Hackneyed gender tropes tell usthat men are hyper-competitive, egoistic warriors who will fight to the death. Women, on the other hand, are deeply concerned about interpersonal relationships, so they’re more likely to work together as a unit than get in fist-fights and pissing matches.

But how does intra-gender politics actually work in the real-world, beyond the lazy stereotypes? Do women really collaborate together more than men?

According toa new studypublished inCurrent BiologybyJoyce Benenson, a psychology professor at Emmanuel College and an associate at Harvard University’s Department of Human Evolutionary Biology, and two of her colleagues, the world of academia doesn’t conform to these clichés. Female and male full professors at 50 university psychology departments across North America actually “were equally likely” to collaborate with same-gender, equal-rank colleagues on research papers.

But when hierarchy is folded into the equation, behavior shifts radically.

The researchers used “numbers of co-authored peer-reviewed publications” between 2008 to 2012 to measure the robustness of senior faculty members’ cooperation with their younger colleagues. They calculated the expected number of publications among full professors and their same-gender, junior colleagues, based on chance and the position and gender compositions of the departments. The team discovered that female full professors came in well below the random mark for co-authorship with their younger, same-gender colleagues, while theirmale counterparts exceeded expectations:

There were significantly fewer publications co-authored by one senior female with one junior female than by one senior male with one junior male than would be expected. … In contrast, analysis of co-authored publications between senior and junior co-authors of the other gender yielded no difference…. These results show that high-ranked male professors co-published morethan high-ranked female professors with same-gender low-ranked faculty.

Our results are consistent withobservations suggesting that socialstructure takes differing forms forhuman males and females. Males’tendency to interact in same-gendergroups make them more prone tocooperation with asymmetrically rankedmales. In contrast, females’ tendency torestrict their same-gender interactionsto equally ranked individuals makethem more reluctant to cooperate withasymmetrically ranked females.

The female preference for cooperation with equals has also been observed in other studies of chimps and human infants, adolescents, and adults. Benenson suspects there’s an evolutionary basis for the behavior. “Males benefit from cooperating with groups to defeat other groups. Females invest more in kin and not in unrelated individuals, except a best friend,” she explained in an email.

This lack of cross-rank cooperation seems particularly discouraging for young female professors, who already have to compete in a system full of institutionally- and societally-imposed handicaps. Recognizing and reversing this trend may be one of many ways to work against the gender imbalanceamong tenured professors.

Given the findings, I asked Benenson whether she wished she’d collaborated on her paper with a junior female faculty member rather than two male colleagues. She replied: “It is not so easy to do!”