Here is a very interesting paper on sex differences in brain size and intelligence, notable for linking people’s brain scans with their detailed intelligence test results. It has been accepted for publication in Intelligence.
Sex differences in brain size and general intelligence (g)
Dimitri van der Linden, Curtis S. Dunkel, Guy Madison
Utilizing MRI and cognitive tests data from the Human Connectome project (N = 900), sex differences in general intelligence (g) and molar brain characteristics were examined. Total brain volume, cortical surface area, and white and gray matter correlated 0.1 – 0.3 with g for both sexes, whereas cortical thickness and gray/white matter ratio showed less consistent associations with g. Males displayed higher scores on most of the brain characteristics, even after correcting for body size, and also scored approximately one fourth of a standard deviation higher on g. Mediation analyses and the Method of Correlated Vectors both indicated that the sex difference in g is mediated by general brain characteristics. Selecting a subsample of males and females who were matched on g further suggest that larger brains, on average, lead to higher g, whereas similar levels of g do not necessarily imply equal brain sizes.
Men’s brains are bigger than women’s, even when controlling for bigger body size, which means they should have higher intelligence, though the evidence for that is conflicting. Most researchers find no notable differences overall, saying that different strengths and weaknesses balance each other out, but Lynn and Irwing (2002, 2004) argued that adult males are almost 4 IQ points brighter than adult females. The authors of the present paper have found one of the largest MRI samples available, each scanned person having done 10 cognitive tests, which is what makes this study particularly interesting. The tests included: Penn progressive matrices, Peabody vocabulary, reading recognition, working memory, pictorial episodic memory, spatial orientation, card sorting, verbal episodic memory, and the Flanker task of inhibition and sustained attention.
First, here are the correlations between brain measures and overall mental ability.
The tests were used to create an overall g score. Correlations with this overall g measure and brain measures are not large, but for both males and females the highest correlations are with gray matter volume. It seems that Agatha Christie’s fictional Belgian detective Hercule Poirot was right, when he said that crime detection and problem solving depended on “the little gray cells”.
Here are the scores for the individual mental ability tests:
Once again, I recommend that men pay close attention to the largest sex difference, which plays out in their favour: spatial orientation, in which they have a 6 IQ points advantage. I recommend that women play close attention to Episodic memory in which they have an advantage of 4 IQ points, giving women the upper hand when remembering male transgressions. Those particular findings hold up even when you control for g, so they are very real cognitive sex differences, and are mostly across the board of the abilities measured.
The spatial male advantage shows up in the first year of life.
The authors conclude:
there was a significant sex difference in g in this sample, with an effect size of one quarter of a standard deviation. This corresponds to approximately 3.75 IQ points, which is a similar sex difference in general intelligence as reported in previous large population studies (Lynn & Irwing, 2004) and meta-analyses (e.g., Madison, 2016; Irwing & Lynn, 2005).
Analyses of the cognitive subsets used to extract g showed that the sex differences were not related to extremely high scores of males on a limited number of particular tasks, but tends rather to reflect a more general pattern.
the central point of the present study, is that the various statistical methods applied seem to suggest that sex differences in brain characteristics indeed mediate sex differences in g. Direct support for this notion came from the mediation analyses, indicating that brain volume measures could account for roughly half of the sex differences in g.
It is an interesting observation that in the nineteenth century the consensus was that sex differences in brain size exists, leading to a slightly higher average of males in general intelligence (e.g., Darwin, 1871). However, improved psychometric and brain imaging techniques have led to a new wave of studies and have reactivated the debate on this topic. Regarding this, the present study may contribute to this field by applying a combination of newer and more traditional methods. Overall, we agree with the conclusion of Burgaleta et al. (2012) and Escorial et al. (2015) that within subgroups or at the individual level, larger male brains do not necessarily have to be accompanied with higher general intelligence. Nevertheless, the present study also clearly indicates that, at the group level, there is a sex difference in g and that differences in brain size likely play a relevant role in this. Given that those conclusions were based on the results of one of the larger MRI studies available, it can be expected that the effect sizes provide reliable estimates of the relations and can be regarded as benchmarks in the literature in this area.
This study supports the minority position of Lynn and Irwing, that men are about 4 IQ points brighter than women, an across-the-board advantage, plus better spatial ability, and that part of this difference may be attributed to brain size. Here is Prof Richard Lynn lecturing about it:
You will see that male advantage shows up in the Wechsler standardisation samples, though Wechsler will not agree to this being acknowledged in print, though they have passed on the results privately. An odd situation, to say the least.
As usual, a small difference in means has larger consequences at the extremes. If one assumes a 4 point difference straddling the mean, then women will be 98 to men’s 102. Keeping the standard deviations to 15 for both sexes, and setting the cutoff point at IQ 130 then 3.1% of men and 1.6% of women pass the threshold, meaning 65% of the brightest people will be men.
Since this sample showed no sex difference in standard deviations (to the author’s and my surprise) that is as far as I will go with the calculations. However, even without putting in the usually found smaller standard deviation for women, the implication is clear: 65% of intellectually demanding occupations will be taken by men, if entry to those occupations is based only on mental ability. If the bar is set higher at IQ 145, then 70.5% of such posts will be taken by men.
As MRI samples go, this is an excellent sample, and will get better as more results are released. The totals available for study have just increased to 1200 scanned persons, and I have already seen a conference abstract for next month looking at other features of this larger sample. As epidemiological samples go this is a small sample, so whole populations as achieved in birth cohorts and national examinations are better, though these do not have any brain scan results. It is the combination of good numbers of scans and concomitant ability testing which make this study so informative.
Let us see whether the increased sample confirms or changes these results. I hope to see preliminary results soon, though there will be the usual year before they get published. I saw the results of this current paper a year ago, but could only hint at it until it was published.
When will academic publishing catch up with the pace of modern research? In the meantime, the current suggestion is that the biggest study of brain scans supports a sex difference in intelligence of about 4 points, probably due to larger male brains.