Abstract
BACKGROUND: Deficits in social cognition could be involved in the pathogenesis of delusions in psychotic disorders (Bentall et al., 2009). Childhood trauma (CT) has been associated with an increased risk for psychosis (Varese et al., 2012). Neurocognitive and social cognition deficits could mediate in the association between CT and psychosis (Mansueto et al., 2019). Social cognition and childhood trauma have been understudied so far in delusional disorder (DD). We aimed to assess social cognition in a sample of patients with delusional psychoses (i.e., DD and schizophrenia) and healthy controls (HC) and to explore the potential effect of childhood trauma on social cognition and delusion. METHODS: This cross-sectional, transdiagnostic study included 69 patients with a DSM-IV-TR-confirmed diagnosis of DD (mean age 44.06 ± 11.39 years, 53.6% female), 77 with DSM-IV-TR-confirmed schizophrenia (mean age 38.12 ± 9.27 years, 27.3% female), and 63 HC (mean age 43.6 ± 13.0 years, 68.3% female). Attributional bias was assessed with the “Internal, Personal, and Situational Attributions Questionnaire.” Theory of Mind (ToM) performance was assessed with the “Reading the Mind in the Eyes Test” and the “Faux Pas Recognition Test.” Childhood trauma was measured with the “Childhood Trauma Questionnaire.” Neuropsychological functioning was measured with a comprehensive battery assessing attention, verbal learning, working memory, and executive function. We used ANCOVAs and linear regression analyses to assess the association between the three measures of social cognition and i) diagnosis, ii) dimensional measures of delusion proneness (Peters Delusion Inventory, PDI) and intensity (Maudsley Assessment of Delusion Schedule, MADS), and iii) childhood trauma; after controlling for potential confounders (age, sex, socioeconomic status, and estimated premorbid intelligence quotient). RESULTS: Patients with DD showed significantly poorer performance on the “Eyes Test” than HC (Cohen’s d=-0.44, p=0.037), after controlling for potential confounding variables. The difference was no longer significant after controlling for verbal memory. Patients with schizophrenia (d=-1.54, p<0.001) and DD (Cohen’s d=-0.60, p=0.002) showed significantly poorer performance than HC on the “Faux Pas Test,” after controlling for potential confounders. The difference between patients with schizophrenia and HC remained significant after controlling for neuropsychological functioning (Cohen’s d=-1.09, p<0.001), while differences between patients with DD and HC were no longer significant after controlling for executive function and working memory performance (Cohen’s d=-0.23, p=0.596). No significant differences were found between diagnostic groups in externalizing or personalizing attributional bias. In the fully adjusted models, intensity of the delusional idea was significantly associated with performance in the “Faux Pas Test” in DD, and with externalizing and personalizing attributional bias in schizophrenia. A positive history of CT was significantly associated with lower performance on the “Faux Pas Test” (Cohen’s d=-0.40, p=.022) and higher delusional proneness scores in the delusional psychosis samples (Cohen’s d=-0.49, p=.006), but not in HC. DISCUSSION: Social cognition deficits are associated with delusional intensity in delusional psychoses. Childhood trauma could increase the risk of psychosis through its effect on social cognition.