Research suggests that papers by women tend to get cited less than those by men; however, the factors driving this gendered citation behaviour are not clear. Here, the authors used a probabilistic algorithm to assign genders to the first and last authors of 54,225 papers from five neuroscience journals since 1995 and found that papers with male first and last authors (man/man written papers) were cited 11.6% more than would be expected if randomly drawn references were cited, whereas woman/woman papers were cited 30.2% less than expected. The authors found that this imbalanced citation behaviour seems to be gradually increasing over time, and that it is mostly driven by the citation practice of man/man author teams.