Sensorimotor beta power reflects the precision-weighting afforded to sensory prediction errors

Area of Science:

• Neuroscience
• Motor Control
• Computational Neuroscience

Background:

• Accurate motor control is theorized to involve Bayesian inference, integrating sensory data with prior knowledge.
• Beta oscillations (12-30 Hz) in sensorimotor cortex, measured by electroencephalography (EEG), are implicated in Bayesian inference parameters.
• Previous research linked post-movement beta synchronisation to prediction error, with newer studies suggesting a role in representing uncertainty.

Purpose of the Study:

• To investigate the neurophysiological basis of uncertainty in Bayesian updating during visuomotor adaptation.
• To directly modulate sensory uncertainty and assess its impact on sensorimotor beta power.
• To correlate behavioral learning parameters derived from the Hierarchical Gaussian Filter (HGF) with electroencephalography (EEG) data.

Main Methods:

• Visuomotor adaptation task in healthy human participants.
• Direct modulation of sensory uncertainty.
• Electroencephalography (EEG) recordings over sensorimotor cortices.
• Behavioral data modeling using the Hierarchical Gaussian Filter (HGF).

Main Results:

• Sensorimotor beta power demonstrated a significant correlation with inverse uncertainty associated with sensory prediction errors.
• This correlation was observed both before and after movement execution.
• The findings suggest beta power reflects relative uncertainty rather than prediction error itself.

Conclusions:

• Sensorimotor beta oscillations appear to encode latent uncertainty parameters within the sensorimotor system.
• This provides a potential neurophysiological mechanism for how the brain represents and utilizes uncertainty in motor control.
• Understanding this neural signature is vital for comprehending adaptive behavior and motor learning.