Cross-Frequency Coupling as a Neural Substrate for Prediction Error Evaluation: A Laminar Neural Mass Modeling...

Area of Science:

• Computational neuroscience
• Neurobiology
• Systems neuroscience

Background:

• Predictive coding theories propose hierarchical error minimization as a core neural computation.
• The precise neural mechanisms underlying predictive coding remain largely undefined.
• Cross-frequency coupling (CFC) is a potential candidate for implementing these computations.

Purpose of the Study:

• To propose and model cross-frequency coupling (CFC) as a neural mechanism for predictive coding.
• To investigate the roles of signal-envelope coupling (SEC) and envelope-envelope coupling (EEC) in hierarchical inference.
• To explore the implications of disrupted CFC in Alzheimer's disease and psychedelic drug action.

Main Methods:

• Development of a laminar neural mass model (LaNMM).
• Simulation of signal-envelope coupling (SEC) for prediction-error generation.
• Simulation of envelope-envelope coupling (EEC) for precision weighting and gating.
• Modeling of perturbations to CFC mechanisms to simulate disease and drug states.

Main Results:

• SEC and EEC were shown to instantiate a hierarchical comparator mechanism.
• SEC was demonstrated to generate prediction-error signals.
• EEC was shown to implement gating for precision weighting.
• Modeling revealed aberrant prediction error signaling in Alzheimer's disease and altered sensory attenuation with psychedelics.

Conclusions:

• Multiscale cross-frequency coupling (CFC) is implicated as a key computational mechanism for predictive coding.
• Disruptions in CFC contribute to the pathophysiology of Alzheimer's disease.
• Altered CFC dynamics may explain the effects of serotonergic psychedelics on sensory processing.