Untangling cross-regional cross-frequency coupling in dynamic neural oscillations

Area of Science:

• Neuroscience
• Computational Neuroscience
• Signal Processing

Background:

• Brain networks utilize long-range phase coupling of low-frequency oscillations (LFOs) for communication.
• Cross-frequency coupling (CFC), where LFO phase modulates high-frequency power, is crucial for neural activity regulation.
• The mechanisms of cross-regional CFC and methods to differentiate local from remote LFO influences are not well understood.

Purpose of the Study:

• To develop and validate a novel method, the partial modulation index (PMI), to accurately measure cross-regional CFC.
• To differentiate true cross-regional CFC from apparent coupling caused by simultaneous phase coupling.
• To assess the PMI's effectiveness in simulated data and human intracranial recordings.

Main Methods:

• Developed the partial modulation index (PMI) based on Pearl's do-calculus, a derivation of the modulation index (MI).
• Tested PMI on simulated datasets to assess its ability to distinguish biased from unbiased CFC.
• Evaluated PMI using intracranial local field potentials from thalamus and cortex in a patient with essential tremor.

Main Results:

• PMI successfully differentiated between genuine focal CFC and biased cross-regional phase coupling in simulated data, unlike conventional MI.
• Analysis of human thalamocortical data indicated that observed CFC was partially biased, as revealed by PMI.
• Simulated data results validated PMI's capability to remove mathematical bias inherent in CFC measurements.

Conclusions:

• The novel PMI method provides a mathematically rigorous approach to characterize residual CFC.
• This facilitates accurate investigation of brain-wide LFO contributions to CFC.
• Improved understanding of CFC mechanisms can enhance insights into neurological disorders.