Abstract
Humans typically display inequality aversion in social situations, which manifests itself as a preference for fairer distributions of resources. However, people differ in the degree to which they dislike being worse off [disadvantageous inequality (DI) aversion] or better off [advantageous inequality (AI) aversion] than others. Competing models explain such behavior by focusing on aversion to payoff differences, maximization of total payoff or reciprocity. Using functional near-infrared spectroscopy, we asked which of these theories could better explain dorsolateral prefrontal cortex (dlPFC) activity while participants accepted or punished fair vs unfair monetary transfers in an anonymous norm compliance task. We found that while all participants exhibited DI aversion, there were substantial differences in preferences for AI, which were strongly predicted by dlPFC activation. Model comparisons revealed that both punishment behavior and prefrontal activity were best explained by a model that allowed for AI seeking rather than imposing aversion. Moreover, enhancing this model by taking into account behavioral response times, as a proxy for choice difficulty, further improved model fits. Our data provide evidence that the dlPFC encodes subjective values of payoff inequality and that this representation is richer than envisaged by standard models of social preferences.