Dynamics of Human Cooperation in Economic Games

Human decision behaviour is quite diverse. In many games humans on average do not achieve maximal payoff and the behaviour of individual players remains inhomogeneous even after playing many rounds. For instance, in repeated prisoner dilemma games humans do not always optimize their mean reward and frequently exhibit broad distributions of cooperativity. The reasons for these failures of maximization are not known. Here we show that the dynamics resulting from the tendency to shift choice probabilities towards previously rewarding choices in closed loop interaction with the strategy of the opponent can not only explain systematic deviations from 'rationality', but also reproduce the diversity of choice behaviours. As a representative example we investigate the dynamics of choice probabilities in prisoner dilemma games with opponents using strategies with different degrees of extortion and generosity. We find that already a simple model for human learning can account for a surprisingly wide range of human decision behaviours. It reproduces suppression of cooperation against extortionists and increasing cooperation when playing with generous opponents, explains the broad distributions of individual choices in ensembles of players, and predicts the evolution of individual subjects' cooperation rates over the course of the games. We conclude that important aspects of human decision behaviours are rooted in elementary learning mechanisms realised in the brain.