Dopamine neurons encode trial-by-trial subjective reward value in an auction-like task

Area of Science:

• Neuroscience
• Decision Science
• Behavioral Economics

Background:

• The dopamine reward prediction error signal is recognized as subjective but has primarily been studied in aggregate choices.
• Individual choices fluctuate, offering a more direct measure of instantaneous subjective reward value.

Approach:

• Utilized the Becker-DeGroot-Marschak (BDM) mechanism, an auction-like method, to elicit accurate bids reflecting subjective reward value in experienced male rhesus monkeys.
• Recorded midbrain dopamine neuron activity during BDM bidding for juice rewards, where bids varied trial-by-trial despite constant reward amounts.
• Employed Support Vector Regression to assess the predictive power of dopamine neuron activity on animal bids.

Key Points:

• Dopamine neuron responses tracked trial-by-trial variations in monkey bids for identical rewards.
• Dopamine responses were comparable for similar bids made for different physical reward amounts.
• A small number of dopamine neurons (as few as twenty) could accurately predict animal bids.

Conclusions:

• The phasic dopamine reward signal dynamically encodes instantaneous subjective reward value.
• This finding provides a neural basis for understanding subjective valuation in decision-making.
• The study highlights the role of dopamine in reflecting real-time subjective preferences.