A popular paleoanthropological notion is that early on in evolution men specialized in obtaining high value but uncertain supplies of protein by hunting which they exchanged for more certain carbohydrates gathered by women (protein was probably exchanged for sex, too). This suggested that evolution may have optimized each sex for these activities. Leeds University (U.K.) psychologist Gijsbert Stoet tested this hypothesis using computer games that simulated gathering tasks. The results published in the journal Evolution and Human Behavior found that men were in fact better at gathering than women. As the abstract explains [PDF]:
The hunter–gatherer theory of sex differences states that female cognition has evolutionarily adapted to gathering and male cognition to hunting. Existing studies corroborate that men excel in hunting-related skills, but there is only indirect support for women excelling in gathering tasks. This study tested if women would outperform men in laboratory-based computer tests of search and gathering skills. In Experiment 1, men found target objects faster and made fewer mistakes than women in a classic visual search study. In Experiment 2, participants gathered items (fruits or letters presented on screen), and again, men performed significantly better. In Experiment 3, participants' incidental learning of object locations in a search experiment was studied, but no statistically significant sex differences were observed. These findings found the opposite of what was expected based on the hypothesis that female cognition has adapted to gathering. Alternative interpretations of the role of object location memory, female gathering roles and the division of labor between the sexes are discussed.
The division of labor is the crucial issue here. As Stoet notes in the discussion section of his article:
There can be different reasons for a division of labor in a society, and it is not necessarily the case that both genders need to be optimized for the tasks they are doing. It could simply have been the case that a division of labor was driven solely by the fact that men were good at hunting. Women might have chosen to do the gathering, not because they were adapted to it, but because it was the task that remained to be doing. Given that there is no apparent evidence for women being excellent gatherers, this must be considered a plausible scenario.
I found it curious that Stoet did not expliciitly mention the concepts of comparative advantage and the gains from trade. Journalist Matt Ridley offers a nice explanation of how comparative advantage and trade might have worked in early human evolution in his excellent book, The Rational Optimist: How Prosperity Evolves.
Take the case of two cavemen Adam and Oz. Clumsy Adam takes three hours to make a fish hook and four hours to catch fish (7 hours total). Handy Oz can make a fish hook in two hours and catch a fish in an hour (3 hours total). Since Oz is better at both tasks, does Oz really have anything to gain from Adam? Yes. If Adam makes two hooks in six hours and trades one to Oz for a fish that allows Oz to now spend two hours catching two fish. By specialising and trading both Adam and Oz gain an hour of leisure. (A nice illustrated version of Ridley's description of comparative advantage can be found about 5 minutes into this video of his TED talk).
The same sorts of gains from trade would have been operating with regard to the paleoanthropological division of food acquisition labor between men and women.
See also my Reason.tv interview with Ridley here.