Skip to content

← Psychology

Myths about ability

Chapter 20 · How people differ

In this chapter 5 sections
  1. Learning in your own style
  2. Many kinds of clever
  3. Music and a quick boost
  4. Training the brain
  5. Believing you can grow

In 2012 a team of researchers gave 242 schoolteachers in Britain and the Netherlands a list of 32 statements about the brain and asked which were true. These were keen teachers, the kind who read about the science of learning. One statement was believed by more of them than any other: “Individuals learn better when they receive information in their preferred learning style.” In Britain 93 teachers in 100 said it was true. In the Netherlands it was 96.

It isn’t, as far as anyone has been able to show. And it has company. Ideas about learning in your own style, about many kinds of clever, about Mozart, about brain games and about believing you can grow have all traveled fast through schools and newspapers, and each of them grew out of something true. That is why they are so easy to believe, and why each one deserves to be told properly: what was really found, what happened when other people checked, and what is left standing. Usually something is left. It is just smaller, and different, from the story people tell.

Learning in your own style

The idea goes like this. Some people are visual learners, who learn best from pictures and diagrams. Some are auditory, and learn best by listening. Some are kinesthetic, and learn best by moving and doing. Find out which you are, with a questionnaire, and then you should be taught, and should study, in your own style.

Part of this is plainly true. People do have preferences. Ask a class whether they would rather read a chapter or hear it read aloud, and they will tell you, and they will tell you differently. People also differ in some fairly specific abilities: one person is better at picturing shapes, another at remembering words.

The claim that does the work is a different one: that teaching matched to your preference makes you learn more. And that claim can be tested. In 2008 four psychologists were asked to look at the evidence, and they said what a fair test would need. Sort students by their style. Then give each group, at random, one way of teaching or another. Then give everyone the same test at the end. If styles matter, the results should cross: the visual learners should do best with pictures, and the listeners best with listening. If instead one way of teaching works better for everyone, styles don’t matter, however real the preferences are.

They found almost no studies that had done this, and of the few that had, several found the opposite of the matching idea. Their verdict was that there was no adequate evidence for using learning-styles tests in schools. Many versions had simply never been tested, they added, so nobody could say they had all been ruled out.

Since then, people have run the test. In 2015 a team in the United States gave adults a passage to read or to hear, after sorting them as visual or auditory learners. Then, in 2020, they did the same with 125 children of 10 and 11 at a school in rural Pennsylvania. Neither time did a person’s style tell anything about whether they understood better by reading or by listening. If anything, the children sorted as visual learners did better at both. In another study, of 426 university students learning anatomy, most didn’t study in their supposed style anyway, and whether they did made no difference to their marks. What went with good marks was what they actually did, such as working with the virtual microscope, whatever their style.

So why does it feel so true? One experiment caught a reason. People who said they preferred words felt surer they had learned a list of words, and people who preferred pictures felt surer about pictures. But when they were tested, their preference made no difference to what they remembered. A style seems to change how learning feels, not how much of it happens.

The ways of studying that do help, testing yourself and spreading your study out, help nearly all learners alike: they are in the chapter on memory, and there are small habits built on them.

Many kinds of clever

In 1983 the American psychologist Howard Gardner, of Harvard, published a book called Frames of Mind with an idea that spread widely among teachers. He argued that intelligence isn’t one thing but several, each fairly independent of the others. He put it this way himself: instead of one all-purpose computer in your head, you have a set of separate ones.

He named seven: linguistic (words), logical-mathematical (numbers and reasoning), spatial (pictures and shapes), musical, bodily-kinesthetic (using the body, as a dancer or a surgeon does), interpersonal (understanding other people) and intrapersonal (understanding yourself). Later he added an eighth, the naturalist’s, and he has said there may be a few more. To decide what counted, he set eight criteria drawn from several fields, among them whether damage to one part of the brain can take away that ability alone, and whether there are people with one extraordinary gift and little else.

The idea spread through schools around the world, Nepal’s among them. CDC’s curricula, in subject after subject, ask teachers to keep their students’ बहुबौद्धिकता, their multiple intelligences, in mind when they teach and when they assess. Its appeal is easy to see. It says that everyone can be clever at something, and that a child who struggles with sums may shine somewhere else.

The trouble is the evidence. Gardner’s idea has been heavily criticized for having little. Yet it does predict something checkable: if the intelligences are really separate, being good at one shouldn’t tell you much about the others. In 2006 three psychologists tried it. They gave 200 adults two tests for each of Gardner’s eight intelligences, built from his own descriptions. The tests of the mainly mental abilities, words, logic, space, nature and even understanding other people, all went together, and strongly. They shared a large general factor, the g that Charles Spearman found in schoolchildren a century before (see the chapter on intelligence). Only the tests that leaned on the body, the senses or personality, such as the musical and bodily ones, stood more apart. Within most of the eight, the two tests did share a little beyond g, some modest support for the idea that each area hangs together.

Gardner has answered his critics that they misunderstand and oversimplify the theory, and the argument goes on. But in this test of 200 adults, the plain prediction, that the mental intelligences don’t go together, didn’t hold.

What survives? People do have real strengths and weaknesses, and a test of general intelligence leaves out a great deal, music and movement among them. And Gardner himself, tired of hearing his idea confused with learning styles, wrote in 2013 that they are different things: an intelligence is the power of a part of the mind, whatever sense its information arrives through, while “visual learner” and “auditory learner” were, he said, labels that were unhelpful at best. His advice to teachers was to teach each important thing in several ways, through stories, pictures, diagrams and acting it out, and to drop the word “styles” altogether. That advice doesn’t depend on the theory.

Music and a quick boost

The Mozart story began with a short letter in the journal Nature in 1993. Frances Rauscher and two colleagues at the University of California, Irvine, had tested 36 college students. Each student was tested three times: once after 10 minutes of Mozart’s Sonata for Two Pianos in D major, once after 10 minutes of a tape of relaxation instructions, and once after 10 minutes of silence. Each test was a spatial puzzle from an intelligence test: folding and cutting paper in their heads, or finding the missing piece of a pattern.

After the Mozart, they did better. Turned into the scale of an IQ test, their scores came to 8 or 9 points higher than after the tape or the silence. The gain lasted no longer than the 10 or 15 minutes of the puzzle.

That was all: one piece of music, one kind of puzzle, college students, a quarter of an hour. Newspapers reported that Mozart makes you smarter. In 1998 the governor of the American state of Georgia proposed that every newborn baby there go home with a classical music CD, and Florida passed a law making state-funded day-care centers play classical music for at least an hour a day. Rauscher herself later called the belief that Mozart raises general intelligence a misconception, and said her study had given no sign that the effect was special to Mozart.

He wasn’t. In 1999 two Canadian psychologists ran the test again with a twist. Mozart and a piece by Schubert both beat silence. But when the alternative to Mozart was a short story by Stephen King, read aloud, the music lost its lead: people did better after whichever of the two they preferred. Something you like seems to wake you up for a while, and being a little stirred up helps with that kind of puzzle.

In 2010 three researchers in Vienna gathered nearly 40 studies, with over 3,000 people, including unpublished theses that had never made the news.

Mozart, against no musicOther music, against no musicMozart, against other musicno gain0.20.40.6size of the gain (0.2 small, 0.5 medium)
Figure 20.1 How much better people did on a spatial puzzle just after listening, pooled from nearly 40 studies. Each dot is the average gain and each bar its likely range. Mozart beat no music by about as much as any other music did, and beat other music by little. Pietschnig and colleagues, 2010.

Listening to Mozart did give a small lift over hearing no music, whether silence or something else, and listening to other music gave about as much. Mozart against other music hardly differed. The studies also showed signs that small, disappointing ones had gone unpublished, which makes the lifts look bigger than they are, and the studies done by Rauscher’s own group found noticeably larger effects than everyone else’s. The researchers concluded that there was little evidence left for any special Mozart effect.

So a small real thing survives: a few minutes of music you enjoy can briefly help with one sort of puzzle. It doesn’t make a baby, or anyone, more intelligent. Whether years of learning to play an instrument change a child is a separate question, with its own arguments.

Training the brain

Brain-training games make a bigger promise. Play these puzzles for a few minutes a day, the advertisements say, and your memory, attention and reasoning will sharpen, at school, at work, into old age.

Here, too, the true part comes first. Practice makes you better at whatever you practice. Nobody doubts that you get faster at a game by playing it. The question is whether the gain spreads to things you didn’t practice.

In 2010 a team of British scientists tested this with the help of a BBC science program. Its viewers were invited to train online for six weeks, at least 10 minutes a day, three times a week, and 11,430 of them finished the tests before and after. One group trained on puzzles of reasoning, planning and problem solving, a second on memory, attention, shapes and mathematics, the kind of tasks commercial games use. A third group did no training at all: they answered obscure general-knowledge questions, looking up the answers anywhere they liked online. Before and after, everyone did four tests of mental ability.

The trainees got much better at the games they played. But on the four tests, all three groups improved by about the same small amount, the people looking up trivia included. Six weeks of training had made people good at the training.

Others have found the same. A review of 23 studies that trained working memory, the mind’s small notepad, found real gains on tasks like the ones trained, which mostly faded, and no convincing spread to reading words, arithmetic or reasoning. In 2014 two groups of scientists published open letters that contradicted each other, one of more than 70 saying brain games had no good scientific basis, one of 133 saying they did. So in 2016 a team went through the published studies that the leading brain-training companies cited on their own websites as their evidence. There was plenty of evidence that people improved on the trained tasks, less that they improved on similar tasks, and little that the training helped with different tasks or with everyday life. None of the studies met every standard of a fair test.

The same year, the American government’s consumer watchdog, the Federal Trade Commission, charged Lumos Labs, the maker of the Lumosity games, with deceiving customers by claiming, without the evidence to support it, that its games improved performance at school and work and held off dementia. The company agreed to pay $2 million to settle the charges, which were never proved in court. The head of the commission’s consumer protection bureau said the company “simply did not have the science to back up its ads”.

What does seem to spread is broader and slower: years of school, which raise scores on intelligence tests. A puzzle game is fun, and that is a perfectly good reason to play one.

Believing you can grow

The last idea is the most hopeful, and its story is the most argued over.

The American psychologist Carol Dweck studied what children believe about ability. Some think intelligence is fixed, something you have or lack. Others think it grows with effort and good teaching. She called these a fixed mindset and a growth mindset. In a famous set of six studies in 1998, she and a colleague praised fifth graders for their work, some for being clever and others for working hard. Then the children met a failure. Those praised for being clever enjoyed the tasks less afterward, gave up sooner and did worse than those praised for effort, and they were more likely to describe intelligence as fixed. The story spread from there: praise effort, not talent, and teach children that the brain is like a muscle that grows stronger with use.

Part of it is certainly true. Ability can grow: schooling raises intelligence-test scores, and practice builds skill. And the American Psychological Association, among others, says that what students believe about their ability can affect how they learn. The hard question is whether a short lesson that teaches the belief changes anything that shows up in a report card.

A large, carefully run test was published in Nature in 2019. Its plan was written down and published before the results came in, and the data were collected by people independent of the scientists. It took 12,490 ninth graders, in 65 schools chosen to stand for the whole of the United States, and gave half of them, chosen at random, a growth mindset lesson online: two short sessions, under an hour in all. The other half were the comparison. For the students who were already doing well, grades didn’t change. For those in the lower half of their school, grades rose by about a tenth of a grade point on a scale that runs to about 4, small but real. More students overall signed up for harder mathematics. And the gains were largest in schools where classmates themselves were willing to take on hard work.

Reviews of all the other studies tell a cooler story. One in 2018 pooled 43 studies of mindset lessons and found the effect weak overall, with signs that it might help poorer or struggling students. Another, in 2023, gathered 63 studies, with nearly 100,000 students, and found an average effect so tiny that it could no longer be told apart from nothing once they allowed for studies that had found nothing and gone unpublished. In the best-designed studies there was no clear effect at all. And studies by authors with money at stake in good results reported larger effects than the rest. A series of studies in China with 624 children aged 9 to 13, beginning with a careful repeat of the 1998 praise studies, found little of what the original had reported.

The other side answered that averaging every study together misses the point. A lesson aimed at struggling students shouldn’t be expected to help students who are already doing well, and a second review of the same year, looking at the students expected to benefit and at lessons delivered as planned, found small real gains there. The two sides still disagree, mostly about how to add the studies up.

So a short lesson in believing you can grow does little for most students on average, and the early promises were far too big. For students who are struggling, in a school where trying hard is respected, a short lesson has done a little good, and it costs almost nothing. The belief at its heart is the true part. Ability does grow. What makes it grow is the work, the teaching and the next attempt, which is why the “not yet” habit is about changing what you do, not only what you say.

Put the five side by side and a pattern shows. Each started from a real finding, small and particular: people have preferences, abilities have some parts, music can wake you up, practice improves the thing practiced, beliefs can matter. Each then grew into a large and simple promise, because a large and simple promise is easy to repeat, flatters the person hearing it, and in some cases sells well. And each shrank again when people tested it carefully, on many people, and compared fairly.

None of that makes a teacher who used these ideas foolish. It means that the way to find out whether something helps people learn is the same as for anything else in this book: try it on many people, compare it fairly, and believe the comparison more than the feeling. Your mind is not sorted into a style or a type. It is one mind, with strengths and weaknesses of its own, and it grows the plain way, through learning, teaching and practice.

Found a mistake? Please tell us.