For a trait so highly heritable, intelligence has been awfully reluctant to give up its genes.

    There is wide agreement that cognitive ability at least partly reflects the influence of DNA: Dozens of studies of
    thousands of twins have shown identical twins, who share the same genes, tend to have more-similar IQs than do
    other sibling pairs, and children match the IQ of their biological more than their adoptive parents.

    Together, these studies imply genes account for about 50 percent of the difference in intelligence from one person
    to the next. That's a high enough "heritability" that you'd think genome labs would be practically spitting out genes
    related to intelligence.

    But they're not. And therein may lie an important clue to the biology of what Robert Plomin, a professor of
    behavioral genetics at King's College London, calls "the most complex -- and most controversial -- of all complex
    traits."

    Intelligence has many meanings, but what scientists call general cognitive ability seems to reflect memory skills,
    verbal and spatial abilities, and abstract reasoning. Usually, if you're good at one, you're good at the others.
    Although that correlation may reflect not "brain quality" alone but something nonphysiological, such as differences
    in motivation, it has inspired a search for genes that make better brains.

    Prof. Plomin and his colleagues were the first to identify a suspect. In 1998, they reported that one form of a gene
    called insulin-like growth factor-2 receptor was present in 32 percent of children with high IQs, but in 16 percent
    of kids with average IQs. It was also especially frequent in people with exceptional math or verbal talents.
    Experiments in other labs had shown the gene is active in regions of the brain devoted to learning and memory.
    But when the King's team tried to replicate its finding, it failed: The "smart" gene showed up in 19 percent of
    high-IQ children ... and 24 percent of those with average IQ.

    That didn't deter biologists. Since 2000, teams have identified at least four more genes associated with
    intelligence. Two studies fingered genes for an enzyme called catechol O-methyltransferase. Others identified
    cathepsin D, CHRM2, or cystathionine beta-synthase as having variants that are more common in people with
    high IQs.

    As with all such studies, you have to watch out for a chopsticks effect. Just because a genetic variant shows up
    more often in people adept at using chopsticks doesn't mean it causes manual dexterity: It may simply be more
    prevalent in Asian populations. Similarly, purported IQ genes may cluster, by chance, in groups whose culture
    values education, yet not actually make a brain smarter. There's another problem. Neuroscientists can't find any
    fundamental brain processes that distinguish Einstein from the rest of us -- not speed of neuronal transmission,    not the ability to form synapses, not the quantity and quality of neurons, Prof. Plomin says. That makes it             less likely that genes for those basic characteristics (even if scientists find them) have a significant effect on intelligence.

    Even if the newly suspect intelligence genes hold up, they will surely turn out to be only the tip of a huge iceberg. It
    looks more and more as if intelligence reflects the complex interaction of scores of genes with each other and the
    environment. No one gene makes more than a tiny difference. Different forms of CHRM2, for instance, account
    for a spread of only three or four IQ points, while CTSD may account for perhaps 3 percent of the variation
    between people.

    The heritability of intelligence may, paradoxically, reflect the importance of environment. If Susie is born with a
    slightly better brain than Mary, she will like school, receive more praise from her teachers, haunt the library, take
    more demanding courses. In short, she will bootstrap her way to greater intelligence.

    That explains why the measure of heritability of intelligence rises with age, from 40 percent in childhood to 60
    percent in adulthood. It isn't that genes grow stronger. Instead, says James R. Flynn of the University of Otago,
    Dunedin, New Zealand, a slight genetic edge at birth snowballs by nudging people to choose
    intelligence-enhancing experiences. The result is "a potent multiplier," he says in Current Directions in
    Psychological Science.

    Prof. Flynn discovered that IQ soared in recent decades. Since 1950, scores on one IQ subtest have risen 18
    points per generation in the Netherlands, Belgium, Israel and Argentina; between 1948 and 1989, Americans
    gained the equivalent of 20 IQ points. The genes we have don't change fast enough to explain this "Flynn effect" --
    but which genes are turned on might. Perhaps growing up with enough leisure time to play chess and even
    videogames, or living in smaller and more affluent families that can indulge children's intellectual curiosity, turns up
    the activity of genes related to intelligence. For that reason, says Prof. Plomin, the holy grail in this field is
    identifying what experiences turn on genes that influence intelligence.

Even before that happens, it's already clear that, with so many genes involved in IQ, genetic engineering for it isn't in the cards. If we care about intelligence, we must seek ways to nurture it not in the genes we pass on to our
kids, but in the world we make for them.

    ©2003 Associated Press