Preservation of Partially Mixed Selectivity in Human Posterior Parietal Cortex across Changes in Task Context

Area of Science:

• Neuroscience
• Brain-Computer Interfaces
• Neural Decoding

Background:

• Partially mixed selectivity describes how the posterior parietal cortex (PPC) represents multiple effectors and cognitive strategies.
• Understanding the generalizability of these neural representations across different task contexts is crucial for brain-machine interface (BMI) applications.

Purpose of the Study:

• To investigate if the structure of neural representations in the PPC is preserved when transitioning from an open-loop motor imagery task to a closed-loop BMI control task.
• To determine the robustness and generalizability of partially mixed selectivity across different task contexts.

Main Methods:

• Recorded neural activity from the PPC of a human tetraplegic patient using a 4x4 mm electrode array during a clinical BMI trial.
• Analyzed representations of left/right hand imagined/attempted movements.
• Compared neural representation structure between an open-loop training phase and a closed-loop online control task using a 1D cursor control paradigm.

Main Results:

• The structure of neural representations in the PPC was largely maintained between the training (open-loop) and online control (closed-loop) phases.
• Individual BMI control performance for different hand movements was compared, demonstrating the accessibility of mixed variables.
• Partially mixed selectivity proved to be a robust property, conserved across changes in task context.

Conclusions:

• Neural representations in the PPC exhibit a conserved structure, maintaining partially mixed selectivity across open-loop and closed-loop tasks.
• Decoding mixed variables from a small cortical area in the PPC is feasible for individual BMI control.
• The findings support the potential for robust and generalizable neural decoding in brain-machine interfaces.