Extensive research documents the existence of egocentric biases in the perception and application of justice norms. The origin of these biases remains poorly understood. We investigated both inter- and intra-individual differences in egocentric justice biases. Participants played an ultimatum game presumably with different anonymous players (simulated by a computer) in which they contributed differentially to the joint production of the initial endowment. We examined how contributions (low vs. high) affect proposers' offers and responders' acceptance decisions, as well as their fairness judgments and their emotional reactions to different types of offers (equal, equitable, unfair, and hyperfair). An egocentric bias in proposers' offers (indicating more flexible preferences) was found only in individualists and not in prosocials, suggesting differences in the motivations (or cognitions) underlying their choice of justice norms. Responders also showed egocentric biases in their judgments of fairness and in their emotional reactions to equal and equitable offers, but not in their acceptance decisions. Such dissociation might suggest that some form of emotion regulation occurred. Responders may evaluate offers on valence dimensions (e.g., goal conduciveness/outcome favorability and norm compatibility/justice) that are multiply interacting and potentially conflicting. The individual's acceptance/rejection decision reflects the relative weight attributed to competing appraisals. For this overt behavioral decision, the (personal) appraisal of outcome favorability that drives (analytical) acceptance of goal-conducive outcome seems to be stronger than the (social) appraisal of outcome fairness, which may trigger covert (emotional) rejection of offers that are incompatible with justice norms. Our data show that the emotional reaction patterns provide a more fine-grained readout of the overall evaluation of the proposer's action, the underlying emotional dynamics of which may, in real life, strongly determine future interactions with specific partners. Further research on the relationship between emotion and behavior in economic games is needed to explore potential dissociations and long-term effects.