Neural representations for multi-context visuomotor adaptation and the impact of common representation on multi-task...

Area of Science:

• Neuroscience
• Cognitive Science
• Motor Control

Background:

• Human motor adaptability relies on context representations and structural learning.
• Direct evaluation of these mechanisms in sensorimotor tasks is limited.

Purpose of the Study:

• Distinguish neural representations of visual, movement, and context levels in multi-context visuomotor adaptation.
• Investigate the link between representation commonality and adaptation performance.

Main Methods:

• Used functional magnetic resonance imaging (fMRI) and multivariate decoding analysis.
• Focused on three contexts: -90° rotation, +90° rotation, and mirror-reversal.
• Analyzed decoding accuracy for visual, movement, and task-context representations.

Main Results:

• Visual and movement representations decoded in expected brain regions (occipital and visuomotor areas).
• Task-context representations were distinguishable and overlapped with visual/movement encoding regions.
• Higher task performance correlated with greater task-context representation commonality across contexts.

Conclusions:

• Neural encoding of visual and movement directions depends on context.
• Structural learning efficiently extracts commonalities across diverse task contexts.
• Brain's efficient learning mechanisms support versatile motor abilities.