In reciprocal altruism, unrelated individuals engage in behaviors that benefit others at some cost to themselves. While Trivers’theory of reciprocal altruism explains how natural selection can favor cooperation (or reciprocity) even among unrelatedindividuals, the mechanisms underlying this behavior remain elusive. Studies using the Iterated Prisoner’s Dilemma (iPD) havedemonstrated reciprocal altruism in monkeys and recently, with positive and negative reinforcement along with a well-definedreward contrast, in rats, which exhibit high levels of mutual cooperation. Here, we ask whether this learning can be explainedby long-term reward maximization, without any social interaction. We replaced the live opponent rat with a simulated oneincapable of errors. Interestingly, rats still exhibited high levels of mutual cooperation. This suggests that social interaction,while potentially important, may not be strictly necessary for rats to learn reciprocal altruism within the iPD framework.