Individual Differences in Motor Noise and Adaptation Rate Are Optimally Related

Area of Science:

• Neuroscience
• Motor Control
• Computational Neuroscience

Background:

• Individual differences in motor adaptation correlate with movement variability (motor noise).
• Previous findings on motor noise and adaptation rate lack consistent replication.
• Motor noise may comprise distinct planning (brain) and execution (periphery) components.

Purpose of the Study:

• To investigate the differential impact of planning noise and execution noise on motor adaptation rate.
• To test the hypothesis that motor adaptation is tuned to approximate optimal learning, akin to Kalman filters.
• To apply a novel Bayesian fitting procedure to individual adaptation data using a state-space model.

Main Methods:

• Visuomotor adaptation experiment with 69 participants.
• Novel Bayesian fitting procedure to apply the state-space model to individual data.
• Quantification of planning noise and execution noise from individual adaptation parameters.

Main Results:

• Adaptation rate positively correlated with planning noise (β = 0.44).
• Adaptation rate negatively correlated with execution noise (β = -0.39).
• Steady-state Kalman gain positively correlated with adaptation rate (r = 0.54).

Conclusions:

• Motor adaptation appears tuned for optimal learning, supporting the optimal control framework.
• The cerebellum, implicated in motor adaptation, shows sensitivity to both planning and execution noise.
• Findings reconcile previous inconsistencies by differentiating noise components in motor adaptation studies.