Impaired reward prediction error encoding and striatal-midbrain connectivity in depression

Area of Science:

• Neuroscience
• Psychiatry
• Computational Psychiatry

Background:

• Anhedonia and negative bias are key features of major depressive disorder (MDD).
• These symptoms may arise from aberrant reinforcement learning processes.
• Previous research suggests blunted reward prediction error (RPE) signaling in the striatum in MDD, but findings are inconsistent.

Purpose of the Study:

• To investigate neural correlates of reward and punishment learning in unmedicated MDD.
• To examine functional connectivity between the ventral tegmental area (VTA), striatum, and habenula during learning.
• To explore the relationship between learning deficits and illness severity in MDD.

Main Methods:

• fMRI data acquired from 25 unmedicated MDD patients and 26 healthy controls during a monetary instrumental learning task.
• Computational modeling used to quantify reward prediction error (RPE) and punishment prediction error (PPE) signals.
• Analysis of striatal activation, habenular activation, and functional connectivity between VTA-striatum and VTA-habenula.

Main Results:

• MDD patients exhibited impaired reward learning and blunted striatal RPE signals compared to controls.
• Striatal RPE signal blunting worsened with increased number of major depressive episodes (MDEs).
• Reduced VTA-striatal connectivity to feedback was observed in MDD; no group differences in PPE signaling or VTA-habenula connectivity were found.

Conclusions:

• MDD is characterized by impaired reward learning and disrupted RPE signaling in the striatum.
• Reduced VTA-striatal connectivity contributes to reward processing deficits in MDD.
• Findings underscore the role of reward-related learning deficits and their neural underpinnings in MDD pathophysiology.