Related Experiment Video
Updated: Dec 29, 2025
04:44
Inter-Brain Synchrony in Open-Ended Collaborative Learning: An fNIRS-Hyperscanning Study
Published on: July 21, 2021
4.8K
Nonlinear interaction decomposition (NID): A method for separation of cross-frequency coupled sources in human brain
Mina Jamshidi Idaji1, Klaus-Robert Müller2, Guido Nolte3
1Department of Neurology, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany; Machine Learning Group, Technical University of Berlin, Berlin, Germany; International Max Planck Research School NeuroCom, Leipzig, Germany.
Neuroimage
|February 9, 2020
Summary
We developed Nonlinear Interaction Decomposition (NID) to detect nonlinear brain oscillations in EEG/MEG data. This novel method reliably identifies cross-frequency coupling interactions, even with low signal-to-noise ratios.
Area of Science:
- Neuroscience
- Computational Neuroscience
- Signal Processing
Background:
- Cross-frequency coupling (CFC) is crucial for integrating brain information across different spatial and spectral scales.
- Existing methods for detecting CFC in electroencephalography (EEG) and magnetoencephalography (MEG) have limitations.
Purpose of the Study:
- To introduce a novel framework, Nonlinear Interaction Decomposition (NID), for detecting nonlinear cross-frequency coupling (CFC) in EEG/MEG data.
- To demonstrate the efficacy of NID in identifying nonlinearly interacting neuronal oscillations based on statistical properties of linear mixtures.
Main Methods:
- NID leverages the non-Gaussian distribution of linear mixtures of nonlinearly coupled brain oscillations.
- Analytical evaluation was performed for phase-phase and amplitude-amplitude coupled oscillations.
- Extensive validation using simulated EEG with realistic head modeling and real resting-state EEG data from 81 subjects.
Main Results:
- NID successfully extracted nonlinearly interacting components with high reliability, even at signal-to-noise ratios as low as -15 dB.
- Application to resting-state EEG revealed widespread phase-phase coupling between alpha and beta oscillations.
- Identified interactions were localized to temporal, parietal, and frontal brain regions, indicating diverse local and distant nonlinear interactions.
Conclusions:
- NID provides a robust and sensitive method for uncovering nonlinear cross-frequency interactions in EEG/MEG.
- The findings highlight the prevalence of complex nonlinear interactions in resting-state brain activity.
- The publicly available code facilitates further research into neural oscillations and brain connectivity.
