Thought disorder in schizophrenia is hypothesized to reflect disrupted maintenance of internal models within hierarchical predictive processing. Beta-range oscillations have been proposed as a carrier of top-down predictive signals. Prior resting-state studies largely assessed continuous spectral beta power, which can miss transient, high-amplitude events — so-called beta bursts — that may be crucial for stabilizing and reinstating internal representations, particularly within the default mode network (DMN). This study examined whether transient beta-burst dynamics, rather than conventional power metrics, reveal spatially dissociable abnormalities related to thought disorder in schizophrenia.
Resting-state magnetoencephalography (MEG) data were acquired from 25 patients with schizophrenia (7 females) and 25 age- and sex-matched healthy controls (8 females) between May 2024 and October 2025. Source localization produced cortical time series from which beta-band (15–30 Hz) bursts were identified. The analysis quantified multiple burst characteristics: burst rate, burst duration, burst power, and inter-regional burst synchrony. The spatial focus included anterior versus posterior cortical positions and nodes of the DMN. Within-patient associations between burst features and objective measures of speech organization (computationally derived semantic and syntactic features) were assessed using principal-component regression.
When using conventional continuous spectral measures, the study found no significant group differences in beta power after correction. In contrast, transient beta-burst metrics produced statistically significant effects. Specifically, there were significant group-by-anterior-posterior-position interactions for burst rate (p = 0.018), burst power (p = 0.0014), and burst duration (p < 0.001). These interactions indicate that burst abnormalities in schizophrenia were spatially heterogeneous along the anterior–posterior axis rather than uniform changes in mean beta power.
Within the DMN, specific regional burst abnormalities were observed in patients compared with controls. The inferior parietal cortex in patients showed significantly shorter beta-burst duration (t(46.0) = -2.74, FDR-p = 0.0343). In contrast, measures of burst synchrony were increased in prefrontal regions: within the left prefrontal cortex (t(31.0) = 2.81, FDR-p = 0.0343) and between prefrontal cortices (t(29.8) = 2.73, FDR-p = 0.0423). These findings point to a dissociation between posterior reductions in burst duration and anterior increases in burst synchrony.
Within the patient group, principal-component regression identified burst-derived components that predicted an objective measure of speech organization (overall model R-squared = 0.47, F(3,19) = 5.66, p = 0.006). The association was driven primarily by a component reflecting inferior parietal burst duration (β = -0.55, p = 0.0036), such that shorter inferior parietal burst duration related to worse speech organization. Burst components did not predict overall clinical severity in this sample according to the reported analyses.
The authors note several limitations that constrain interpretation and generalizability. The patient sample size was modest, resting-state recordings were relatively short, and the study used a cross-sectional design, which limits causal inference about dynamics and symptoms. The MEG-derived measures lack laminar specificity, restricting mechanistic claims about microcircuit processes that could underlie burst generation. Finally, the reported associations between burst features and speech organization require replication in independent cohorts before firm conclusions can be drawn.
The primary conclusion is that schizophrenia is associated with spatially dissociable anomalies in transient beta-burst dynamics along the anterior–posterior axis. Posterior regions (inferior parietal cortex) showed reduced burst duration, whereas anterior prefrontal regions showed increased burst synchrony. These dissociable changes tracked objective, computational measures of disrupted speech organization — a hallmark of thought disorder — suggesting impaired hierarchical coordination of predictive beta signalling as a candidate mechanism linking intrinsic network dysfunction to disorganized thought. The absence of group differences in conventional beta power underscores the value of transient burst analysis for revealing clinically relevant dynamics.
Declared funders included Mitacs, the Wellcome Trust, the Quebec Bioimaging Network (QBIN), the Canada Foundation for Innovation, the McConnell Brain Imaging Centre (Magnetoencephalography in Psychosis), and a Canadian Institutes of Health Research fellowship. One author (Lena Palaniyappan) reported personal fees and honoraria from several industry and professional sources and investigator-initiated educational grants outside the submitted work; all other authors reported no biomedical financial interests or potential conflicts related to this study.