Related Experiment Video
Updated: May 8, 2026
Correlating Behavioral Responses to fMRI Signals from Human Prefrontal Cortex: Examining Cognitive Processes Using Task Analysis
Published on: June 20, 2012
Reward expectation and prediction error in human medial frontal cortex: an EEG study
Massimo Silvetti1, Elena Nuñez Castellar, Clémence Roger
1Department of Experimental Psychology, Ghent University, Henri Dunantlaan 2, 9000, Ghent, Belgium; GIfMI (Ghent Institute for Functional and Metabolic Imaging), Ghent University Hospital, De Pintelaan 185B, 9000 Ghent, Belgium.
The human medial frontal cortex (MFC) tracks reward expectations before choices and updates them based on prediction errors after outcomes. This brain region rapidly adjusts reward anticipation using prediction error signals.
Area of Science:
- Neuroscience
- Cognitive Neuroscience
- Decision Science
Background:
- The medial frontal cortex (MFC) is crucial for reward-based decision-making in mammals.
- In primates, the MFC predicts upcoming rewards and calculates reward prediction errors (RPEs).
- Limited human data exists on the temporal dynamics of MFC activity during reward prediction and RPE computation.
Purpose of the Study:
- To investigate the time course of human medial frontal cortex activity during reward prediction and prediction error.
- To examine the relationship between reward expectation and prediction error signals in the human MFC.
Main Methods:
- Electroencephalography (EEG) was recorded during a reinforcement learning task in human participants.
- Analysis focused on neural activity during cue (reward prediction) and outcome (prediction error) periods.
Main Results:
- Human MFC activity during the cue period correlated with reward prediction.
- MFC activity during the outcome period reflected reward prediction error.
- Reward expectation showed differential correlations with prediction error in error versus correct trials, suggesting adaptive updating.
Conclusions:
- The human MFC processes reward predictions during the cue phase and prediction errors during the outcome phase.
- MFC activity demonstrates rapid updating of reward expectations based on prediction errors, mirroring findings in nonhuman primates.
- This study elucidates the temporal dynamics of MFC involvement in reinforcement learning and adaptive decision-making in humans.
