FEATURE: MUSIC LISTENING AND SOFTWARE DEVELOPERS

The Sound found possible detrimental repercus-

sions of music in specific areas, such

of Software
as reading comprehension, 5 and that
the effects of music on a person may
vary based on personality type.5

Development Today, software developers often

work in open-floorplan offices. In
these environments, the use of head-
phones is a common phenomenon,

Music Listening Among providing an opportunity for indi-

viduals to choose to listen to music
while working. Given the potential
Software Engineers for music listening to be inspirational
for productivity, creativity, and con-
centration, 6 we were interested in
Laura Barton and Gülipek Candan, University of British Columbia how professional software develop-
Thomas Fritz, University of Zürich ers engage with music during work.
With two large-scale surveys, we
Thomas Zimmermann, Microsoft investigated the music-listening hab-
Gail C. Murphy, University of British Columbia its of professional software devel-
opers while at work and the effects
of this activity. These inquiries pro-
// Listening to music is a common phenomenon vide insights about music listening
among software developers in today’s work during software development work
that go beyond existing laboratory-
environments. Music can reduce stress, improve
based3 and limited-scope in-depth
happiness, and even increase performance. We field studies.4 The first survey, which
conducted two surveys with 2,242 professional involved 1,445 software developers,
delved into when and why music lis-
software developers and found that between tening occurs, gathering responses
63 and 88.2% of participants listen to music about developers’ frequency of and
reasons for listening to music and
at work at least some of the time often when
the situations and activities in which
writing code or doing repetitive tasks, and that they do so. In this study, we also in-
these listeners tend to be more extroverted.// vestigated what kinds of music they
choose, as an exploratory step to-
ward understanding whether there
is sufficient variation to enable fol-
low-up explorations into subtleties
in music listening6 and the effects of
music type on mood and arousal.7
The second survey, which in-
volved 797 software developers, in-
MUSIC CAN HAVE a positive impact shown that listening to music can, for corporated questions about music
on a human’s performance and per- instance, improve performance, en- listening and personality traits into a
ception of certain tasks. Studies have hance mood, and reduce stress dur- broader inquiry on developer person-
ing physical exercise,1 work tasks, 2 ality and development practices. This
Digital Object Identifi er 10.1109/MS.2019.2906312
and software development efforts.3,4 second survey avoided participation
Date of current version: 12 February 2020 At the same time, several works have bias and allowed us to determine if

78 I E E E S O F T WA R E | PUBLISHED BY THE IEEE COMPUTER SOCIE T Y 0 74 0 -74 5 9 / 2 0 © 2 0 2 0 I E E E

Authorized licensed use limited to: b-on: UNIVERSIDADE DO MINHO. Downloaded on May 31,2021 at 08:19:18 UTC from IEEE Xplore. Restrictions apply.
 trends from the first study were seen that accompanies this article in IEEE were taken from the International Per-
in a larger, unbiased group. Xplore). We additionally asked about sonality Item Pool,9 a repository of 50
The analysis of the responses demographics, such as gender, age, ed- survey questions used to measure per-
shows that most software develop- ucational background, and office size. sonality in the Five Factor or so-called
ers regularly or continuously listen We received a total of 1,445 responses Big Five model: openness, conscien-
to music during work (63% in one
survey and 88.2% in the other),
and that developers often do so to
cut down on background noise in It is inconclusive whether listening
the office (70.9%) and reduce dis-
tractions more generally, something
to music directly improves coders’
commonly found especially in open- productivity, although the two are
office work spaces. The results also generally correlated.
show that music listening helps de-
velopers to regulate their emotions
by lifting their mood (61.4%), relax-
ing them (54.6%), or making them to survey 1, which we distributed tiousness, extroversion, agreeableness,
feel more energetic (48.3%). through two channels. and neuroticism. Participants were re-
The top reason for not listening to cruited through personal email invita-
music while working was that it dis- • Using personalized email, the tions within the same large software
turbs the developer’s focus (65.8%). questionnaire was sent to 1,200 company used for survey 1; we received
This apparent contradiction is partly developers in a large software 797 responses to survey 2 (out of 3,000
explained by our examination of the company; 486 responded (a invitations, a response rate of 26.6%).
activities correlated with listening response rate of 40.5%), most Both surveys were anonymous
to music: while less mentally taxing of them from the United States and extensively piloted before we
tasks, such as writing or testing code, (95.3%). opened them for participation. Par-
are very likely to be accompanied by • Using social media, we adver- ticipants could enter a drawing of
music, cognitively demanding work, tised for the survey more globally Amazon.com gift cards as apprecia-
such as code reviews and learning among professional software tion for their time.
new programs, was rated as being developers, and we received 959
less appropriate for music listening. responses, most of whom were Prevalence of Listening
from Germany (76.2%), Switzer- to Music at Work
Methodology land (10.4%), and Austria (6.4%). Between 63% (survey 2) and 88.2%
To examine software developers’ (survey 1) of the participants re-
music-listening behavior, we con- The focus of survey 2 was on how ported listening to music at work at
ducted two studies—survey 1 and many software developers listen to least some of the time. The difference
survey 2—with a total of 2,242 pro- music and their personality traits. To between these two rates could be ex-
fessional software developers. prevent a self-selection bias of the sur- plained by a possible self-selection
The focus of survey 1 was to inves- vey population toward music listeners, bias of survey 1, which was an invi-
tigate the why and how of software we integrated our question in an in- tation to developers to participate in
developers’ music-listening behav- dependent and general survey on de- a survey regarding listening to mu-
ior. In this study, we asked 20 ques- veloper personality and development sic at work. Although the invitation
tions that addressed the reasons for practices.8 In particular, we added one asked to “take the survey regard-
listening or not listening to music, the question to the survey asking about less of whether you listen to music
work activities during which develop- whether participants listen to music or not,” it was more likely to attract
ers are more or less likely to listen to during work. The survey further con- those who do listen to music. Still,
music, headphone usage, and the fa- tained questions on other development it is safe to say that listening to mu-
vorite music during work (see the ques- practices and on personality traits. sic at work is common, as a signifi-
tionnaire and supplemental material The questions about personality traits cant majority of developers in both

M A R C H /A P R I L 2 0 2 0 | I E E E S O F T WA R E 79

Authorized licensed use limited to: b-on: UNIVERSIDADE DO MINHO. Downloaded on May 31,2021 at 08:19:18 UTC from IEEE Xplore. Restrictions apply.
 FEATURE: MUSIC LISTENING AND SOFTWARE DEVELOPERS

surveys reported practicing it. In sur- options from related studies and had 1. “can’t concentrate/distracts me”
vey 1, only 171 people out of 1,445 a text box for participants to provide (65.8%)
(11.8%) indicated that they never lis- additional reasons. See the supple- 2. “decreases my productivity”
ten to music while they work; 141 of mental material in IEEE Xplore for a (39.8%)
those 171 had tried listening to music list of related studies and additional 3. “reduces the quality of my
while working but had stopped. reasons provided by participants. work” (26.5%)
In survey 1, there was a higher For listeners of music at work, the 4. “I have too many interruptions”
likelihood of listening to music among reasons stated for listening, in their (26.5%)
younger developers, especially those order of popularity, are as follows: 5. “too much overhead” (19.6%)
32 years of age or under (91.7% ver- 6. “I’m in too many meetings”
sus 83.8%). Those with a private of- 1. “cuts down background noise” (13.9%)
fice (83.8%) were less likely to listen (70.9%) 7. “it is not allowed at my work”
to music than those who shared an of- 2. “lifts my mood” (61.4%) (12.0%)
fice space (89.4%). 3. “increases my focus” (56.2) 8. “indicates to coworkers that I’m
In survey 2, we compared the mu- 4. “reduces distractions” (55.2%) available” (7.5%)
sic-listening behavior of developers to 5. “relaxes me” (54.6%) 9. “lowers my mood” (1.9%).
their scores on the Big Five personality 6. “makes me more energetic” (48.3%)
traits. We followed a common practice 7. “motivates me” (48.1%) Interestingly, while 65.8% of those
of normalizing the mean to 25 and the 8. “improves my productivity” who do not listen to music at work said
standard deviation of the personality (44.6%) they do not do so because it is distract-
scores to five. This facilitates compari- 9. “signals coworkers to not dis- ing, 56.2% and 55.2%, respectively,
son across different populations and turb me” (22.6%) of those who do listen at work stated
groups. Higher scores for a dimension 10. “increases the quality of my that music “increases their focus” and
mean that a person exhibits that per- work” (18.8%) “reduces distractions.” This appar-
sonality trait more. For the five person- 11. “gives me a sense of control over ent contradiction is likely a reflection
ality traits, we found three differences my aural environment” (16.2%). of the fact that not all activities are a
that were statistically significant at p good match for musical accompani-
< 0.05 (a false discovery rate less than The popularity of “cuts down on ment, such as learning something new,
8%): software developers who listened background noise” (number 1) and “re- and that developers are not constantly
to music while working scored higher duces distractions” (number 4) shows listening to music throughout the day
on extroversion (25.34 versus 24.43) that developers seek ways to control and choose to turn it off for specific
and openness (25.31 versus 24.47) their personal environment and man- reasons, as examined in the follow-
and lower on conscientiousness (24.78 age distractions, especially with the ing section. It could also be that not
versus 25.38). Neuroticism and agree- prevalence of shared offices (69.7% of all music causes equal reactions in
ableness had no statistically significant participants shared the office with oth- the listener; software professionals in
correlation with music habits. In a lo- ers). While “gives me a sense of control our study specified volume, character,
gistic regression model, the personal- over my aural environment” was not a tempo, and the absence of lyrics (at
ity traits on their own were not reliable popular affirmative response (number least in a language they understand) as
predictors of whether someone listened 11), it is reasonable to extrapolate from factors that made music a support for
to music while working. the former two high-ranking responses their work rather than a hindrance.
that this is exactly what developers are Additionally, music provides a
Reasons for Listening or doing—by blocking out background way to self-regulate affect (i.e., mood
Not Listening to Music noise and other nearby distractions, and emotion), encouraging the de-
In survey 1, we asked participants to software personnel are trying to regu- veloper to stay motivated or aid-
choose one or more reasons why they late their auditory inputs so that their ing relaxation when work becomes
listen to music on the job (if self-iden- focus is not disrupted. stressful. “Lifts my mood” was the
tified as a music listener) or why they The reasons stated for not listen- second-most-popular reason for lis-
do not (if not self-identified as a mu- ing to music at work, in their order tening to music (61.4%), with “relaxes
sic listener). We identified the possible of popularity, are: me” (number 5) and “makes me more

80 I E E E S O F T WA R E | W W W. C O M P U T E R . O R G / S O F T W A R E | @ I E E E S O F T WA R E

Authorized licensed use limited to: b-on: UNIVERSIDADE DO MINHO. Downloaded on May 31,2021 at 08:19:18 UTC from IEEE Xplore. Restrictions apply.
 energetic” (number 6) both ranking activities, in order of the most likely Not surprisingly, tasks that are
high as well. For sample responses to be accompanied by music, are: largely independent (writing code) or
to the questionnaire, see “Selected not especially mentally taxing (repetitive
Quotes From Respondents.” • writing code (85.5%) tasks) are the most likely matches for
• repetitive tasks (76.8%) music. There is a drop in the frequency
Activities During • building code (64.8%) of music listening when the tasks be-
Which Developers • testing code (60.2%) come verbally expressive (writing docu-
Listen to Music • writing documentation (48.7%) mentation, code reviews, management
Focusing on only those coders in sur- • debugging (45.4%) tasks, emails), or when they require cre-
vey 1 who listen to music at work, • email (32.3%) ative problem solving and open-ended
we examined which activities were • management tasks (29.5%) thinking (debugging, program under-
most often accompanied by music. • code reviews (25.7%) standing, learning), i.e., more cogni-
Participants chose from a list of ac- • program understanding (25.6%) tively demanding activities that involve
tivities regarding whether they were • multitasking (23.9%) different modes of processing.
less or more likely listen to music— • learning (19.9%) Many developers switch the type
or if there was no difference. The • memorization (14.4%). of music based on the task they are

SELECTED QUOTES FROM RESPONDENTS

WHY DO YOU LISTEN TO MUSIC?
• “Sometimes, a task is so dull, music makes it easier to bear.”
• “The holy place to be while developing is right in the flow—music helps entering it.”
• “To reduce boredom. Sometimes I need to write a ton of code that isn’t really that complex. It requires minimal thinking,
but it’s still a ton of work. By listening to music there are only two things happening in my world: writing code and humming
along to the music. I don’t worry about people walking past my office, squirrels running in trees, or birds flying around. You
can argue that I listen to music to escape from this wonderful work environment.”
• “Typing rhythm is faster when listening to music.”
• “Cutting down background noise really is one of the most important reasons. When other colleagues around me in the same
office are on the phone or discussing topics unrelated to my work, it’s hugely distracting.”
• “If a task is repetitive and does not require a significant amount of concentration, music distracts me from the repetitive-
ness of it and allows me to complete more before getting worn out.”

WHY DO YOU NOT LISTEN TO MUSIC?

• “I must use all my resources to deliver high-quality code. Music distracts me—it changes my mind while I concentrate on
many things. Music attracts my attention, it generates thoughts and feelings. Coding while listening to music results in more
bugs than average—especially conceptional failures are increasing, and those are the most productivity-killing ones.”
• “Osmotic Communication. Sometimes you overhear relevant things only by accident.”
• “I only listen while doing rote-type tasks. I find that if I have a task that requires significant thinking, music reduces my abil-
ity to think clearly.”
• “While about everyone has headphones in the office, it is frowned upon to use them continuously throughout the day. One of
the fundamental ideas behind team rooms is that information flows more easily because everyone can absorb discussions
that they are not involved in. Headphones detract from this goal.”
• “If I’m trying to focus on a new problem or learn a new technology, I tend to turn off the music because it can be distracting.
However, all of the noise in the open space is also distracting, so it’s kind of a lose-lose situation.”

M A R C H /A P R I L 2 0 2 0 | I E E E S O F T WA R E 81

Authorized licensed use limited to: b-on: UNIVERSIDADE DO MINHO. Downloaded on May 31,2021 at 08:19:18 UTC from IEEE Xplore. Restrictions apply.
 FEATURE: MUSIC LISTENING AND SOFTWARE DEVELOPERS

working on (39%). For example, listening to music for pleasure, they more likely to listen to music without
one developer wrote, “When writing are more likely to listen to rock, metal, lyrics. Those who listen to music to
code, I like music which helps me to and pop than at work. These data may increase the quality of their work fa-
concentrate, such as Trance or any correspond to research that suggests vor instrumental and classical music.
other electronic genre. Otherwise, that cognitive tasks benefit from mod- Software professionals who want to
when I’m stressed or angry about erate levels of arousal.6 be more energetic gravitate toward
something, it goes more in the metal/ More than 50% of developers pre- rock, pop, and electronic music.
rock direction, something that helps ferred music genres that typically do
me to cool down.” not include prominent vocals, e.g., Related Work
electronic dance music, classical, and A multitude of studies has exam-
Music-Listening film or video game soundtracks. Sixty- ined the effects of music listening on
Preferences eight survey participants explicitly a variety of activities, showing, for
We also asked software developers stated that they preferred listening to instance, that self-selected and mo-
about the music they listen to while music without lyrics (or lyrics in a lan- tivational music can lead to a per-
working and whether it is the same as guage they do not understand). “Songs formance increase during physical
what they listen to for pleasure. This with lyrics distract me when I need to exercise,10 and that slower-tempo
information provides initial informa- focus on writing something,” one de- background music can affect shop-
tion for others to build on to explore veloper wrote. Because even rote cod- ping behavior and lead to an increase
personal preferences in music listen- ing requires linguistic processing, we in gross sales in supermarkets.11
ing and their effect on work,6 such as suggest that developers recognize the When it comes to work, studies
considering whether different musical distraction caused by the compet- on the effects of music have yielded
styles might lead to different arousal ing information of lyrics in music and mixed results, with an overall posi-
levels. We contrast the music genres therefore try to limit this interference. tive trend suggesting that music is
that are popular for work and plea- We also linked the music genre to beneficial to work performance,
sure in the word cloud in Figure  1(a). the motivation for listening to mu- helps to reduce stress, and improves
Developers are more likely to listen to sic; see the word cloud in Figure 1(b). individuals’ happiness and satisfac-
classical, electronic, instrumental, and Developers who want to signal co- tion. For instance, Lesiuk 3 found
no-lyrics music while at work. When workers to not disturb them are that listening to music lowered

1. Lifts My Mood
2. Relaxes Me
3. Motivates Me
4. Cuts Down Background Noise
5. Reduces Distractions
6. Signaling Coworkers to Not Disturb Me
7. Sense of Control Over Aural Environment
8. Improves My Productivity
9. Increases My Focus
10. Increases the Quality of My Work
11. Makes Me More Energetic

(a) Work (b)

FIGURE 1. Comparison clouds showing the music-listening preferences of software developers. (a) The word cloud shows which
music genres developers are more likely to listen to at work (red) versus music they listen to for pleasure (blue). (b) The word cloud
links the music genre to the motivation for music listening; the music genre is prefixed with a number representing the motivation (for
example, 6 represents the motivation “Signaling coworkers to not disturb me”). Not all motivations could be linked to music genres
(“Lifts my mood,” “Cuts down background noise,” and “Increases my focus”).

82 I E E E S O F T WA R E | W W W. C O M P U T E R . O R G / S O F T W A R E | @ I E E E S O F T WA R E

Authorized licensed use limited to: b-on: UNIVERSIDADE DO MINHO. Downloaded on May 31,2021 at 08:19:18 UTC from IEEE Xplore. Restrictions apply.
 developers’ stress and anxiety with- introverts are in danger of becoming should I challenge this?” on Stack-
out detracting from the logic and overloaded by the extra stimulation. exchange.com). An interesting ques-
syntax of their code. Surgeons had The experimental tasks of previ- tion is whether music is necessary or
reduced autonomic reactivity and ous studies are highly varied in both whether white or pink noise, which
performed arithmetic calculations type of cognitive demand and com- blocks out other sounds without it-
faster and more accurately while plexity (for example mental arithme- self being engaging, might achieve
listening to their own preferred tic,12 reading comprehension,5,15 or similar goals. In fact, five of our sur-
soundtrack,12 memorization5) and are not always vey participants indicated that they
Most studies on the effects of mu- representative of real technology- listen to white noise to mask office
sic at work have focused on a specific workplace scenarios. Only a few stud- sounds without being distracting in
task, often in a laboratory setting. ies focus on actual coding and typical and of itself, and 68 coders explicitly
The study by Lesiuk4 monitored de- developer tasks.3,4 Software develop- wrote that they prefer music without
velopers engaged in their typical jobs ment can run the gamut from repeti- lyrics (or lyrics in languages they do
over several weeks: a baseline control tive tasks to extremely demanding not speak). Several more stated that
week, two weeks in which they were and innovative problem solving, and familiar music is best; for example,
encouraged to listen to music, a week it is likely that music is not a good two of their comments were “The
when they were not allowed to listen, match for every situation. Most stud- better I know a song, the better I can
and a return to listening privileges at ies use silence as the control, whereas focus on my task” and “Led Zeppe-
the end of the study. The self-reported a typical software office will, in fact, lin is pretty good for me, because I
quality of work was highest during be open plan; the choice for a real- know every song by heart, so I don’t
the music-listening weeks, and devel- world developer is not between silence pay much attention to it.” Software
opers spent longer than they intended and music but between background professionals seem to have an inher-
to complete their tasks during the noise and music. ent understanding that music should
week without music. not be so stimulating as to dominate
The effect of music listening has also Discussion their central cognitive focus.
been shown to be affected by a vari- The participants’ two main motiva- Music listening may be especially
ety of factors, such as a person’s music tions for listening to music were to shut important for those who have an in-
preference, the habituation to work- out the distractions of the office around troverted personality and are easily
ing with music, personal traits, and them and to regulate their mood when overstimulated, helping them to find
task type.12–14 For instance, music and they need to stay energized or manage a way to mitigate the effects of office
noise were inimical to performance on stress. These habits appear to confirm noise.14,15 While we found that in-
a reading comprehension test,14 both the previous research on workplace troversion has some correlation with
for introverts and extroverts, although music listening: in short, music can choosing not to listen to music, using
introverts suffered a greater impact. In help regulate focus and mood, but not white noise or other noise-masking
another study, introverts were nega- everyone is assisted by the extra stim- sound may remove the distractions
tively impacted by background mu- ulus, and, furthermore, music can be of background chatter without irri-
sic during reading comprehension and distracting when coupled with cogni- tating introverted developers. More
memorization, while extroverts showed tively demanding tasks. study is needed to understand how
no significant difference in task perfor- Open-plan offices may not be personal choices and personality
mance.5 Daoussis and McKelvie15 came avoidable in continuously expand- traits impact appropriate choices for
to a similar result for reading recall ing and evolving technology com- individuals and companies alike to
tasks and pointed to Eysenck’s theory of panies. As we have reported in this provide productive workplaces.
cortical arousal as an explanation: in- article, some developers use music as Apps such as MyNoise.net, Brain.
troverts have a lower optimum arousal a means of coping with these office fm, and Focus@Will promise improved
threshold, whereas extroverts require situations. Anecdotal reports on the concentration through so-called scien-
higher stimulation to reach their opti- web suggest that the use of music is tifically designed music that is tailored
mum arousal level for working. Because not universally accepted across all to different tasks and sometimes even
music provides stimulation, extro- companies (for example, see “Head- generated by artificial intelligence.
verts can find it energizing, whereas phone use has been banned, how There may also be merit in developing

M A R C H /A P R I L 2 0 2 0 | I E E E S O F T WA R E 83

Authorized licensed use limited to: b-on: UNIVERSIDADE DO MINHO. Downloaded on May 31,2021 at 08:19:18 UTC from IEEE Xplore. Restrictions apply.
 FEATURE: MUSIC LISTENING AND SOFTWARE DEVELOPERS
ABOUT THE AUTHORS

LAURA BARTON is a software develop- THOMAS ZIMMERMANN is a senior

ment engineer at Amazon, working researcher at Microsoft. He analyzes
with human resource data analytics, software data for a living and works on the
management, and retention compliance. productivity of Microsoft’s software de-
Barton received her bachelor’s degree in velopers and data scientists. Zimmermann
computer science from the University of received a Ph.D. in computer science from
British Columbia, Canada, and had previ- Saarland University. He is a distinguished
ous postsecondary training in music and member of ACM and a Senior Member of
pedagogy. Contact her at lauraelbarton@ the IEEE and the IEEE Computer Society.
gmail.com. Contact him at tzimmer@microsoft.com or
visit http://thomas-zimmermann.com.

GÜLIPEK CANDAN is a software developer at Avigilon, a GAIL C. MURPHY is a professor of

Motorola Solutions Company, Vancouver, Canada. Candan computer science and vice president for
received her B.Sc. from the University of British Columbia, research and innovation at the University
Canada. Contact her at candan.gulipek@gmail.com. of British Columbia, Canada. Her research
interests are in improving the productiv-
ity of software developers and knowledge
workers by providing the necessary tools
to identify, manage, and coordinate the in-
formation that matters most for their work.
Murphy received her Ph.D. in computer sci-
ence from the University of Washington. She
is a fellow of the Association for Computing
Machinery and the Royal Society of Canada
and a member of the IEEE Computer Soci-
ety. She is a cofounder and on the board of
Tasktop Technologies Inc. Contact her at
murphy@cs.ubc.ca.

THOMAS FRITZ is an associate professor at

the University of Zürich, Switzerland. His re-
search interests include empirically studying
software developers and on using personal
and biometric data to improve developers’
productivity. For his research, he works
with more than 20 international companies,
including ABB, Microsoft, and Logitech. Fritz
received his Ph.D. from the University of
British Columbia, Canada. He is a Member of
the IEEE and the Association for Computing
Machinery. Contact him at fritz@ifi.uzh.ch.

new noise-management technolo- Kickstarter. Thirty-one of our partici- could be removed and the music type
gies, such as headphones that adjust pants mentioned that they select their could be adjusted to avoid any task
to the listener’s cognitive state using music based on their mood or the task distraction, especially for cognitively
electroencephalogram sensors, similar at hand. By adjusting the music to the demanding work.
to Mindset headphones (www.think developers’ cognitive load, the over- Given that we saw a very wide
mindset.com), a recent project on head of selecting and managing music range of musical preferences in our

84 I E E E S O F T WA R E | W W W. C O M P U T E R . O R G / S O F T W A R E | @ I E E E S O F T WA R E

Authorized licensed use limited to: b-on: UNIVERSIDADE DO MINHO. Downloaded on May 31,2021 at 08:19:18 UTC from IEEE Xplore. Restrictions apply.
 study—from black metal to techno performance and productivity under 7. W. F. Thompson, E. Glenn Schel-
to Vivaldi concerts—music selec- some but not all conditions. Addition- lenberg, and G. Husain, “Arousal,
tion is personal and should be left ally, music may not always be inher- mood, and the Mozart effect,”
to the individual. Not only do de- ently desirable, but is a solution to the Psychol. Sci., vol. 12, no. 3, pp.
velopers match their music to their problem of ambient open-office noise. 248–251, 2001.
mood and task, but very strong Software developers adjust their 8. E. K. Smith, C. Bird, and T. Zim-
preferences can come into play; one workplace listening—the genre, vol- mermann, “Beliefs, practices, and
developer could love dance music, ume, or whether they engage in it at personalities of software engineers: A
whereas another would find it es- all—to match their current activity, survey in a large software company,”
pecially annoying. It is inconclusive mood, and personality type. Adap- in Proc. Int. Workshop Cooperative
whether listening to music directly tive technologies could therefore learn and Human Aspects Software Engi-
improves coders’ productivity, al- from what developers are already in- neering, 2016, pp. 15–18.
though the two are generally cor- tuitively doing by tailoring music or 9. L. R. Goldberg, et al., “The Interna-
related. 6 However, it does seem other auditory stimulation in the envi- tional Personality Item Pool and the
clear that developers are happier ronment to benefit the user. future of public-domain personality
when they are able to listen to mu- measures,” J. Res. Personality, vol.
sic; this positive affect is more likely Acknowledgment 40, no. 1, pp. 84–96, 2006.
to be what improves performance.4 We thank everyone who responded to our 10. C. I. Karageorghis and D.-L. Priest,
Therefore, to see benefits, software surveys. “Music in the exercise domain: A re-
professionals need to be allowed to view and synthesis (part I),” Int. Rev.
choose their own music to listen to References Sport Exercise Psychol., vol. 5, no. 1,
over headphones (versus, say, an of- 1. B. L. Copeland and B. D. Franks, “Ef- pp. 44–66, 2012.
ficewide stereo). fects of types and intensities of back- 11. R. E. Milliman, “Using background
ground music on treadmill endurance,” music to affect the behavior of super-
Limitations J. Sports Med. Phys. Fitness, vol. 31, market shoppers,” J. Marketing, vol.
A limitation of any empirical research no. 1, pp. 100–103, 1991. 46, no. 2, pp. 86–91, 1982.
is the generalizability of the results, 2. W. A. Smith, “Effects of industrial 12. K. Allen and J. Blascovich, “Effects
that is, different contexts may lead to music in a work situation requiring of music on cardiovascular reactiv-
different results. Our findings might complex mental activity,” Psychol. ity among surgeons,” J. Amer. Med.
also have been skewed by the outsize Rep., vol. 8, no. 1, pp. 159–162, 1961. Assoc., vol. 272, no. 11, pp. 882–884,
presence of the large company in our 3. T. Lesiuk, “The effect of music lis- 1994.
population. However, the participants tening on a computer programming 13. C. Etaugh and P. Ptasnik, “Effects of
from the large company responded task,” Comput. Inf., vol. 40, no. 3, studying to music and post-study re-
differently than other participants on pp. 50–57, 2000. laxation on reading comprehension,”
only a few questions, so we do not be- 4. T. Lesiuk, “The effect of music listen- Perceptual Motor Skills, vol. 55,
lieve that it dominated the results. To ing on work performance,” Psychol. pp. 141–142, Aug. 1982.
support replication, we provide the Music, vol. 33, no. 2, pp. 173–191, 14. A. Furnham and L. Strbac, “Music is
questionnaire (see the questionnaire 2005. as distracting as noise: The differential
that accompanies this article as supple- 5. A. Furnham and A. Bradley, “Mu- distraction of background music and
mental material in IEEE Xplore). sic while you work: The differential noise on the cognitive test performance
distraction of background music on the of introverts and extraverts,” Ergo-

L
cognitive test performance of introverts nomics, vol. 45, no. 3, pp. 203–217,
istening to music while working and extraverts,” Appl. Cogn. Psychol., 2002.
is a common practice, but it is vol. 11, no. 5, pp. 445–455, 1997. 15. L. Daoussis and S. J. Mc Kelvie,
done thoughtfully by software 6. A. B. Haake, “Individual music “Musical preferences and effects of
developers based on their current mood listening in workplace settings: An music on a reading comprehension
and task focus. Many of their natu- exploratory survey of offices in the test for extraverts and introverts,”
ral choices are supported by research, UK,” Musica Scientia, vol. 15, no. 1, Perceptual Motor Skills, vol. 62,
suggesting that music can improve pp. 107–129, 2011. no. 1, pp. 283–289, 1986.

M A R C H /A P R I L 2 0 2 0 | I E E E S O F T WA R E 85

Authorized licensed use limited to: b-on: UNIVERSIDADE DO MINHO. Downloaded on May 31,2021 at 08:19:18 UTC from IEEE Xplore. Restrictions apply.