Kids... they spend all their time playing video games and texting each other. Where's it ever going to get them?

Phyllida Brown

A CLICHE it may be, but you have to admit it's true. Most kids are much better than their parents at programming the video or setting up the new computer. Almost certainly they can thrash Mum and Dad at video games. And they probably think they are smarter than their parents. But here's the surprise: those kids may be right. If IQ tests tell us anything at all, today's young people really are smarter than their parents. Studies from numerous countries suggest that IQ scores have been rising fast since at least the 1950s-enough to mean that someone with an IQ classed as average then could be labelled as having low intelligence today. Humans, it seems, are getting cleverer and cleverer with each passing generation.
But not so fast. Intelligence is a slippery concept at the best of times. There's a serious argument among psychologists over just what the upward trend in IQ scores really means. Some researchers argue that we're not getting more intelligent overall, just getting better at a particular sort of problem-solving. Others, though, say they have found echoes of rising IQ scores in "real world" measures of intelligence. And to confuse things still more, there are hints that the trend may be about to grind to a halt.

Dumb and dumber Does it really matter if IQ scores go up or down? WHATEVER it is that IQ tests measure, we've been getting progressively better at doing them for several decades. Political scientist James Flynn identified the trend in the 1980s, and since then the evidence for rising IQs has grown as relentlessly as the test scores themselves. Alas, nothing lasts forever. This week we report the first evidence that the era of booming IQ growth is over (see p 24). Indeed, a pessimist scanning the latest figures from Denmark, a country noted for its detailed IQ records, might conclude that average IQs have not just peaked but are heading back down (see Diagram). Should we be alarmed? It will be several more years before we can be certain of the new trend, but even if the dip is sustained for decades to come it won't spell catastrophe for intellectual life. Far from it. Take the IQ rise of recent decades at face value, and the average child today should be as bright as the near genius of yesteryear. So where is the unparalleled renaissance this should have brought about? The fact that there hasn't been one powerfully exposes the limitations of IQ as a proxy for intelligence. Kids today may well be geniuses at solving the sort of abstract on-the-spot problems that dominate IQ tests. But the way the world has shaped up indicates this to be a narrow, perhaps even trivial, aptitude. True, recent decades have seen ever younger chess grandmasters and greater scientific productivity. But rising IQs are unlikely to be fuelling such trends. The financial rewards for exceptional talent in any sphere, including chess, are much greater than they were. And today's scientists are under far more pressure to publish. Even if we wanted to intervene to keep IQ scores on an upward track it is not obvious what we could do. That's because the cause, or causes, of the IQ rise remain a mystery. Better nutrition? Children born into the severe Dutch famine of 1944 did not miss out on the rising IQ curve. Better education? The gains are missing or small on the kind of IQ tests closest to school-taught material. In recent years the IQ rise has been variously attributed to visual stimulation from cartoons, crowded computer screens, video games, even fast-food place mats and cereal boxes smothered in word games and mazes. But all this is speculation. Nobody has a shred of empirical evidence that any of these influences actually boost IQ. A century ago, scientists developed simplistic ideas about racial differences in IQ scores and ended up providing ammunition for the Nazis and other eugenicists. This century's IQ experts must take care they do not manufacture undesirable ammunition of a different kind. In an ever more complex world, it is inevitable that brain power should become more highly valued and perhaps increase in certain ways. Fast food chains and the consumer electronics industry neither need nor deserve the credit for bringing it about.

What should we make of all this? The obvious person to ask is James Flynn, professor of political science at the University of Otago in Dunedin, New Zealand. He first showed that IQ scores were rising in the mid-1980s and became famous after others dubbed what he observed the "Flynn effect". The findings raised a whole new set of questions about what IQ tests can and cannot tell us.
In a landmark paper published in 1987, Flynn made detailed comparisons across 14 countries of trends in the results of tests that measure people's reasoning power. The most complete data came from countries where military service had continued into the 1980s, such as the Netherlands and Belgium. Virtually every young man (and in some countries, every young woman) had to sit a test set by a draft board, so the results were representative of the population. The tests vary from country to country but their results can be converted into IQ scores for comparison. For other countries, Flynn had to rely on IQ test data. Overall, the results he published in Psychological Bulletin (vol 101, p 171) showed a real growth in scores of up to 25 points in a generation.

Still, though, the crude overall scores didn't tell Flynn enough. To find out more, he looked at scores for each type of reasoning IQ tests set out to measure, such as verbal, numerical and visuo-spatial. This last group usually includes Raven's matrices-sequential patterns with a piece missing (see below). The candidate has to choose from a set of alternatives which shape to put in the empty space. When Flynn took a closer look at the scores, the pattern was clear. The strongest gains in almost every case were on Raven's matrices and similar tests, while verbal and arithmetical scores showed more modest rises. Since the 1987 paper, Flynn has looked at more countries, including Israel and, most recently, Argentina. He finds the same upward trend, with abstract and visuo-spatial skills responsible for most of the climb.

Flynn is sure his eponymous effect is meaningful, but does not believe we're all turning into brain-boxes. "Our grandparents were not retarded, and we are not geniuses," he says. More likely we have simply developed the skills and habits of mind that make us better at solving abstract problems-and, just as important, to take such problems seriously.
"People really have got better at some sorts of problem solving," says Flynn. The reason, he argues, is that, over time, our society has been pulling the triggers that make people develop these skills. When our grandparents were growing up, people valued different mental skills, such as doing mental arithmetic or building up a good vocabulary. Their reaction to some of the problems that we now take so seriously would have been: "What's the point?".

Flynn believes that in societies where intelligence testing has become something of an obsession, abstract problem-solving skills develop faster than other skills. Likewise, demands on people's visuospatial abilities have grown because of TV, computers and car driving, so successive generations would be expected to develop their skills in this direction more than in others. "I don't think intelligence advances all at once," he says. Most psychologists accept that IQ scores have risen. But there are different views about what might account for the increase (New Scientist, 21 April 2001, p 44).

'WHEN IT COMES TO PROBLEM SOLVING AND VISUO-SPATIAL SKILLS, YOU ARE SMARTER THAN YOUR PARENTS AND SMARTER STILL THAN YOUR GRANDPARENTS'

Some, such as experimental psychologist Robert Howard at the University of New South Wales in Sydney argue that our overall intelligence really has risen, helped by better nutrition, smaller families, wider access to education and other environmental changes. And John Rust at Goldsmiths College, University of London, suggests that it's not just an increased emphasis on visuo-spatial activities that is pushing up IQ scores. Among several factors, he thinks that as society becomes more complex, people are having to think harder about all sorts of problems. Also, he points out, if one advanced individual first understands a concept, such as Einstein grasping the theory of relativity, it may be easier for all of us to think in a more sophisticated way: collectively, we may be getting used to dealing with more complex ideas.

But some researchers are looking beyond people's performance in IQ tests and have begun searching for "real-world" evidence that humans are getting more intelligent. In a study last year, Howard analysed a basket of indicators including scientific productivity and performance in intellectual games such as chess, bridge and go. From this he tentatively concluded that people really could be getting progressively cleverer. For example, the age of the youngest chess grandmaster has fallen four times since 1991 having stayed the same since the 1950s. Admittedly, none of the measures taken alone is compelling evidence of a rise in intelligence. An increase in scientific productivity, for example, could simply reflect the fact that scientists are under ever-increasing pressure to publish papers. And some measures, such as the number of patents awarded, did not show clear trends (Personality and Individual Differences, vol 30, p 1039).

Shape up:scores in tests such as the ones shown above,which are designed to assess visuo-spatial reasoning,have risen consistently since the 1960s. Some researchers argue this is evidence that humans are becoming more intelligent. Others say it's simply a product of our environment becoming more visually challenging.

But despite these reservations, Howard reckons there really is something going on, especially with visuo-spatial skills, such as those that good chess players need. "Kids are getting much more visually smart, because they deal with visual things all the time," he says. And, like Flynn, he believes this is pushing up their IQ scores.
Whatever the underlying cause, it looks as if the Flynn effect is real. When it comes to problem solving and visuo-spatial skills, you are probably smarter than your parents, and smarter still than your grandparents. But can it carry on? Would someone with average intelligence today be thought of as a dunce in 2050? Perhaps not.

Howard's latest research has sent him on a different line of enquiry. Controversially, he now believes that the rise in intelligence might have peaked in the industrialised countries. His conclusions come, in part, from surveys of schoolteachers. "If intelligence is rising, teachers should be reporting it," he says. No one had asked them systematically, so Howard surveyed high-school teachers in Sydney who had been working since at least 1979. Most of those who replied said that children's general intelligence was not rising.
Next, Howard gave the same questionnaire to experienced teachers in Australian primary schools, and primary schools in Singapore and South Korea. The difference was striking. Primary-school teachers in Australia said that children had not got any brighter over the years. But the primary teachers in Singapore and South Korea overwhelmingly said that children there were getting cleverer. "It's absolutely taken for granted that kids today are smarter than they were," Howard says. He believes the contrast with Australian teachers' viewpoints is not an artefact. "My interpretation is that the rise in general intelligence in industrialised countries stopped quite a long time ago."

He speculates that in industrialised societies the social changes that stimulated a rise in overall intelligence have now slowed down, with only visuo-spatial skills continuing to improve. In contrast, the Asian tiger economies have developed into modern societies in about 40 years and rapid change continues.
To support his view, Howard cites data from Denmark. The Danish military still summons virtually all young men before a draft board, even though few now have to do military service against their will. Since 1957, all 17-year-olds have sat exactly the same intelligence test - including tests similar to Raven's matrices. Candidates must also be able to identify the basic geometric shapes that make up a more complex shape. There are also verbal questions, such as "Sun is to day as Moon is to?", and number series questions, such as 2 3 5 8 ?.

Results in these tests have been carefully analysed by Thomas Teasdale, a neuropsychologist at the University of Copenhagen, and David Owen, now at The City University of New York. During the 1960s and 1970s, scores showed gains on a scale similar to those of other countries. But by the 1990s, the rate of improvement slowed right down to about 1 IQ point over the decade (Intelligence, vol 28 p 115) What's more, the scores were improving only on the visuo-spatial questions. Verbal and numerical performance was flattening out.
That wasn't all. Scores from 1999 onwards show an overall downturn, Teasdale has recently reported. He adds that 1999 was also the first year on record in which the number of Danish teenagers being admitted to post-16 academic education fell. Why was this happening? One reason might be that there is some limit to people's potential in the tests, and that Danish teenagers are now reaching it. There might be some absolute ceiling to people's abilities to perform these tasks, no matter how much the prevailing culture encourages them. But according to Teasdale, that can't be the answer "It wouldn't explain why scores are starting to go down," he says. If a ceiling had been reached, the test results should show scores pushing against it, not falling.

The great leveller
Howard thinks that the levelling off of IQ scores in Denmark could be down to a problem with motivation. From his surveys of Australian schoolteachers, he received a strong signal: most thought their students were no brighter or dumber than before, but the vast majority complained that students were less interested in school work. Teasdale isn't convinced by this either: The teachers' views could just be an example of adults' tendency to think the worst of youngsters, he suggests. The trouble is, even in South Korea, where teachers thought their kids were getting brighter, secondary-school teachers reported that students' motivation seemed to be lower than in the past.

Yet the biggest and most awkward missing piece in this jigsaw is schooleavers' academic results. Worldwide, are results improving, getting worse, or staying more or less the same? Unfortunately, it's hard to tell. Tests have changed over the years, and because the numbers of students taking them also change it is difficult to get a sense of how well candidates are doing. The best guess is that overall performance is staying flat, if not actually falling. In the US, where a school-leavers' test has stayed relatively unchanged, the results since the 1950s show at best a plateau. In Britain, the number of students achieving high exam grades at age 18 has increased sharply since 1990. But Rust thinks that this represents a lowering of examination standards to allow more people to enter higher education. "If you change access from 5 to 30 per cent, the only way to do so is to lower entry standards."

Flynn himself is keeping a close eye on the findings. He's not jumping to conclusions until data from more countries come in, but he does wonder whether we've reached a point where students are not willing to be pushed any harder. Could the conditions that have led to such a dramatic improvement in ability disappear as societies grow more and more affluent? "In human history, affluence often leads to decadence," he says. "Look at what happened to the Romans: they got lazy and hired Greeks to do their thinking for them." So far there is no reason to think that industrialised societies are heading that way. But maybe, just maybe, we should be worried.

Author

Phyllida Brown is a science writer living in Exeter