The authors explore the interfaces between different dimensions of moral cognition, bridging economic, Bayesian and reinforcement learning perspectives. The human aversion to harming others cuts across these different interfaces, influencing decisions, judgments, and inferences about morality.