A Generalized Coherence Framework for Detecting and Characterizing Nonlinear Interactions in the Nervous System.

Area of Science:

• Neuroscience
• Signal Processing
• Systems Biology

Background:

• Existing coherence measures often fail to capture the full spectrum of nonlinear neural dynamics.
• Characterizing nonlinear interactions is crucial for understanding complex brain function.

Purpose of the Study:

• Introduce a generalized coherence framework, cross-spectral coherence (CSC), for detecting and characterizing nonlinear interactions in the nervous system.
• Apply CSC to investigate nonlinear stimulus-response interactions in the human proprioceptive system.

Main Methods:

• Verified CSC performance using model simulations of static and dynamic nonlinear systems.
• Applied CSC to analyze electroencephalogram (EEG) responses to periodic wrist perturbation during isotonic wrist flexion.

Main Results:

• CSC successfully detected various nonlinear interactions, including harmonic and intermodulation coupling, in simulations and human data.
• High-order nonlinearities were evident in the proprioceptive system, with subharmonic coupling observed in some subjects.

Conclusions:

• CSC provides a comprehensive tool for characterizing nonlinear neural dynamics and input-output relationships.
• The study reveals complex nonlinear dynamics in the human proprioceptive system, including the novel finding of subharmonic coupling.