Socio-cognitive alterations, including difficulties in mentalising, are observed across many psychiatric conditions. The authors set out to determine whether atypical neural activation in the Theory of Mind (ToM) network aligns with categorical diagnostic groups or with dimensional measures of psychopathology. They evaluated whether symptom severity measured across individuals provides greater explanatory power for ToM-related brain responses than traditional diagnostic labels.
The study analyzed data from 313 participants: 91 healthy controls and 222 psychiatric inpatients. The patient cohort was distributed across five diagnostic clusters: substance use disorders, schizophrenia spectrum disorders, affective disorders, anxiety disorders, and personality disorders. To elicit ToM-related processing, the investigators used a naturalistic movie paradigm during functional magnetic resonance imaging (fMRI) and focused analyses on established ToM regions of interest (ROIs).
The source abstract does not report specific ROI names, scanner or acquisition parameters, preprocessing steps, or details of the movie stimulus; those methodological details were not provided in the abstract content available here.
Using frequentist statistical approaches, the authors compared ToM-selective activation between psychiatric inpatients and healthy controls. Across all predefined ToM ROIs, psychiatric patients showed decreased ToM-selective activity relative to controls. However, when comparing activation between the five diagnostic clusters, no significant differences were observed. Thus, categorical diagnosis did not differentiate ToM activation within the inpatient sample in these analyses.
The investigators examined relationships between ROI activation and dimensional symptom measures using the Adult Self-Report (ASR) scales. ROI activation correlated negatively with ASR Total Problems, Internalising, and Externalising scales. In other words, higher reported symptom burden on these dimensional measures was associated with lower ToM-selective neural responses across the examined ROIs.
To formally compare the predictive utility of categorical diagnoses versus dimensional symptom measures, the authors conducted Bayesian analyses of covariance (ANCOVAs). Across these Bayesian models, results consistently favored models that excluded traditional categorical diagnoses. Models including dimensional measures—particularly Total Problems and Internalising scores from the ASR—showed substantially greater support as predictors of ToM-related regional brain activation. The Bayesian approach provided evidence that dimensional symptom severity better explains variance in ToM neural responses than diagnostic category membership in this sample.
Taken together, the frequentist and Bayesian findings lead the authors to conclude that atypical mentalising, as indexed by reduced ToM-selective activation, may represent a shared transdiagnostic vulnerability across psychiatric disorders rather than disorder-specific neurobiological signatures. Dimensional symptom measures appear to capture variance in social-cognitive neural function that categorical labels do not.
This interpretation challenges reliance on traditional categorical classifications in neurobiological research and supports integrating dimensional assessment of symptom burden when studying neural correlates of social cognition.
This summary is based on the preprint abstract. Specific methodological and statistical details that are important for evaluation were not reported in the abstract: exact ROI definitions and coordinates, full demographic breakdown of diagnostic clusters, scanning and preprocessing parameters, effect sizes, p values and Bayes factors, control for potential confounders (for example medication status or comorbidities), and task or movie characteristics. The article is a preprint and has not been peer reviewed.
In a large transdiagnostic sample, psychiatric inpatients showed reduced ToM network activation compared with healthy controls, and dimensional ASR symptom scores—especially Total Problems and Internalising—provided greater predictive utility for regional ToM activation than categorical diagnoses. The authors interpret these results as supporting a transdiagnostic, symptom-dimension–based perspective on atypical mentalising in psychiatric populations. Funding was declared from Paracelsus Medical University. Because the source text is the abstract of a preprint, readers should consult the full manuscript and peer-reviewed reports for complete methodological and statistical information.