Cerebellar circuits anticipate dopamine rewards

Area of Science:

• Neuroscience
• Computational Neuroscience
• Reward Prediction

Background:

• The cerebellum is known for motor control and reward prediction.
• Disentangling reward prediction from consummatory actions is crucial for understanding its role.

Purpose of the Study:

• To investigate the cerebellum's role in reward prediction independent of consummatory actions.
• To differentiate the neural signals for reward anticipation versus motor execution.

Main Methods:

• Mice were trained to perform a pushing task for delayed dopamine rewards.
• Two-photon imaging was used to record activity in cerebellar granule cells (GrCs) and climbing fibers (CFs).
• Generalization of neural activity between dopamine and water rewards was assessed.

Main Results:

• Cerebellar granule cells (GrCs) exhibited sustained predictive activity anticipating dopamine reward delivery.
• GrC activity temporally scaled to match reward intervals, linking actions to expected reward times.
• Climbing fibers (CFs) primarily fired just after reward delivery.
• Neural activity generalized across different reward types (dopamine vs. water).
• Mice learned to associate CF stimulation with reward, showing predictive GrC activity without consummatory actions.

Conclusions:

• Cerebellar circuits contribute to reward anticipation even when no physical action is required.
• This suggests a broader role for the cerebellum in brain reward prediction networks.
• The findings help to disentangle reward prediction from motor control functions of the cerebellum.