Intelligent Women Get Punished by Teammates — Men Don't. A Study Across 826 People Shows the Penalty Is Real but Small.

Across two samples in the US and China, women who scored higher on cognitive ability tests were rated more hostile and received less help. Men faced no such penalty — and in China, were treated better.

Imagine you're playing a cooperative board game and one player is clearly the strongest strategist at the table. If that player is a guy, the group leans in — asks for advice, follows the lead, shares resources. If that player is a woman, something subtler happens: the group starts withholding information, offering less help, and rating her as 'difficult.' Nobody decided to do this. The board state just shifted. That's the core mechanism in Nitya Chawla's study, published this year out of UCL School of Management. The committed claim: women's cognitive ability triggers a perception of hostility in teammates that men's cognitive ability does not, and this perceived hostility mediates real interpersonal costs — less help, more undermining. The study ran two samples: 432 US business students working in real consulting teams, and 394 full-time employees at a Chinese financial firm. Intelligence was measured via standardized reasoning tests used in hiring. Teammates never saw each other's scores. The theoretical scaffold is social role theory — the idea that people carry implicit expectations about gendered behavior (men as assertive and capable, women as warm and supportive), and that a woman displaying high cognitive ability violates the warmth expectation. Colleagues don't consciously decide she's hostile; they infer it from the mismatch between what they expect and what they observe. The study tested whether the women were simply behaving differently. Higher-scoring women were rated as somewhat more domineering, but dominance did not explain the reduced help or increased undermining. The pathway ran through perceived hostility, not actual behavior. The effects were small, and the study is honest about its limits. This is correlational, not causal — it shows a link between cognitive ability and teammate treatment, not proof that perception caused the behavior. Participants self-reported help and undermining, which introduces reporting bias. The US student sample's gender difference lost statistical significance when two extreme low-scorers were included, which is a fragility flag. Race and ethnicity were not examined. The China sample showed a stronger and more consistent pattern, with high-scoring men actually receiving better treatment — but the authors did not formally test whether cultural differences in gender norms drove the divergence. The ladder here matters. This is not the first study on gender bias in workplace competence — the backlash literature (Rudman 1998, Heilman 2004) has shown for decades that women who display agentic traits face social penalties. What's newer is the specific pathway: cognitive ability → perceived hostility → interpersonal mistreatment, measured in real teams over time rather than in vignette experiments. The ecological validity of using real team projects and real employees is a genuine step up from the typical 'read this profile and rate it' design. But the effect sizes are modest, and the mediation chain is long enough that each link adds noise. Chawla's practical recommendations are sensible but generic: managers should track interpersonal treatment, organizations should run check-ins, individuals should ask whether they'd react the same way to a male colleague. The research contribution is stronger than the applied contribution. What the study actually gives the field is a measured pathway from a specific individual difference (cognitive ability) through a specific perception (hostility) to specific outcomes (help withdrawal, undermining), with gender as the moderator — and it does this in two cultural contexts. The study does not answer the question most readers will want answered: what interventions actually reduce this penalty? That's the successor experiment. The finding is diagnostic, not therapeutic. It names the mechanism but doesn't test how to disrupt it. For now, the value is in the naming itself — women who've experienced this pattern now have data confirming they weren't imagining it.