Brain Storm

Can we talk about sex differences in math and science aptitude without yelling?


At a recent academic conference, MIT biologist Nancy Hopkins walked out in the middle of a talk by Harvard President Lawrence Summers. That might seem rude, but it turns out she had a medical excuse.

"When he started talking about innate differences in aptitude between men and women," Hopkins told The New York Times, "I just couldn't breathe because this kind of bias makes me physically ill." Had she remained in the room, she explained to The Boston Globe, "I would've either blacked out or thrown up."

Although Hopkins' reaction was especially dramatic, her queasiness was shared by several other women in the audience, who were described as "outraged" and "deeply offended." Those feelings quickly spread to Harvard and elsewhere, prompting angry criticism of Summers and an apology in which he said "I deeply regret the impact of my comments."

This controversy is ostensibly about the ability of women to excel in math and science. But it says more about the ability of academics to engage in rational debate when confronted by views that contradict their cherished assumptions.

Speaking at a conference sponsored by the National Bureau of Economic Research, Summers, an economist and former treasury secretary, suggested three factors that may help account for the scarcity of women on the math, physical science, and engineering faculties of leading universities. In addition to discrimination (the explanation favored by Hopkins) and the reluctance of mothers to put in the long hours required by top math and science positions, he mentioned sex-related differences in ability.

Summers' remarks may have failed the Hopkins Nausea Test, but they hold up better when judged by more scientific standards. Decades of testing have shown that boys and men tend to do better than girls and women on tasks that require spatial reasoning (e.g., mentally rotating objects) and advanced mathematical abilities. These differences are especially pronounced at the upper end of the distribution, where future scientists and mathematicians congregate.

"It has been fashionable to insist that these differences are minimal, the consequences of variations in experience during development," wrote University of Western Ontario psychologist Doreen Kimura in a 1992 Scientific American article. "The bulk of the evidence suggests, however, that the effects of sex hormones on brain organization occur so early in life that from the start the environment is acting on differently wired brains in girls and boys."

Since then, the evidence has become stronger. "A variety of data collected throughout the 1990s show that gonadal hormones…have demonstrable effects on the cognitive abilities of women and men," wrote psychologists Diane Halpern of California State University in San Bernardino and Mary LaMay of Loma Linda University in a 2000 Educational Psychology Review article. "Converging evidence from a variety of sources supports the idea that prenatal hormone levels affect patterns of cognition in sex-typical ways."

The precise contributions of early brain structure and subsequent experience are still a matter of controversy. Halpern and LaMay, for instance, suggest initial differences in aptitude may be magnified by their impact on interest, encouragement, and self-esteem. But Summers never implied the matter was settled; to the contrary, he called for further research and debate.

His critics took it personally. A Harvard senior told the Times "it's disconcerting that the man who is supposed to have your best interest in mind and is the leader of your education community thinks less of us."

Yet average group differences in ability do not imply a judgment about any particular individual, since there is still much overlap between the sexes. Although men predominate in the upper echelons of math and science, that doesn't mean the women who make it are any less qualified. The situation could change, of course, if the demand for gender balance leads universities to select faculty members based on their sex.

Given the implications for attempts to achieve faculty "diversity" (a goal to which Summers pledges allegiance), it's not surprising that the subject of sex differences in math and science aptitude provokes strong feelings among academics. But that is not all it should provoke.

"I think if you come to participate in a research conference," Georgia State University economist Paula Stephan told the Times, "you should expect speakers to present hypotheses that you may not agree with and then discuss them on the basis of research findings." Surely that is not demanding too much of people who consider themselves scientists.