Imaging the changing role of feedback during learning in decision-making

Area of Science:

• Neuroscience
• Cognitive Psychology
• Decision Making

Background:

• Learning from decision outcomes influences future choices and reactions.
• Neural responses to feedback are crucial for adapting behavior.

Purpose of the Study:

• Investigate neural responses to choice outcome feedback before and after learning a strategy.
• Examine how learning affects brain activation patterns related to feedback processing.

Main Methods:

• Functional magnetic resonance imaging (fMRI) in healthy subjects.
• A task with unavoidable losses, requiring learning a specific response strategy.
• Analysis of neural activation during feedback processing across different learning phases.

Main Results:

• Neural activation during feedback processing was higher in the exploratory learning phase than in the proficiency phase.
• Reduced activation was observed with learning, both within subjects and between subjects with different learning speeds.
• Brain regions involved in working memory, uncertainty, and stimulus-outcome contingencies showed decreased activation post-learning.
• Activation patterns for gains changed more with learning than for losses, suggesting differential sensitivity to reward predictability.

Conclusions:

• Controlled and automatic processing are characterized by differing neural activation levels in task-relevant areas.
• Positive prediction errors are more sensitive to increased reward predictability than negative prediction errors.
• Learning modulates neural responses to feedback, with significant changes in brain networks involved in cognitive control and learning.