Math is Hard, Barbie Said

If I ever again hear the word "hard-wired" used to describe anything other than an electrical system—the human brain, for instance—I'm going to scream. Even allowing for the unfortunate fact that old ideas in science tend not to die out until the mandarins who hold those ideas are in their graves, the dogma of the hard-wired brain has endured for an inexcusably long time given the evidence against it. The motor cortex is supposedly hard-wired, its left half controlling the right side of the body and its right half controlling the left side. But therapy developed for stroke patients can coax the left motor cortex to move the left side of the body, taking over for the stroke-damaged right motor cortex. Even our visual cortex, which you'd think would be as hard-wired as hard-wired can be given the centrality of vision, can change jobs: when people spend a week blindfolded and receive intense tactile stimulation (feeling Braille dots), the visual cortex switches from processing what the eyes send to what the fingertips send, scientists led by Harvard's Alvaro Pascual-Leone reported in August. Something similar happens in people who are blind from birth. So much for hard-wired.

If not even a structure as fundamental as the visual cortex is hard-wired, can we please retire the claim that boy brains are hard-wired for math and girl brains are not?

There is no denying that, at the elite levels of math, men vastly outnumber women. Women received 27 percent of the Ph.D.s in math awarded by American universities from 1993 to 2002, edging up to a still-woeful 29 percent last year. They make up only 19 percent of the tenure-track faculty in math departments. No woman has ever won a Fields Medal, the "math Nobel." The Study of Mathematically Precocious Youth, which looks at kids younger than 13 who score 700 or above on the math part of the SAT, found a 13-to-1 boy-girl imbalance, implying what the researchers called "superior male mathematical ability."

Now for the "however" part. That 13-to-1 ratio was true in 1983. In 2005 it fell to 2.8 to 1. Nothing in the brain that is "hard-wired" can change that quickly. Cross-cultural data on young people with off-the-scale math ability are even more telling, as researchers will report in next month's issue of the Notices of the American Mathematical Society. They combed the last 20 years of results of the world's premier math competitions, such as the International and U.S. Mathematical Olympiads and the Putnam Competition (with 12 problems so difficult, most of the 3,500 or so undergrads who take it get none right; the top 25 get five or so right). What was striking was where top-scoring girls come from. Of the 11 girls among the top 25 Putnam scorers in the past 16 years, eight were foreign-born.

The International Mathematical Olympiad, a nine-hour, six-problem exam, showed the same pattern. The highly ranked Bulgarian, East German/German and U.S.S.R./Russian teams have included 21, 19 and 15 girls, respectively, compared with three on U.S. teams. Over the years, 20 percent of the members of Russia's Olympiad teams were girls, as were one quarter of some other teams. U.S. teams had no girls for 23 straight years; Japan had one in 19 years. Since 1988, Bulgarian girls have won twice as many medals in the international Olympiad as American girls; Russian girls have won three times as many. Countries whose girls excel in the Olympiad have rigorous national math curricula and cultures that encourage girls as well as boys who excel in math. "Whether mathematical ability is identified depends on social, cultural and other environmental factors," says Janet Mertz of the University of Wisconsin, who led the study. Not only "identified," but "nurtured": all the top Putnam scorers received extracurricular training in problem solving. In addition to being mathematically gifted, these kids had to choose to spend free time learning problem-solving strategies. Countries whose girls lag behind boys tend to see math as for nerds only, which drives away many U.S. girls (who are more sensitive to social status than boys).

For anyone still grasping for biological explanations for the math gender gap, consider neighboring countries with a common gene pool, such as the former East Germany and West Germany, or Slovakia and the Czech Republic. The first of each pair regularly sent many more girls to the Olympiad—by margins of 5 to 0 and 3 to 1, respectively. (NOTE: The original version of this column incorrectly stated these numbers.) It's hard to see that as anything but the result of the starkly different social and other environmental forces in each country, not intrinsic biology. How powerfully do social forces affect brain function? In a 2007 study, girls reminded of the girls-are-spatially-challenged stereotype did worse on a test of spatial ability than those who were not, and brain imaging showed why: they had higher activity in the anterior cingulate, the site of negative emotions such as anger and sadness, and lower activity in high-order visual areas and complex working memory areas, found Maryjane Wraga of Smith College. Anxiety triggered by social forces had muted activity required for spatial reasoning. Scale that up to years of messages telling girls they're intrinsically inferior and then try to argue that a hard-wired brain rather than the messages society sends explains the math gender gap.

© 2008