Reward processing dominates the brain during feedback evaluation: Electrophysiological evidence.

Abstract

During reinforcement learning, people learn based on both positive and negative feedback. Researchers have revealed the reward positivity (RewP) that reflects positive feedback evaluation, however, no electrophysiological indicator has been found to explain negative feedback processing. In reinforcement learning, people would like to expect the positive feedback after a choice, which might explain why previous studies usually found neural responses for reward processing, while the negative feedback processing seems absent. However, no study has designed a task to separate positive and negative feedback processing to measure the corresponding electrophysiological responses. Thus, the present study designed gain processing-advantaged (GA) and loss processing-advantaged (LA) contexts using a modified gambling task to measure specific neural responses to gain processing and loss processing. Electroencephalography (EEG) data were processed using time-domain and time-frequency analyses. The results revealed the RewP and delta oscillation following gain feedback in the GA context, while no negativity or other components specific to loss processing was found, even in the loss condition of LA context. The current results indicate that reward processing dominates the brain during feedback evaluation, whereas loss processing can't be captured by electrophysiological signal.