Both biosociological and psychological models, as well as animal research, suggest that testosterone has a key role in social interactions. Evidence from animal studies in rodents shows that testosterone causes aggressive behaviour towards conspecifics. Folk wisdom generalizes and adapts these findings to humans, suggesting that testosterone induces antisocial, egoistic, or even aggressive human behaviours. However, many researchers have questioned this folk hypothesis, arguing that testosterone is primarily involved in status-related behaviours in challenging social interactions, but causal evidence that discriminates between these views is sparse. Here we show that the sublingual administration of a single dose of testosterone in women causes a substantial increase in fair bargaining behaviour, thereby reducing bargaining conflicts and increasing the efficiency of social interactions. However, subjects who believed that they received testosterone-regardless of whether they actually received it or not-behaved much more unfairly than those who believed that they were treated with placebo. Thus, the folk hypothesis seems to generate a strong negative association between subjects' beliefs and the fairness of their offers, even though testosterone administration actually causes a substantial increase in the frequency of fair bargaining offers in our experiment.