Evidence against a single coordinate system representation in the motor cortex.

Area of Science:

• Neuroscience
• Motor Control
• Computational Neuroscience

Background:

• The coordinate system used by motor cortical cells to encode movement direction remains a fundamental unresolved question.
• Previous research proposed extrinsic (Cartesian), intermediate (shoulder-centered), or intrinsic (joint-angle) frames.

Purpose of the Study:

• To investigate and compare movement direction encoding across Cartesian (CA), shoulder-centered (SC), and joint-angle (JA) coordinate systems in the primary motor cortex.
• To determine if any single coordinate system dominates neural representations of movement direction.

Main Methods:

• Simultaneous multi-unit recordings from the primary motor cortex during movement tasks.
• Analysis of directional tuning invariance across workspace sub-regions.
• Mutual information analysis comparing neuronal firing rate and movement direction in different coordinate frames.
• Population vector decoding to assess encoding bias.

Main Results:

• A broad range of directional invariance was observed across all three coordinate systems (CA, SC, JA), with no single system showing strong dominance.
• Mutual information analyses supported the lack of a dominant coordinate system.
• A small but significant bias towards the shoulder-centered (SC) coordinate frame was detected, further supported by population vector decoding.
• Similar findings were observed for hand/torque force direction encoding.

Conclusions:

• The motor cortex does not exclusively rely on a single, simple coordinate system for movement encoding.
• Results suggest the motor cortex may utilize a more complex, yet undiscovered, coordinate system or that it may not be substrate for a single coordinate representation.