In this book, Kimura gives an unbiased overview of sex differences in cognition, and provides scientific data of how hormones act prenatally and perinatally on brain development and organization. The main bulk of the book deals with differences between men and women (and boys and girls) in cognitive abilities. The introductory chapters focuses on topics ranging from evolutionary psychology to early hormonal development and the various effects of androgens, and following this there are chapters focusing on specific subjects such as sex differences in motor skills, spatial abilities, mathematical aptitude, verbal abilities, and so on. Seasonal cycles of testosterone levels and fluctuations in estrogen levels across the menstrual cycle are also discussed in a later chapter, highlighting the finds presented earlier in the book. Further there are chapters dealing with studies on brain mechanisms in normal and in damaged brains respectively, and finally a chapter on body asymmetry and cognition. The book is magisterially written and lucidly argued. It is also refreshing to read a book by a female professor who completely eschews the grammarian’s term “gender” - there is certainly more to sex differences than grammar, to say the least - as Kimura repeatedly shows throughout this book. She writes with regard to feminist critics in the Introduction: “Why should sex differences be treated differently from other kinds of data?” - Why indeed. She goes on to show that the (much touted) environment, to the degree that it is influential, works on brains that are already differently wired from the outset.
Kimura also brings to attention something I wasn’t at all aware of until now: that “items on tests like the SAT (and many other standardized tests) change over the years, because some items go out of fashion, or even because items performed better by one sex were deleted!” (p. 69) - And further there’s a discussion of the verbal half of the Wechsler Adult Intelligence Scale (WAIS), where she writes as follows: “In originally devising his IQ test, Wechsler omitted tests that yielded large sex differences (such as mental rotation tests), the intent being to equate IQ scores obtained between men and women. So it is despite this aim that there is a slight edge for men on the Verbal IQ, indicating that, popular beliefs and claims to the contrary, women are not more verbally intelligent than men.” (p. 95) – It is disturbing to know that even standard tests like these are in fact devised to conceal sex differences.
Some known sex differences are not dealt with in this book, such as the greater variance in male scores on IQ tests, with men occupying far more of the higher end of the IQ scale than women - though this is a later find I believe. On the whole, this is an excellent introduction to the field; the book is well organized with a Summary at the end of each chapter and there’s even an Appendix dealing with statistical methods.




This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License. ( )

Author Doreen Kimura was a psychology professor at Simon Fraser University. Sex and Cognition is a condensation of her life’s work on gender and mental differences. There’s probably a lot of political backstory that Ms. Kimura doesn’t cover; she was founder of the Society for Academic Freedom and Scholarship, and her introductory chapter tries to preempt some of the potential criticisms of her work:


* There are no biological differences in male and female behavior; apparent differences are all the result of socialization.


* Because some apparent differences only appear after childhood, they must be due to socialization rather than biology.


* All differences must be considered due to socialization until proven otherwise.


* All people would be equal in behavior if they were exposed to the same social influences.


* Because there is a great deal of overlap in behavior between men and women, there are no real differences.


* Research on sex differences in behavior should be held to different standards than other research.


* Research on sex differences should be done at all because it is potentially harmful.


Ms. Kimura, of course, has answers to all of these, none of which are particularly harsh (although she does call a criticism “absurd”. Once.).


For the rest of the book, Kimura qualifies every statement with lists of potential confounding factors not yet accounted for; inclusion of studies that seem to give contradictory or null results; disclaimers that more research is necessary on a particular area before an solid conclusion can be drawn; repeated cautions that causation does not necessary follow from correlation; and, frequently, that maybe observed differences really are due to socialization. This tends to make for less than zingy text, but it’s obviously necessary to fend off criticism. Following that, you must assume that all those qualifications apply in my summary below; it would be tedious to repeat them. I can pretty much guarantee that for every comment made suggesting a nonbiological explanation, “She thought of that and accounted for it” will be the response.


Kimura first step is to define what it means to be a man or a woman. As is usual in science, the interesting things happen in the borderline states. Basically human embryos start out with structures that can develop into either testis or ovaries (and their associated plumbing). If the embryo has at least one Y chromosome (and the cells with the Y chromosomes are in the embryonic areas that are in charge of sexual differentiation), a series of changes takes place that suppresses the female stuff and develops the male; otherwise the development continues as female. Since female is more or less the default condition, and there are a number of steps that have to proceed in sequence for male development, it’s more likely for things to go wrong in male development than female. Kimura also considers (where data is available) homosexuals and transsexuals of both genders. This is chapter is an excellent summary of what goes on in the development of gender.


The conventional wisdom is women are better at “verbal” tasks and men are better at “spatial’ tasks. Kimura generally confirms this but notes it’s a lot more subtle and involved. The single largest effect – and the only one where there’s at least one standard deviation between the sexes – is “targeting” – throwing an object to hit a target; straight men are better at this than straight women (apparently there are no data for gay women). (Interestingly, male chimpanzees throw things more often than females – but aren’t any more accurate; gay men and straight women were equal in targeting; and the effect is observable by age five).


For women, the largest superiority (although not as large as male superiority in targeting) was fine motor control of the extremities (not, interestingly enough, a “verbal” ability) in various tests involving manipulation of small objects. (Here’s one you can try at home; hold out a hand, palm down, and bend each finger 90° downward at the middle joint. Women find this easier than men). The popular truism that women can’t read maps turns out to be false but there is a difference; women process geographic information differently from men, relying more on recognition of landmarks and less on direction and distance. When women were tested in the lab with maps that included landmark information they were able to learn routes as quick as or quicker than men. This difference goes far down the evolutionary tree; male rats learn to run mazes faster when there is geometric information (Kimura doesn’t explicitly state what “geometric information” is; I assume distinctive angles or lengths in the route but it could be something else); female rats learn faster when there is “landmark” information – pictures on the walls of the maze or objects scattered through it. This brings up another observation; it’s generally true that the sex differences only apply under time constraints; if allowed unlimited time, both genders do equally well (for example, in tasks requiring mental rotation of an object to match a choice of pictures). And another interesting discovery is that practice helps, but in a perhaps counterintuitive way. If repeatedly offered different versions of the same general test – mental object rotation or word selection or color matching, and so on – both sexes improved their scores; however, not by actually improving their performance. Instead they developed test-taking strategies that allowed them to eliminate obvious wrong answers, gave up more quickly and moved on if the problem was too difficult, and so on.


All fascinating stuff. Kimura, presumably attempting to diffuse one of the obvious criticisms, notes that for all the tests – even for the targeting test – the within-group differences are larger than the between-group differences (i.e., the difference between the best performing man and the worst performing man was much larger than the difference between the average man and the average woman, and vice versa). The flip side, of course, is that none of the noted differences make much difference to men or women or society in general.


There are a couple of downsides; one is Kimura’s tendency to give adaptationist explanations for differences. For example, women do better than man on a test showing a picture of objects in an array – either geometric symbols or actual things, then being asked to identify the differences between that and a second picture. Well and good, but then Kimura suggests this might be because women need to be better at identifying if objects in the household have been disturbed by vermin or a stranger. Could be but like most adaptationists explanations it’s impossible to think of a way to falsify it. Another annoyance is Kimura’s explanation of why women are less likely than men to pursue scientific and technical careers – rather than sexism, women are more “person-oriented”. I’m confounded by this generalization, especially since Kimura has been so careful to qualify everything elsewhere; she provides no explanation of (A) what “person-oriented” means, (B) how you would measure how “person-oriented” somebody was, or (C) that women really are more “person-oriented”, given that you can establish (A) and (B).


The last chapter seems unnecessary (and Kimura apologizes in her introduction); how body symmetry differs between men and women. This is probably research she was doing at the end of her career and it’s interesting (women and gay men tend to have more fingerprint ridges on fingers on their left hands; straight men have more on their right) but it’s only convolutedly related to the rest of the text. There’s also an appendix explaining what normal distributions, means, standard deviations, variance and correlation coefficients are; if you don’t know those already you probably shouldn’t be reading the book in the first place.


Food for thought, though. As might be expected, heavily referenced, usually from primary sources (journal articles rather than books). Examples of many of the tests are provided; it would have been interesting (although it would have made for a much longer book, obviously) to follow the spirit of books that test IQ and give full length versions so the reader could test themselves and figure out what their gender was. ( )