In addition to core gender identity and sexual orientation, several other human behaviours, sometimes called gender role behaviours, show sex differences.
They include:
(a) Childhood play,
(b) Aggression,
(c) Interest in parenting or nurturing,
(d) Certain cognitive abilities, and
(e) Manifestations of neural asymmetry, such as hand preferences for writing and neural asymmetries for language processing.
(a) Childhood Play:
Several aspects of juvenile play behaviour, including toy, activity, and playmate preferences, show sex differences. In regard to toys, boys prefer vehicles, weapons, and building toys, while girls prefer dolls, kitchen accessories, and cosmetics and dress-up toys. Sex differences in toy preferences appear early in life, at least by the age of 12 months.
The size of the sex difference depends on the means of assessment (e.g., questionnaire versus direct observation) and the specific toys compared. Although no meta analysis is available, sex differences in toy choices can be large (d > 0.80).
In addition to this sex difference in toy preferences, boys also tend to be more active than girls are and to engage in more rough, active play, including rough-and-tumble interactions that involve playful aggression and overall body contact. Meta analytic findings suggest that the sex difference in activity level begins prenatally and is moderate in size.
No meta analysis is available for the sex difference in rough-and-tumble play, but individual studies suggest that it is moderate in size. Finally, boys and girls differ in preferred play partners with approximately 80-90% of chosen partners being of the same sex.
(b) Aggression:
There also are sex differences in aggression, including its expression in fantasy, imitation of aggressive models, verbal insults, and responses on paper-and-pencil questionnaires. Across cultures and from childhood through adulthood, males are more aggressive than females.
Meta analytic results suggest that the sex difference is moderate in size (d = 0.50) and may be larger in young children than in adults (d = 0.58 versus 0.27), although this apparent age difference could relate to the use of different measures of aggression at different points in the life span.
(c) Interest in Parenting:
It is widely assumed that women have more interest in parenting or nurturing than men do. Additionally, in most cultures, and in most families within our culture, women spend more time caring for children than do men. Interest in infants also shows sex differences and this too can be measured using questionnaires.
In fact, questionnaires regarding interest in infants show bigger sex differences than are seen when behaviour is actually observed, and part of the sex difference on this and other self-report personality measures may be artifactual.
That is, because females and males know that certain characteristics, such as interest in infants, are more valued in a particular sex, they may respond to questionnaires in a way that is more sex-typed than their actual behaviour. The sex difference in interest in infants also is not constant over the life span. It appears to be largest in young adults, perhaps because issues related to parenting are most salient at that age.
(d) Cognitive Abilities:
There is no sex difference in general intelligence. This is not surprising, given that intelligence tests have been specifically designed to avoid sex differences. However, some measures of specific cognitive abilities show sex differences. These include specific subtypes of visuospatial, mathematical, and verbal abilities, and perceptual speed and accuracy.
(i) Visuospatial Abilities:
Mental rotations or the ability to rotate two- or three-dimensional stimuli in the mind rapidly and accurately shows a sex difference favouring males. The difference is present in children, and adults, and may increase with age, although it is hard to be certain because different tasks are used with different age groups.
In contrast to mental rotations and spatial perception tasks, measures of a third aspect of visuospatial ability, spatial visualization, do not show appreciable sex differences. These tasks require complex, sequential manipulation of spatial information and typically have more than one solution strategy.
Measures include tests that require the identification of simple figures within complex designs, the construction of specified shapes from three-dimensional blocks, and imagining what unfolded shapes would look like when folded to form three dimensional objects. Sex differences on these types of tasks are negligible
(ii) Mathematical Abilities:
The overall sex difference in mathematical abilities is negligible. However, measures of problem solving show small sex differences favouring males, particularly among older, highly selected samples, such as college students. Some standardized tests, again used with highly selected samples, also favour males.
This is true of the Mathematics subtests of the Scholastic Aptitude Tests and the Graduate Record Exam, which are used in the United States to select students for Bachelors and Doctoral degree programs, respectively.
In contrast, in childhood, tests of computational skills show small sex differences favouring females, and there are no sex differences in computational skills in adults or in understanding of mathematical concepts at any age.
(iii) Verbal Abilities:
There is a negligible female advantage for general verbal ability in children as well as adults. However, sex differences vary for different types of verbal tasks. Males show a negligible advantage on analogies, and females show a small advantage on speech production and a moderate advantage on verbal fluency (e.g., the ability to generate words that begin with specified letters.
Are sex differences in other cognitive abilities also declining? In some cases the answer appears to be yes. Feingold presents data on various measures of specific cognitive abilities from the 1940s to the 1980s. Sex differences on all of them decline linearly over the decades, with the exception of algebraic problem solving, such as on the SAT, which continues to favour males especially at the upper end of the distribution.
Another exception is three-dimensional mental rotations, on which the sex difference favouring males has been found to be stable from the 1970s to the 1990s. The reduction in the sex differences on some tests does not seem to be caused by test revision in an attempt to remove sex differences, since the decline is linear, not stepwise.
However, despite the declines, which might be attributable to social and educational changes, sex differences in performance have not disappeared, and the portion that remains could be related, at least in part, to other factors, including hormones.
(e) Neural Asymmetries:
(i) Hand Preferences:
Most people, both male and female, are right-handed. However, there are more men than women among the minority who are left- handed. In addition, when the degree of preference for the right hand is assessed across a range of skilled manual tasks, women are found to show stronger or more consistent right hand preferences than men.
(ii) Language Lateralization:
The specialization of the two cerebral hemispheres for language and speech also shows a sex difference. Again, most people show left hemisphere dominance, but more members of one sex than the other appear in the minority, which does not. However, in contrast to handedness, where men are more likely to be atypical, women are more likely than men to show atypical cerebral dominance for language.
Evidence of this comes from studies of speech and language disruption following neural injury and from studies of normal language function. In regard to the former, both men and women are more likely to show disruption following injury to the left than the right hemisphere, but the impairment has been reported to be less severe in women.
This is thought to occur because language and speech are less exclusively focused in the left hemisphere in females. In the intact brain, language lateralization is assessed by simultaneous presentation of verbal stimuli to each cerebral hemisphere separately.
The number of stimuli accurately identified by each hemisphere is then calculated. Data from these procedures also suggest that women show less dramatic language lateralization than men, although meta-analytic results suggest that the sex difference for these types of procedures overall may be negligible.