Nucleus accumbens dopamine release reflects Bayesian inference during instrumental learning.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Behavioral Neuroscience

Background:

• Dopamine release in the nucleus accumbens is theorized to signal reward prediction error, crucial for reinforcement learning.
• Understanding the brain's mapping of sensory information to reward predictions is vital, especially with ambiguous environmental cues.

Purpose of the Study:

• To investigate how probabilistic beliefs influence instrumental task performance and dopamine signaling.
• To test if dopamine release in the nucleus accumbens reflects Bayesian updating of beliefs about the environment's hidden state.

Main Methods:

• Mice performed a two-armed bandit task with a repeatedly switching hidden state.
• Choice behavior was recorded, alongside dopamine release in the nucleus accumbens core using dLight signals.
• Cognitive models were compared using behavioral data to identify the best-fitting belief computation strategy.

Main Results:

• Behavioral data strongly favored models employing Bayesian updating of probabilistic beliefs over non-Bayesian alternatives.
• The same Bayesian models provided a superior quantitative match to the recorded dopamine measurements.
• This indicates that probabilistic belief computation is integral to task performance and reflected in dopamine signaling.

Conclusions:

• Probabilistic belief computation plays a significant role in instrumental learning and decision-making in mice.
• Mesolimbic dopamine signaling in the nucleus accumbens appears to reflect these underlying probabilistic computations.
• The findings support a model where dopamine signals integrate belief updating for adaptive behavior.