Genetics and intelligence

There’s a new genetic study (stuff.co.nz has the Associated Press article) claiming that variation in intelligence is about 50% genetic.  That claim is not new, but previous studies got estimates like that by studying the IQ of close relatives,  and this study actually measures genes.  The researchers studied 3500 unrelated people in Scotland and England,   measuring half a million genetic variants on each person and relating them to two different types of intelligence test, and found that while they couldn’t identify specific genetic variants that affected intelligence, they did find evidence that there were hundreds of variants with some effect. 

The analysis had two components. The first was to look at genetic variants one at a time to see if they were associated with the intelligence test results.  Although none of the associations were strong enough to pass the stringent threshold for statistical significance, a lot more associations were close to the threshold than would be expected by chance.   More convincingly, when they combined all these weak associations and looked at an independent sample of people from Norway, they found that the weakly-associated variants from England and Scotland did predict intelligence test results in the Norwegians.

The second component of the study looked at genetic similarity between people. People who, as far as they know, are unrelated will still share some genetic variants, and some pairs of people share more variants than others.  If there are a lot of genes affecting intelligence, you would expect that people with similar intelligence test results would share more genetic variants than people with very different intelligence test results. And they did.

One of the unfortunate aspects of results like these is that certain people want to interpret them to say something about differences between racial or ethnic groups.  The study doesn’t say anything about this.  Not only was the study restricted to British people of European ancestry, but comparisons were made only within the participating cohort studies.  Manchester participants were compared to other Manchester participants; people from Aberdeen were compared to other people from Aberdeen.  In addition, the researchers tried to estimate and adjust for any other differences in ancestry, using techniques that seem to have been pretty reliable in other large genetic studies.   So, the study doesn’t say anything about racial or ethnic differences, nor is there any logical or biological reason to expect it to generalize that way.

The study also doesn’t tell us anything about intelligence being determined and unchangeable. We already know that intelligence test results also have a strong cultural component, most dramatically from the Flynn effect, the steady world-wide increase in IQ first pointed out by Dunedin researcher James Flynn.  Other, less controversial traits with strong genetic heritability, like eyesight and height, are certainly very responsive to social changes — I’ve drawn the short straw genetically when it comes to eyesight, but some inexpensive pieces of curved plastic give me even better distance vision than the average person.

So what does this study actually get us?  One positive result is that it shows the family-based estimates of heritability for intelligence are not totally bogus, which was not at all clear before.   Other than that, the study is basically negative: there are a huge number of influences on different aspects of intelligence, and genetics is unlikely to be a good way to disentangle them.