Learning Task-Related Activities From Independent Local-Field-Potential Components Across Motor Cortex Layers

Area of Science:

• Neuroscience
• Computational Neuroscience

Background:

• Motor cortical microcircuits integrate diverse inputs for voluntary movement.
• The precise roles of these inputs in motor learning remain unclear.

Purpose of the Study:

• To investigate the contribution of independent components of motor cortical activity to motor learning.
• To explore the roles of specific frequency bands (slow gamma, fast gamma, theta) and cortical layers in motor control.

Main Methods:

• Analysis of local field potential (LFP) activity from rat motor cortex during reward-motivated lever movements.
• Extraction and analysis of independent components (ICs) across different cortical depths.
• Training a recurrent neural network model using identified ICs to simulate motor learning.

Main Results:

• Oscillations in slow and fast gamma bands correlate with distinct behavioral states (preparation and execution).
• A novel IC, prominent in deep cortical layers, shows enhanced slow gamma activity during movement execution.
• The four major ICs differentially contribute to task-related activities and motor learning, with overlapping roles.

Conclusions:

• Independent components of motor cortical activity play distinct yet overlapping roles in motor learning.
• A novel deep-layer component with specific oscillatory dynamics is identified during movement execution.