Motor adaptation and generalization of reaching movements using motor primitives based on spatial coordinates

Area of Science:

• Neuroscience
• Motor Control
• Computational Neuroscience

Background:

• The brain employs diverse coordinate systems for sensory and motor processing, from retinal to body-centered frames.
• Understanding the reference frames in the motor cortex is crucial for explaining action guidance.

Purpose of the Study:

• To investigate the allocentric reference frame in the motor cortex using spatial coordinates.
• To examine physiological and psychophysical evidence for coordinate systems in motor control.

Main Methods:

• Extending a computational framework based on spatial vectors and cross products to model motor adaptation.
• Analyzing how coefficients of velocity- and acceleration-dependent cross products adapt to perturbations.

Main Results:

• Each cross product exhibits a distinct reference frame, influencing generalization to untrained locations.
• Velocity-dependent cross products adapt to force fields (intrinsic frame), while acceleration-dependent ones adapt to visuomotor rotations (extrinsic frame).
• The model predicts specific under- or overgeneralization patterns in visuomotor adaptation based on workspace location.

Conclusions:

• A unified computational model based on cross-product bases explains distinct generalization patterns in visuomotor and force-field adaptation.
• Kinematic and dynamic motor adaptation may not require separate neural substrates.