Sex Differences In Cognition

May 4, 2008 at 3:05 pm (feminism, neat things, sexism)

I’ve been working pretty constantly for the last two days on my big final paper for a class of mine. It’s a research paper on sex differences in humans, specifically differences in problem solving and on the performances of various mental tasks. What a huge, exhausting topic! Ugh.

I’m posting it here in order to make the big amount of work feel vaguely more satisfying (it’s an assignment and a post!), and because, as you may have observed, I tend to post my longer, more interesting papers. This paper will be edited by my peers and my professor this week, so I may post the revision if I feel it has improved substantially. Also to look forward to next week: my final essay for another class, Writing for Arts and Culture, about sexual imagery in the work of Jeff Mangum, in which I will try to tease out the commonalities in the sexual references in the lyrics of In The Aeroplane Over The Sea, On Avery Island, and some live recordings including Live at Jittery Joe’s, a show at Aquarius Records, and a show in Auckland, New Zealand, in 2001, hopefully developing an idea of some kind of over-arching meaning, attitude, intention, something. That one should be fun.

Anyway, this paper. Almost entirely below the fold, of course, ’cause it’s ten damn pages long. As a taste-test to help you figure out if you have any interest in reading it, here are some things I learned that I wasn’t expecting to (in addition to a whole bunch of stuff I was expecting to learn): little girls are about 2% more likely to engage in “cross-gender behavior” than little boys (van Beijsterveld 650), but despite this, boys are something like six times more likely to get referred to doctors to be treated for GID (van Beijsterveld 655), which shows just how huge the imbalance is between the social acceptance of female masculinity (in kids) and male femininity. Also, it’s fairly apparent, I think, that a combination of classism, racism, weird sexist ideas, and macho culture are undermining a lot of boys’ academic success. These two things have convinced me even more that we desperately need a men’s equivalent of feminism, not to shuck off the oppressive regime of women or to overturn feminist achievements or anything like that, obviously — die-hard feminist here — but to do for boys and men what feminism is doing for girls and women: to expand opportunities and remake gender roles. We’ve begun the work of making it socially acceptable for girls to excel in math and science; as we continue that work, we need to begin the process of encouraging boys in the humanities. As we continue the work of expanding our conceptions of femininity and femaleness, we need to do the parallel work of expanding our ideas of masculinity and maleness. The gender system will not die until we until we begin to attack it from every side. And for fuck’s sake, that thing must die.

On that note, here is my paper.

Sex Differences In Cognition:
Disparities, Similarities, and Explanations

There are documented differences in the average intellectual strengths and weaknesses of women and men. Some postulate that these dissimilarities are innate, the product of an evolutionary history that placed different pressures on the sexes. Others account for the variation with the strong forces of culture and socialization, asserting that the divergences are the product of the different treatment of young boys and girls.

At this time, it is not clear whether the demonstrable differences in the behavior of women and men are caused primarily by biological or environmental forces — there is support for both claims. What is clear, however, is that the sex of an individual is not an indicator of his or her particular abilities. There is a tremendous amount of overlap between the sexes, a the tremendous amount of variation within each sex, and a tremendous number of forces that influence behavior. Though there are observable trends along gender lines, they are just that: trends, which cannot reveal information about any given individual. Furthermore, “variation between men and women tends to be smaller than deviations within each sex” (Kimura 34) — regardless of certain tendencies, women and men are far more alike than they are different, and sex is just one of many axes on which persistent differences can be observed.

The Differences: A Brief Overview
Beyond the obvious, anatomical sex differences, men and women tend to differ in their mental faculties. These differences occur “in patterns of ability rather than in overall level of intelligence” (Kimura 33). On average, in laboratory conditions, women outperform on verbal exercises, precision manual tasks, and tests that ask subjects to match like items as quickly as possible (Kimura 34). Men in laboratory conditions tend to do better than women at spacial tasks such as navigating routes, aiming objects, and matching lines of the same slope (Kimura 34).

Men and women may also tend to use different parts of their brains for the same activities. Amongst patients who sustained damaged to the anterior of the brain, women were more than twice as likely to develop aphasia, the loss of the ability to speak or to comprehend speech (Kimura 37). Amongst patients who had experienced damage to the posterior region, however, the situation is reversed, and even more extreme: 60% of men developed aphasia, compared to just 12% of women (Kimura 37). This suggests that women tend to use the anterior of the brain for speech-related tasks, while men tend to use the posterior.

Adolescent males are more likely to engage risky alcohol consumption — binge drinking has been found to be about 50% more prevalent amongst teenage boys than teenage girls (Kuntsche et al. 670). In some populations, young girls greatly outperform young boys on reading comprehension tests (Entwisle 116); there is also a more general trend of female students doing better than their male cohorts overall (Whitelaw 88).

Nurture: The Effects of Environment
Some of these differences, such as brain structure, are unlikely to have been caused by culture. Others, however, such as school performance, appear to be largely the result of social forces.

Girls and boys have roughly the same scores on reading comprehension tests in the first grade (Entwisle 116). Over the next few years, however, the scores begin to diverge, with girls scoring 18 points higher than their male peers by the fifth grade (Entwisle 116).

A closer look reveals a more complicated trend. Amongst students in the higher strata of income — as defined by not receiving meal subsidies — girls’ and boys’ test scores continue to be the same, with an average score of 529 for both groups (Entwisle 116). Amongst fifth-grade students receiving meal subsidies, the average score was 458 for boys, 477 for girls (Entwisle 116). In this case, inherent, biological sex differences are clearly not the cause of the disparity — boys from higher-income families keep pace with their female peers, so there is no reason to suspect that male students cannot do as well in school. Further, while the gender gap between lower-income students is 19 points, the gap between lower-income girls and their higher-income cohorts is 52 points: a rift nearly triple the size of the gender gap. Class, not sex, is the more significant factor here.

One study explored the gender gap in school achievement by surveying students about their attitudes, seeking to determine the relationships amongst gender, age, and “pupils’ views of factors relating to academic success” (Whitelaw 93). The study found a very clear trend amongst boys: while the younger male students were overall ambivalent about statements such as “clever boys are popular with other boys” and “it’s cool to do well in school,” by the time they reached the higher secondary grades, boys had reached a consensus that smart boys are not popular, and that doing well in school is not cool (Whitelaw 94). Amongst girls, however, there was no relationship between age and whether they considered it cool to succeed academically (Whitelaw 96). This suggests that amongst boys, but not girls, an increasingly anti-academic culture emerges as they get older: “it is clear that as many boys progress through the school their attitude towards academic success deteriorates” (Whitelaw 99). It is likely that these attitudes, rather than innate sex differences, greatly influence boys’ declining school performance over the course of their academic careers.

There are several reasons why such outlooks might appear among boys, but not girls, namely the attitudes and expectations of parents, teachers, and the culture at large. Studies have shown that the grades that students receive are based as much on teachers’ subjective perceptions as on objective achievement, and that grades are as predictive of students’ future learning as they are indicative of current performance (Entwisle 116) — so if teachers tend to give one sex better grades, that could cause the favored gender to actually learn more in school (Entwisle 116). Classroom conduct is a major contributor to the marks children receive, especially early in school, and can account for as much as half of the deviation in grades amongst pupils (Entwisle 116). Parents tend to expect better behavior in school from daughters than sons (Entwisle 116), possibly causing a feedback loop in which parental expectations lead to different behaviors, leading to different grades, leading to different levels of learning, leading to an even larger gap in grades.

As for the differences along socioeconomic lines, parents of higher income tend to have higher expectations for their children in school (Entwisle 116) and, perhaps more importantly, are more likely to engage in a wide range of activities that help children succeed academically. Higher-income parents are more likely to read to children and to encourage their children to borrow library books (Entwisle 116). Middle-income families “advocate” for students at school more than their lower-income peers (Entwisle 116), such that middle- and higher-income students may receive better treatment than lower-income pupils even at the same school. And because most teachers are from middle-income backgrounds, they are probably operating with social norms more similar to those of other middle-income families (Entwisle 117).

Nature: The Effects of Biology
Many of the differences between the behaviors of women and men can be explained by such social conditions. But biology also has a role to play.

One study attempted to determine the role of androgens — male sex hormones — in one particular cognitive strength: mental rotation task (“MRT”) performance (Peters 251), or success at spatial-reasoning tasks that involve imagining what an object would look like if it were rotated. This kind of spatial reasoning is one area in which men tend to outperform women (Peters 251). The researchers tested subjects’ proficiency at MRTs, and compared their scores with several factors that are indicators of exposure androgens: sex, since male bodies contain more androgens than female bodies; height, which can be influenced by androgens; finger-length ratio, because prenatal exposure to androgens affects the proportions of one’s fingers; and sexual orientation, because there is mounting evidence that prenatal exposure to androgens is connected to sexual orientation, with homosexual women exposed to higher levels than heterosexual women (as revealed by finger-length ratio), and the inverse for men (Peters 251).

The study’s findings were as predicted (Peters 251). Though there was great overlap between men’s and women’s scores, with the two groups being approximately equally likely to score 3 out of 6 possible points, overall, men scored higher than women across all ethnic groups (Peters 254). As expected, homosexual and bisexual women — who are more likely to have been exposed to higher levels of androgens — did better than heterosexual women, while bisexual men received lower scores than their heterosexual peers, and homosexual men’s scores were still lower (Peters 255). Further corroborating the assertion that the different scores are the product of sex hormones are the data from the other measures, which reveal that taller people performed better, and that average heights varied across sexual orientations (Peters 256). Heterosexual men had the highest average height, with homosexual and bisexual men averaging a centimeter shorter (Peters 256) Correspondingly, homosexual and bisexual women averaged about a centimeter taller than heterosexual women (Peters 256). The results for digit ratio were parallel, with those with smaller ratios — indicating greater exposure to androgens — scoring higher (Peters 257).

These findings demonstrate that there is a relationships between sex hormones spatial reasoning ability, present in both women and men. In the case of mental rotation task performance, it is clear that the gender gap has its routes in biology, not culture.

Men and women have very different hormonal constitutions. Both men and women produce androgens, a percentage of which are converted to estrogens, female sex hormones (Federman); “androgens such as androstenedione and testosterone are synthesized in both the ovary and testes and then partially converted to the estrogens estrone and estradiol” (Federman). Though the process is the same, the proportions are very different, with men typically producing 7000 micrograms of testosterone, the primary androgen, each day, and converting less than 1% to estradiol, the primary estrogen (Federman). Women produce an average of 300 micrograms of testosterone, and convert half of it estradiol (Federman).

These levels and ratios vary in women throughout the menstrual cycle. Some studies have shown that women’s performance at certain tasks differs with this hormonal fluctuation (Kimura 36). During times of especially high estrogen production, women’s abilities were shown to be more extreme manifestations of the general trends, with even lower scores on relevant spatial tasks, and even higher performance on verbal tasks and tests of manual dexterity (Kimura 36). These findings suggest that, just as increased androgens lead to greater spatial reasoning abilities, even small escalations in estrogens lead to increased linguistic fluency and manual precision.

Although this suggests that adults respond strongly to sex hormones, it is also likely that many of the effects of hormones are hardwired early in development. Male and female bodies begin producing very different hormonal cocktails at puberty, but some research has shown that the differences between females and males are present even in prepubescent children: studies have shown that boys as young as three are better than their female cohorts at aiming objects and mental rotation tasks (Kimura 35). Correspondingly, preadolescent girls have been shown to outperform their male peers in tests of verbal memory (Kimura 35). These early divergences may be the result of exposure to different hormones in the womb.

Variation Within the Sexes
As demonstrated by the MRT performance study, there is considerable variation within each sex, which is heavily influenced by hormones, and is related to sexual orientation. Other studies have produced similar results. For example, part of the hypothalamus — the part of the brain connected to gendered sexual behavior — is typically smaller in women than in men (Kimura 33); both sexual orientation and gender identity have been found to be related to the size of this region (Kimura 33). Research has shown that another region of the hypothalamus is smaller in male-to-female transsexuals than in male-bodied persons who are not transsexual (Kimura 33), suggesting that gender identity has biological routes and is part of a spectrum of anatomical variety within the sexes.

A study of Dutch twins attempted to assess the prevalence of “cross-gender behavior,” and to estimate the roles of biology and environment in its genesis (van Beijsterveld 647). The study found that, amongst seven-year-olds, 3.4% of boys and 5.2% of girls were said by their parents to act “like [the] opposite sex,” with 1.0% of boys and 1.7% of girls described as wishing “to be of [the] opposite sex,” for a total of 3.7% of boys and 5.7% of girls falling into the category of gender-variant (van Beijsterveld 650). By age ten, the total number of children exhibiting cross-gender behavior went down to 2.7% for boys and 3.6% for girls (van Beijsterveld 650). It is unclear whether greater frequency of cross-gender behavior in girls is caused by greater tolerance for gender-atypical behaviors in girls, or by a a truly larger level of gender variance in females (van Beijsterveld 654). It is likely that social factors play a major role in the disparity, however, because despite the fact the girls apparently exhibit more cross-gender behavior, boys are up to six times more likely to be referred to doctors for treatment for Gender Identity Disorder (van Beijsterveld 655), the clinical name for transgenderism. This suggests that feminine behaviors in young boys are much less socially acceptable than masculine behaviors in young girls.

By comparing the rates of concordance for monozygotic and dizygotic twins, the researchers concluded that gender variance is a “highly heritable trait” (van Beijsterveld 655), with about 70% of the variation accounted for by genetic factors in both girls and boys (van Beijsterveld 655).

Conclusion: Many Causes, Little Relevance
It is clear that differences do exist in the average abilities of men and women, in a variety areas. It is also apparent that both biology and culture are significant contributing factors to these disparities, that neither women nor men comprise uniform groups, and that, though some gender gaps are large, many are small, and many are smaller than the gaps between other social groups.

Interestingly, some of the evidence suggests that, while men and women tend to use different techniques to solve problems, they may be comparably successful at actually solving them. For example, though men tend to be better at mathematical reasoning, women tend to be superior at mathematical calculation (Kimura 34). Similarly, one study of rats demonstrated that females used landmarks, such as images on walls, to navigate mazes, whereas males tended to use geometrical cues, such as the shapes of rooms (Kimura 35) — these findings are similar to other studies’ findings in humans (Kimura 34), revealing two different strategies that, in the real world, are likely to be similarly useful. Add to this the fact that men and women, though they tend to have different strengths, have the same level of overall intelligence (Kimura 33), and there emerges a picture of sex differences that may have very little relevance to daily life. This is especially true if gender roles continue to expand and diversify, slowly diminishing the culturally induced differences.


Works Cited

Entwisle, Doris R., Karl L. Alexander, and Linda S. Olson. “Early Schooling: The Handicap of Being Poor and Male.” Sociology of Education 80.2 (2007): 114-138

Federman, Daniel D. “The Biology of Human Sex Differences.” The New England Journal of Medicine 354.14 (2006): 1507-1515

Kimura, Doreen. “Sex Differences in the Brain.” Scientific American May 2002: 32-37

Kuntsche, Emmanuel, et al. “Disentangling gender and age effects on risky single occasion drinking during adolescence.” European Journal of Public Health 16.6: 670–675

Peters, Michael, John T. Manning, and Stian Reimers. “The Effects of Sex, Sexual Orientation, and Digit Ratio (2D:4D) on Mental Rotation Performance.” Archives of Sexual Behavior 36 (2007): 251–260

Van Beijsterveldt, C. E. M., James J. Hudziak and Dorret I. Boomsma. “Genetic and Environmental Influences on Cross-Gender Behavior and Relation to Behavior Problems: A Study of Dutch Twins at Ages 7 and 10 Years.” Archives of Sexual Behavior 35 (2006): 647-658

Whitelaw, Sarah, Lena Milosevic, and Sandra Daniels. “Gender, behaviour and achievement: A preliminary study of pupil perceptions.” Gender and Education 12.1 (2000): 87-112



  1. Daran said,

    A closer look reveals a more complicated trend. Amongst students in the higher strata of income — as defined by not receiving meal subsidies — girls’ and boys’ test scores continue to be the same, with an average score of 529 for both groups (Entwisle 116).

    My analysis of the Connecticut Mastery Test Scores suggested otherwise. Boys at the higher strata school fell proportionately behind their female peers about the same as boys from the lower strata school did theres, though this was less harmful to the higher strata boys because most of them achieved passing grades anyway.

  2. Daisy said,

    Hey Daran! I will read your analysis; I haven’t seen it before.

    In the case of the study I was looking at (I can send you a PDF of the paper if you’d like), the numbers were unambiguous: the average score of 5th graders not receiving meal subsidies was 529 for both girls and boys (though I admit that was the extent of my exploration, since I was just trying get the paper done on time). Also, as I pointed out here, the gap between higher-income and lower-income girls was triple the gap between lower-income girls and boys, which suggests to me that class is, indeed, the primary factor.

  3. A Just Gender Culture, Or, To End Sexism, We May Need More Gender, Not Less | Revolutionary Act said,

    […] a big range in terms of what a given culture codes as one or the other. The existence of some observable differences between women and men (that are probably biological in origin) at the large scale doesn’t negate this. (And, about […]

Comments are closed.

%d bloggers like this: