Ten Minutes of Mozart

In 1993 Nature published a one-page letter by Frances Rauscher, Gordon Shaw and Katherine Ky of the University of California, Irvine. Thirty-six college students were tested on spatial reasoning tasks from a standard intelligence test, and three conditions were compared: ten minutes of Mozart's Sonata for Two Pianos in D major, K. 448; ten minutes of relaxation instructions; ten minutes of silence. After the Mozart their scores were higher, by the equivalent of eight or nine points on the test's spatial scale. The effect had gone within ten to fifteen minutes.

That is the whole finding. It concerns adults, one kind of task, and a quarter of an hour.

What happened next is a case study in how a result travels. The phrase the Mozart effect was taken up by the press and then by a popular book of that title. The adults became babies, the spatial tasks became intelligence, and the quarter of an hour became a childhood. Shops filled with classical recordings for infants. In January 1998 Zell Miller, the governor of Georgia, asked the state legislature for 105,000 dollars to give every newborn in the state a classical recording, and played the legislators an excerpt of Beethoven's Ninth to make the case.

Meanwhile other laboratories tried to reproduce the original result, with mixed success. In 1999 Christopher Chabris pooled the available studies in Nature and found the effect small, by his own description about one and a half IQ points, and confined to one kind of paper-folding task. The same year Kenneth Steele and colleagues published a careful failure to replicate. In 2010 Jakob Pietschnig, Martin Voracek and Anton Formann pooled about forty studies with some three thousand participants and found a small effect, signs of publication bias, and a telling pattern: studies run by Rauscher and her associates found noticeably larger effects than everybody else's.

The more interesting work asked why any effect appeared at all.

The answer that has held up is arousal and mood. In 2001 William Thompson, Glenn Schellenberg and Gabriela Husain set the same Mozart sonata, which is quick and bright, against a slow, sad adagio. Mozart beat silence; the adagio did not; and listeners after the Mozart reported feeling more aroused and in a better mood. When those differences were held constant statistically, the Mozart advantage disappeared. Two years earlier, Kristin Nantais and Schellenberg had found that when the comparison was a story read aloud rather than silence, Mozart had no advantage at all, and listeners did best after whichever of the two they preferred.

The best demonstration used the British public. In 1996 the BBC programme Tomorrow's World, advised by the psychologist Susan Hallam, arranged for schools across the country to hear ten minutes of Mozart on Radio 3, pop songs including Blur's Country House on Radio 1, or a discussion of the experiment on Radio 5, and then to take two spatial tests. Some eight thousand ten- and eleven-year-olds took part. When Schellenberg and Hallam analysed the results, the pop group did best on the paper-folding test, and on the other test the music made no difference. Their paper's subtitle was The Blur Effect.

Nobody concluded that Blur is good for the brain. The children liked the pop more, it lifted them more, and they did slightly better for a few minutes. That is the arousal account again, with a different playlist.

None of this means music does nothing for a mind. Whether learning to play an instrument, over years, has measurable effects on other abilities is a separate question, with its own evidence and its own disputes. It is simply not the question the Mozart effect was about, and blurring the two together is part of what the phrase achieved.

The lesson travels well beyond music. A small, specific, short-lived effect in one population became a large, general, permanent effect in another, and nobody had to commit fraud for it to happen. Each retelling dropped one qualification, and after enough retellings there were none left.