In their article, Wood and colleagues speculate about the evolutionary basis for this sex difference. Perhaps, since women are the primary child-rearers, their brains have become programmed to develop a larger SG, to prepare them to be sensitive nurturers. Prenatal sex hormones are known to alter behavior and certain brain structures in other mammals. Perhaps such hormones—or sex-specific genes—may enhance the development of females’ SG (or dampen the development of males’) leading to inborn differences in social cognition.

The best way to test this hypothesis is to look at children. If the sex difference in the SG is present early in life, this strengthens the idea that it is innately programmed. Wood and Nopoulos therefore conducted a second study with colleague Vesna Murko, in which they measured the same frontal lobe areas in children between 7 and 17 years of age.

But here the results were most unexpected: they found that the SG is actually larger in boys ! What’s more, the same test of interpersonal awareness showed that skill in this area correlated with smaller SG, not larger, as in adults. The authors acknowledge that their findings are “complex,” and argue that the reversal between childhood and adulthood reflects the later maturation of boys’ brains, compared to girls. (Adolescents’ brains undergo a substantial “pruning” or reduction in gray matter volume during adolescence, which happens about two years earlier in girls, compared to boys.)

However, in both studies, Wood and colleagues added another test that reminds us to be cautious when interpreting any finding about sex differences in the brain. Instead of simply dividing their subjects by biological sex, they also gave each subject a test of psychological “gender:” a questionnaire that assesses each person’s degree of masculinity vs. femininity—regardless of their biological sex—based on their interests, abilities and personality type. And in both adults and children, this measure of “gender” also correlated with SG size, albeit in just as complicated a way as the correlation between “sex” and SG size. (Larger SG correlated with more feminine personality in adults but less feminine personality in children.)

In other words, there does seem to be a relationship between SG size and social perception, but it is not a simple male-female difference. Rather, the SG appears to reflect a person’s “femininity” better than one’s biological sex: women who are relatively less feminine show a correspondingly smaller SG compared to women who are more feminine, and ditto for men.