Skip to content Skip to menu

The evolution of cooperation in humans

Understanding the evolution of human hyper-sociality


No other species on Earth cooperates with non-relatives to the extent and over the range of contexts that humans do. Also, humans are everywhere – which makes them quite easy to find and experiment on. Our research asks how punishment and reputation-based partner choice stabilise cooperation. We also explore how culture shapes variation in social behaviour. Finally, a new strand of our research is exploring how mental disorders, such as psychosis, affect social cognition. We use online crowdsourcing websites, lab-based experiments, large online datasets and real-world methods to identify the mechanisms that promote cooperation – and the contexts and settings that undermine it. Our work has produced some of the following key findings:


  • Reputation-based cooperation: Men compete with one another when donating to online fundraising pages, especially when giving to attractive female fundraisers. This study, using > 11,000 fundraising pages set up for the 2014 London Marathon, provides the first empirical evidence of competitive helping in the real world. In other work, we explore when individuals might hide their helpful actions from others. Other work has shown that a fair reputation can be prized over other signals of quality, such as wealth, when selecting partners for social interactions.
  • Punishment and cooperation: Does punishment promote cooperation? These experiments show that, even when we incorporate power asymmetries between players, punishment has a detrimental effect on cooperation. Read our latest review on the topic here.
  • The reputation of punishers: When can harming someone make you look good? We find that hird-party punishers are rewarded  and preferentially chosen as partners on the basis of their actions and this article explores the reasons why.
  • Motives underpinning punishment: Why do we punish other people? This experimental work teases out the desire for revenge from the desire to level the playing field, and explores how these motives might vary across cultures.
  • Watching-eye effect on cooperation: Do images of eyes really promote cooperative behaviour? Our large, online study casts doubt on this finding, showing that people were more generous when they saw images of flowers rather than images of eyes.