Tonic dopamine and biases in value learning linked through a biologically inspired reinforcement learning model

Area of Science:

• Neuroscience
• Computational Psychiatry
• Reinforcement Learning

Background:

• Biased future predictions are a key feature of psychiatric disorders.
• Understanding the neural mechanisms of value learning is crucial for developing effective treatments.

Purpose of the Study:

• To investigate the mechanisms underlying biased value learning.
• To model how synaptic plasticity and basal ganglia circuits contribute to prediction biases.

Main Methods:

• Utilized reinforcement learning models.
• Incorporated recent findings on synaptic plasticity.
• Examined opponent circuit mechanisms in the basal ganglia.

Main Results:

• Variations in tonic dopamine shift the balance of learning from positive and negative reward prediction errors.
• Dopamine receptor (D1 and D2) dose-occupancy curves and affinities explain biased value predictions.
• The model successfully explains biased value learning in both mice and humans.

Conclusions:

• Tonic dopamine levels significantly modulate learning processes.
• The proposed mechanism provides a foundation for understanding basal ganglia function.
• This research offers insights into the neurobiological underpinnings of psychiatric disorders.