# Frontal and parietal planning signals encode adapted motor commands when learning to control a brain–computer interface

## TL;DR

The study shows how frontal and parietal brain areas adapt together when learning to control a brain-computer interface.

## Contribution

It reveals that both brain regions encode adapted motor commands during BCI learning, even in neurons not directly involved.

## Key findings

- Frontal and parietal areas better reflect adapted motor commands than visual feedback during movement.
- Adaptive changes in planning activity transfer more strongly to movement corrections in the frontal cortex.
- Both local and remote neurons show adaptive responses during BCI-based motor learning.

## Abstract

Perturbing visual feedback is a powerful tool for studying visuomotor adaptation. However, unperturbed proprioceptive signals in common paradigms inherently co-varies with physical movements and causes incongruency with the visual input. This can create challenges when interpreting underlying neurophysiological mechanisms. We employed a brain–computer interface (BCI) in rhesus monkeys to investigate spatial encoding in frontal and parietal areas during a 3D visuomotor rotation task where only visual feedback was movement-contingent. We found that both brain regions better reflected the adapted motor commands than the perturbed visual feedback during movement preparation and execution. This adaptive response was observed in both local and remote neurons, even when they did not directly contribute to the BCI input signals. The transfer of adaptive changes in planning activity to corresponding movement corrections was stronger in the frontal than in the parietal cortex. Our results suggest an integrated large-scale visuomotor adaptation mechanism in a motor-reference frame spanning across frontoparietal cortices.

How do parietal and premotor areas in the brain adapt when controlling a brain-computer interface (BCI)? This study shows that frontal and parietal brain areas co-adapt during BCI-based motor learning, offering insights into large-scale visuomotor adaptation and informing future developments in neuroprosthetics and brain-computer interfaces.