Most recently, the controversy incited by the 1994 book The Bell Curve has revived the nature-nurture debate: Is intelligence malleable or genetically programmed? Is a human mind born or made? After decades of research, most intelligence experts have come to decide that it's both, in roughly equal measure.

So it was perhaps inevitable that someone somewhere would begin the search for "smart" genes. That someone is Robert Plomin, a 25-year veteran of intelligence research who is currently stationed at the Institute of Psychiatry in London. Last year, Plomin published the first evidence of a gene linked to high IQ. This year, he reported the locations of three more smart genes. And within the next several months, he expects to find at least two dozen of the most important genetic determinants of intelligence. Already, his work has provoked visions—and fears—of DNA doctors tinkering with the gears of cognition. "I knew that nobody else would be crazy enough to do it," he says.

Plomin's quest is one of the most audacious in the field of behavioral genetics, a discipline dedicated to finding hereditary factors that influence human behavior. One type of study traces patterns of inheritance by comparing identical twins reared together and apart. Another method compares the traits of adopted children with those of their biological and adoptive parents and siblings. Done well, these studies determine the heritability of a trait: to what extent the differences among individuals are due to genes, rather than to environmental forces such as upbringing, nutrition, and schooling. Once a heredity pattern for a trait is established, researchers can home in on the genes responsible. 

    But the search gets complicated when the genes in question are for smarts. On one hand, intelligence as measured by IQ tests is a reliable and stable component of human behavior. It doesn't change much over a lifetime, and different tests tend to produce the same results. Intelligence is also the most highly inheritable mental attribute known: Twin and adoption studies suggest that 30 to 70 percent of the differences among people's IQ scores can be attributed to genes. Many experts, including Plomin, think 50 percent is the most likely figure. (Physical attributes such as height and weight can be up to 90 percent heritable.) 

But intelligence is a complex phenomenon, governed by hundreds or even thousands of genes. So patterns of inheritance aren't obvious. Nor is the discovery of any one gene for intelligence likely to be earthshaking, Plomin concedes. "We don't know how many genes are involved for any complex trait in any plant or animal," he says. But any of the majority of smart genes probably accounts for less than 1 percent of the heritability of intelligence.
    Scientists link genes to traits by using DNA markers: stretches of DNA whose positions on chromosomes have been precisely plotted. The sequence of each marker can vary, just as the gene for blue eyes varies slightly from the gene for brown eyes. These different versions of each marker, called alleles, correspond to different versions of nearby genes. If people with a particular allele of a DNA marker possess a trait and people without the allele don't, then a gene for that trait is likely close to, or even the same as, the marker.
    Using this approach, Plomin compared two groups of children: 51 with an average IQ of 103, and 51 with an average IQ of 136. He used 37 markers on chromosome, which was chosen because it has already been implicated in reading disabilities. His tests revealed differences in one site located in a gene for a hormone receptor that may be active in learning and memory. Plomin thinks variations in this gene or another nearby could account for 1 to 2 percent of the variance in IQ scores. 

To detect even smaller genetic contributions across the entire human genome, Plomin would need a lot more people and a thousand times as many DNA markers. Such large-scale studies seemed impractical until Plomin's colleague Michael Owen suggested scrapping the laborious genetic profiling of individual subjects. Owen proposed pooling subjects' DNA instead and combing the pooled DNA from each study group just once with the markers. Although DNA pooling can't determine with precision the frequency of alleles in a study population, says Plomin, it can point to differences between the normal- and high-IQ groups.       

Plomin has already used DNA pooling to identify three more sites linked with high IQ, this time on chromosome 4. By year's end, he and his colleagues will have screened hundreds of volunteers with 3 000 markers. Based on his results so far, he expects to find 20 to 30 more genes.

And then what? It's almost inevitable, he says, that the genes for intelligence will be used to determine people's genetic IQ. But such tests wouldn't turn up much new information. "By measuring the parents' IQ, we can already predict a kid's IQ tremendously better than we will ever be able to predict it with a DNA test," Plomin explains." And if schools were going to select kids on the basis of ability, they'd still do better by administering IQ tests. I think the implications for science are much greater than the social implications."

The scientific implications could include a better understanding of the neural pathways involved in global reasoning, learning, and memory. Identifying and tracing the products of smart genes could help researchers understand the origins of these abilities, with potentially far-reaching consequences.

"If we knew more about the developmental aspects of [intelligence] genetics, we'd have a more realistic appraisal of what strands of development are modifiable, to what extent, and when," says Craig Ramey, a psychologist at the University of Alabama, Birmingham, who designs intervention programs to raise the IQs of disadvantaged infants and toddlers. "We're now using more of a shotgun approach."

Still, some people may be unable to resist the idea of a genetic fix.

"What will rouse the imagination for most people is: Can we remediate low intelligence or enhance normal intelligence through genetic manipulation or intervention?" says Linda Gottfredson, a psychology researcher at the University of Delaware. "There's certainly a big push to develop drug-enhancement strategies. So why not genetic enhancement?"

Although it's true that each smart gene has a small effect, Gottfredson says, even tiny boosts in IQ are significant. The first gene that Plomin found, for example, accounted for up to 2 percent of the variance in IQ scores. That translates into four IQ points—not a sizable gain for an individual. But tweaking just a few more genes of comparable effect could give a person a 15-point leg up on the smart scale. "And 15 IQ points is a considerable advantage," Gottfredson says. "That's the difference between doing OK in high school, but no more, and doing great in college. Fifteen points will get you a different kind of job, get you into a different kind of neighborhood, with different friends, a different kind of life. I'd take it."

If his latest project goes as planned, Plomin's years of quiet gene hunting may soon provoke another unquiet round in the intelligence debate.

(1 409 words)

(From Discover, October 1999）