Volume VII, Issue 1, Winter 2000

Table of Contents

~ ~ ~

The “Mozart Effect”: A Small Part of the Big Picture

To The Point
The Mozart Effect Doesn’t Increase General Intelligence! Best way to long-term benefits is by music study and music making
Music Linked to Reduced Criminality
Musical Brain — Special Brain Area Discovered for Reading Music Scores

Matters of Opinion
The Mozart Effect: Public Enchantment and Public Confusion

~ ~ ~

 

“The Mozart Effect”: A Small Part of the Big Picture
Copyright © 2000 Norman M. Weinberger
and the Regents of the University of California. All Rights Reserved.

 

Studies have shown that music learning and practice also benefit many mental and behavioral processes, including cognitive development, language learning, reading ability, creativity, motor skills, and personal and social adjustment. In contrast to these effects of continual involvement in music, merely brief exposure to some music of Mozart is thought by the public to increase intelligence. Although this “Mozart Effect” has generated a lot of excitement, it is also the most misunderstood aspect of music research. This article reviews the Mozart Effect and explains why it is only a very small part of the very large field of music research. As the Mozart Effect actually does not increase general intelligence and lasts only a few minutes, it does not provide a substitute for music study and practice. Substantial, long lasting effects require deeper and more sustained involvement in music study and music-making.

Studies have shown that music education and music-making have positive effects on many mental and behavioral factors that are themselves not part of music. We have reviewed many of the findings in previous issues of MRN. Reported benefits include the enhancement of general cognitive development (see issues of Spring, ’95; Fall, ’96), language development, reading-readiness and reading (Fall, ’94; Fall, ’95), verbal abilities (Fall, ’96), abstract conceptual thinking and reasoning (Fall, ’94, Fall, ’96), creativity and originality (Fall, ’94; Fall, ’95; Fall, ’96; Spring, ’98; Fall, ’98), memory (Fall, ’96), play improvisation (Fall, ’96), motor development and coordination (Fall, ’96), positive attitudes toward school, improved personal and social adjustment (Fall, ’95) as well as stress reduction (Fall, ’97), and increased quality of life including advanced age (Spring, ’96). [The relevant articles can be found directly at the MuSICA web site (http://www.musica.uci.edu) by going to the MRN cumulative Subject Index and selecting “Benefits of Music”.] These effects appear to result from a considerable investment of time and guided study of music. Such exciting findings show that we are now on the brink of beginning to understand the full importance of music for mental processes and behavior.

In contrast to these findings, public knowledge about music research is currently dominated by the “Mozart Effect”. As widely understood in the public mind, the Mozart Effect is that a few minutes of listening to Mozart increases intelligence. This article describes the Mozart Effect, summarizes the current state of research and places it within the larger context of music research. It also suggests why the Mozart Effect commands an excessive amount of interest. We’ll begin with the last topic first.

Cognitive Bargain Hunting

The mass media have played a major role in starting and maintaining public excitement about the Mozart Effect. Reports have been sensationalized and oversimplified. However, the continued and even growing belief that a few minutes of Mozart improves intelligence also requires a public that is too ready to accept simplistic solutions to complex problems. A couple of examples are relevant.

In 1513, Ponce de Leon “discovered” Florida, trying to find the legendary “Fountain of Youth”, a drink from which was reputed to reverse the aging process. In a 1989 movie, “Indiana Jones and the Last Crusade”, Harrison Ford sought the Holy Grail; one sip from this Chalice of Jesus would confer immortality. Both stories, fact and fiction, reflect a deeply-rooted but generally unspoken human impulse, to “get something for nothing”, or in its more common form, to “get a lot for a little”. In a helpful form, this is seen in shopping as bargain hunting. Unfortunately, people also use this approach for “cognitive” bargain hunting. This is nowhere more obvious than in the public’s fascination with the Mozart Effect, for which the chance to “buy” increased intelligence at bargain prices, the cost of a CD and a bit of listening time, has proven irresistible.

The Mozart Effect — Background

    The First Study — This story began in 1993 when Frances Rauscher, Gordon Shaw and Katherine Ky published a brief paper in the prestigious journal Nature, launching an innovative line of research. They asked whether brief exposure to certain music could increase a cognitive ability. Thirty-six college students were divided into three groups which spent ten minutes in one of three conditions: listening to (1) a piano sonata by Mozart (sonata for two pianos in D, K 448), (2) a tape of relaxation instructions or (3) silence. Immediately after, they were tested on spatial/temporal reasoning. The measurements of spatial/temporal (S/T) reasoning were obtained using subtests from a standard intelligence battery (group) of tests, the Stanford-Binet Test. The important subtest was a paper folding and cutting task (PF/C). The subject has to imagine that a single sheet of paper has been folded several times and then various cut-outs are made with a scissors. The task is to correctly predict the pattern of cut-outs when the paper is unfolded.

The authors found significantly higher scores for the Mozart group than for the relaxation or the silence groups. The differences translated into spatial IQ scores for the Mozart group that were 8-9 points higher than the other two groups. However, the effect was very brief; it did not last beyond 10-15 minutes.1

The authors did not claim that the effects would be limited to Mozart’s music but did think that the benefits to S/T reasoning would require complex rather than repetitive music. However no further definitions of complexity were presented. Also, the authors did not claim that the effects would be found for other aspects of intelligence, such as verbal reasoning or short-term memory, but suggested that these be tested.

The findings were widely reported in the mass media. This was not the first report that music and spatial reasoning were found to be related. Marianne Hassler and her colleagues at the University of Tubingen in Germany had previously discovered a significant relationship between spatial ability and music; students (9-24 years of age) who scored higher on tests of musical ability also scored higher on a test of spatial visualization, i.e., the ability to mentally “see” and “rotate” objects.2 Although this study had revealed a relationship, it was a correlation and not about whether music actually was the cause of better spatial abilities. The report by Rauscher, Shaw and Ky suggested that listening to music actually caused the brain to perform better in spatial reasoning, at least for a few minutes.

As is often the case with public reports of scientific findings, emphasis was placed on the most “sensational” aspects of the findings, i.e., the apparent 8-9 point “boost” in IQ. Two critical and highly limiting factors were generally ignored — first, that the effects were not on general intelligence but only for a test of spatial abilities; second, that the effects lasted only a few minutes. As widely disseminated, the Mozart Effect was born as the idea that listening to Mozart increases intelligence. There was an immediate run on recordings of Mozart’s sonata for two pianos, K. 448, which are still in short supply to this day.3

    The Followup Study — The next stage of this story was provided by a more extensive study by Rauscher, Shaw and Ky, published in 1995.4 In this experiment, 79 college students were tested for their S/T abilities, using the PF/C test. The experiment covered five days. On day 1 they were tested for spatial ability and then divided into three groups that had the same average spatial ability. On days 2-5 they had one of three listening experiences and then were immediately tested again. The three groups received 10 minutes of (1) Mozart’s sonata for two pianos, (2) something different every day; (minimalist music by Phillip Glass one day, a story another day and dance music a third day), or (3) silence every day.

The authors obtained the Mozart Effect. The benefit was greatest for students who were the worst at the PF/C task on Day 1, before the listening treatments were started. However, the effect was found only for the first day of listening; there was no difference between the Mozart and silence groups on days 3, 4 or 5.5 Importantly, the effect was limited also to the PF/C task; the Mozart group showed no superiority on a test of short-term memory.

Once again, the findings were viewed by the media as showing that listening to Mozart increased intelligence. The restriction to the S/T task and the lack of effect on memory were generally ignored, as was the brief duration of the Mozart Effect.

    The Give and Take of Science — Naturally, the Mozart Effect reports also stirred interest in the scientific community. The first attempt to replicate the Effect, published in 1994 by Stough and colleagues in New Zealand, failed to find an effect. However they used a different test than did Rauscher and colleagues.6 This was quickly followed by two more failures to find the Mozart Effect.7 However, both of these studies also used tests that were different from those used by Rauscher and colleagues, who have argued that the tests used don’t actually measure S/T abilities. 8 Steele and colleagues also failed to find a Mozart Effect but tested short-term memory rather than S/T processes.9 As Rauscher et al had specifically found no effect on memory in their 1995 paper, this failure was not surprising.10

Meanwhile, the Mozart Effect was found by some other laboratories. In 1996, Rideout and Laubach compared the effects of a relaxation tape to the effects of listening to the Mozart sonata and reported better performance for the Mozart group on the PF/C task. The following year, Rideout and Taylor obtained the Mozart Effect in a similar study.11 In neither case were the effects very large but they did not occur by chance, suggesting a reliable, if small, positive effect. Rideout, Dougherty and Wernert not only replicated the Mozart Effect, comparing the sonata to a relaxation tape, they extended it to a composition by the contemporary composer Yanni.12 The latter’s music, which is certainly regarded as extremely different from that of Mozart, was selected by a musician to be similar to the Mozart sonata in “tempo, structure, melodic and harmonic consonances, and predictability”. This finding emphasizes the need to have systematic studies to discover the special features of music that may be effective. In 1997, Wilson and Brown did find the Mozart Effect using maze problems to test spatial abilities, compared to relaxation and silence treatments. However, they also found that the relaxation tape produced more maze solutions than the silence condition.13

    Very Recent Developments — The year 1999 saw several publications that have been critical of the Mozart Effect. Steele and colleagues attempted a faithful replication of the Rauscher et al studies, using the PF/C task. They failed to find the Effect in two experiments.14 A combined report of similar studies from three universities (Steele et al) also failed to obtain the Effect, using the PF/C task.15 Chabris performed a quantitative analysis of all Mozart studies and concluded there was no reliable Mozart Effect.16 However, Rauscher argued, once again, that many of the studies used in Chabris’ analysis used tasks that do not measure S/T abilities.17 There still exists some dispute concerning which tests are best at measuring these abilities but Rauscher and Shaw had previously argued that replicated positive Mozart Effect findings used the same or highly similar tasks whereas negative findings used different tests (see footnote # 8). However, negative reports by Steele and colleagues did use the same PF/C task as Rauscher, Shaw and Key (see footnotes 14 and 15).

In trying to reconcile positive and negative results from studies all of which used the same PF/C task, it becomes obvious that apparently small differences between experiments may have large effects. Thus, although Steele and associates used the same PF/C task, examination of their data shows that their subjects were much better at the task than the subjects of Rauscher et al before the Mozart music was played. In fact, their baseline performance was about as good as the improved performance reported by Rauscher. Therefore, the Steele studies may have failed to find an effect because their subjects were already doing as well as they could do with this task. This possibility needs to be studied directly.18

A recent experiment did replicate the Mozart Effect, but with an interesting twist. Nantais and Schellenberg first replicated the Mozart Effect, both with the Mozart sonata and also a piano piece by Schubert, compared to a condition of silence. Next, they ran an experiment comparing Mozart or Schubert to listening to a story. The Mozart Effect was found only for those students who preferred Mozart to the story. The authors argue that the Mozart Effect, while real, is caused when any preferred experience is paired with a less preferred experience. In short, they argue that the Mozart Effect is caused by a more pleasant mood.19 This is highly interesting but it leaves one puzzled about why the Effect should show up, when it does, on a spatial-temporal task and not on other tasks, for which mood benefits might be expected to help as well.

    Summary — Within the scientific community there is no agreement about whether or not the Mozart Effect is genuine. Moreover, the effective features of music that has yielded the Effect are not understood, so systematic analyses of musical characteristics are needed. It is now obvious that an apparently simple novel question, i.e., whether being exposed briefly to certain music can boost a cognitive ability, is not in fact such a simple question. Meanwhile, as the scientific controversy continues, the Mozart Effect has been extended to clinical applications. Thus, Johnson et al reported an improvement in spatial-temporal reasoning in a case of Alzheimer’s disease.20 Furthermore, the Mozart sonata apparently can reduce brain seizures.21

It is important to distinguish the Mozart Effect discussed above from other studies of the effects of music, including Mozart, on behavior, both in children and adults.22 Many people believe that the Mozart Effect described here applies to children but such studies have never been done with young children. Also, the effects of long term listening to music and long term involvement in music lessons are quite another matter from the Mozart Effect reviewed here.

In any event, the larger context and implications of the Mozart Effect have been mainly ignored within the scientific community and grossly misrepresented and distorted within the public domain. It is now time see how the Mozart Effect fits into the “big picture”.

The Mozart Effect in the Big Picture of Music and Behavior

Setting aside current controversies, which will eventually be resolved, let us assume that the Mozart Effect is genuine and that 10 minutes of exposure to Mozart (and some other music) does increase the ability to perform S/T tasks for a period of 10-15 minutes. How important is this finding? The importance of a finding depends on the question being asked. If the question is whether or not brief exposure to certain music can be used to understand aspects of cognition and brain function by linking music listening to S/T reasoning, then the answer may be positive; The Mozart Effect could provide an important scientific tool in this research.

However, if the question is whether or not brief exposure to certain music can produce long term improvements in intelligence, either limited to spatial/temporal abilities or to more general intelligence, then the answer is no. As currently understood, the most optimistic views of the Mozart Effect are that it is very brief and also largely restricted to certain S/T tasks. Thus, its practical implications seem minimal. Whether or not long term listening to music would have long term effects remains to be studied.

In any event, the field of music research is very extensive and has spanned the entire Twentieth Century. So let’s consider the Mozart Effect within the large domain of music research. The Table below shows how two factors are related: the “Duration of Some Effects of Music on Cognition and Behavior” and the “Amount of Effort, Involvement” with music. This Table is not intended to be comprehensive, only illustrative.

 

Amount of Effort, Involvement

Duration of Some Effects of Music on Cognition and Behavior

Minutes

Years

Passive Listening
for 10 minutes

“Mozart Effect”
(increases ST reasoning)

none

Educated Listening
in music classes for one or more school years

none

Understanding and appreciating musical forms, genres, meanings and performances in historical, social and cultural context

Instrumental or vocal lessons and regular practice for several years

none

Reading musical notation, integrating sight, sound, touch and movements to perform and express self musically, solo, in cooperative group or both

 

The Mozart Effect requires only 10 minutes of exposure (not necessarily even attentive listening) to music. This is certainly at the most minimal levels of Involvement and Effort with music. As for Duration of Effect, it is only a few minutes. This is certainly at the briefest of durations. So the Mozart Effect is placed in the upper left box of the table. Are there any known long lasting effects, that is years, of exposure to 10 minutes of music? Not to the best of my knowledge — so the upper right box shows “none”.

What about the effects of educated listening or music lessons? Do these last only a few minutes? Obviously not, hence “none” summarizes these non-existent states of affairs.

Now consider the other parts of the Table that concern effects that last over years. The examples given, while not named as formal “effects” are nonetheless very well-established. Educated listening and study require a lot of effort, so much more than ten minutes of exposure to music that it might as well be infinitely more. How long do the effects of education about music last? Certainly years. It is not the remembering of a particular fact from music class that is important — do you really remember Brahm’s birthdate, I don’t — but rather the ability to hear and think about music in many ways at different levels of understanding. How can the music of Johann Sebastian Bach, Ludwig van Beethoven, Duke Ellington and John Lennon all sound so different and yet all be so sublime? Does the educated listener “hear more” in a composition than a naive listener? Certainly.

Finally, look at the duration of effects of instrumental and vocal instruction and practice. These activities within music itself require even more effort over a longer period of time than does learning about music. And commensurately, the effects last years, even a lifetime. The issue here is not how well one sings or performs over ten, twenty or fifty years, but that one does them at all.

Lessons for the Future

This is not the occasion to discuss how and why hard work and concentrated effort in music over months and years produce long lasting changes in our brains, but they do. They literally change us, for we are our brains. But it is the time to revisit Ponce de Leon, the “immortality sip” from the Holy Grail and other manifestations of our desire to get something for nothing, i.e., increased intelligence from brief exposure to music.

Given all else we know about the realities of life, why would anyone believe in this mental “get rich quick” scheme? It can’t be because of Rauscher, Shaw and others who have reported positive Mozart Effect findings, because none of them claimed a meaningful (i.e., long lasting) effect on intelligence.23 Well, as P.T. Barnum said “There’s a sucker born every minute”. Yes, folks will continue to seek not only monetary bargains but also the Easy Street to cognitive benefits. But the Mozart Effect “bus” doesn’t go there. I don’t agree with P.T. Barnum. While a potential sucker is born with every human birth, our “sucker potential” doesn’t have to become real because a rational human being also comes with every birth. An informed public can certainly exercise judgement and caution.

The best ways to gain long term benefits from music are still to devote guided effort and hard work to its study and performance. This bus does go to “Brain Street”.

— N. M. Weinberger

1  Rauscher, F.H., Shaw, G.L. and Ky, K.N. (1993) Music and spatial task performance. Nature, 365:611.

   The authors believe that the PF/C task involves two mental processes: spatial imagery and the temporal ordering of spatial components. For their discussion of this issue, see Rauscher and Shaw, cited in footnote #8.

   This experiment was an extension of previous research done by Dr. Shaw concerning patterns of brain activity and mental processes. Although discussion of these aspects of the research lie outside the scope of this article, we can note that the basic idea was that there are different patterns of brain activity in which brain cells work together across time in various brain regions to accomplish various mental tasks; the music of Mozart was thought to “jump start” patterns that would be best for spatial-temporal reasoning, such as used in the paper-folding task.

   As a friendly colleague of Dr. Shaw’s for many years at the University of California Irvine, I have been well aware of his research program and findings, having discussed many aspects of this work over a period of years. We also jointly established MuSICA (first called the Music and Brain Information Center). Given this close past association, I have made special efforts to comment on Dr. Shaw’s research in a standard neutral manner.

2  Hassler, M., Birbaumer, N. and Feil, A. (1985). Musical talent and visual-spatial abilities: a longitudinal study. Psychology of Music, 113:99-113.

3  As I have made clear elsewhere (MRN, Spring, 1998, Matters of Opinion, “On the Importance of Being Accurate”), the indiscriminate use of the term “The Mozart Effect” to refer to any situation in which music in any form has positive effects on cognition or behavior, is both wrong and misleading. In this article, I use this term to refer specifically to studies concerned with the effects of brief exposure to Mozart (and some other compositions) on tests of mental abilities of young adults.

4  Rauscher, J.H., Shaw, G.L. and Ky, K.N. (1995). Listening to Mozart enhances spatial-temporal reasoning: towards a neurophysiological basis. Neuroscience Letters, 185:44-47.

5  The “something different” group was reported to be significantly worse than the Mozart and silence groups on days 3, 4 and 5. The authors conjectured that the Mozart Effect reached the maximum possible level after only one 10 minute period of listening (on Day 2), so they could not have done better on days 3, 4, 5; they also suggested that the silence group caught up to the Mozart group because of learning and practice on the task. To the best of my knowledge, these hypotheses have not been directly tested.

6  Stough, C., Kerkin, B, Bates, T. and Mangan, G. (1994). Music and spatial IQ. Personal. Individ. Diff., 17:695. However, the methodology of this particular apparent failure to replicate has been criticized by Kenealy and Monsef; see footnote #18.

7  Carstens, C.B., Huskins, E. and Hounshell, G.W. (1995). Listening to Mozart may not enhance performance on the revised Minnesota Paper Form board Test., Psychol. Rpt., 77:111-114. Newman, J., Rosenback, J.H., Burns, K.L., Latimer, B.C., Matocha, H.R. and Vogt, E.R. (1995). An experimental test of “the mozart effect”., Percept. Mot. Skills, 81:1379-1387.

8  Rauscher, F.H. and Shaw, G.L. (1998), Key components of the Mozart effect, Percept. Mot. Skills, 86:835-841.

9  Steele, K.M., Ball, T.N. and Runk, R. (1997). Listening to Mozart does not enhance backward digit span performance. Percept. Mot. Skills, 84:1179-1184.

10  A brief report found no significant differences between listening to Pachelbel’s Canon in D vs. Bartok’s Music for Strings, Percussion and Celeste, Movement 2. The authors considered the Pachelbel to be “highly structured” but apparently not the Bartok. Of the several tests used, none were like the PF/C task. Also, no non-music control group was used. This study is not very illuminating and the title (see following), which suggests an effect, is misleading. Cash A.H., El-Mallakh, R.S., Chamberlain, K., Bratton, J.Z. and Li, R. (1997). Structure of music may influence cognition., Percept. Motor Skills, 84:66.

11  Rideout, B.E. and Laubach, C.M. (1996). EEG correlates of enhanced spatial performance following exposure to music. Percept. Mot. Skills, 82:427-432. These authors also recorded brain waves, the results of which are not critical for study of the replication of the Mozart Effect.

     Rideout, B.E. and Taylor, J. (1997). Enhanced spatial performance following 10 minutes exposure to music: a replication. Percept. Mot. Skills, 85:112-114.

12  Rideout, B.E., Dougherty, S. and Wernert, L. (1998). Effect of music on spatial performance: a test of generality. Percept. Mot. Skills, 86:512-514.

13  Wilson, T.L. and Brown, T.L. (1997). Reexamination of the effect of Mozart’s music on spatial-task performance. J. Psychol., 1997, 13:365-370. These authors also found that the order in which Mozart was presented (before or after silence or relaxation instructions) was important, suggesting that the circumstances for obtaining the effect may be somewhat limited.

14  Steele, K.M., Brown, J.D. and Stoecker, J.A. (1999). Failure to confirm the Raushcer and Shaw description of recovery of the Mozart effect. Percept. Motor Skills, 88:843-848. Steele, K.M., Bass, K.E. and Crook, M.D. (1999). The mystery of the Mozart effect: failure to replicate. Psychol Science, 10:366-369.

15  Steele, K.M., Dalla-Bella, S., Peretz, I., Dunlop, T., Dawe, L.A., Humphrey, G.K., Shannon, R.A., Kirby, J.L. and Olmstead, C.G. (1999). Nature, 400:827. Details given are insufficient to evaluate the particular methods used.

16  Chabris, C. (1999). Nature, 400:826-827. This quantitative “meta-analysis” included both S/T and non-S/T tasks, the latter may be considered irrelevant. However, according to Chabris, the S/T tasks alone generally showed only insignificant effects.

17  Rauscher, F.H. (1999). Nature, 400:827-828.

18  A brief letter published in 1994 had reported a failure to find the Mozart Effect using the PF/C task. However, insufficient details were provided to permit an assessment of the validity of the negative findings. Kenealy and Monsef (1994). The Psychologist, 7: 346.

19  Nantais, K.M. and Schellenberg, E.G. (1999). The Mozart effect: an artifact of preference. Psychol. Science, 10:370-373. The Shubert piece was Fantasia for Piano, Four Hands in F minor (D940). If the “mood change” explanation is correct, then listening to a pleasing story also should have produced better performance on the paper folding/cutting task, compared to Mozart. While the authors found this tendency, the effect did not reach statistical significance. Rauscher (personal communication) has found that students preferred a Mendelson piece to Mozart but the S/T improvement was only found with the Mozart music.

20  Johnson J.K.; Cotman, C.W.; Tasaki, C.S. and Shaw, G.L. (1998). Enhancement of spatial-temporal reasoning after a Mozart listening condition in Alzheimer’s disease: a case study. Neurol. Research, 20:666-672.

21  Hughes, J.R., Fino, J.J. and Melyn, M.A. (1999). Is there a chronic change of the “Mozart effect” on epileptiform activity? A case study. Clinical Electroenceph. 30:44-45. Hughes, J.R., Daaboul, Y., Fino, J.J. and Shaw, G.L. (1998). The “Mozart effect” on epileptiform activity. Clinical Electroenceph. 29:109-119.

22  For other aspects of the effect of the Mozart sonata on behavior and on brain activity, see Shaw, G.L. (1999). “Keeping Mozart in Mind”, New York: Academic Press.

23  Chabris (personal communication) has argued that the original report (see footnote #1) implied an effect on general intellectual abilities because of the way that the tests were described. I did not get this impression because the authors’ discussion distinguished S/T from other abilities. He has also correctly pointed out that the specific restriction to S/T tasks was not clarified by Rauscher and Shaw until their 1998 paper where the PF/C task was shown to exhibit the Mozart Effect whereas other tasks did not (see footnote #8).

~ ~ ~

 

To The Point
Copyright © 2000 Norman M. Weinberger
and the Regents of the University of California. All Rights Reserved.

 

This issue inaugurates a new feature of MRN, “To The Point”, which replaces “Recent Publications.” “To the Point” provides a rapid way to get up-to-date information on important findings, in a reader-friendly format. It will also provide a capsule summary of a lengthy MRN article, as in the present issue which contains “The ‘Mozart Effect’: A Small Part of the Big Picture” We hope you will find “To The Point” interesting and helpful. We’d be pleased to hear from you.

— N. M. Weinberger

 

THE MOZART EFFECT DOESN’T INCREASE GENERAL INTELLIGENCE!
BEST WAY TO ACHIEVE LONG-TERM BENEFITS IS BY MUSIC STUDY AND MUSIC-MAKING

The “Mozart Effect”, reported as an increase of intelligence after briefly listening to Mozart, is by far the most well-known and sensational finding in music research in the public mind. At the same time, it is the most misunderstood. Trumpeted in the mass media as a way to increase general intelligence, it does no such thing, as shown in a detailed article in this issue of MRN.

Francis Rauscher and Gordon Shaw first reported in 1993 that 10 minutes of listening to a Mozart sonata for two pianos increased the ability of college students to solve spatial-temporal problems. For example, subjects had to imagine that a single sheet of paper had been folded several times in certain ways and then various cuts made with a scissors. They had to then predict what the pattern of cut-outs would look like when the paper was unfolded. This test requires mentally keeping track of the sequence of events in space. Rauscher and Shaw specifically pointed out that the effect lasted only 10-15 minutes and worked only for spatial-temporal problems, not for other aspects of intelligence.

The public’s belief of long term effects in general intelligence was fed by simplistic reports in the media combined with its tendency to believed in “cognitive bargains”, i.e., getting a big boost in mental processes without expending much effort. The Mozart Effect is more important for researchers than for any practical applications. Long term benefits from music are best achieved by intensive study and music-making.

MUSIC LINKED TO REDUCED CRIMINALITY
Martin Gardiner of Brown University recently reported, at a national arts education meeting in Oklahoma City, the results of analyzing a large scale data base. The data base included information gathered over a period of many years for more than a thousand residents of Rhode Island. Tracking people from birth through the age of thirty, Gardiner checked the relationship between arrest records of teen-agers and their degree of involvement in music.

Gardiner found that the greater the involvement in music, the lower the arrest record. Teens who had music education were less likely to get into trouble than students who didn’t. However, those who also were involved in playing a musical instrument had even fewer brushes with the law. Those who had the most experience, including good sight-reading ability, had a negligible arrest record. This research, still in progress, was funded by the International Foundation for Music Research (IFMR).

MUSICAL BRAIN — SPECIAL BRAIN AREA DISCOVERED FOR READING MUSIC SCORES
Music is wrongly considered to be mere entertainment and often regarded as an educational frill. However, research has shown that humans are born with musical capabilities, so music is part of human nature. This is particularly evident in research that has shown how the human brain processes music. Recently, neuroscientists have discovered an area in the brain that is devoted to reading music scores.

T. Nakada and his co-workers at the University of California and the Niigata Brain Research Institute in Japan applied brain-imaging techniques to people who could read a musical score. They compared brain activity during score reading and also during reading language. Reporting in the journal NeuroReport (1998), Nakada and colleagues found regions of the brain that were involved in both types of reading. However, most importantly, they also discovered a brain area that was activated only during reading musical scores. This is near the visual part of the brain in the right hemisphere. The findings reveal that the human brain is specialized for music and therefore the human brain is a “musical brain”.

~ ~ ~

 

Matters of Opinion
Copyright © 2000 Norman M. Weinberger
and the Regents of the University of California. All Rights Reserved.

 

The Mozart Effect: Public Enchantment and Public Confusion

The following opinions about music are intended to provoke thought and sometimes perhaps even argument, but ultimately to energize and enlarge conceptions and inquiry about music.

The dominance of the Mozart Effect in the public eye prompts the dedication of this issue to the topic. MuSICA receives innumerable inquiries about this phenomenon, some of which reveal that very serious actions are being based on the false belief that briefly listening to Mozart, or perhaps other classical music, will result in a meaningful increase in intelligence. In one such case, a women had given a Mozart CD to a teenage girl who had given birth to two children. The well-meaning women believed the music would raise the girl’s IQ and thus forestall more pregnancies. However, the good Samaritan had recently read in the paper that the Mozart Effect was not real. She was confused and worried. What was the truth, she wanted to know?

Such highly oversimplistic views of the effects of music on behavior cause real problems, perhaps harm. In the present case, the teenage mother needed another type of help. Harm is done to the extent that belief in a non-effective treatment delays psychological counseling or medical help. There are other sorts of bad effects such as the “negative halo effect” in which anything similar to the failed treatment takes on the same negative aura. So we get “I read that the Mozart Effect isn’t real … I’m not going to waste my time and money having my kids listen to classical music anymore.” What rotten logic! It says, “If I can’t get instant gratification, then I won’t help my children on a longer term basis.”

Although the Mozart Effect may not live up to the unjustified hopes of the public, listening to ten minutes of Mozart could get someone interested in listening to more unfamiliar music, opening up new vistas. And many children get interested in playing a musical instrument after hearing any new music, including Mozart. At the very least, and beyond the fact that the Mozart Effect may serve as a research tool, it has brought widespread interest in music research to the public. This interest now needs to be sustained by drawing attention to the many other aspects of music’s effects on human thought and behavior and to the need for more research on the long term effects of music study and music performance.

— N. M. Weinberger

* * *


—>    MuSICA Home Page     Cumulative Article Index     Subject Index     Previous Issue     Next Issue