LOS ANGELES (AP) — Scientists who hunt for "intelligence genes" used to think there were fewer than half a dozen of them.
In recent years, they have determined that there may be at least 1,000 — each with just a tiny effect on the differences in people's IQs. A study released Tuesday finds new evidence that many genes play a role in intelligence, but scientists still couldn't pinpoint the specific genes involved.
"It's been kind of a shock to the system that it hasn't worked," said Eric Turkheimer, a psychologist at the University of Virginia who had no role in the study. "We can't find the effects of any individual genes that are large enough to seem worth worrying about."
Previous work involving twins and adopted children has found that genes have a significant influence on IQ scores, producing about half the difference between adults in general. The influence of genes on IQ appears to grow from childhood to adulthood.
Scientists have come to realize that, as with height, differences in intelligence come not from a few genes, but rather the overall effect of many genes, each with only a tiny influence. That makes them hard to tease out.
The new DNA study, reported online Tuesday in the journal Molecular Psychiatry, came to similar conclusions. Many genes work together to shape intelligence much like the different instruments of an orchestra shape music. Unless there's a soloist playing, it's often difficult to decipher the contributions of individual instruments.
The new work was done by I.J. Deary of the University of Edinburgh in Scotland and colleagues in several countries. The team wanted to find out "whether genetic differences that we could test on people's DNA could explain some of the reasons that people have different intelligence-test scores," Mr. Deary said in an email.
Researchers didn't identify any genes affecting IQ. But they estimated that they found a genetic influence that accounts for at least 40 percent to 50 percent of the differences on intelligence tests in the 3,511 unrelated adults in their study who were tested on knowledge and problem-solving skills.
They focused on more than 500,000 places in the participants' DNA, looking for evidence that IQ-influencing genes lay close to those places. They concluded that the overall effect was coming from many scattered genetic differences, each of only small influence.
The latest work adds to evidence that even the most powerful of these has only weak influence. Mr. Deary said that future studies probably will need to involve millions of people to detect the genetic effects.
Robert Plomin of the Institute of Psychiatry in London, who has looked for intelligence-related genes for 15 years but didn't participate in the new study, isn't surprised by the latest findings.
"We've got a century of twin and adoption studies," such as those comparing twins reared in different families, that support the notion that about half of IQ differences come from DNA, he said.