Content uploaded by E. Glenn Schellenberg
Author content
All content in this area was uploaded by E. Glenn Schellenberg on Jan 23, 2019
Content may be subject to copyright.
Music and Cognitive Abilities
E. Glenn Schellenberg
University of Toronto, Mississauga, Ontario, Canada
ABSTRACT—Does music make you smarter? Music listening
and music lessons have been claimed to confer intellectual
advantages. Any association between music and intellec-
tual functioning would be notable only if the benefits apply
reliably to nonmusical abilities and if music is unique in
producing the effects. The available evidence indicates
that music listening leads to enhanced performance on a
variety of cognitive tests, but that such effects are short-
term and stem from the impact of music on arousal level
and mood, which, in turn, affect cognitive performance;
experiences other than music listening have similar effects.
Music lessons in childhood tell a different story. They are
associated with small but general and long-lasting intel-
lectual benefits that cannot be attributed to obvious con-
founding variables such as family income and parents’
education. The mechanisms underlying this association
have yet to be determined.
KEYWORDS—music cognition; intelligence; cognitive develop-
ment
People’s eagerness for quick fixes can be seen in the seemingly
irrational appeal of crash diets, get-rich-quick schemes, and get-
smart-fast programs. Is the claim of music as a route to enhanced
intelligence yet another self-improvement fantasy? In the pages
that follow, I outline the origins of the claim and evaluate the
available evidence.
Intellectual benefits of exposure to music would be noteworthy
if (a) they extend to nonmusical abilities, (b) they are systematic
and reliable, and (c) they are more likely to result from hearing or
playing music than from other activities. Unlike in other cul-
tures, where virtually everyone participates in music making,
musical experiences in Western society typically involve lis-
tening and, only in some cases, lessons and performing. Music
listening is ubiquitous, both purposefully (e.g., listening to the
radio) and incidentally (e.g., background music in stores and
restaurants). By contrast, relatively few individuals take music
lessons for several years. The consequences of music listening
are likely to differ quantitatively and qualitatively from those of
music lessons.
MUSIC LISTENING
Widespread interest in the potential benefits of music listening
was sparked by the publication of an article (Rauscher, Shaw, &
Ky, 1993) that reported superior spatial abilities for participants
who listened to a recording of music composed by Mozart com-
pared to those who sat in silence or listened to relaxation in-
structions. The finding, known as the ‘‘Mozart effect,’’ was
publicized widely in the popular media. Although the effect was
found to be brief (lasting 10–15 minutes) and the participants
were undergraduates, the news captured the public imagination
and led to social-policy changes. These included the distribu-
tion of a CD of Mozart’s music to every baby born in Georgia and
the formation of a cottage industry of music recordings designed
to make infants smarter. Presumably, the underlying rationale
was that if the short-term effect is reliable, long-term exposure to
music in infancy—when brain plasticity is greatest—might fa-
cilitate neural connections that could have long-term impact.
Subsequent replication attempts met with mixed success
(Chabris et al., 1999), however, which could be attributable to the
weakness of the effect or the reliance on group testing
(Schellenberg, in press). The original authors proposed that passive
listening to music composed by Mozart primed spatial abilities in
particular, and they attributed the replication failures to the wrong
music or the wrong task. But their proposal of cross-modal priming
between two unrelated domains is at odds with the available re-
search. Priming is a relatively robust psychological phenomenon
that occurs between stimuli with an obvious link. In language, for
example, cross-modal priming effects are evident for subsequent
presentations of the same word, a related word, a homonym, and the
sentential structure implied by the word (i.e., repetition,semantic,
phonological,andsyntactic priming, respectively).
The meta-analysis in Chabris et al. (1999) motivated specu-
lation that the Mozart effect, when evident, could be explained
as an artifact of arousal. Optimal levels of arousal (i.e., physical
and mental activation) have widespread, facilitative effects on
performance. In line with this view, a colleague and I (Nantais
and Schellenberg, 1999) replicated the Mozart effect (compared
to sitting in silence) but we also found a ‘‘Schubert effect’’ of
Address correspondence to Glenn Schellenberg, Department of Psy-
chology, University of Toronto at Mississauga, Mississauga, Ontario,
Canada, L5L 1C6; e-mail: g.schellenberg@utoronto.ca.
CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE
Volume 14—Number 6 317Copyright r2005 American Psychological Society
identical magnitude. When listening to Mozart was contrasted
with listening to a narrated story, the effect disappeared (see
Fig. 1) but performance interacted with preferences. Listeners
who preferred Mozart performed better after listening to Mozart
than to the story. Listeners who preferred the story showed the
opposite pattern (a ‘‘Stephen King effect’’).
Thompson, Husain, and I (Thompson, Schellenberg, & Hu-
sain, 2001) subsequently formulated the arousal-and-mood
hypothesis: Listening to Mozart is one example of a stimulus that
influences the perceiver’s arousal level and mood, which can
affect performance on a variety of cognitive tasks. Our partici-
pants heard a fast-tempo piece by Mozart in a major (happy-
sounding) key, or a slow-tempo piece by Albinoni in a minor
(sad-sounding) key. As predicted, we observed a Mozart effect
(compared to silence) but no ‘‘Albinoni effect’’ on a test of spatial
abilities. We also found that arousal and mood were higher and
more positive after listening to Mozart than after listening to
Albinoni. The effect size of the Mozart advantage on the spatial
test was virtually identical in magnitude to the Mozart advantage
in arousal and mood. When changes in arousal or mood were
held constant by statistical means, the Mozart advantage on the
spatial test disappeared. In another study (Husain, Thompson, &
Schellenberg, 2002), the tempo (fast or slow) and mode (major
or minor) of the same Mozart piece were manipulated before
listeners completed a spatial task. The manipulations led to
different arousal levels and moods across conditions, which, in
turn, accounted for the majority of the variance in spatial abil-
ities. Yet another study tested nonspatial abilities after under-
graduates listened to Mozart or to Albinoni (Schellenberg,
Nakata, Hunter, & Tamoto, in press). When the listening expe-
rience resulted in differences in arousal and mood, performance
on a test of processing speed was also better following Mozart
than it was following Albinoni.
Paper-and-pencil measures of arousal and mood are not
available for children, but the available findings reveal en-
hanced cognitive performance after listening to music that is
thought to be arousing and pleasant for the age group under
investigation. For example, Hallam and I (Schellenberg and
Hallam, in press) reported a ‘‘Blur effect’’ for 10- and 11-year-
olds, who performed better on a spatial test after listening to pop
music (by Blur and other bands) compared to music composed by
Mozart or a scientific discussion. In a test of creativity among
younger children (Schellenberg et al., in press), 5-year-olds
drew with crayons after listening to Mozart, Albinoni, or familiar
children’s songs, or after singing familiar songs. Drawing times
were longer, and the drawings were judged to be more creative,
for the children exposed to familiar songs (a ‘‘children’s play
song effect’’). The effects did not differ between the listening and
singing groups (see Fig. 2).
In sum, music listening (or singing) can lead to enhanced
performance on a variety of tests of cognitive ability. These ef-
fects are mediated by arousal and mood and are unlikely to differ
from those that arise as a consequence of exposure to nonmusical
stimuli that have similar emotional impact (e.g., giving partici-
pants a cup of coffee or a small bag of candy; see Isen, 2000;
Smith, Osborne, Mann, Jones, & White, 2004). Listening to
music composed by Mozart does not have unique or special
consequences for spatial abilities. Rather, upbeat, age-appro-
priate music can improve listeners’ arousal level and mood, at
least for short periods. In turn, effects of arousal and mood ex-
tend beyond measures of spatial ability to tests of processing
speed and creativity. In principle, similar short-term cognitive
benefits might be evident among infants, whose arousal level is
altered by exposure to maternal singing (Shenfield, Trehub, &
Nakata, 2003). It is well established that infants perform best in
the laboratory when they are alert and content.
Mozart-Silence Schubert-Silence Mozart-Story
8
9
10
11
12
13
14
15
16
Mean Paper-Folding-and-Cutting Score
Experimental Comparison
Music
Control
Fig. 1. Mean scores on the paper-folding-and-cutting spatial test in the
music-listening and control conditions (Nantais & Schellenberg, 1999).
The figure shows a Mozart effect (left), a Schubert effect (middle), and,
when the control condition involved listening to a story, no effect (right).
Mozart Albinoni Familiar-
Listening
Familiar-
Singing
2.5
3
3.5
4
4.5
Mean Rating of Children's Drawings
Musical Experience
Fig. 2. Mean adult ratings of children’s drawings (Schellenberg et al.,
2005). Children drew after one of four musical experiences: listening to
Mozart, listening to Albinoni, listening to familiar children’s songs, or
singing familiar songs. Adult raters (blind to group membership) rated
the drawings. Higher ratings indicate more favorable appraisals of the
drawings relative to a baseline (no music) drawing.
318 Volume 14—Number 6
Music and Cognitive Abilities
MUSIC LESSONS
We turn now to the issue of whether music lessons confer non-
musical benefits. I (Schellenberg, 2004) conducted the only
controlled experiment to date that included random assignment
of individual children to music lessons or comparison condi-
tions. One hundred and forty-four 6-year-olds were administered
an entire standardized IQ test (the Wechsler Intelligence Scale
for Children-III, or WISC-III) before entering first grade (at age
6) and again between first and second grade (at age 7). In the
interim, two groups of children received 36 weeks of keyboard or
vocal instruction. Two control groups received drama lessons or
no lessons. All four groups had reliable increases in full-scale IQ
from the first to the second testing session. Such increases are a
known consequence of attending school. The two music groups
did not differ in this regard, nor did the two control groups, but
the increase in IQ was greater for the music groups than for the
control groups (see Fig. 3). This difference was not a conse-
quence of elevated performance on a specific subset of intel-
lectual abilities (e.g., verbal or spatial). Compared to the control
groups, the music groups had larger increases across the four
main areas of intellectual ability measured by the WISC-III (i.e.,
the four index scores, see Fig. 3). An incidental finding was that
the drama group had increases in adaptive social skills that were
larger than those in the other three groups.
This experiment provided evidence that music lessons cause
improvements in intellectual ability. My finding of broad intel-
lectual benefits of music lessons is also consistent with the lit-
erature as a whole (Schellenberg, in press), which includes
reports of positive associations between music lessons and
reading, mathematical, verbal, and spatial abilities. Would
music lessons of longer duration be accompanied by larger in-
tellectual benefits? Because longer-term experimental studies
would likely suffer from differential attrition across conditions,
I (Schellenberg, 2005) tested this question correlationally. In
contrast to previous research, I also measured confounding
variables such as family income and parents’ education, which
were held constant in the statistical analyses.
The participants were 147 children and 150 undergraduates.
For the children, outcome measures included the entire WISC-
III as well as a standardized test of educational achievement and
grades in school. The number of months of music lessons had a
modest but positive association with each outcome variable that
remained reliable when family income, parents’ education, and
involvement in nonmusical activities were held constant. Once
again, the effects were broad, extending across the four index
scores and the different areas of academic achievement (math,
spelling, reading) but not to social skills. Involvement in non-
musical out-of-school activities was not predictive of IQ, aca-
demic achievement, or social behavior. For the undergraduates,
the outcome measure was an entire adult IQ test (the Wechsler
Adult Intelligence Scale-III, or WAIS-III). The association be-
tween years of playing music regularly and IQ was smaller than
the one observed in childhood but it was statistically reliable
even after accounting for individual differences in family in-
come and parents’ education. These correlational findings ex-
tend those of the experimental study by showing that real-world
effects of musical training on intellectual abilities are (a) larger
with longer periods of training, (b) long lasting, (c) not attrib-
utable to obvious confounding variables, and (d) distinct from
those of nonmusical out-of-school activities.
What are the underlying mechanisms driving the association
between music lessons and intellectual ability? One possibility
is that because music lessons are school-like, the intellectual
benefits of attending school are exaggerated by the positive
impact of additional schooling on IQ. From this perspective,
drama lessons might be insufficiently school-like (e.g., they
include pretending and dressing up), although other out-of-
school activities such as reading or chess lessons ought to have
benefits similar to music. Music lessons would be special only
because they represent a school-like activity that many children
enjoy and choose to do on a regular basis.
A second possibility is that the association stems from the con-
stellation of abilities that music lessons train and improve—abil-
ities including focused attention and concentration, memorization,
reading music, fine-motor skills, expressing emotions, and so on.
Although the association could be a consequence of improvement
in one of these abilities or a particular subset, the diffuse nature of
the association implicates the contribution of multiple factors. The
particular type of music lessons (e.g., the instrument or teaching
method) might have more specific effects on outcome measures
other than IQ. For example, keyboard lessons are as good as drama
lessons in improving children’s ability to decode the emotions
conveyed by prosody in speech, but voice lessons have no bene-
ficial effect (Thompson, Schellenberg, & Husain, 2004).
FSIQ VC PO FD PS
0
2
4
6
8
10
12
14
16
18
20
Mean Difference Score
WISC-III Outcome
Keyboard
Vocal
Drama
No Lessons
Fig. 3. Mean difference scores on the Wechsler Intelligence Scale for
Children-III (WISC-III) outcome measures for groups of children re-
ceiving keyboard lessons, vocal lessons, drama lessons, or no lessons
(Schellenberg, 2004). The music groups had larger increases than the
drama-lessons and no-lessons (control) groups. (FSIQ 5Full-Scale IQ, VC
5Verbal Comprehension Index, PO 5Perceptual Organization Index,
FD 5Freedom from Distractibility Index, PS 5Processing Speed Index.)
Volume 14—Number 6 319
E. Glenn Schellenberg
A third possibility is that music promotes intellectual develop-
ment because of its inherently abstract nature. For example, a tune
is defined solely by relational information. A listener can identify a
particular tune (e.g., ‘‘Yankee Doodle’’) when it is played fast or
slow, at a high or a low pitch level, on a piano or a guitar, and so on.
In other words, tunes are abstractions. Listeners’ representations
must generalize even farther to patterns that have similar but not
identical relational information (e.g., variations on a theme). To
illustrate, the opening bars of Beethoven’s Fifth Symphony have a
repeated motive that varies in absolute pitch (different starting
tone) and in relative pitch (major third then minor third), yet lis-
teners hear the second motive as a repetition because of its iden-
tical rhythm and melodic contour. Learning to think abstractly and
to recognize musical similarities across contexts could facilitate
intellectual development more generally.
Finally, learning a musical language could have cognitive
benefits similar to those evident in bilingual children (Craik &
Bialystok, 2005). Although this view has intuitive appeal be-
cause music and language are both auditory communication
systems, the positive effects of bilingualism are evident for fluid
intelligence (i.e., executive control) but not for crystallized in-
telligence (e.g., knowledge acquired through experience, such
as vocabulary), whereas the effects of music lessons appear to
extend to both domains.
CONCLUSION
Does music make you smarter? The answer is a qualified yes.
Music listening and music lessons can lead to short-term and
long-term cognitive benefits, respectively. This positive answer
is qualified because the short-term benefits of music listening do
not appear to differ from those associated with other stimuli that
optimize arousal level or elicit mild positive affect. The mech-
anisms driving the intellectual benefits of music lessons remain
more obscure. Similar benefits could be derived from other out-
of-school activities that are similarly school-like. Nonetheless,
music lessons might be special in this regard because (a) they are
a school-like activity that many children enjoy, (b) multiple
skills are trained in music lessons, (c) music is a domain that
improves abstract reasoning, or (d) acquiring musical knowledge
is similar to acquiring a second language. Future research could
identify the underlying mechanisms more clearly.
Is music a quick fix to the problem of intelligence? The answer
is most definitely no. Short-term positive benefits of music lis-
tening dissipate rapidly as listeners’ arousal level and mood
fluctuate with time and experience. Although music lessons have
relatively long-term benefits, the effort involved (weekly lessons,
daily practice) can hardly be construed as a quick fix. Depending
on the perspective of the child and his or her parents, the benefits
of a few IQ points may or may not be worth the costs in time,
money, and effort. Moreover, different extracurricular activities
(e.g., drama lessons) have salutary effects in other domains (e.g.,
social skills) that could be as valuable as the modest increase in
IQ attributable to music lessons. The simplest take-home mes-
sage is that extracurricular activities benefit child development.
Recommended Reading
Isen, A. (2000). (See References)
Schellenberg, E.G. (2004). (See References)
Schellenberg, E.G. (in press). (See References)
Acknowledgments—Funded by the Natural Sciences and
Engineering Research Council of Canada and the International
Foundation for Music Research. Craig Chambers, Kate McLean,
and Sandra Trehub provided helpful comments on earlier ver-
sions of the manuscript.
REFERENCES
Chabris, C.F., Steele, K.M., Dalla Bella, S., Peretz, I., Dunlop, T., Dawe,
L.A., Humphrey, G.K., Shannon, R.A., Kirby, J.L. Jr., Olmstead,
C.G., & Rauscher, F.H. (1999). Prelude or requiem for the ‘Mozart
Effect’? Nature,400, 826–828.
Craik, F., & Bialystok, E. (2005). Intelligence and executive control:
Evidence from aging and bilingualism. Cortex,41, 222–224.
Husain, G., Thompson, W.F., & Schellenberg, E.G. (2002). Effects of
musical tempo and mode on arousal, mood, and spatial abilities.
Music Perception,20, 151–171.
Isen, A.M. (2000). Positive affect and decision making. In M. Lewis &
J.M. Haviland-Jones (Eds.), Handbook of Emotions (2nd ed., pp.
417–435). New York: Guilford.
Nantais, K.M., & Schellenberg, E.G. (1999). The Mozart effect: An
artifact of preference. Psychological Science,10, 370–373.
Rauscher, F.H., Shaw, G.L., & Ky, K.N. (1993). Music and spatial task
performance. Nature,365, 611.
Schellenberg, E.G. (2004). Music lessons enhance IQ. Psychological
Science,15, 511–514.
Schellenberg, E.G. (2005). Long-term positive associations between
music lessons and IQ. Manuscript submitted for publication.
Schellenberg, E.G. (in press) Exposure to music: The truth about the
consequences. In G.E. McPherson (Ed.), The child as musician: A
handbook of musical development. Oxford, U.K.: Oxford University
Press.
Schellenberg, E.G., & Hallam, S. (in press). Music listening and cog-
nitive abilities in 10 and 11 year olds: The Blur effect. Annals of the
New York Academy of Sciences.
Schellenberg, E.G., Nakata, T., Hunter, P.G., & Tamoto, S. (in press).
Exposure to music and cognitive performance: Tests of children
and adults. Psychology of Music.
Shenfield, T., Trehub, S.E., & Nakata, T. (2003). Maternal singing
modulates infant arousal. Psychology of Music,31, 365–375.
Smith, B.D., Osborne, A., Mann, M., Jones, H., & White, T. (2004).
Arousal and behavior: Biopsychological effects of caffeine. In A.
Nehlig (Ed.), Coffee, tea, chocolate, and the brain: Nutrition, brain,
and behavior (pp. 35–52). Boca Raton, FL: CRC Press.
Thompson, W.F., Schellenberg, E.G., & Husain, G. (2001). Arousal,
mood and the Mozart effect. Psychological Science,12, 248–251.
Thompson, W.F., Schellenberg, E.G., & Husain, G. (2004). Decoding
speech prosody: Do music lessons help? Emotion,4, 46–64.
320 Volume 14—Number 6
Music and Cognitive Abilities
