Information theoretic analysis of proprioceptive encoding during finger flexion in the monkey sensorimotor system

Area of Science:

• Neuroscience
• Computational Neuroscience
• Motor Control

Background:

• Debate exists on whether neural information is coded by firing rate or spike timing.
• Understanding neural coding is crucial for deciphering brain function and developing advanced brain-machine interfaces.

Purpose of the Study:

• To investigate how sensorimotor cortex, deep cerebellar nuclei, and dorsal root ganglia code information during a finger flexion task.
• To compare the coding capabilities of neural firing rate, discharge irregularity, and spectral power.

Main Methods:

• Recorded single-unit spike discharge from macaque monkeys performing a trained finger flexion task.
• Utilized information theory to analyze coding of task conditions in spike rate, irregularity, and 15-25 Hz spectral power.
• Examined coding redundancy and independence between different neural measures and simultaneously recorded unit pairs.

Main Results:

• All three measures (rate, irregularity, power) successfully coded task information across all tested brain areas.
• Coding between irregularity and 15-25 Hz power was largely independent (60%), while rate and irregularity showed moderate redundancy (56%), and rate and power showed high redundancy (93%).
• Most simultaneously recorded unit pairs (86%) coded information independently using the same measure, and combining measures often provided additional task information.

Conclusions:

• Sensorimotor systems employ both rate and temporal codes to represent information related to finger movement tasks.
• Spike irregularity is a significant coding measure, and its inclusion in algorithms could improve brain-machine interface decoding accuracy.