Rhesus monkeys learn to control a directional-key inspired brain machine interface via bio-feedback

Area of Science:

• Neuroscience
• Biomedical Engineering
• Machine Learning

Background:

• Brain-machine interfaces (BMIs) enable direct neural control of external devices.
• Existing BMIs often use complex bio-mimetic algorithms, but bio-feedback approaches leveraging user learning are gaining traction.
• The accessibility of chronic recordings and widespread use of novel appendages like computer cursors support user learning in bio-feedback paradigms.

Purpose of the Study:

• To test the feasibility of a simple bio-feedback brain-machine interface (BMI) approach that relies on user learning.
• To implement and evaluate a firing-rate-to-motion correspondence rule for cursor control.
• To explore the potential of bio-feedback control for novel, non-biological appendages.

Main Methods:

• Implemented a simple firing-rate-to-motion correspondence rule for a 2D cursor.
• Assigned groups of neurons to virtual "directional keys" for control.
• Utilized a bio-feedback paradigm where two Rhesus monkeys performed a center-out cursor movement task over multiple sessions.

Main Results:

• Monkeys achieved better performance on the cursor control task after approximately one week of training.
• Neuronal signal patterns exhibited group-level changes, indicating successful neural learning.
• The study demonstrated the feasibility of the proposed bio-feedback control paradigm.

Conclusions:

• A simple bio-feedback BMI approach leveraging user learning is feasible and effective for cursor control.
• This paradigm shows promise for controlling novel appendages and warrants further research in multi-dimensional applications.
• The findings suggest that user adaptability can be effectively harnessed in BMI systems.