Detecting electrophysiological alterations in psychiatric disorders through event-related microstates: a systematic review

Andrea Perrottelli¹Francesco Flavio Marzocchi¹Giorgio Di Lorenzo^2,3*Chiara D'Amelio⁴Noemi Sansone¹Luigi Giuliani¹Pasquale Pezzella¹Edoardo Caporusso¹Antonio Melillo¹Giulia Maria Giordano¹Paola Bucci¹Armida Mucci^1*Silvana Galderisi¹

• ¹Department of Mental and Physical Health and Preventive Medicine, University of Campania "Luigi Vanvitelli", Naples, Italy
• ²Laboratory of Psychophysiology and Cognitive Neuroscience, Department of Systems Medicine, Tor Vergata University of Rome, Rome, Italy
• ³Santa Lucia Foundation (IRCCS), Rome, Lazio, Italy
• ⁴Department of Applied Clinical Sciences and Biotechnology, University of L'Aquila, L'Aquila, Abruzzo, Italy

Introduction: Event-related potentials (ERPs), recorded through electroencephalography (EEG) during sensory and cognitive tasks, have been consistently employed to investigate electrophysiological correlates of psychiatric disorders. However, traditional peak component analysis of ERPs is limited by the a priori selection of time windows and electrodes. Microstate analysis, a data-driven approach based on identifying periods of quasi-stable scalp topographies, has been applied to ERP data, offering a valuable tool for understanding the temporal dynamics of large-scale neural networks. This review aims to provide a comprehensive summary of studies examining event-related microstates in individuals with psychiatric disorders. Methods: A systematic review of English-language articles indexed in PubMed, Scopus, and Web of Science (WoS) was conducted on May 1, 2024. Studies were included only if they applied microstate analysis to ERP data and analyzed data from at least one group of patients with psychiatric disorders in comparison to healthy controls. Results: Of the 1,115 records screened, 17 studies were included in the final qualitative synthesis. The majority of these studies (n=8) included patients with schizophrenia, using various tasks focusing mainly on visuospatial processing (n=6) and face processing (n=6). Regarding the microstate methodology, the primary clustering approach employed was the k-means clustering algorithm (n=8), while the cross-validation criterion (n=10) was the most commonly used measure of fit. Sixteen of the 17 studies reported at least one significant difference in microstate features between patients and healthy controls, mainly in the temporal and topographic characteristics of microstates and the sequence of their occurrence. Conclusions: This review highlights the value of event-related potential microstates analysis in identifying spatiotemporal alterations in brain dynamics associated with psychiatric disorders. However, the limited number of studies and the heterogeneity of experimental paradigms constrain the generalizability of the findings.