The restart effect in social dilemmas shows humans are self-interested not altruistic

Do economic games show evidence of altruistic or self-interested motivations in humans? A huge body of empirical work has found contrasting results. While participants routinely make costly decisions that help strangers, consistent with an evolutionary novel form of altruism, participants also typically learn to pay fewer costs with experience, consistent with self-interested individuals adapting to an unfamiliar environment. Key to resolving this debate is explaining the famous ‘restart effect’, a puzzling enigma whereby failing cooperation in experiments can be briefly rescued by a surprise restart. Here we show that this canonical result, which is often assumed to be evidence of altruism, can be entirely removed, replaced, or even reversed depending on experimental design. Specifically, the restart effect (1) disappears when reputational benefits to cooperation are fully removed, consistent with strategically motivated, self-interested, cooperation; (2) can be replaced by an irrational restart that benefits no-one if individuals are grouped with computers, consistent with confusion; and (3) can even be reversed, so that contributions gradually increase rather than decrease towards the end of the game, if the contributions of the computerized groupmates are programmed accordingly. These results show that the restart effect is driven by a mixture of self-interested and irrational beliefs about the game’s payoffs and not altruism. Consequently, our results suggest that economic games have often been measuring self-interested but confused behaviours and reject the idea that conventional theories of evolution cannot explain the results of economic games.