This study explores the effects of feedback on cooperation in the iterated prisoner’s dilemma game (PDG). Four sources of feedback were identified: peer, buyer, market and cultural feedback. Peer and buyer feedback were intrinsic to the PDG, for they were analyzed, but not manipulated. Market and cultural feedback comprised independent variables and their effects were measured on players’ and group cooperation (dependent variables). Twenty-seven participants played a PDG, divided in 9 groups of 3 players each. Cooperation was measured as rates of individual players’ cooperative X choices, and as aggregate products within groups. At the molecular (moment-to-moment) level, there was a significant within-subjects main effect of the market feedback F(1, 28) = 6.50, p = .02, ?p2 = .19. At the molar level, there was no significant effect of the market feedback, nor of the cultural feedback. It was not possible to establish a metacontingency between recurrent group cooperation and positive contingent group consequences. Players displayed sub-optimal choice behavior, seeking to maximize relative earnings within their group (defecting) over absolute earnings (cooperating). These results are discussed in light of how the source of feedback may sustain cooperation or defection in the PDG, and their implications in organizational settings. Reinforcing cooperative behaviors can be key to the maintenance and development of any organization, for informative performance feedback may not suffice. This study contributes to the understanding of economic decisional behavior in groups from a cultural selectionist perspective.Keywords: choice, cooperation, feedback, metacontingency, prisoner’s dilemma game