Sex Differences in Brain and Behavior On the Interaction of Structure, Hormones, and Social Influences

BALTIMORE, MARYLAND, FEBRUARY 10—By middle age, men are losing frontal lobe brain tissue almost three times as fast as women of the same age. Trying to compensate for the loss, men tend to overdrive their neurons, “riding hard on what’s left … with limited success,” according to APS member Ruben C. Gill of the University of Pennsylvania Medical Center, speaking here at the annual meeting of the American Association for the Advancement of Science (AAAS).

Women, however, seem to be able to reduce their rate of neuronal activity in proportion to the tissue they lose, Gur said.

This may account partially for the fact that women survive about a decade longer, on average, than men, Gur believes. If you overdrive cells, according to Gur, cytotoxic effects may begin to act on the active cells due to the inability of wastes to be cleared out quickly enough to avoid a toxic buildup that kills cells. His findings augment those published last year (see January 27, I 995, Science) that showed marked metabolic differences between men and women in the regions of the brain controlling motor functions and emotional responses.

Men’s IQ scores appear to decline more than women’s IQs with aging, Gur said in response to questions. He made this finding by using IQ tests and “other measures that are not included in the typical IQ test,” he said. He used Positron Emission Tomography scans to measure glucose metabolism rates in the frontal lobes of the 24 women and 37 men in his study, and Magnetic Resonance Imaging to measure their brain volumes. (Males begin life with larger brains on average than women, primarily because males’ bodies are larger, in general.)

Gur joined four other behavioral scientists to discuss recent findings on sexual differences in brain and behavior in humans and in other mammals and birds at the AAAS symposium. Their aim was to shed some initial light on how various factors might interact.

Academic Achievement, Sexual Stereotyping

Jacquelynne Eccles, an APS fellow, presented findings on sexual stereotyping and differential treatment by parents and teachers. Along with colleagues at the Institute for Social Research of the University of Michigan, Eccles conducted research using two longitudinal studies of a total of 2,600 boys and girls from kindergarten to high school. Eccles said that when sixth graders in one of the studies got high grades in mathematics, investigators asked their parents how they accounted for that success: whether it was because “my child is very talented” or because “my child worked very hard.”

The investigators found that “if parents are talking about a son, they say both of those things are important, but if they are talking about a daughter, they say it’s because she works hard, not because she is smart.” The same effect was documented with teachers, too, though it proved to be independent of any measure of real aptitude differences between the boys and girls, Eccles said.

“So, are girls really working harder?” Eccles asked with a shrug of the shoulders. “Well, if you ask girls, they will tell you they are working harder. But if you compare the actual number of minutes they spend working on math homework, in fact there is no difference. And if you do not tell teachers that you are studying gender, and you ask them how hard ‘Johnny’ works and how hard ‘Mary’ works, they tell you there is no difference. Yet, the teachers nevertheless believe the girls are working harder,” said Eccles.

Overtime, these attributions of success, according to Eccles, undermine “girls’ self-confidence, so girls come to believe ‘I am doing well because I work hard, and when the material gets difficult I will fail. ‘ They are not willing to project [their current success] overtime and say instead ‘Why wouldn ‘t I do well next year, too?’ or ‘Why wouldn’t I do well in engineering as a career?'” Whereas boys, Eccles stated, absorb a very different message from their educational experiences: “I ‘m doing well both because I’m working hard and because I’m talented.”

Hormones and Sexual Differentiation

A third member of the panel, APS member Melissa Hines of the University of London and University of California-Los Angeles, discussed variations in sex and gender behavior of humans exposed to abnormal endocrine environments during early development.

“We are now beginning to get beyond the question of whether or not hormones have an influence,” Hines said. “We can now begin to identify the nature of the influence and sort out critical periods for the influences and even begin to identify how hormonal influences might interact with other factors such as social and cultural influences to shape human behavior.”

Hines’ described some of her own research focused on women exposed prenatally to DES, the synthetic estrogen that was prescribed for millions of pregnant women in the 1950s because it was thought to prevent miscarriage. She and her colleagues compared the sexual profiles of women exposed to DES prenatally and those who were not exposed. The researchers hypothesized that the prenatally DES-exposed women would show more masculine cognitive profiles in visual/spatial ability, and the stronger right-ear dominance in verbal tests that men generally tend to exhibit.

Hines and her colleagues found no evidence to support their hypothesis regarding visual/spatial ability differences but some evidence that DES women differed from non-exposed women with respect to right-ear dominance. Hines said studies by other investigators have reported that DES-exposed women are likely to be bolder in personality and more likely to be bisexual than unexposed women.

She suggested that different behaviors may be influenced by different hormonally influenced mechanisms and “that hormones contribute to our sexual differentiation but that each characteristic is influenced in its own idiosyncratic way. One type of influence that could differ is in the timing … of exposure. DES exposure is always prenatal. It may be that differentiation of language [right ear/left ear] lateralization is prenatal. Development of sexual orientation may also be prenatal. But visual/spatial performance depends more on cortical development that occurs during the postnatal period.”

Brain and Sex

Other speakers, both from the University of California-Los Angeles, were Arthur P. Arnold, who chaired the symposium, and Roger A. Gorski, professor of neurobiology in the School of Medicine.

Arnold reported that studies of sexual differentiation in songbirds point to direct genetic influences rather than mediation by hormonal secretions. Arnold said that “attempts to apply the traditional model of hormone control of brain sexual differentiation to birds and marsupials have not been very successful. “In zebra finches, it is possible to cause masculine patterns of development in young females by giving them estrogen, but, so far, attempts to prevent masculine patterns of development in genetic males [e.g., by reducing estrogen action or synthesis] have not been successful,” he said. Reporting on recent research at the Brain Research Institute, Arnold said that testicular secretions are not sufficient to induce masculine patterns of neural development in this species. The results raise doubts about the universality of the traditional model of hormonal regulation and modulation of such differentiation, he concluded.

But stepping up the evolutionary ladder, structural sex differences in the human brain as well as implications for sexual orientation were the subject of Gorski’s discussion. “The possibility that sex differences in cognitive function and/or sexuality have an underlying basis in…brain structure is exciting yet controversial, but amenable to further study,” he said. As to whether such structural differences are related to the process of the sexual differentiation of the human brain, Gorski said this prospect “is even more controversial and much less amenable to experimental study.”

Gorski first laid out a context of scientific findings to help make more understandable his discussion of structural sex differences: In humans and other mammals, nature’s “default program, or blueprint,” according to Gorski, is female in terms of both the internal reproductive organs [excluding the gonads, which are determined genetically] and external genitalia. This is also true of brain function and structure in rats and other mammals, he claims. A number of “structural sex differences in the rat brain undergo the process of testicular hormone-dependent sexual differentiation or masculinization,” said Gorski. Some of these are related to reproduction, but others are not. In addition, a number of structural sex differences exist and are determined during the process of sexual differentiation and presumably underlie functional sex differences.

The list of structural sex differences in human brains is much shorter than that of the rat, and the differences are less pronounced as well as more controversial. Gorski waded through a number of human studies that both did and did not indicate structural differences in different regions (e.g., corpus callosum and its subregions including the anterior commissure, splenium, massaintermedia, and isthmus). Many such studies did not match subjects for age (the corpus callosum’s size changes with age) or failed to account for subject’s handedness! Within the hypothalamus, the preoptic area has been found to exhibit a sexually dimorphic nucleus (larger in the male), though not in Gorski’s studies. His research has, however, revealed two other sexually differentiated areas in other hypothalamic regions.

It is clear, Gorski emphasized, “all putative sex differences in brain structure must be replicated consistently before being … accepted.” And, of course, once the differences are demonstrated conclusively, there is the issue of uncovering functional significance. Gorski said the functional differences remain obscure but could underlie reported sex differences in cognitive abilities.

In addition, reported differences in brains of apparently heterosexual and homosexual men still require replication. “Since in homosexual individuals there is usually no discordance between perceived brain sex and phenotypic sex, just an atypical selection of a sexual partner, it is more difficult to interpret structural differences in the brain,” explained Gorski. In the rat, the known structural differences that correlate with sexual orientation are the volumes of the suprachiasmatic nuclei and an area in the hypothalamus, and the midline area of the anterior commissure. The process of sexual differentiation in the rat has multiple, independent components, and it may be possible that changes in just one could lead (or predispose) to homosexuality, according to Gorski.