Disclaimers: this primer deals with sex differences in cognitive tasks, not with the obvious differences in sexual behaviour. Today, the hype word for sex is "gender", which is taken (wrongly, according to grammarians) to mean the "socially constructed" aspect of sex, that which does not spring from biological differences. But here I use the word "sex", because very few experiments actually studied the subjects' gender independently of their biological sex: there is no experiment involving transsexuals, and only a handful control for homosexuality.

In an nutshell, sex differences in cognition are real, but quite shallow. Men are better at throwing darts (this was shown in an experiment that controlled for sports practice, but, sadly not for pub attendance), mentally rotating objects, and some other tasks related to spatial cognition. This makes it unsurprising that men (compared to women) should be better at geometry than at calculus. This difference is the most important and most consistently observed. Men proved more efficient in tasks concerned with gambling and risk-taking (with correlatively different activations in the orbitofrontal cortex, a brain area involved in making emotions-laden choices - se here). Women are better at all tasks involving precise and rapid motion (such as the purdue-pegboard test, tapping all of one's fingers again one's thumb as rapidly as possible, etc.), at deciphering facial expressions, and (as Tania Singer's beautiful neuroimaging work revealed) at empathizing with others' pain.

There is heated dispute over the question of functional lateralization in men and women; some studies claim that brain activity is more lateralised in men than it is in women, but these results are contested, even omre so since they are associated to the "2 brains" hypothesis, an outdated theory which associated the left brain with (masculine) reason and planification, and the right brain with (feminine) affectivity and imagination.

Moral reasoning and its development seem different in men and women: little girls lay much emphasis on care, protection, and avoidance of suffering, whereas little boys put a premium on fairness and retribution.

Some of the studies adressing the question of sex differences in cognition have come under severe attacks, and not only because of their perceived political impact. There are indeed several methodological concerns that plague the science of sex differences, the most serious of which is the complete lack of cross-cultural controls. Anyway, all these differences are simple refinements over general capacities shared by both sexes. No cognitive faculty has been identified so far that one sex would possess, but not the other (though there are some women who possess 4 kinds of retinal cones- instead of 3 - allowing for some kind of "quadrichromic" vision).

Another problem with sex differences is the fact that they can be provoked by presenting the subjects with the notion that their sex is better (or worse) at such and such tasks, an effect called stereotypes priming. Women primed with the idea that they typically perform worse than men at maths effectively do so, but not if they are primed with the inverse idea (see Hugo's post).

Moderate though as they can be, sex differences do not loose their interest for cognitive science. What little sexual dimorphism there is in humans tells us many things about our evolution; sexually dimorphic traits are typically provoked by sexual selection (the kind of natural selection that is performed on animals by animals themselves when they have to choose their mating partners), which, roughly, vary in strength and scope with the degree of polygyny of a species. Fossil teeth, differences in life length, sexual organs, and many other morphological cues let us think that our evolutionary history was one of moderate to weak polygyny. This seems difficult to reconcile with the position held by some very clever scientists (like David C Geary, Geoffrey Miller, or Charles Darwin himself) according to whom the human mind was born through sexual selection.

Other theorists view male and female cognition as mixts of male and female characteristics: each one of us exemplifies in various dosages both typically male and typically female cognitive trait, depending on our prenatal exposure to certain hormones, and on our upbringing; only some of us are extremely male or female. This is the gist of Simon Baron Cohen's hypothesis, according to which autism is the typical "extremely male" mind. This explains the strongly biased sex-ratio of autism, but does not tell us what an extremely female brain is like.

Sexual cognitive dimorphisms are a touchy matter, not so much because it is difficult in itself, not so much because it's a dauntigly complex complex one, or because the research is approximately conducted, but because it opens a political Pandora's box. Sorry, I am not supposed to mention Pandora - this greek myth conveys sexually discriminativs stereotypes;o)... Let us say that it awakens a pandemonium of opinionated people shouting for their cause. The view that hardwired cognitive differences exist and are not caused chiefly by culture is typically viewed by its adversaries as justifying male opression and unequality between sexes. But the actual data certainly do not argue for male supremacy in intellectual domains; moreover, the idea that women's mind are, in a way, special, can be made into a powerful feminist argument. Indeed, it was one of the XXth century suffragists' chief reasons for allowing women to vote: because their perception of political issues was deemed to be unique, it could not be adequatly represented by men.

Links

See the debate between Steven Pinker and Elizabeth Spelke on this subject.

A bibliography on sex differences in cognition (by Pinker).

My favorite paper: A Cross-Cultural Analysis of the Behavior of Women and Men: Implications for the Origins of Sex Differences. Wendy Wood & Alice H. Eagly. Psychological Bulletin 2002, Vol. 128, No. 5, 699–727

Doreen Kimura sums up the evidence for sex differences here.