Continuous Force Decoding from Local Field Potentials of the Primary Motor Cortex in Freely Moving Rats

Area of Science:

• Neuroscience
• Biomedical Engineering
• Motor Control

Background:

• Local field potential (LFP) signals from the primary motor cortex offer rich information for decoding motor functions.
• While kinematic parameters like position and velocity are decodable from LFP, continuous force decoding in freely moving animals remains challenging.

Purpose of the Study:

• To investigate the feasibility of decoding continuous force magnitude using LFP signals in freely moving rats.
• To identify the contribution of different LFP signal components to force decoding accuracy.

Main Methods:

• Three rats were trained to press a force sensor for a water reward.
• A 16-channel micro-wire array was implanted in the primary motor cortex to record LFP signals.
• Decoding models were trained using LFP signals to predict continuous force sensor readings.

Main Results:

• The decoding model achieved an average coefficient of correlation (r) of 0.66 and R² of 0.42 between decoded and actual force signals.
• LFP signals within the gamma frequency band (30-120 Hz) demonstrated the highest contribution to the decoding model.
• The study successfully demonstrated continuous force decoding using a limited number of LFP channels.

Conclusions:

• Continuous force decoding from LFP signals in freely moving animals is feasible.
• Gamma frequency band LFP signals are critical for accurate force magnitude decoding.
• This approach holds promise for real-life Brain-Machine Interface (BMI) applications.