Adaptive Asymmetric Adjustment in Affective Evaluation and Reciprocity for Altruistic Behaviors from Exogenous Uncertainty to Certainty

Direct reciprocity requires the beneficiary’s real-time evaluation of others’ altruistic behaviors under exogenous uncertainty, i.e., environmentally imposed uncertainty that can be resolved upon the disclosure of outcomes (e.g., uncertainty in others’ cost to rescue oneself in a natural disaster). However, the neurocognitive mechanisms underpinning this dynamic adjustment to exogenous uncertainty fluctuations remain unexplored. Combining interpersonal tasks simulating exogenous uncertainty-to-certainty transitions with one fMRI experiment applying multivariate pattern analyses and three behavioral experiments, we uncover an adaptive asymmetric adjustment in the beneficiary’s affective evaluation and reciprocity in response to altruistic behaviors: the beneficiary’s gratitude and ensuing reciprocity intensify when the final benefactor-cost (or self-benefit) exceeds the expectation under exogenous uncertainty; however, a parallel reduction in benefactor-cost (or self-benefit) does not elicit equivalent decreases. This asymmetric adjustment, perceived as morally superior by third parties, challenges classical theories of gratitude, decision-making, and social learning involving uncertainty. We resolve this paradox by proposing a mechanism of prosocial information integration: guided by the adaptive goal (e.g., gathering social acceptance), the beneficiary tends to asymmetrically weigh prior information (e.g., the cost the benefactor willing to undertake under uncertainty) against posterior information (e.g., final benefactor’s cost) when evaluating benefactor’s intention, contributing to the observed adaptive asymmetry in gratitude and reciprocity dynamics. This process is supported by neural representations within the theory-of-mind system, particularly the dorsomedial prefrontal cortex. By reframing direct reciprocity as a dynamic process shaped by prosocial information integration, this work extends the theoretical framework of cooperation under uncertainty, offering new insights into human social adaptation.