Real-time EEG-based brain-computer interface to a virtual avatar enhances cortical involvement in human treadmill...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Rehabilitation Technology

Background:

• Non-invasive brain-computer interface (BCI) technologies enable neural decoding of gait intent and kinematics.
• The cortical dynamics during BCI-controlled human walking remain largely unexplored.
• Understanding cortical involvement is crucial for developing advanced BCI applications.

Purpose of the Study:

• To investigate changes in cortical involvement during treadmill walking with and without closed-loop BCI control.
• To identify specific brain regions and neural activity patterns associated with BCI-assisted gait.

Main Methods:

• Participants walked on a treadmill while their brain activity was recorded using non-invasive methods.
• Source localization techniques were employed to analyze cortical network activity during BCI control of a walking avatar.
• Analysis focused on changes in alpha/mu and gamma frequency bands in specific cortical regions.

Main Results:

• Significant differences in cortical network activity were observed between BCI-controlled and unassisted walking.
• Sustained alpha/mu suppression in the Posterior Parietal Cortex and Inferior Parietal Lobe indicated increased cortical engagement during BCI control.
• Increased Anterior Cingulate Cortex (ACC) activity in the low-frequency band suggested involvement in error monitoring and motor learning, with low gamma modulations potentially linked to enhanced voluntary gait control.

Conclusions:

• Closed-loop BCI control of a walking avatar modulates cortical activity, increasing involvement in areas related to spatial processing and attention.
• The findings highlight the role of the ACC in error monitoring and motor adaptation within a BCI-assisted locomotion context.
• This research advances the development of BCI-based training paradigms for rehabilitation, promoting a top-down approach to motor recovery.