Social surplus determines cooperation rates in the one-shot Prisoner's Dilemma

We provide evidence on how cooperation rates vary across payoff parameters in the Prisoner's Dilemma (PD), using four one-shot games that differ only in the payoffs from mutual cooperation. In our experiment, participants play only the PD game, and play the game once and only once, so there are no potential confounds or methodological issues. Our results show that higher monetary payoffs from cooperation are associated with substantially higher cooperation rates, which increase monotonically from 23% to 60%. Participants' beliefs about cooperation rates track closely actual cooperation rates: higher cooperation is expected from others when mutual cooperation payoffs are higher. This is true also for participants who, in a follow-up experiment, only make guesses about the choices of others.