Dopaminergic computations for perceptual decisions.

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Neuroscience

Background:

• Dopamine signals are extensively studied in value-based decision-making.
• Recent research reveals dopamine's role in perceptual decision-making, influenced by feedback and perceptual confidence.
• Understanding dopamine's function in perception is key to understanding learning and decision-making.

Purpose of the Study:

• To review recent studies on dopamine signals in perceptual decision tasks.
• To explore how reinforcement learning models can explain dopamine's role in perceptual learning.
• To bridge the gap between dopamine research in value-based and perceptual decision-making.

Main Methods:

• Review of existing literature on dopamine signaling in perceptual decision-making across species.
• Analysis of reinforcement learning models applied to perceptual decision tasks.
• Discussion of dopaminergic circuits and their involvement in learning.

Main Results:

• Dopamine signaling is pervasive in perceptual decision-making, depending on feedback and sensory statistics.
• Reinforcement learning models successfully account for learning dynamics and dopaminergic involvement in perceptual tasks.
• Heterogeneous dopamine responses can be explained within a unified framework.

Conclusions:

• Reinforcement learning offers a promising framework for understanding dopamine's role in perceptual decision-making and learning.
• This framework helps unify findings across different decision-making paradigms and dopamine response patterns.
• Future research can leverage RL to elucidate dopamine's function from novice to expert performance.