« Pork eating not allowed | Gene Expression Front Page | Shibboleths can matter »
August 05, 2004

Keeping with your own kind might not be that bad....

Sometimes commentator Gregory Cochran has asserted several times on this blog's message boards that the fertility rate of the affluent and educated dropped below the rate of the less affluent and educated sometime in the late 19th century. Though I havenít found any evidence to confirm or refute this assertion in the context of that time period, it seems that most of the contemporary data tends to support the assumption that the professional classes are reproducing at a lower rate than the working classes in the Western nations.

This leads to the perception of a possible dysgenic trend for the genes that effect intelligence, and allelic combinations that "should" be less fit are actually advantaged in the replication race in comparison to "superior" allelic combinations. By this, I mean that individuals who possess a genotype that leads to a greater likelihood of expressing a high IQ phenotype are less likely to reproduce than individuals who possess a genotype that is likely to express a low IQ phenotype. Therefore, the alleles that confer upon individuals the potential for a high IQ seem to reduce reproductive fitness of the individual carrying those genes, and so those genes should decrease in frequency within the population.

But the assumption of a century long dysgenic trend seems to fly in the face of the reality that technological advancement and creativity have not decreased as the century has passed (at least my subjective perception). One can dispute the metrics, but at the higher ends of intellectual achievement the system seems to be working. Because of raised expectations some are being "left behind," the engines of modern civilization place high demands on individual capital.

Let me neglect the issue of the Flynn Effect for now, and move on to another point: assortive mating (the preferential mating of individuals who share common phenotypic tendencies, at least one, but perhaps many). In The Bell Curve Murray & Hernstein suggested that we were witnessing the emergence of a multi-tiered meritocracy. With the acceleration of mating based on educational attainment and professional success (especially in an age where males can evaluate women directly in these criteria instead of using indirect proxies like their fathers & brothers) many worry about the chasms that are developing between various segments of the American population. I won't examine their points, there are many writers and thinkers who ruminate on this issue, rather I want to point out one of the many possible benefits of the educational and intellectual stratification of our society: more super-geniuses.

Keep in my mind that my model is a simplistic thought-experiment. My conjecture is this: I suspect that the increased population substructure of modern societies along the axis of intellect might be resulting in an increase in the absolute numbers of highly creative and intellectually impressive individuals. The analogy I have in mind is selective breeding: when one selects a parental generation with an average phenotype shifted over from the mean, the offspring generation regresses toward the mean (because many of the parents with impressive phenotypes expressed their traits because of non-genotypic factors), but nevertheless the new mean is shifted over from the old mean of the original population (selection response). If you iterate the procedure over many generations in the end you have may have a phenotypic mean that is extremely shifted over from the original population. Not only might the mean be shifted over from the original population, within this peculiar descendent population there would likely be individuals that express extreme phenotypes not found in the original population. This is the result of novel and rare allelic combinations (I am assuming an additive polygenic continuous trait) which emerge out of matings between individuals who already possess a genotype that is characterized by an overabundance of low frequency alleles. In other words, though many pairings between geniuses do not result in geniuses (because of regression toward the mean), a genius is much more likely from such a mating. A hyper-exceptional child is far more plausible coming from the pairing of two very bright individuals whose genes may recombine and resegregate in a felicitous fashion.

If my suppositions are correct the population wide increase in alleles that tend to express a lower IQ might be compensated by the increased substructure on the "g" locii, which might result in changes in the character of the "smart fraction." The importance of assortive mating and population substructure might also explain the paradoxes that one is confronted by in the low IQ of South Asians and the high achievement of members of the Diaspora and the historical importance of South Asian contributions transculturally (eg; Buddhism). What matters is not the mean IQ of South Asians, but their smart fraction.

Just a thought I'd been kicking around....

Related: See Griffe on the Smart Fraction.

Posted by razib at 06:04 PM