Abstract / Summary
Background Prognostic markers of transition from a clinical high-risk (CHR) state to psychosis are important for intervention. Although electroencephalography (EEG) and magnetoencephalography (MEG) abnormalities are associated with psychotic disorders, their prognostic utility in CHR populations remains unclear. This systematic review examined EEG and MEG differences between CHR individuals who transitioned to psychosis (CHR-T) and those who did not (CHR-NT). Methods Following PRISMA guidelines, PubMed, APA PsycINFO, APA PsycArticles, CINAHL Ultimate, Scopus and Web of Science Core Collection were searched for studies published from January 1990 to July 2026. Backward citation searches of included publications and reviews supplemented database searches. Eligible studies compared baseline EEG or MEG measures between CHR-T and CHR-NT groups. Findings were narratively synthesised because of heterogeneity, study quality was assessed using Quality in Prognosis Studies (QUIPS), and event-related potential (ERP) data were meta-analysed where comparable. Results Out of 613 records screened, 52 publications met inclusion criteria. Random-effects meta-analyses were conducted by ERP subtype. The clearest pooled differences were observed for duration MMN ( k = 9, Hedges' g = 0.49, 95% CI [0.14, 0.83]) and P3b ( k = 5, g = −0.61, 95% CI [−0.97, −0.25]), both reduced in CHR-T compared to CHR-NT. Other subtype analyses included few studies and produced mixed findings. Discussion Achieving reliable predictive sensitivity from EEG biomarkers is difficult during the clinically and diagnostically heterogeneous CHR period. Inconsistency across EEG methodologies and outcome definitions likely contributes to variability in reported prognostic effects. Recommendations are provided to improve methodological consistency and support future biomarker research.