Human claustrum neurons encode uncertainty and prediction errors during aversive learning

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Computational Neuroscience

Background:

• Flexible behavior requires continuous updating of internal models.
• Neural circuits for this process are poorly understood.
• The claustrum's extensive cortical connections suggest a key role.

Purpose of the Study:

• Investigate neural responses in the human claustrum during aversive learning.
• Compare claustrum activity with anterior cingulate cortex (ACC) and amygdala.
• Determine if claustrum neurons encode higher-order cognitive variables.

Main Methods:

• Single-neuron recordings from human claustrum, ACC, and amygdala.
• Aversive learning task design.
• Analysis of neural responses related to stimulus onset, outcomes, and latent variables (uncertainty, prediction error).

Main Results:

• Claustrum and ACC neurons showed structured, task-related responses.
• Distinct subpopulations encoded stimulus onset and action outcomes.
• Both claustrum and ACC encoded uncertainty and prediction error, but with different temporal dynamics.
• Amygdala showed minimal latent-variable modulation.

Conclusions:

• Human claustrum neurons track unobservable higher-order cognitive variables.
• Claustrum and ACC have dissociable roles in tracking latent task states.
• These findings advance understanding of neural circuits supporting flexible behavior.