Scientists Gregory Cochran, Jason Hardy, and Henry Harpending wrote a paper explaining how Ashkenazi Jews could have evolved IQs about a standard deviation higher than the European and Middle Eastern populations they split off of, in such a short period of time (less than a 1000 years).  They argue that because of the historic correlation between genetic fitness and wealth and because of the correlation between wealth and IQ, there was rapid selection for IQ.  They write:

Assume, for example, that the correlation between income and IQ is 0.4 (about the correlation in the United States today) and that individuals in the top 10% of income have twice the average fitness. The mean wealth of parents would be .16 standard deviations above the population average and the mean IQ of parents would be 0.4×0.16 or 0.064 IQ standard deviations, that is 1 IQ point above the population mean. This is the selective differential, and with a heritability of 0.8 IQ would increase by 0.8 points per generation. In 500 years—20 generations—average IQ would increase by 16 points.

Of course all this assumes relatively independent genetic effects (a reasonable but unproven assertion), but aside from an overly generous estimate of narrow sense heritability, the scenario they paint sounds plausible at first glance.

But then I wondered, if you can get a 16 IQ point increase in just 500 years from such a minor selection effect, then why do we only see about a 15-20 point IQ gap between blacks and whites?  Afterall, the ancestors of whites and blacks have had different selection pressures for at least 50,000 years, and presumably, the challenges of adapting to cold winters likely selected for higher IQ in non-Africans during all that time, so either the cognitive demands of cold winters were very small, or, there’s something wrong with the breeder’s equation.

I think part of the explanation is that the breeder’s equation has diminishing returns.  You might start out with a 0.4 correlation between IQ and income among incipient Ashkenazem, but with each generation of selection, the cognitive variance of this population shrinks slightly, and with the reduced variance, the correlation between IQ and incomes shrinks slightly (in the restricted population) and the parent-offspring IQ correlation also shrinks slightly, because range restriction tends to reduce correlations.

The rest of the explanation might be Charteruse’s claim of environment dependent genetic effects.  In other words, a population might be hyper-selected for genes that make them smart in medieval Europe (and similar societies like modern America), but when the environment changes dramatically, those same genes stop enhancing IQ, or even reduce it.  Perhaps only when a population is repeatedly selected for IQ in many different kinds of environments does it evolve environment independent high IQ genes.

Because if selection for IQ were really as easy as the breeder’s equation implies, then you would expect different populations to differ in IQ (and other polygenetic traits) by much greater amounts than the evidence shows.