Behavioral Neuroscience

2006, Vol. 120, No. 6, 12111217

Copyright 2006 by the American Psychological Association

0735-7044/06/$12.00 DOI: 10.1037/0735-7044.120.6.1211

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Illusory Changes in Head Position Induced by Neck Muscle Vibration Can

Alter the Perception of Elbow Position
Joanna Knox

Paul Cordo

University of Queensland

Oregon Health & Science University

Rachel Skoss

Shane Durrant

Curtin University of Technology

Oregon Health & Science University

Paul Hodges
University of Queensland
Acuity for elbow joint position sense (JPS) is reduced when head position is modified. Movement of the
head is associated with biomechanical changes in the neck and shoulder musculoskeletal system, which
may explain changes in elbow JPS. The present study aimed to determine whether elbow JPS is also
influenced by illusory changes in head position. Simultaneous vibration of sternocleidomastoid (SCM)
and the contralateral splenius was applied to 14 healthy adult human subjects. Muscle vibration or passive
head rotation was introduced between presentation and reproduction of a target elbow position. Ten out
of 14 subjects reported illusions consistent with lengthening of the vibrated muscles. In these 10 subjects,
absolute error for elbow JPS increased with left SCM/right splenius vibration but not with right SCM/left
splenius vibration. Absolute error also increased with right rotation, with a trend for increased error with
left rotation. These results demonstrated that both actual and illusory changes in head position are
associated with diminished acuity for elbow JPS, suggesting that the influence of head position on upper
limb JPS depends, at least partially, on perceived head position.
Keywords: joint position sense, proprioception, upper limb

However, it is possible that previously reported errors in locating the elbow following changes in head position (Knox &
Hodges, 2005) are a consequence of local biomechanical effects of
the head movement or are due to reflex changes in muscle activity.
For instance, muscles such as trapezius and levator scapulae can
exert torque directly on the neck and upper limbs, so changes in
head position and posture could affect the position of the arm,
thereby introducing error in an arm proprioception task. Moreover,
changes in head position may also affect the ability to develop
torque around the elbow, possibly because of tonic neck reflexes
(Deutsch, Kilani, Moustafa, Hamilton, & Hebert, 1987; Kasai,
1991). Kasai (1991) reported errors in the reproduction of a right
elbow position toward extension with left head rotation, consistent
with the effect of tonic neck reflexes. An alternate way to investigate the relationship between head position and upper limb
position sense, while avoiding the associated biomechanical or
reflex effects, is to use illusory movements of the head. Alterations
in the perceived position of the head using the return phenomenon
have been shown to introduce deviations in straight ahead pointing
movements (Guerraz, Navarro, Ferrero, Cremieux, & Blouin,
2006; Mars, Honore, Richard, & Coquery, 1998), however this
method involves sustained rotation of the head. Although we have
previously shown that illusory head movements induced with
galvanic vestibular stimulation (GVS), in the absence of any head
rotation, are associated with error in the reproduction of an elbow
position (Knox, Coppieters, & Hodges, 2006), it is not possible to

Both the accuracy of pointing to a remembered target (Berger et

al., 1998; Fookson et al., 1994) and the acuity of elbow proprioception (Kasai, 1991; Knox et al., 2006) have been shown to be
affected by changes in head position. One explanation for these
parallel observations is that the central nervous system (CNS)
relies on accurate sensory information about the position of the
head to interpret the positions of upper limb segments (Knox &
Hodges, 2005; Roll, Roll, & Velay, 1991). Thus, alterations of
head position during an upper limb task could affect the interpretation of arm position and misdirect the movement.

Joanna Knox and Paul Hodges, Division of Psychotherapy, University

of Queensland, Brisbane, Australia; Paul Cordo and Shane Durrant, Neurological Sciences Institute, Oregon Health & Science University; Rachel
Skoss, School of Physiotherapy, Curtin University of Technology, Perth,
Australia.
Funding for Joanna Knox was provided by a University of Queensland
Graduate School Research Travel Award. Paul Cordo is supported by
Grant AR31017 from the National Institutes of Health. Paul Hodges is
supported by the National Health and Medical Research Council of Australia. We would like to thank Chad Smith, Mark Chapman, and Victor
Gurfinkel for their assistance with experimental planning, preparation of
equipment, and data collection.
Correspondence concerning this article should be addressed to Paul
Hodges, Division of Physiotherapy, The University of Queensland QLD,
4072 Australia. E-mail: p.hodges@uq.edu.au
1211

KNOX, CORDO, SKOSS, DURRANT, AND HODGES

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

1212

exclude the possible effect of vestibular reflexes. To further explore the mechanisms underlying errors in elbow proprioception
that occur with changes in head position, we used neck muscle
vibration to introduce illusory changes in head position without an
actual change in head position.
Muscle vibration was used to induce illusory rotation of the
head in seated subjects. Mechanical vibration over a muscle belly
excites muscle spindles and the associated afferent nerves (Brown,
Engberg, & Matthews, 1967), which is perceived by the CNS as
joint rotation and movement (Goodwin, McCloskey, & Matthews,
1972), as if the vibrated muscles were lengthening, provided the
vibration rate is greater than 30 Hz (Cordo et al., 1995). Although
vibration has been frequently used to induce movement illusions in
the limb muscles, few previous studies have documented kinesthetic illusions of head movement as a result of neck muscle
vibration (Gurfinkel & Levik, 1991; Roll, Vedel, & Roll, 1989;
Taylor & McCloskey, 1991). Taylor and McCloskey (1991) used
concurrent vibration of sternocleidomastoidus (SCM) and the contralateral splenius to achieve the strongest illusion of head rotation.
The aim of the present study was to evaluate the effect of illusory
changes in head position, induced by mechanical vibration of the
neck muscles, on the reproduction of elbow joint position. We
included conditions to assess neck muscle vibration with the head
fixed in a neutral position and with the head free to move around
a vertical axis. Both conditions were included because greater
illusions are induced in the limbs when vibration is applied to a
restrained limb (Rabin & Gordon, 2004), but this has not been
investigated with vibration to the neck muscles.

Method
Fourteen healthy human subjects, 8 male and 6 female ranging in age
from 21 to 45 years, participated in this study. Twelve of the subjects were
right hand dominant and two were left hand dominant. None of the subjects
had a history of any neuromuscular disorders affecting the cervical spine or
upper limb. Prior to inclusion in the study, each subject provided informed
consent following the procedures of the Oregon Health & Science University Institutional Review Board.

Experimental Setup
Subjects performed a proprioceptive task with the elbow while seated in
a comfortable chair and with the elbow resting on a foam cushion. The
forearm rested in a U-shaped cuff, and the hand gripped a handle (Figure
1). The shoulder was abducted 50, the elbow was free to move through
its normal range, and the wrist was held in a midrange position. Beneath
the table, a DC motor (Cleveland Motion Controls, 3509C, Cleveland, OH)
controlled the angular position of the elbow via a manipulandum. The
subject wore a helmet, which was attached to a head-fixation device that
could be either secured to prevent head movement (head fixed) or released
to allow rotation of the head around the vertical axis (head unconstrained).
When fixed, the head faced forward.
Vibration was applied to the neck muscles with the eyes closed to induce
the illusion of head rotation. Custom-built vibrators (frequency: 59 64 Hz,
amplitude: 23 mm), mounted on an aluminium shoulder harness, were
positioned bilaterally and symmetrically over the SCM and splenius muscles, between the levels of the C5 and C7 spinous processes (Figure 1a).
Each vibrator consisted of a DC motor (Maxon Motor M98748, Burlingame, CA) with a cam mounted at the end of the motor shaft to deliver
mechanical pulses via a probe. The probe, constructed of nylon, was
mounted at the end of a stainless steel rod that was orthogonally oriented
with respect to the motor (Figure 1b). Vibration was applied simulta-

head fixation device

splenius

C6
vibrators

sternocleidomastoid

shoulder harness

manipulandum

c
start
rest
target

Figure 1. Experimental setup. The subject was seated in a comfortable

chair with the head fixed at a midline position. The right arm was placed
in a manipulandum to measure elbow position. Four vibrators were
mounted on a custom-made shoulder harness. Inset a shows a cross section
through the level of C6 to indicate the position of the vibrators. Inset b
shows the design of the vibrators. Inset c indicates the start and target
positions (solid lines). The rest positions were either the same as the start
position or located at 3, 6, or 9 from the start position (dashed lines).

neously to either the right or left SCM and the contralateral splenius. Pilot
trials indicated that the most consistent illusion of left rotation was induced
with vibration to the left SCM/right splenius and the most consistent
illusion of right rotation was induced with vibration to the right SCM/left
splenius. To control for any distracting audible noise from the vibrators, we
conducted trials with vibrators positioned over, but not in contact with, the
lateral aspect of the neck. During these control trials, the subjects did not
report illusory head movement. Vibration was applied with the head in the
midline position, either fixed or unconstrained.
Prior to the main experiment, the subjects perception of head position
following passive head rotation and during neck muscle vibration was
assessed. With the eyes closed, the subjects operated a pointer mounted on
a board to indicate the horizontal orientation of the head. Each subject
indicated the position of their head three to five times for each of the
following conditions: (a) head in the midline, (b) head rotated 30 to the
left, (c) head rotated 30 to the right, (d) vibration of the left SCM/right
splenius with the head unconstrained, (e) vibration of the left SCM/right
splenius with the head fixed, (f) vibration of the right SCM/left splenius
with the head unconstrained, and (g) vibration of the right SCM/left
splenius with the head fixed. Smooth head rotation (15/s) was manually
induced via a lever designed to lock at the 30 (left and right) position.

Experimental Procedure
A proprioceptive task was performed that measured the subjects ability
to reproduce a previously presented elbow position. Subjects kept their
eyes closed throughout each elbow proprioceptive trial. The manipulandum
passively moved the subjects elbow from a constant start position (70
flexion) to a constant target position at 25 of flexion relative to the start
position. The elbow movement was not pure flexion, and involved some
shoulder rotation, but was constant for all trials. The target position was
held for 3 s, and the arm was then returned, at 10 s1, to a rest position.
Four rest positions were used, 0, 3, 6, and 9 flexed from the start
position (Figure 1c, dashed lines). After 5 s in the rest position, the
manipulandum released the arm by unclutching (Inertia Dynamics SL30,
Torrington, CT) it from the actuator to allow the subject to actively
reproduce the target position. An optical encoder (HEDS, 5500 E05, US

PERCEPTIONS OF HEAD AND ELBOW BY NECK VIBRATION

a
Head position ()

80

For the trials in which the perception of head position was assessed, the
indicated position of the head following 30 rotation or during muscle
vibration was quantified, in degrees, relative to the indicated position of the
head when it remained at the midline. The absolute error (absolute difference between the target angle and the reproduced angle), constant error
(relative difference between the target angle and the reproduced angle), and
variable error (standard deviation of the constant error) for the JPS task
were measured for each trial as the difference, in degrees, between the
target position and the reproduced position. A positive constant error
represented overshooting the target and a more flexed elbow position,
whereas negative constant error represented undershooting the target and
a more extended elbow position. Trials in which the absolute error was
greater than two times the standard deviation from the mean (3.2% of
trials) were excluded from further analysis as the error in these trials was
not likely to represent true acuity for joint position. Two-way repeated
measures analyses of variance were used to compare error for elbow JPS
between the direction of real or illusory head rotation (three head rotation/
vibration conditions) and between rest positions (four rest positions). Hand
dominance for each subject was included in the analyses of variance as a
categorical covariate. Post hoc comparisons were made with Bonferronis
test. Significance was set at p .05.

Results
Range of Perceived Head Rotation
Following rotation of the head 30 to the left or right, the
subjects indicated an average of 26 left head rotation or 33 right
head rotation, respectively (Figure 2a). When vibration was applied with the head unconstrained, a consistent illusion of head
rotation was perceived in 10 out of 14 subjects. In 9 of these 10
subjects, left SCM/right splenius vibration produced an illusion of
left head rotation with an average extent of 21 (Figure 2b). Right
SCM/left splenius vibration produced an illusion of right rotation
in the same 9 subjects with an average extent of 28 (Figure 2b).
The remaining subject indicated the opposite illusions. When
vibration was applied with the head fixed, only 8 out of 14 subjects
experienced illusions of head rotation. In 5 of the 8 subjects, an
illusion of left rotation was reported with left SCM/right splenius
vibration, and an illusion of right rotation was reported with right
SCM/left splenius vibration (Figure 2c). The remaining 3 subjects
experienced the opposite direction of illusions. With the head
fixed, the average indication of illusory left rotation was 17 and
illusory right rotation was 19.

Perception of Elbow Position Following Head Rotation

When the head was rotated 30 to the left or right, there was an
increase in absolute error for elbow joint position sense (JPS):

33

0
-26
-40

Head position ()

80
40

Neck muscle vibration with the head unconstrained

L SCM/R sp
R SCM/L sp
28

0
-21
-40
-80

Data Analysis

40

Passive head rotation

30 left
30 right

-80

80
Head position ()

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Digital, Vancouver, WA) registered the position of the elbow with a

precision of 0.01. The position of the elbow was first converted to an
analog signal with an electronic circuit and then digitized at 500 Hz by a
data acquisition system (Cambridge Electronic Designs, Power, 1401,
Cambridge, UK).
Each subject reproduced the elbow position under the seven experimental conditions of passive head rotation or neck muscle vibration in which
the perception of head position was assessed. In each condition, the head
was positioned at the midline for the presentation of the target position.
Neck muscle vibration or passive head rotation was applied during the rest
period and maintained while the subject reproduced the target position.
Within each condition, eight trials were conducted in a random order, two
for each rest position.

1213

Neck muscle vibration with the head fixed

L SCM/R sp
R SCM/L sp

40
19
0
-17
-40
-80

Figure 2. Indication of perceived head position. Each panel shows the

head position indicated by the subjects during (a) passive head rotation, (b)
vibration with the head unconstrained, and (c) vibration with the head
fixed. Negative values represent perceived left rotation, dashed lines represent the group average. L SCM/R sp left sternocleidomastoid/right
splenius, R SCM/L sp right sternocleidomastoid/left splenius.

head rotation main effect, F(2, 26) 4.5, p .05, (Figure 3, left
panels). Post hoc tests demonstrated that 30 head rotation to the
right significantly increased absolute error for elbow JPS ( p
.05), but there was no effect with left head rotation ( p .3). Head
rotation also influenced constant error for elbow JPS: head rotation
main effect, F(2, 26) 3.5, p .05. Post hoc tests indicated that
the reproduction of elbow position with head rotation to the right
was more flexed (toward the trunk) than the reproduction of elbow
position in control trials ( p .05). There was no difference in
constant error between head rotation to the left and control ( p
.9). For both directions of head rotation, and trials with no rotation,
there was a tendency to overshoot the target elbow position.
Variable error was not affected by head rotation: head rotation
main effect, F(2, 10) 0.2, p .8. There were no main effects or
interactions for hand dominance ( p .3).

Perception of Elbow Position With Neck Vibration

When the head was unconstrained and all subjects data were
included in the analysis, there was a trend for neck muscle vibration to affect absolute error: vibration main effect, F(2, 24) 2.8,
p .08; and no effect of neck muscle vibration on constant error
for elbow JPS: vibration main effect, F(2, 24) 0.7, p .5; or
variable error for elbow JPS: vibration main effect, F(2, 16)
0.05, p .9. When only the 10 subjects who experienced illusory
head rotation were included in the analysis, vibration with the head
unconstrained increased the absolute error for elbow JPS: vibration
main effect, F(2, 16) 6.2, p .01 (Figure 3, middle panels). In
this analysis, data were grouped according to the direction of
illusory rotation reported by each subject. For example, the 9
subjects who reported illusory left rotation with left SCM/right
splenius vibration and the 1 subject who reported illusory left
rotation with right SCM/left splenius vibration were grouped. Post

KNOX, CORDO, SKOSS, DURRANT, AND HODGES

1214

Absolute error ()

p = 0.3

p < 0.05

p > 0.9

4
3
2
1
0

Constant error ()

p < 0.05

p >0.9

p < 0.05

5
4
3
2
1

Variable error ()

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

0
5
4
3
2
1
0
30 left

Control

30 right

Passive head rotation

Left

Control

Right

Vibration with the head unconstrained

Left

Control

Right

Vibration with the head fixed

Illusory rotation

Figure 3. Errors in elbow joint position sense. The top three panels show changes in absolute error, the middle
panels show constant error, and the bottom panels show variable error with actual and illusory rotation of the
head. The target position is represented by 0 error. The grey fields indicate the 95% confidence intervals for the
control trials.

hoc testing demonstrated that there was an increase in absolute

error for elbow JPS for conditions in which subjects reported
illusory left rotation ( p .05), but no change was seen for
conditions in which subjects reported illusory right rotation ( p
.9). There was no change in constant error for elbow JPS following
neck muscle vibration with the head unconstrained in subjects who
had reported illusory rotation of the head: vibration main effect,
F(2, 16) 2.1, p .2. For variable error, there was an interaction
between the effect of illusory rotation and rest position: Vibration Rest Position, F(6, 30) 3.9, p .005. There was greater
variable error for conditions with illusory right rotation ( p .05)
compared with the control trials when the rest position was the
same as the start position (Figure 4). Under the same conditions,
illusory left rotation did not induce greater variable error ( p .2).
Variable error was not affected by illusory head rotation for any
other rest position ( p .1). In subjects who did not experience

Figure 4. Interaction between vibration and rest position for variable

error. Variable error was greater in muscle vibration conditions with
illusory left and right rotation when the rest position was the same as the
start position (black lines). Variable error was not affected by illusory head
rotation for the other three rest positions (grey lines). Error bars represent
95% confidence intervals.

illusory changes in head position during neck muscle vibration

with the head unconstrained, there was no effect on absolute error:
vibration main effect, F(2, 6) 0.4, p .7; constant error: vibration
main effect, F(2, 6) 1.4, p .3; or variable error: vibration main
effect, F(2, 4) 1.4, p .3.
Elbow JPS was assessed during neck muscle vibration with
the head fixed in all 14 subjects and no changes in absolute
error: vibration main effect, F(2, 260 0.1, p .9; in constant
error: vibration main effect, F(2, 26) 1.1, p .3; or in
variable error: vibration main effect, F(2, 18) 0.5, p .6,
were present. When only the 8 subjects who experienced illusory head rotation were included in the analysis, vibration with
the head fixed had no effect on absolute error for elbow JPS:
vibration main effect, F(2, 14) 0.03, p .9; constant error for
elbow JPS: vibration main effect, F(2, 14) 1.0, p .4; or
variable error for elbow JPS: vibration main effect, F(2, 8) 1.3,
p .3 (Figure 3, right panels). In subjects who did not experience
illusory changes in head position during neck muscle vibration
with the head fixed, there was no effect on absolute error: vibration
main effect, F(2, 10) 0.2, p .9; constant error: vibration main
effect, F(2, 10) 0.8, p .5; or variable error: vibration main effect,
F(2, 8) 0.3, p .8.
Across all trials, rest position affected constant error for elbow
JPS during passive changes in head position: rest position main
effect, F(3, 39) 8.6, p .001; with vibration when the head was
unconstrained: rest position main effect, F(3, 36) 14.2, p
.001; and with vibration when the head was fixed: rest position
main effect, F(3, 39) 9.5, p .05 (Figure 5). A linear regression
equation was fitted to the data (y 0.184x 1.25, r2 0.99) and
shows that the constant error for elbow JPS increased by 0.2 for
every 1 that the rest position was closer to the target position. That
is, when the rest position was closer to the target (more flexed), the
angle of reproduction was more flexed. Despite this effect, there
was no interaction between rest position and the effects of either

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

PERCEPTIONS OF HEAD AND ELBOW BY NECK VIBRATION

Figure 5. Effect of rest position on elbow joint position sense. From left
to right arranged by rest position, constant error is shown for elbow JPS
across all trials for passive head rotation and neck muscle vibration. Error
bars represent 95% confidence intervals.

head rotation or vibration on absolute or constant error for elbow

JPS (all ps .2).

Discussion
This study demonstrated that vibration of SCM and the contralateral splenius induces illusory head rotation in 70% of subjects. Illusory left head rotation was associated with decreased
accuracy in the reproduction of a previously presented elbow
position. These findings suggest that the influence of head position
on the accuracy of upper limb movement (Knox & Hodges, 2005)
cannot be explained entirely by the biomechanical effects of head
movement on proximal body segments. It may be inferred, therefore, that the CNS requires accurate knowledge of head position to
accurately execute upper limb movement.

Neck Muscle Vibration Induces Illusory Changes in Head

Position
The illusions of head movement in this study agree with the
findings of Taylor and McCloskey (1991) and Roll et al. (1989)
who reported kinesthetic illusions of head movement during neck
muscle vibration in the direction of muscle lengthening. Variability in the reports of illusory head movement observed in the
present study (i.e., only 10 out of 14 subjects) is also consistent
with previous data (Taylor & McCloskey, 1991). Other studies
have investigated illusory motion of a visual target in response to
neck muscle vibration and also have reported variation between
subjects (8 out of 10 in Biguer, Donaldson, Hein, & Jeannerod,
1988; 9 out of 17 in Karnath, Sievering, & Fetter, 1994). The
absence of illusions of head movement in some individuals may be
due to a decreased weighting of neck proprioceptive information in
the presence of conflicting somatosensory input regarding neck
position. Consistent with this hypothesis, illusions were less common during vibration with the head fixed, when additional somatosensory information from the head restraint was available. In
studies of vibration to limb muscles, illusions of movement are
greater with the limb restrained compared with vibration without
restraint. This is thought to be explained by differences in the
mismatch between antagonist versus agonist muscle stretch. In an
unconstrained task, the tonic vibration reflex in limb muscles
causes the vibrated muscle to shorten, thus unloading the muscle

1215

spindles in the vibrated muscle and stretching the antagonist

(Cody, Schwartz, & Smit, 1990; Goodwin, McCloskey, & Matthews, 1972), and the degree of resulting flexion is underestimated.
However, in the restrained position there is no change in muscle
length, which prevents unloading of the vibrated muscle spindle
and the antagonist stretch, thus a stronger illusion is perceived.
Several factors may explain the different results for vibration to the
neck muscles. First, as there was no visible head movement in the
present study, the extent to which the tonic vibration reflex induced muscle contraction, and associated changes in muscle
length, is likely to be small. Thus, the difference between restrained and unrestrained conditions would be less than that experienced in the limbs. Second, recent data suggested that illusory
limb movement can be reduced by touching a stable surface (Rabin
& Gordon, 2004). These authors argued that a priori knowledge of
the stability of an external support can influence the interpretation
of sensory inputs and attenuate the illusion during muscle vibration. As subjects were aware that the head support was fixed and
stabilized the head, this is likely to have led to attenuation of any
illusory changes in head position.

Illusory Changes in Head Position Affect Upper Limb

Movement
Previous data suggested that neck muscle vibration induces
deviations in pointing toward a visual target (Biguer et al., 1988;
Han & Lennerstrand, 1999). In these studies, it was proposed that
the deviation in pointing was due to an illusory visual displacement of the target location. One previous study has shown that
neck muscle vibration does not influence the perception of elbow
position, which would suggest that neck muscle vibration can
affect only upper limb movements with a visual target (Kasai,
1991). However, in that study no information was provided about
the presence or absence of illusory head movement during muscle
vibration. In contrast, the results of the present study demonstrate
that illusory displacement of the head, induced by neck muscle
vibration, can influence error in a proprioceptive task with no
visual component. In subjects that did not report illusory changes
in head position, there were no changes in error for elbow JPS.
These findings are supported by data from our laboratory showing
increased errors in elbow JPS during GVS, but only in the subjects
who experienced illusory changes in head position during GVS
(Knox, Coppieters, & Hodges, 2006). Therefore, disruption of the
perception of head position can influence both the visual and
proprioceptive components of upper limb movement.
In the present study, passive head rotation to the right was
associated with larger absolute error for right elbow JPS than
passive left rotation. However, with neck muscle vibration, only
illusory left rotation was associated with greater absolute error for
right elbow JPS. These results contrast with those of Kasai (1991),
who found changes only in right elbow JPS with passive left head
rotation and no changes in elbow JPS with neck muscle vibration.
However, in that study, head rotation was to the end of range, and
the parameters of neck muscle vibration were not described. Moreover, head rotation and vibration were sustained throughout the
proprioceptive trial, not introduced after presentation of the target
position. However, the results of both studies indicate that the
effect of actual rotation of the head is greater than neck muscle
vibration. This may be because muscle vibration activates only a

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

1216

KNOX, CORDO, SKOSS, DURRANT, AND HODGES

subpopulation of proprioceptive afferents (i.e., muscle spindle 1a

afferents). However, it has also been demonstrated that illusory
changes in head position induced with vestibular stimulation have
a consistent effect on the perception of elbow position (Knox et al.,
2006). The observation that illusory changes in head position
induced with GVS have a more consistent effect on elbow JPS
than illusory changes induced with neck muscle vibration provides
insight into the weighting of sensory information during an elbow
proprioceptive task. In both situations, the illusory head movement
is induced by stimulation of a specific population of afferents
(either vestibular or proprioceptive). However, in both situations,
there are also conflicting sensory inputs to the CNS regarding the
actual static position of the head. As the effect of illusory head
movement on elbow JPS is consistent only during GVS, it is
possible that the CNS is more affected by inaccurate vestibular
information than inaccurate neck muscle proprioceptive information during this task. It is therefore suggested that vestibular
information is given greater weighting.
The effect of neck muscle vibration on elbow JPS could also be
due to small changes in gaze direction associated with vibratory
stimuli. Whereas we did not measure gaze direction in the present
study, previous studies show that in all subjects tested, neck
muscle vibration induces horizontal ocular torsion (Han & Lennerstrand, 1995), which redirects gaze. Although visual input was
not required for the task used in the present study, gaze direction
behind the closed eyelids can potentially influence the accuracy of
an upper limb task. For example, Voisin and Chapman (2003)
showed that the somatosensory perception of an object without
vision (haptic perception) is more accurate when the head is turned
toward the object than when it is turned away. In the present study,
errors in elbow JPS did not occur in all subjects. Thus, it is
unlikely that changes in gaze direction, which appear to occur
universally with vibration, can explain errors in upper limb JPS.
The absence of a change in elbow JPS with illusory right
rotation was not related to the degree of illusion, as the illusions of
right rotation were larger, on average, than left rotation. In a
previous study using passive head rotation, the effect on elbow JPS
had the same magnitude for both left and right rotation (Knox et
al., 2006). Similarly, illusory lateral flexion of the head, either to
the left or right, induced with GVS, is associated with errors in
elbow JPS (Knox et al., 2006). The reason for the absence of an
effect of illusory right rotation on right elbow JPS is therefore
unclear but may be related to the strength of the illusion. Illusions
of head rotation during vibration with the head fixed did not
influence the error for elbow JPS. This finding may be due to the
small number of subjects included in this analysis (n 8), the
inconsistency of the illusions both within and between subjects, or
the smaller range of illusory head rotation.
The effect of rest position on constant error for elbow JPS
provides some insight into the strategy used by subjects to determine the target location. In this study, the target location and start
position for the presentation of target position remained constant,
but there were four different rest positions preceding the reproduction of the target position. There was a linear association
between rest position and the degree of target overshooting, however with a slope well below unity. Therefore, this result indicates
that the subjects reproduced the target position by using information about the perceived movement path (including movement
distance, movement velocity, and movement time) and the actual

target position. These findings are consistent with previous studies

in which the constant error for reproduction of a target position
was related to the start position (Bevan, Cordo, Carlton, & Carlton,
1994; Lonn, Crenshaw, Djupsjobacka, Pedersen, & Johansson,
2000). Together these results show that the length of the movement
path, both to the target position and from the target position back
to the rest position, may be used to reproduce the target, in addition
to the perception of target location itself. Therefore, it is interesting
that illusory head rotation influenced only variable error in the
trials for which the rest position was the same as the start position.
Although these trials had two identical movement paths that
should have made that task simpler, they were the most affected by
the muscle vibration stimulus.

Conclusions
Consistent illusions of head rotation may be elicited in the
majority of subjects when vibration is applied to the SCM and the
contralateral splenius. Both actual and illusory head rotation are
associated with increases in the error during the reproduction of a
target position at the elbow joint. It is important to note that the
relationship between head position and the planning and performance of upper limb movement cannot be entirely explained by
biomechanical effects of head movement. These findings add to
the growing body of evidence that knowledge of head position is
an important component of the organization of intrinsic sensory
information for upper limb position.

References
Berger, M., Lechner-Steinleitner, S., Kozlovskaya, I., Holzmuller, G.,
Mescheriakov, S., Sokolov, A., & Gerstenbrand, F. (1998). The effect of
head-to-trunk position on the direction of arm movements before, during, and after space flight. Journal of Vestibular Research, 8, 341354.
Bevan, L., Cordo, P. J., Carlton, L., & Carlton, M. (1994). Proprioceptive
coordination of movement sequences: Discrimination of joint angle
versus angular distance. Journal of Neurophysiology, 71, 18621872.
Biguer, B., Donaldson, I. M. L., Hein, A., & Jeannerod, M. (1988). Neck
muscle vibration modifies the representation of visual motion and direction in man. Brain, 111, 14051424.
Brown, M. C., Engberg, I., & Matthews, P. B. C. (1967). The relative
sensitivity to vibration of muscle receptors of the cat. Journal of Physiology, 192, 773 800.
Cody, F. W., Schwartz, M. P., & Smit, G. P. (1990). Proprioceptive
guidance of human voluntary wrist movements studied using muscle
vibration. Journal of Physiology, 427, 455 470.
Cordo, P. J., Bevan, L., Gurfinkel, V. S., Carlton, L., Carlton, M., & Kerr,
G. K. (1995). Proprioceptive coordination of discrete movement sequences: Mechanism and generality. Canadian Journal of Physiology
and Pharmacology, 73, 305315.
Deutsch, H., Kilani, H., Moustafa, E., Hamilton, N., & Hebert, J. P. (1987).
Effect of head-neck position on elbow flexor muscle torque production.
Physical Therapy, 67, 517521.
Fookson, O., Smetanin, B., Berkinblit, M., Adamovich, S., Feldman, G., &
Poizner, H. (1994). Azimuth errors in pointing to remembered targets
under extreme head rotations. Neuroreport, 5, 885 888.
Goodwin, G. M., McCloskey, D. I., & Matthews, P. B. C. (1972, March
24). Proprioceptive illusions induced by muscle vibration: Contribution
by muscle spindles to perception? Science, 175, 13821384.
Guerraz, M., Navarro, J., Ferrero, F., Cremieux, J., & Blouin, J. (2006).
Perceived versus actual head-on-trunk orientation during arm movement
control. Experimental Brain Research, 172, 221229.

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

PERCEPTIONS OF HEAD AND ELBOW BY NECK VIBRATION

Gurfinkel, V. S., & Levik, Y. S. (1991). Perceptual and automatic aspects
of the postural body scheme. In J. Paillard (Ed.), Brain and space (pp.
147162). Oxford, England: Oxford University Press.
Han, Y., & Lennerstrand, G. (1995). Eye movements in normal subjects
induced by vibratory activation of neck muscle proprioceptors. Acta
Ophthalmologica Scandinavica, 73, 414 416.
Han, Y., & Lennerstrand, G. (1999). Changes of visual localization induced by eye and neck muscle vibration in normal and strabismic
subjects. Graefes Archive for Clinical and Experimental Ophthalmology, 237, 815 823.
Karnath, H. O., Sievering, D., & Fetter, M. (1994). The interactive contribution of neck muscle proprioception and vestibular stimulation to
subjective straight ahead orientation in man. Experimental Brain Research, 101, 140 146.
Kasai, T. (1991). An empirical note on tonic neck reflexes: Control of the
upper limbs proprioceptive sensation. Perceptual and Motor Skills, 72,
955960.
Knox, J. J., Beilstein, D. J., Charles, S. D., Aarseth, G. A., Rayar, S.,
Treleaven, J., & Hodges, P. W. (2006). Changes in head and neck
position have a greater effect on elbow joint position sense in whiplash
associated disorders. The Clinical Journal of Pain, 22, 512518.
Knox, J. J., Coppieters, M. W., & Hodges, P. W. (2006). Do you know
where your arm is if you think your head has moved? Experimental
Brain Research, 173, 94 101.
Knox, J. J., & Hodges, P. W. (2005). Changes in head and neck position
affect elbow joint position sense. Experimental Brain Research, 165,
107113.

1217

Lonn, J., Crenshaw, A. G., Djupsjobacka, M., Pedersen, J., & Johansson,
H. (2000). Position sense testing: Influence of starting position and type
of displacement. Archives of Physical Medicine and Rehabilitation, 81,
592597.
Mars, F., Honore, J., Richard, C., & Coquery, J. M. (1998). Effects of an
illusory orientation of the head on straight-ahead pointing movements.
Current Psychology of Cognition, 17, 749 762.
Rabin, E., & Gordon, A. M. (2004). Influence of fingertip contact on
illusory arm movements. Journal of Applied Physiology, 96, 15551560.
Roll, J. P., Roll, R., & Velay, J.-L. (1991). Proprioception as a link between
body space and extra-personal space. In J. Paillard (Ed.), Brain and
space (pp. 112132). Oxford, England: Oxford University Press.
Roll, J. P., Vedel, J. P., & Roll, R. (1989). Eye, head and skeletal muscle
spindle feedback in the elaboration of body references. Progress in
Brain Research, 80, 113123.
Taylor, J. L., & McCloskey, D. I. (1991). Illusions of head and visual target
displacement induced by vibration of neck muscles. Brain, 114, 755
759.
Voisin, J., & Chapman, C. E. (2003, November). An egocentric frame of
reference for haptic processing of macrogeometric shape. Paper presented at the annual meeting of the Society for Neuroscience, New
Orleans, LA.

Received January 20, 2006

Revision received August 21, 2006
Accepted August 24, 2006