The Mozart effect

Products such as the Baroque-a-bye Baby CD claimed that slow Baroque music at 60 beats per minute would calm and stimulate babies.

In 1998 the governor of Georgia, Zell Miller, announced $105,000 a year so that every newborn in the state could receive a classical music recording.

The New York Times reported that the findings could give college-bound students an edge on the SAT.

The claim began with a short Scientific Correspondence, "Music and spatial task performance", by Frances Rauscher, Gordon Shaw and Catherine Ky. Nature published it on 1 October 1993, in volume 365 on page 611.

The participants were college students, not babies or children.

They did better on standardized tests of spatial ability after listening to a Mozart sonata than after listening to relaxation instructions or sitting in silence.

The piece was Mozart's Sonata for Two Pianos in D major, K. 448, written in 1781. The sonata typically runs 8 minutes and 24 seconds.

In 1999 Kenneth Steele, Karen Bass and Melissa Crook set out to repeat one of the original positive experiments. They followed the procedures the original authors had listed as necessary.

They found neither a statistically significant effect nor an effect size of practical importance. They concluded that there was little evidence for basing intellectual intervention programs on the Mozart effect.

Their result matched earlier independent work. Several labs had been unable to confirm the effect, even though the original lab had reported two positive results.

Rauscher and Shaw argued in 1998 that an exact replication of the original design was needed to see the effect.

In a 2001 study in Psychological Science, William Thompson, Gabriela Husain and Glenn Schellenberg compared a lively Mozart sonata with a slow, sad Albinoni adagio and with silence. Spatial performance improved after music only for people who heard Mozart.

The improvement tracked mood and arousal. When mood and arousal were the same in the music and silence conditions, Mozart had no effect on spatial ability.

The authors argued that the original studies lacked appropriate control conditions. Enjoyable stimuli raise positive mood and arousal, which in turn bring modest gains on a range of tasks.

In 2010 Jakob Pietschnig, Martin Voracek and Anton Formann of the University of Vienna published a meta-analysis in the journal Intelligence. It covered about 40 studies, published and unpublished, with 104 independent samples and more than 3,000 participants.

Compared with no stimulus or a non-musical one, Mozart's K. 448 gave an effect size of d = 0.37. Other music gave d = 0.38, about the same.

The authors concluded that there is only little support for a specific Mozart effect. The University of Vienna's release said the evidence offered no support for gains in spatial ability specifically due to listening to Mozart.

The analysis found strong signs of publication bias in studies comparing the Mozart sonata with non-musical or silent conditions, where published effects were larger than unpublished ones. For Mozart against no stimulus, effects were about three times larger in the Rauscher or Rideout labs than in other labs.

The lead author said: "I recommend listening to Mozart to everyone, but it will not meet expectations of boosting cognitive abilities."

For at least 10 minutes after music that lifts mood stops, performance can improve on some tests, those most relevant to music.

The authors of the 2010 analysis suggested that more arousing music may boost spatial performance more, which could explain the small gap between Mozart and other music.

Learning music is a different matter from listening to it. In a 2008 study, nonmusicians who learned a piano sequence over two weeks showed more auditory cortex enlargement than those who only listened and made judgments.