Single-unit activity, threshold crossings, and local field potentials in motor cortex differentially encode reach...

Area of Science:

• Neuroscience
• Computational Neuroscience
• Neural Engineering

Background:

• Extracellular electrodes record diverse neural signals, including action potentials and local field potentials (LFPs).
• The information content and underlying neural phenomena of these distinct signal types remain incompletely understood.
• Current research often focuses on specific signal types, leading to differing interpretations of neural activity.

Purpose of the Study:

• To compare the information content of four simultaneously recorded extracellular signal types.
• To investigate whether these signals capture similar or different neural phenomena and kinematic information.
• To assess the implications for brain-machine interface (BMI) applications.

Main Methods:

• Simultaneous recording of single-unit action potentials, multiunit threshold crossings, and LFPs (two frequency bands) using multielectrode arrays in primary motor cortex.
• Quantification of signal tuning to kinematic parameters of reaching movements.
• Comparative analysis of information integration across individual electrodes and multiple electrodes.

Main Results:

• Multiunit threshold crossings are not a reliable proxy for single-unit action potential activity.
• Threshold crossing activity shows greater similarity to high-frequency LFP (100-300 Hz) than to single-unit activity on individual electrodes.
• Threshold crossing activity and LFPs integrate neural information at different spatial scales across multiple electrodes.
• Beta-band LFP power (10-40 Hz) reliably indicates movement onset but lacks moment-to-moment kinematic encoding.

Conclusions:

• Diverse extracellular signals provide distinct and complementary neural information.
• These signal types may originate from partially distinct biological phenomena.
• Utilizing these varied signals can enhance the performance of brain-machine interfaces.