The Reward Positivity: A Goal Prediction Error

The Reward Positivity (RewP) is an electroencephalographic feature that emerges following reward receipt and is sensitive to positive prediction errors and other experiences of hedonic value. Here we address two unresolved issues regarding the RewP. First, we clarify the biophysical signature of the RewP. Contrary to the dominant perspective, we argue that the RewP is an independent EEG feature that selectively responds to positive prediction errors and is superimposed over the obligatory evoked background. We distinguish this additive characteristic of the RewP from the modulation of the obligatory N2 component, which is influenced by negative prediction errors and other signals related to control demands. Second, we propose a refined theoretical account of the computational process reflected by the RewP that highlights its long-standing sensitivity to goal achievement, namely, that the RewP reflects a belief update of value triggered by a discrete indication of goal attainment. This process resembles a value-sensitive but action-agonistic critic from an actor/critic dichotomy. This conceptual shift – from viewing the RewP as a generic signal of reward receipt to a more nuanced signal of goal attainment -- emphasizes its dependence on control processes. Taken together, we propose that the critic-like RewP and the actor-like N2 may function as opponent processes that support the updating of goal-directed behavior. This framework opens several avenues for future empirical investigation aimed at further advancing our understanding of this neural signal.