Spatiotemporal characterization of the neural correlates of outcome valence and surprise during reward learning in humans

Abstract

Reward learning depends on accurate reward associations with potential choices. Two separate outcome dimensions, namely the valence (positive or negative) and surprise (the absolute degree of deviation from expectations) of an outcome are thought to subserve adaptive decision-making and learning, however their neural correlates and relative contribution to learning remain debated. Here, we coupled single-trial analyses of electroencephalography with simultaneously acquired fMRI, while participants performed a probabilistic reversal-learning task, to offer evidence of temporally overlapping but largely distinct spatial representations of outcome valence and surprise in the human brain. Electrophysiological variability in outcome valence correlated with activity in regions of the human reward network promoting approach or avoidance learning. Variability in outcome surprise correlated primarily with activity in regions of the human attentional network controlling the speed of learning. Crucially, despite the largely separate spatial extend of these representations we also found a linear superposition of the two outcome dimensions in a smaller network encompassing visuo-mnemonic and reward areas. This spatiotemporal overlap was uniquely exposed by our EEG-informed fMRI approach. Activity in this network was further predictive of stimulus value updating indicating a comparable contribution of both signals to reward learning.