Theory-of-Mind-Related Social Cognition as a Mediator and Bridge Between Neurocognition and Social Functioning in Chinese Patients With Schizophrenia: Latent Variable Modeling and Psychological Network Analysis

Abstract

Background: Social functioning impairments are a hallmark of schizophrenia and a strong predictor of longterm disability. While neurocognition and social cognition are known to predict functioning, the micro-level topological architecture connecting these domains—and the exact extent of mediation after adjusting for measurement error—remains poorly understood. Because the socialcognitive measures used in the present study primarily assessed theory of mind (ToM), this study specifically examined whether ToM-related social cognition statistically medicated the association between neurocognition and social functioning, and identified bridge nodes that may represent potential intervention targets.

Methods: This cross-sectional study recruited clinically stable outpatients with schizophrenia using a consecutive sampling strategy at the Wuhan Mental Health Center, Wuhan, China. Neurocognition was assessed using 4 indices (Digit Symbol Test, Digit Span Test, Continuous Performance Test, and Wisconsin Card Sorting Test categories completed). Social cognition was operationalized primarily through two ToM-focused tasks, the Hinting Task (HT) and Faux Pas Recognition Test (FPRT). Social functioning was evaluated using the 4 domains of the Personal and Social Performance (PSP) scale. Data were analyzed using latent variable structural equation modeling (SEM) and psychological network analysis.

Results: A total of 291 patients (mean [standard deviation (SD)] age, 33.6 [9.6] years; 162 [55.7%] men) were included. Latent variable SEM showed that ToM-related social cognition exhibited a statistically significant indirect effect in the association between neurocognition and social functioning (B = –0.494; 95% bootstrap confidence interval (CI), –1.103 to –0.166), accounting for 38.3% of the total effect. Network analysis revealed a 3-cluster structure wherein the FPRT exhibited the highest bridge strength (0.597), followed by PSP-disturbing and aggressive behaviors (0.517) and HT (0.481), effectively linking the cognitive and functional communities. PSP-personal and social relationships exhibited the highest overall node strength (1.144).

Conclusions: Among Chinese patients with schizophrenia, ToM-related social cognition showed a statistically significant partial mediation pattern in the association between neurocognition and social functioning. ToM-related capacities, particularly faux pas recognition and hint comprehension, emerged as prominent crossdomain bridge nodes in the observed network. These findings suggest that ToM-related processes may be relevant to the association between cognition and functioning; however, because of the cross-sectional design, the results should be interpreted as evidence of indirect association rather than a causal mechanism. Longitudinal and interventional studies are needed to clarify temporal directionality and clinical significance.