Identity prediction errors in the human midbrain update reward-identity expectations in the orbitofrontal cortex

James D Howard,Thorsten Kahnt

doi:10.1038/s41467-018-04055-5

Abstract

There is general consensus that dopaminergic midbrain neurons signal reward prediction errors, computed as the difference between expected and received reward value. However, recent work in rodents shows that these neurons also respond to errors related to inferred value and sensory features, indicating an expanded role for dopamine beyond learning cached values. Here we utilize a transreinforcer reversal learning task and functional magnetic resonance imaging (fMRI) to test whether prediction error signals in the human midbrain are evoked when the expected identity of an appetitive food odor reward is violated, while leaving value matched. We found that midbrain fMRI responses to identity and value errors are correlated, suggesting a common neural origin for these error signals. Moreover, changes in reward-identity expectations, encoded in the orbitofrontal cortex (OFC), are directly related to midbrain activity, demonstrating that identity-based error signals in the midbrain support the formation of outcome identity expectations in OFC.

Highlights

There is general consensus that dopaminergic midbrain neurons signal reward prediction errors, computed as the difference between expected and received reward value
We demonstrate that the human midbrain responds to value-neutral errors in identity predictions, and that these signals are correlated with changes in identity expectations in the orbitofrontal cortex (OFC)
One conditioned stimuli (CS) was paired with a high-intensity unconditioned stimuli (US), and the other CS was paired with the low-intensity US of the same odor identity

Summary

Introduction

There is general consensus that dopaminergic midbrain neurons signal reward prediction errors, computed as the difference between expected and received reward value. Dopamine neurons in rats have recently been shown to respond in situations where value remains the same but the sensory features of the predicted reward change[17] This suggests a much broader role for midbrain error signals beyond cached value to include the learning of a wide range of associations that together form a model of the task state. Throughout the task, participants experienced violations either in expected reward identity while leaving value unaltered, or in expected reward value while leaving identity unchanged Using this approach, we demonstrate that the human midbrain responds to value-neutral errors in identity predictions, and that these signals are correlated with changes in identity expectations in the OFC

Methods

Results

Conclusion