Towards hippocampal navigation for brain-computer interfaces.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Rehabilitation Technology

Background:

• Current brain-computer interfaces (BCIs) for wheelchair control primarily use non-invasive methods and decode simple motor commands.
• These systems lack the ability to interpret higher-order cognitive processes for more intuitive control.
• Severely paralyzed individuals could gain significant autonomy with advanced assistive technologies.

Purpose of the Study:

• To investigate the feasibility of using invasive hippocampal brain signals to decode navigational intent for wheelchair control.
• To explore the potential of decoding higher-order cognitive processes, specifically navigation, for assistive devices.
• To establish a foundation for developing novel invasive BCIs for enhanced mobility.

Main Methods:

• Recorded hippocampal signals from participants during a virtual navigation task.
• Trained a machine learning decoder to classify virtual movement speeds based on neural activity.
• Utilized an invasive neural prosthetic approach for signal acquisition.

Main Results:

• Successfully trained a decoder capable of classifying virtual movement speeds from hippocampal signals.
• Demonstrated that hippocampal activity contains information relevant to navigational speed.
• Established proof-of-concept for decoding navigational intent from the hippocampus.

Conclusions:

• Decoding navigational intent from hippocampal signals is a viable strategy for developing advanced BCIs.
• An invasive hippocampal BCI shows promise for future wheelchair control systems.
• This research is a critical first step toward realizing intuitive, intent-based control for assistive mobility devices.