ars technica: [Researchers] calculated that the genes identified in the first analysis, which should have a stronger effect, account for less than five percent of the variation in intelligence. by John Timmer
'The other point of interest is that many of these areas of the genome had previously been implicated in other aspects of human biology. One of those...is educational achievement. But there were weaker associations with things like autism spectrum, successful smoking cessation, and height. These genes also seem to be somewhat protective against Alzheimer's, depression, schizophrenia, and neuroticism, as well as a low body mass. So these clearly aren't "genes for intelligence," per se; they're genes that influence a broad range of biology, some of which influences how we perform on intelligence tests.
'As for the genes themselves, there's no strong, consistent theme to their function. A couple seem to be involved in the controlled death of neurons that takes place as the brain is developing. But many others were more generally involved in development, either of the brain or other organs. And we're a long way from figuring out how they contribute to intelligence that's often measured decades after the genes may be active.
'It would be nice to think that testing the genetics of nearly 80,000 people might give us a clearer picture of the biology of intelligence. But as the authors point out, the work was meant to be a foundation rather than the last word: "These findings provide starting points for understanding the molecular neurobiological mechanisms underlying intelligence."'
"Genome-wide association meta-analysis of 78,308 individuals identifies new loci and genes influencing human intelligence" by Suzanne Sniekers, et al. here