EEG microstate syntax analysis: A review of methodological challenges and advances

• Published in: NeuroImage (Orlando, Fla.) Vol. 309; p. 121090
• Main Authors: Haydock, David, Kadir, Shabnam, Leech, Robert, Nehaniv, Chrystopher L., Antonova, Elena
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.04.2025; Elsevier Limited; Elsevier
• Subjects: Biomarkers; Brain - physiology; Cognition; Cognition & reasoning; EEG; Electroencephalography; Electroencephalography - methods; Functional magnetic resonance imaging; Humans; Magnetic resonance imaging; Neuroimaging; Resting-state EEG; Signal Processing, Computer-Assisted; Syntax; Time series; Topography; Neuroimaging; Biomarkers; Resting-state EEG; Electroencephalography; Cognition
• ISSN: 1053-8119; 1095-9572; 1095-9572
• DOI: 10.1016/j.neuroimage.2025.121090

•A general overview of the history of EEG microstate syntax analysis is provided.•Three microstate sequence types are defined to facilitate comparisons across studies.•Microstate syntax analysis methods are distinguishable by their approach to sub-sequences.•Analysis methods which only investigate a single length are subject to combinatoric problems.•Microstate syntax investigations would benefit from comparison to continuous signal. Electroencephalography (EEG) microstates are “quasi-stable” periods of electrical potential distribution in multichannel EEG derived from peaks in Global Field Power. Transitions between microstates form a temporal sequence that may reflect underlying neural dynamics. Mounting evidence indicates that EEG microstate sequences have long-range, non-Markovian dependencies, suggesting a complex underlying process that drives EEG microstate syntax (i.e., the transitional dynamics between microstates). Despite growing interest in EEG microstate syntax, the field remains fragmented, with inconsistent terminologies used between studies and a lack of defined methodological categories. To advance the understanding of functional significance of microstates and to facilitate methodological comparability and finding replicability across studies, we: i) derive categories of syntax analysis methods, reviewing how each may be utilised most readily; ii) define three “time-modes” for EEG microstate sequence construction; and iii) outline general issues concerning current microstate syntax analysis methods, suggesting that the microstate models derived using these methods are cross-referenced against models of continuous EEG. We advocate for these continuous approaches as they do not assume a winner-takes-all model inherent in the microstate derivation methods and contextualise the relationship between microstate models and EEG data. They may also allow for the development of more robust associative models between microstates and functional Magnetic Resonance Imaging data.