Related Experiment Video
Updated: Mar 9, 2026
05:55
Modeling the Functional Network for Spatial Navigation in the Human Brain
Published on: October 13, 2023
1.6K
Reliability of graph metrics derived from resting-state human EEG
Karl Kuntzelman1, Vladimir Miskovic1,2
1Department of Psychology, State University of New York at Binghamton, Binghamton, New York, USA.
Psychophysiology
|December 22, 2016
Summary
This study evaluated the reliability of graph theory measures for analyzing brain networks using human electroencephalography (EEG). Findings show reliability varies by frequency band and connectivity method, with phase-based methods best for alpha/beta bands and coherence best for delta/theta bands.
Area of Science:
- Neuroscience
- Brain network analysis
- Computational psychiatry
Background:
- Characterizing large-scale brain functioning requires understanding neuronal synchronization.
- Human electroencephalography (EEG) combined with functional connectivity and graph theory offers tools to study dynamic brain networks.
- Reproducibility of graph theoretical measures in EEG network analysis needs further investigation.
Purpose of the Study:
- To assess the test-retest reliability of a comprehensive set of graph theoretical measures.
- To evaluate these measures on weighted networks derived from high-density resting-state human EEG.
- To identify how frequency bands and functional connectivity methods influence network reliability.
Main Methods:
- Utilized high-density resting-state human EEG data.
- Derived weighted networks using various functional connectivity estimation methods.
- Applied a broad suite of graph theoretical measures to characterize network structure.
- Analyzed reliability across different frequency bands (delta, theta, alpha, beta).
Main Results:
- Overall promising test-retest reliability for graph theoretical measures was observed.
- Reliability was dependent on the chosen frequency band and functional connectivity method.
- Relative phase distribution improved reliability in alpha and beta bands.
- Coherence (phase and amplitude) enhanced reliability in delta and theta bands.
- Substantial variance was noted between different graph metrics.
Conclusions:
- Graph theoretical analysis of EEG-derived brain networks shows good reliability with specific methodological considerations.
- The choice of functional connectivity method is critical and band-specific for reliable network analysis.
- Relative phase is optimal for alpha/beta band analysis, while coherence is better for delta/theta bands.
- Further research is needed to optimize graph metric selection and connectivity methods for robust brain network characterization.
