Perceptual error based on Bayesian cue combination drives implicit motor adaptation.

Area of Science:

• Neuroscience
• Motor Control
• Perception

Background:

• The sensorimotor system recalibrates implicitly, a procedural learning type with an undefined computational mechanism.
• Existing theories based on sensory errors struggle to explain implicit motor adaptation phenomena like over-learning and fast saturation.

Purpose of the Study:

• To propose and validate a new theory where perceptual error, not sensory error, drives implicit motor adaptation.
• To elucidate the computational mechanism underlying sensorimotor recalibration without conscious awareness.

Main Methods:

• Validated increasing sensory uncertainty of visual cues with perturbations via perceptual psychophysics (Experiment 1).
• Predicted and tested learning dynamics of implicit adaptation across perturbation sizes (Experiment 2).
• Examined proprioception changes relative to visual perturbation (Experiment 3) and modulated visual uncertainty (Experiment 4).

Main Results:

• Perceptual error, driven by optimal cue integration, was identified as the primary driver of implicit adaptation.
• The proposed theory accurately predicted trial-by-trial learning dynamics and proprioception changes.
• Experimental manipulation of visual uncertainty confirmed model predictions.

Conclusions:

• The perceptual error framework, supported by Bayesian cue integration, offers a superior explanation for implicit sensorimotor adaptation compared to sensory error models.
• This research clarifies the mechanism of unconscious motor learning and recalibration.