Nisbett’s 2009 book on intelligence, Intelligence and how to get it, is a goldmine of stupid claims that one can quote-mine for introduction and discussion sections of papers. For instance, Nisbett tries to argue that brain size is not causal for intelligence! He writes:

The correlation between cranial capacity and IQ is probably about .30-.40 in the white population (McDaniel, 2005; Schoenemann, Budinger, Sarich, and Wang, 1999). Rushton and Jensen (2005) claim that cranial capacity for blacks is on average smaller than that for whites. A difference between black and white brain size is not always found, however (National Aeronautics and Space Administration, 1978). More important, the correlation found within the white population probably does not indicate that greater brain size causes higher IQ. Within a given family, the sibling with the larger brain has no higher IQ on average than the sibling with the smaller brain (Schoenemann, Budinger, Sarich, and Wang, 1999).

That’s a little surprising given that this association does show up within families when we use the cruder proxy of head size (n = very large). Also because we know the brain uses some 20% of body’s energy while taking up some 2% of the weight, making it 10 times as metabolically expensive as other organs. And human brain size increased enormously during the recent evolution (and in fact continuous doing so in line with FLynn effect). And some species get rid of their brain when they no longer need it to save energy. The large brain is problematic for childbirth since humans have the bizarre design of being bipedal and giving birth between the legs too. Many deaths in childbirth must be related to the unusually large infant head, which means there is a direct selection against head size related to this. If brain size gets larger nonetheless, obviously there must have been strong pressure for it. And so on, our prior that this relationship is causal is nearly 100%. The study in question, thus, seems suspect. If we read the abstract, we immediately see the problem:

Hominid brain size increased dramatically in the face of apparently severe associated evolutionary costs. This suggests that increasing brain size must have provided some sort of counterbalancing adaptive benefit. Several recent studies using magnetic resonance imaging (MRI) have indicated that a substantial correlation (mean r = ≈0.4) exists between brain size and general cognitive performance, consistent with the hypothesis that the payoff for increasing brain size was greater general cognitive ability. However, these studies confound between-family environmental influences with direct genetic/biological influences. To address this problem, within-family (WF) sibling differences for several neuroanatomical measures were correlated to WF scores on a diverse battery of cognitive tests in a sample of 36 sibling pairs. WF correlations between neuroanatomy and general cognitive ability were essentially zero, although moderate correlations were found between prefrontal volumes and the Stroop test (known to involve prefrontal cortex). These findings suggest that nongenetic influences play a role in brain volume/cognitive ability associations. Actual direct genetic/biological associations may be quite small, and yet still may be strong enough to account for hominid brain evolution.

Who the hell tries to establish a null relationship with a sample size of 36??? Of course published in a top journal, PNAS. Top journals routinely publish studies that try to disprove obvious thesis with tiny samples, e.g. this adoption study published in Nature, with a 3 group comparison with n’s 36, 22 and 24. Such tiny samples can produce any given result and so are meaningless to draw conclusions from alone. Nisbett is a reverse thermometer. Whenever he says something controversial, he is always wrong. So, one can simply take this opinion and reverse it to get the truth. Wrong about heritability of IQ, wrong about transracial adoption studies, wrong about brain size’s causal effect on IQ, wrong about race being a social construct in a problematic way, wrong about morality of studying group differences, wrong about genetic contribution to Black White IQ gap, wrong about…

In general, Rushton and Jensen were not amused, so they wrote a very long combined rebuttal and review:

However, unknown to me was that James Lee had also written a pretty thorough and interesting review of the book:

Lee writes:

If Nisbett is truly confident that degree of European ancestry shows no association whatsoever with IQ, he should call for studies employing superior ancestry estimates of the kind displayed in Fig. 3. Note that the increased reliability of ancestry estimation does not obviate the need for a large sample. Even under an extreme hereditarian hypothesis assigning mean genotypic IQs of 80 and 100 respectively to the African and European ancestors of African Americans, we can only expect an increase of .2 IQ points for every percentage increase in European ancestry. The considerable IQ variation among African Americans makes an effect of this size difficult to detect in small samples.

The ultimate test of the hereditarian hypothesis is of course the identification of the genetic variants affecting IQ and a tally of their frequencies in the two populations. Because of their likely small effects, we may have to identify dozens of such variants before we are able to make any confident inferences regarding the overall genotypic means of different populations. Although this task is currently within our technological means, it seems practically out of reach in the very short term. Ancestry estimation is much less costly than gene-trait association research and thus offers the advantage of an immediate increment toward the resolution of this issue.

I believe we have nearly fulfilled this goal with the following studies:

The only possible non-genetic hypothesis (I can think of) that remains is the colorism model, where ancestry is a non-causal confound with skin tone (correlation is ~.50), which then provokes (White) racism which somehow manages to lower IQ scores of some groups, while apparently not harming the others (e.g. US Indians who do very well but are dark skinned), and also leaving unscathed non-IQ traits like self-esteem. This model can be ruled out via sibling studies, or simply by controlling for skin tone. Unfortunately, the dataset used for the admixture study above (PING) does not have a skin tone variable, so one cannot directly rule out colorism in that dataset. I’m pretty confident that colorism can be ruled out in other datasets, however, see the extended discussion of colorism at Human Varieties. I think that in general, colorism does not fit the numbers at all even if real and sizable, so it can be rejected as being the entire explanation solely on numerical grounds.

The issue of using identified variants (usually SNPs) is problematic due to the differential LD decay, which I have discussed at length.