Dissociable cerebellar pathways for reach correction and learning.

Closed-loop perturbations to red nucleus–projecting cerebellar neurons during mouse reaching are compensated with corrective submovements and shift future submovement timing, unlike perturbations to IntA.

• Why it matters: Sensory feedback can correct movements quickly, but it is unclear how self-generated error corrections are computed and how they shape later motor behavior.
• What they did: In mice performing reaching, the team measured corrective submovements (CSMs) that arose spontaneously or after neural perturbations, using closed-loop perturbations targeting red nucleus–projecting neurons versus broader IntA.
• The result: Red nucleus–projecting perturbations were readily compensated and produced learned timing shifts for future submovements, whereas IntA perturbations were not compensated for and yielded no learning, implying distinct cerebellar output roles.

AI summary of the abstract · may contain errors · not medical advice · read the paper.