Reward-driven cerebellar climbing fiber activity influences both neural and behavioral learning

Area of Science:

• Neuroscience
• Computational Neuroscience
• Motor Control

Background:

• The cerebellum is vital for motor coordination and learning.
• Climbing fibers (CFs) may signal reward prediction, suggesting reinforcement learning principles.
• Understanding CFs' role in reward-guided behavior is essential.

Purpose of the Study:

• To investigate if CFs operate via reinforcement learning principles.
• To evaluate the role of reward-related CF activity in learning.
• To determine if CF activity meets reward prediction error (rPE) signal requirements.

Main Methods:

• Two-photon calcium imaging of CFs in Purkinje cells during a Pavlovian task.
• Multi-stimulus experiments to assess rPE signal transfer.
• Optogenetic inhibition of inferior olive neurons using GtACR2.

Main Results:

• CF activity transfer to a conditioned stimulus was impaired when reward was no longer directly experienced.
• Optogenetic inhibition of CFs during reward delivery hindered signal transfer and learning.
• Inhibition led to deficits in anticipatory lick timing, indicating impaired motor learning.

Conclusions:

• CF signals exhibit characteristics of rPEs observed in reinforcement learning.
• The cerebellum utilizes these reward-related CF signals for accurate, timed motor behavior.
• CFs are integral to reward-based motor learning and adaptation.