(iStock)

Everyone knows that certain jobs are more likely to be held by people of certain genders — women are more likely to be teachers, for example, and men are more likely to be computer engineers. These trends contribute to a sizable gender salary gap, but they also force a more difficult question: How do these roles form in the first place?

Over the past several decades, studies have slowly been chipping away at the theory that the two sexes behave differently primarily due to innate biological differences. Those assumptions go back more than a century, but a lot of them are informed by simple correlation-vs.-causation errors. In reality, gender-related expectations probably play a much larger role in gender difference than a lot of people want to admit.

Conventional wisdom tells us that male and female brains are wired in such a way that each sex has advantages over the other when it comes to certain tasks. Men, for example, are often said to be better at math and scientific reasoning while women are better at reading and empathizing. These aren’t just blind assertions; they are claims based on hard employment data.

But new research continues to break down those assumptions. In fact, the internalization of gender sorting may begin as early as first grade, and it could change outcomes. A study published Thursday in the journal “Science” found that 6-year-old girls are less likely than boys to believe that other members of their gender are “brilliant.” This also seems to be the age when they become more likely to shy away from activities described as for children who are “really, really smart,” according to the research.

This might partially explain why other research has found that women are underrepresented in fields where there is a belief that raw, innate “brilliance” is important for success. Another analysis of more than 14 million reviews on RateMyProfessors.com found that the frequency with which college students described professors as “brilliant” or “genius” highly predicted the representation of women in academic fields.

Research has also shown that this sort of thinking is rooted in the home: One famous study, based on an analysis of Google searches, found that parents were more than twice as likely to look up whether their sons were “gifted” or “stupid” compared to their daughters. And what were they more likely to look up regarding their daughters? Whether they were “overweight,” “beautiful” or “ugly.”

Of course, it isn’t just the general population perpetuating the idea that men and women are innately different. In 2005, Lawrence Summers — then president of Harvard University — provoked outrage when he claimed that there were more male geniuses in society than women because men are subject to more variability in terms of IQ. In other words, while men had more people at the tail ends of the IQ bell curve, women were more likely to have average IQ scores.

His theory has been reinforced by data such as standardized math scores — but that doesn’t necessarily mean the phenomenon is biological. If that were true, we would expect to find similar male variability throughout the world and across time. But research shows that this doesn’t seem to be the case: One cross-cultural study examining boys and girls in 40 countries shows that gender gaps in math scores disappear in countries that score better in terms of gender equality (based on measures such as reproductive health and workforce participation). Another paper found that gender gaps among the most mathematically talented in the United States have narrowed considerably since the 1980s.

This also goes beyond intelligence. More and more studies are reinforcing the so-called gender similarities hypothesis — that for the most part, the brains of men and women function with little difference. There are a few crucial exceptions, such as motor performance (like throwing objects) and some aspects of sexuality, but it’s becoming increasingly clear that men and women behave far more similarly than what we’re typically told. Recently discovered is the fact that parts of the brain previously thought to differ in size between men and women actually are not related to sex the way we thought. Even the notion that men are biologically more aggressive than women is ambiguous at best — one analysis of more than 200 studies on domestic violence found that women are just as likely to initiate violence as men.

At the same time, though, we’re inundated with media portraying men and women as vastly different specimens. They think differently. They act differently. They have different interests and desires. This narrative comes with real social costs: Discouraging people with real talent from entering fields in which they might excel.

Read more:

The gender gap in computer science is hurting U.S. businesses

Mammograms deserve more skepticism

Why women in science are lonely — and shouldn’t be