Limb-related sensory prediction errors and task-related performance errors facilitate human sensorimotor learning...

Area of Science:

• Neuroscience
• Motor Control
• Learning

Background:

• The human sensorimotor system adapts to errors, but mechanisms for sensory prediction errors (SPEs) and task performance errors (TPEs) are unclear.
• SPEs result from miscalibrated internal models, while TPEs stem from environmental challenges to task goals.

Purpose of the Study:

• To investigate the distinct neural mechanisms underlying learning from SPEs and TPEs.
• To determine how these error types interact to influence future motor performance.

Main Methods:

• Conducting reach experiments with varied target properties and perturbed visual feedback.
• Manipulating instructions to differentiate between implicit and explicit learning strategies.

Main Results:

• Learning mechanisms for SPEs and TPEs are dissociable.
• TPEs promote explicit, strategy-based learning (action selection), not implicit reach plan refinement.
• SPEs are necessary and sufficient for implicit recalibration.

Conclusions:

• Task performance errors drive volitional strategy selection, while sensory prediction errors drive implicit motor recalibration.
• Combining both learning strategies enhances performance gains, but successful actions without strategy use yield smaller improvements.
• Flexible weighting of strategic action selection and implicit recalibration optimizes learning from errors.