no, because inju-
ries and suffering are not solely due to financial conditions. Aesthetic choices
often take precedence over health considerations. From our analysis, we con-
cluded that having more funds does not necessarily mean that more is spent
on health.
Worthy of note is the case of a choreographer whose assumption of re-
sponsibility for his dancers health was shown in the partial financial support
he provided for their training and therapeutic treatments. The same person,
however, disavowed responsibility by pushing his dancers to extreme limits,
leaving the protection of dancers health up to the rehearsal director and the
dancers themselves. He confided to us: the choreographers intention is to
present something striking on stage. I require dancers who can impel this
idea, and there are no limits to this. As the choreographer I have all the
rights. This is an attitude that even this choreographer considers extreme
and problematic. To us, dance development cannot rest on the acceptance of
excess, unless specific occupational health and safety conditions are imple-
mented. Improved financing is one of the means for preserving both the
dancers health and the choreographers vision; but what is also needed is a
critical examination of the dominant artistic discourse which overvalues the

individual artist exceeding his/her personal limits. Clearly, changes require
that sufficient financial resources be made available; nonetheless, many
transformational strategies depend more on relationship skills or involve
aesthetic compromises that all concerned are unwilling to accept.
Methods of creation as central to aesthetic and health issues
In their creative work, many choreographers asked the collaboration of the

dancers who actively participated in the development of the dance. We asked
if this supports the dancers health and wellbeing. One choreographers point
of view was: I work based on the actual state of the dancers bodies; the
movements are not imposed on them from without. I always tell them that it
is they who must choose their tortures [laughing]. In certain cases, it seems
likely that creative collaborative method means a sharing of health responsi-
bilities between choreographer and dancers. However, often there is a gap
between the egalitarian intentions of the choreographer and the actions that
should, in principle, follow. It must be noted in defense of the choreographers
that sometimes dancers themselves willingly submit to situations that nega-
tively impact their health. As a result of a long and complex process of inte-
grating the dominant discourse in dance, artists have internalized the ways of
thinking, seeing, and acting that lead to an overarching devotion to the art
form. As Rannou and Roharik (2006) note, collaborative methods of creation
may often be even considered as the ultimate form of self sacrifice (p. 181),
which is something artists value in conformity with the dominant dance dis-
course.
Another example of a problematic approach to choreography was re-
vealed in those instances when the creative process involved intentionally
fatiguing the dancers in order to have them reach a state of vulnerability and
exhaustion. Some choreographers and dancers believed that this helped to
bring the dancers subconsciousness to the fore, thus providing richer mate-
rial for the piece. However, this strategy was not always explicitly stated, and
agreed upon by everyone. For health concerns to become an integral part of
collaborative type work, artists need to develop better communication skills.
Silence as a risk factor
In dance, women outnumber men, which results in a competitiveness affect-

ing all aspects of their professional life. Often women have had a longer so-
cialization process than men and they have deeply embodied a culture of
silence. Playing down the experience of pain becomes an implicit norm in a
professional dance career, since expressing pain may have negative conse-
quences. The following comments of an experienced dancer speak directly to
this issue:
At the outset, the choreographer accepted or seemed to accept the situa-

tion [injury during the creative process]. After two weeks, he became
pretty angry and made a big fuss about it. I started to say Ill be able to
do it later, the injury will have healed, but right now I cant do it. I
wanted to explain, but he wouldnt listen to anything from me. The cho-
reographers attitude is really important. Its a constant struggle to have
the freedom to speak your mind. Its not easy because I get sidelined.
Even with all my experience. Emotionally its very difficult.
Our interviews with both choreographers and dancers suggested a virtual

blacklist of choreographers to avoid if a dancer values his/her health. But no
one revealed their names; the milieu is small, and secrets remain kept. As
mentioned, some choreographers profess to be very open-minded, and de-
plore the fact that sometimes it is the dancers who do not want to disclose
their health conditions, but other choreographers maintain an absolute aes-
thetic bias and want to know as little as possible about their dancers injuries.
This is not inherently a problem when the responsibility is passed on to the
rehearsal director, who has been described unanimously by both choreogra-
phers and dancers as being the critical outside eye essential to the work and
the health of the dancers.
All in the name of art
Choreographers, driven by the vision of their gestating choreography, are the

first to admit being more attuned to their artistic vision than to the physical
needs of the dancers. The idea that a choreography takes on a life of its own
and that it is the choreography itself that indicates the necessary creative de-
cisions, could often be discerned from choreographers comments. I am God
when I am creating, said one choreographer. My role is to create a world in
my own image. There are no limits to this. I tell my dancers that I, as the cho-
reographer, have all the rights. I may ask them to jump from the fourth floor
window, and if theyre crazy enough to do it, thats their problem; but that
wont stop me from telling them to do it. Such a view of art as an egotistical
sacred endeavor provides justifications for the aesthetic decisions. Giving
predominance to the dance being created and according a messianic status to
its creator has repercussions on the health of dance artists. In an ideal situa-
tion, both dancers and choreographers share the satisfaction of surpassing

physical and psychological limits. It is not our intention to create a tension
between choreographers and dancers, but the fact remains that it is the
dancer who, to a greater extent, suffers the physical consequences of the risks,
while it is the choreographer who receives the recognition for the works
originality.
It is worth mentioning that for choreographers and dancers, dedication to
their art form can involve both alienation and exhilaration. In our research
we did not focus on the latter, although there was touching commentary
about how the making of dances enriches ones life. In one of our studies, we
assessed the degree to which dance is part of the persons identity, and as
Wainwright et al. (2005) posit, we can say, too, that dance is not just some-
thing that you do, in a very deep sense it is who you are (p. 49).
Health and ethical dimensions
That dancers give life to the choreography is an idea often expressed in the
dance literature; less so is the notion that choreographers participate in
molding dancers bodies (Dantas 2008). During our interviews one choreog-
rapher remarked: I constantly put my dancers in an uncomfortable situa-
tion; so where is the boundary between making an effort and going too far?
Between injuring ones body and being violent to oneself? The question of
exceeding the limits perhaps takes on lesser significance in other art forms
because in dance it is the dancers body that is at stake. As a dancer put it, a
painter will look at his canvas and wonder if the good colors are correct, but
the canvas will not ache the way we dancers do.
Interestingly, a denunciation of human injustice forms the subtext of
many choreographies, but in our observations of different creative processes,
we sometimes noted a lack of consideration for the health and wellbeing of
the performing artists who give tangible form to the work. Jackson (2008)
explains that there is a tendency in the dance world to correlate good prod-
ucts (choreography) with good processes and good people (choreographers,
dancers, rehearsal directors), as if no dichotomy could exist between the
moral dimensions of a choreography and the people that engendered it. If
someone appears to have been mistreated, or if an artist is shown to have
acted immorally, this is viewed as an exception or as necessary for the sake of
art. Health poses a huge challenge for choreographers who have the pressure
to create original work that meets the assessment criteria of grant agencies.
Can health advocates expect that one day grant agencies will require dance
companies to deal comprehensively with aspects of occupational health and
safety, or that soon dance schools and dance organizations will seriously en-
gage in health-promotion? I hope so because, if the dance profession does not
rise to the challenge, it may come to be regarded as irrelevant by education-
ists, decision makers, and the public alike.
Harmonious or obsessive passion
In the dance literature, passion and a sense of calling is frequently used to

explain why dancers accept the physical suffering or psychological distress
they often experience (Turner and Wainwright 2003). Being on stage re-
quires many sacrifices, such as dancing with an injury, as an experienced
dancer explained. Doing this makes no sense at all, but you dont think about
it. Great passion is involved. But from our results, it seems that, more so
than passion, it is the dominant artistic discourse that drives a long sequence
of health-related decisions.
In our analysis, we differentiated the dominant artistic discourse from the
passion with which artists endorse this dominant discourse. In a quantitative
study with pre-professional dancers, we looked at whether dancers exhibit
distinct injury profiles as a function of their passion for dance. We distin-
guished harmonious passionwhich occupies a significant place in ones
identity but is not overwhelming and leaves room for other life pursuits
from obsessive passion, which consumes ones thought and overwhelms ones
identity. We found that obsessive passion for dance was associated with
dance-related injuries and prolonged suffering from chronic injuries, whereas
harmonious passion is associated with sustained and healthful involvement
in dance (Rip et al. 2006). Differentiating harmonious and obsessive passion
could contribute to envisage health in dance as no longer being a threat to the
integrity of the art form. My intention is not to reject the dominant dance
discourse, which would be foolish and unrealistic, but to invite artists to think
critically about the benefits or the cost of participating in different discourses
that inevitably have consequences on their health. Because dancers are de-
voted to and passionate about their art form, and because the dance milieu
features a culture of pain in which dancers are expected to suffer in silence,
helping artists to understand that their passion can be experienced in a man-
ner not detrimental to their health should be a goal of educational programs
informed by research.
Somatic education for better or for worse
Questioned about the younger generation of artists, certain older choreogra-

phers somewhat caustically raised the issue of self-absorbed children lacking
discipline, passion and the will to achieve. They spoke of learning/training

environments that left dancers ill-prepared for the milieus harsh realities,
while others brought up the improved preparation acquired through the in-
creasing number of somatic and anatomic classes.
At one point, I was skeptical. I felt like saying to the dance schools give
more technique classes to your dancers. I was worried that I would have
work with dancers who are too preoccupied by their anatomy, who know
that it is a particular muscle which is sore, who know they shouldnt force
another one [laughing]. But now I see that the young dancers work so
well. I would have loved to have had such knowledge of my body.
Originating from outside the field of dance, a variety of somatic education

practices such as Alexander Technique, Body Mind Centering, or Feldenkrais
Method, have made their way into dance. They are all concerned with lis-
tening to the body and responding to these sensations by consciously altering
movement habits and movement choices (Eddy 2009, p. 7). According to
Shusterman (1992), increasing subjective sensorial experience can help to
reduce the prominence of external agencies on the individual. Somatic edu-
cation thus provides dancers with an alternative discourse to counteract the
fantasy of an ideal body, which is so often removed from the concreteness of
the sensing body.
In our study, we found that dancers integrate the somatic discourse in
three distinctive ways. For one group of dancers, the heightened perceptive-
ness developed through somatic education did not serve to improve their
wellbeing; it was subverted and used to work toward what was important to
themnamely, pushing the limits of their performance. They had fully ap-
propriated the dominant dance discourse and considered it as inescapable
and even essential to building a dance career. A second group of dancers, on
one hand, manifested critical thinking in how they verbalized their reticence
about certain aspects of the dominant discourse; on the other hand, they did
not seem to experience physically the changes that they professed verbally. In
a third group, the experience of somatic education allowed dancers to develop
an internal authority. They made choices based on their intimate self-aware-
ness, respecting the limits of their own bodies. For these dancers, the nor-
mality of pain or the excessiveness of certain artistic practices were no longer
so blindly tolerated or, if so, only under certain conditions and for a limited
period of time.
Markula (2004) appropriately points out that any discourse can be either
emancipatory or oppressive. A somatic-based discourse is no exception, and
this needs to be looked at critically. What really is important when negotiat-

ing between different discourses is knowing how they operate, each having its
own game of truth linked to an accepted consensus about what is sound
knowledge and to the accompanying hegemonic procedures that legitimatize
power relations. It is worth mentioning that power, according to Foucault
(1963), is not exercised through the imposition of external constraints on the
person, but rather is achieved by an internalization of norms by the person.
IMPLICATIONS
Differentiated ways of integrating various discoursesdance or health, domi-

nant or marginalemerged over the course of our studies with dance artists.
Overall our data revealed how:
Artists have integrated the dominant Western health discourse of individ-

ual responsibility and have minimized the importance of external factors
shaping their health.
Health is mostly defined in functional terms with a reluctance to question
the hegemonic norm of pain and suffering of the dominant discourse of
Western theatrical dance.
The process of unionization confronts the dominant dance discourse by
prompting collective mobilization for better occupational health and
safety conditions.
An increase in the arts budget does not necessarily improve artists health.
Methods of creation based on collaboration do not necessarily contribute
to securing health.
A culture of silence prevails in dance, which we consider a risk factor.
Women deal with specific gender challenges impacting their health.
Health accountability is weakened when the artwork prevails over the
artists who give life to the artwork.
Health issues can play a central role in the promotion of shared notions of
ethical behaviors.
Artists participate in the dominant discourse in dance with different types
of passion that have different effects on dance injuries.
Somatic education questions the dominant discourse in dance by lessen-
ing its negative impacts on dancers health.
The above conclusions indicate how important it is for dancers to raise

their consciousness about the competing discourses surrounding them and
the effects on their health and wellbeing. Whereas vocation or passion for the
art form is often presented as an explanation for why artists jeopardize their
health and accept difficult working conditions, we conclude that it is the
dominant performing arts discourse, which values the supremacy of the ar-
tistic work and the surpassing of ones limits, that heads a long sequence of
decisions negatively impacting artists health. Unless all participants in the
dance milieu, individually and collectively, critically address the different
discourses and their embodied truths, changes in dancers health and well-
being will remain limited. Games of truth are unavoidable, but they can be
played out with an ethically-based concern for oneself and for others.
Acknowledgments
I express my heartfelt thanks to my co-researchers Catherine Cyr, Genevive Rail, Pam-

ela Newell, Karen Messing, Martyne Tremblay, Sylvie Trudelle, Robert Vallerand, Adri-
ane Vieira, Blanka Rip, and all the artists who participated in the studies. The research
was supported by a grant from the Social Science and Humanities Research Council of
Canada. Financial help also came from the Services aux collectivits and the Depart-
ment of Dance of the University of Qubec at Montral.
Sylvie Fortin, Department of Dance, University of Qubec at Montral, Case Postale

8888, succursale Centre-Ville, Montral, Qubec H2V 3H7, Canada; Email: fortin.sylvie
@uqam.ca
References
Crawford R. (2006). Health as a meaningful social practice. Health: An Interdiscipli-

nary Journal for the Social Study of Health, Illness and Medicine, 10, pp. 401-420.
Dantas M. (2008). Ce dont sont Faits les Corps Anthropophages: La Participation des
Danseurs la Mise en uvre Chorgraphique Comme Facteurs de Construction
de Corps Dansants chez Deux Chorgraphes Brsiliens. Unpublished doctoral the-
sis, University of Qubec at Montral.
Eddy M. (2009). A brief history of somatic practices and dance: Historical development
of the field of somatic education and its relationship to dance. Journal of Dance
and Somatic Practices, 1, pp. 5-27.
Fortin S. (2008). Danse et Sant. Qubec: Presses de lUniversit du Qubec.
Foucault M. (1963). La Naissance de la Clinique. Paris: Presses universitaires de
France.
Jackson N. and Phim T. S. (2008). Dance, Human Rights and Social Justice. Lanham,
Maryland, USA: Scarecrow.
Laws H. (2005). Fit to Dance 2. London: Dance UK Books.

Lupton D. (1995). The Imperative of Health. London: Sage Publications.
Markula P. (2004). Tuning into ones self: Foucaults technologies of the self and
mindful fitness. Sociology of Sport Journal, 21, pp. 302-321.
Rannou J. and Roharik I. (2006). Les Danseurs. Paris: La documentation franaise.
Rip B., Fortin S., and Vallerand R. J. (2006). The relationship between passion and
injury in dance students. Journal of Dance Medicine and Science, 10, pp.14-20.
Sorignet P. E. (2006). Danser au-del de la douleur. Actes de la Recherche en Sciences
Sociales, 163, pp. 46-61.
Turner B. S. and Wainwright S. P. (2003). Corps de ballet: The case of the injured ballet
dancer. Sociology of Health and Illness, 25, pp. 269-288.
Shusterman R. (1992). L'Art ltat Vif. Paris: Les ditions de minuit.
Wainwright S. P., Williams C., and Turner B. S. (2005). Fractured identities: Injury and
the balletic body. Health: An Interdisciplinary Journal for the Social Study of
Health, Illness and Medicine, 9, pp. 49-66.
Author index
Ackermann, B. 267, 487 Carter, C. 159

Aggett, C. 127 Carvalheiro, S. 179
Ainley, M. 515 Carvalho, S. 255, 559
Akinaga, S. 77 Chaffin, R. 57, 315
Allen, S. E. 217 Clark, T. 97, 573
Almeida, L. S. 139 Clemente, M. P. 501
Altenmller, E. 103, 187, 375, 381, Coimbra, D. 133, 349, 501
387, 389 Collyer, S. 153
Anderson, D. 203 Cormack, J. 25, 31
Angioi, M. 543 Correia, J. S. 547
Arajo, L. S. 139 Crawford, M. 57
Atkins, L. 219 Cruz, J. F. A. 139
Backus, B. C. 225, 459 da Silva, C. F. 349
Bailey, N. J. 461, 583, 585 Dance, S. 459
Bajalica, M. 231 DeClerk, J. 439
Baker, F. 301 Demos, A. 57
Barbosa, A. 501 de Lisle, R. 395
Bangert, D. 403 Dodd, G. 591
Bautista, A. 321 Domenici, C. 451
Barlow, C. 51 dos Santos, R. A. T. 261, 451
Baur, V. 389 Driscoll, T. 267
Bennett, D. 523 Emura, N. 77
Bennetts, R. 359 Ericsson, K. A. 11
Beraut, C. 579 Erskine, S. 515
Bigand, E. 237, 579 Esteban Muoz, E. 409
Bonini Baraldi, F. 237 Evans, Al. 267
Boullet, L. 389 Evans, An. 507
Braga, A. C. 133 Fabian, D. 147
Bravo, A. 243 Ferrari, V. N. 273
Bressman, S. B. 375 Fine, P. 51, 167, 243, 279
Broughton, M. 531 Flossmann, S. 119
Buck, B. 585 Fortin, S. 599
Bull, S. 167 Fortune, J. 487
Burt-Perkins, R. 85, 249 Francis, P. R. 339
Buttkus, F. 389 Friedlander, K. 279
Camp, M.-A. 107 Fujisawa, T. 285
Furuya, S. 339 Lim, V. K. 371
Gaspar, V. 291 Livingstone, S. R. 445
Geeves, A. 415 Lisboa, T. 315, 423
Gerling, C. C. 451 Lockett, D. 231
Ginsborg, J. 51, 85, 173, 481 Loehr, J. D. 159, 445
Goebl, W. 65, 119 Logan, T. 305
Gomez-Pellin, M. 389 Lpez, G. 321
Goto, Y. 293 Loureno, S. 501
Granert, O. 387 MacDonald, R. A. R. 45
Gualda, F. 423 Mackie, C. 113
Hallam, S. 133, 327 MacRitchie, J. 583, 585
Halton, R. 295 Marn Oller, C. 327
Hamm, J. P. 371 Marinho, H. 255, 559
Hagenah, J. 375 Martin, R. 25, 31, 333
Hair, G. 461 McAllister, M. 583
Harper, N. L. 295 McIlwain, D. 415
Henriques, T. 295 McPherson, G. E. 469, 475
Hentschke, L. 261 Mendona, G. 537
Herrojo Ruiz, M. 187, 381 Metsios, G. 543
Hildebrandt, W. 85 Mioshi, E. 301
Hirano, T. 347, 429 Mitchell, H. F. 45
Hodges, J. R. 301 Miura, M. 77, 285, 431
Hsieh, S. 301 Molin, P. 71
Irvine, S. 91 Morita, S. 77
Iwami, N. 285 Mnchau, A. 375
Jabusch, H.-C. 103, 187, 375, 381, Nakahara, H. 339, 347, 429
387, 387 Neskovic, B. 359
Kenny, D. 25, 31, 37, 487 Nuti, G. 341
Kilchenmann, L. 107 Obata, S. 307, 339, 347, 429
Kinoshita, H. 307, 339, 347, 429 Quin, E. 91
Kinou, M. 285, 431 Palmer, C. 65, 159, 359, 445
Kirk, I. J. 371 Peller, M. 387
Klein, C. 375 Pereira, F. 537
Kochman, K. 439 Prez Echeverra, P. 327
Koopmans, E. 159 Pertzborn, F. 133
Koutedakis, Y. 543 Phillips, M. 553
Kubota, Y. 309 Piguet, 301
Lawson, C. 463 Pinho, J. C. 501
Leman, M. 439 Pollitt, J. 523
Lester, G. 173 Pozo, J. I. 321
Pozzo, T. 237, 579 Stevens, C. 173, 531
Pullinger, S. 583 Swayne, L. 461
Rafferty, S. 91 Thompson, J. 395
Redding, E. 91 Thompson, W. F. 359
Reissis, N. 365 Tibbles, J. 463
Renwick, J. M. 469 Trouli, H. 365
Rodrigues, L. X. 179 Twitchett, E. A. 543
Sanders-Pullman, R. 375 Wanderley, M. 159
Santos-Luiz, C. 349 Wasley, D. 85
Schmidt, A. 375 Widmer, G. 119
Schubert, E. 445 Williamon, A. 85, 97, 225, 341, 459,
Senn, O. 107 573
Serra-Dawa, S. 355 Winston, R. 5
Siebner, H. 387 Wu, C. C. 371
Snook, B. 521 Wyon, M. 543
Solomon, A. 341 Xarez, L. 537
Speedy, D. 395 Yanagida, M. 77
Stambaugh, L. A. 567 Yasui, N. 431
Steinmetz, A. 495 Zepidou, G. 459

ISPS2009 Proceedings PDF

Uploaded by

Copyright:

Available Formats

ISPS2009 Proceedings PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ISPS2009 Proceedings PDF

Uploaded by

Copyright:

Available Formats

International Symposium on Performance Science The Author 2009, Published by the AEC

ISBN 978-94-90306-01-4 All rights reserved

AEC, PO Box 805, NL-3500 AV Utrecht, The Netherlands

Published worldwide in The Netherlands by the

The AEC is a European network of institutions in higher music education

Copyright AEC 2009

Cover design by Christopher Tomlin

Preface and acknowledgments

The second International Symposium on Performance Science, ISPS 2009,

Scientific committee page ix

TUESDAY, 15 DECEMBER 2009

WEDNESDAY, 16 DECEMBER 2009

THURSDAY, 17 DECEMBER 2009

FRIDAY, 18 DECEMBER 2009

Author index page 611

Aaron Williamon, co-chair

Performance, science, and society

Imperial College London, UK

The evaluation of music and musicality is of growing interest to scien-

Keywords: brain imaging; physiology; physics; musical perception; in-

Address for correspondence

Discovering deliberate practice activities that

Department of Psychology, Florida State University, USA

Since Sir Francis Galtons book on Hereditary Genius, many scientists

Keywords: deliberate practice; expert performance; innate talent; cogni-

Everyone has experienced the excitement associated with being introduced to

increase the performance further. This assumed inability to influence per-

Figure 1. An illustration of the qualitative difference between the course of improve-

In most domains of professional and leisure activity, the majority of people

of improving some measurable aspect of their performance. For example, a

Overcoming plateaus in chess

ual and extended increases in performance in a wide range of domains, such

Deliberate practice in typing

In sum, the superior speed of reactions by expert performers, such as

Deliberate practice in other domains

In many domains performers and teachers have accumulated knowledge

Address for correspondence

K. Anders Ericsson, Department of Psychology, Florida State University, Campus Box

Abernethy B. (1991). Visual search strategies and decision-making in sport. Interna-

Suffering for ones art: Performance related

Dianna T. Kenny1,2, Justine Cormack2, and Rosemary Martin2

1 Australian Centre for Applied Research in Music Performance,

The study reports the results of an intake questionnaire assessing point

Keywords: performance related musculoskeletal injury; musicians;

Performance related musculoskeletal variables: Performance related mus-

PRMD frequency: assessed on a 10-point Likert scale (0=never, 10=daily)

N Mean SD Max Min F Sig

pel tunnel syndrome, tendonitis, tenosynovitis, and muscle tightness,

N Mean SD Max Min F Sig

Findings confirmed high injury rates and PRMD prevalence of performing

Address for correspondence

Practicing perfection: The physical costs of

Dianna T. Kenny1,2, Rosemary Martin2, and Justine Cormack2

1 Australian Centre for Applied Research in Music Performance,

Despite the many physical demands involved in practice in music and

Keywords: practice; practice routines; musicians; dancers; performance

A comprehensive questionnaire was developed to explore the relationship

Descriptives, practice variables: Music students practiced, on average, 5.83

on average, 156 minutes (SD=73.67, range=45-360 mins) per day, compared