Neural interactions in the human frontal cortex dissociate reward and punishment learning.

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Decision Neuroscience

Background:

• The neural mechanisms underlying how the human brain processes rewards and punishments remain incompletely understood.
• Investigating the interplay between prefrontal and insular cortices is crucial for deciphering reward and punishment learning.

Purpose of the Study:

• To elucidate the interaction patterns between human prefrontal and insular cortices during reward maximization and punishment minimization.
• To determine whether functional specificity arises from local neural representations or interareal interactions.

Main Methods:

• Utilized intracranial recordings in humans to capture neural activity in prefrontal and insular regions.
• Analyzed neural population responses and interareal interaction patterns during reward and punishment learning tasks.

Main Results:

• Prefrontal and insular cortices exhibit non-selective neural populations for both rewards and punishments.
• Context-specific interareal interactions were identified, forming distinct reward (orbitofrontal and ventromedial prefrontal cortices) and punishment (insula and dorsolateral cortex) subsystems.
• The ventromedial prefrontal cortex and insula play driving roles in their respective subsystems.
• Synergistic interactions between these subsystems mediate switching between reward and punishment learning.

Conclusions:

• Functional specificity in reward and punishment learning is better explained by interareal interactions than local neural representations.
• A unified model of distributed cortical interactions supports adaptive learning from both positive and negative experiences.