Modeling Self-Reference in Schizophrenia: General and Specific Mechanisms Underlying Decision-Making are Differentially Associated with Paranoia
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background and Hypothesis: Individuals with schizophrenia (SZ) show difficulties deciding whether social cues are meant for them, which are implicated in paranoia and social functioning. Recent computational modeling findings suggest that such disruptions may stem from inefficient and biased “evidence accumulation”—the process of gathering information to make a decision. However, it is unclear whether these disruptions are specific to social information processing, particularly self-referential processing, or whether they reflect general processing deficits. Study Design: Mechanisms and clinical correlates of decision-making were examined in 39 SZ and 42 controls across three domains: self-referential social processing (eye contact detection task), non-self-referential social processing (facial gender identification task), and non-social general perceptual processing (visual integration task). Drift Diffusion Models characterized whether inefficient and biased evidence accumulation in SZ was specific to self-referential processing or reflected more general deficits. Study Results: Relative to controls, evidence accumulation in SZ was less sensitive to visual cues during eye contact detection and gender identification—not visual integration. Biased evidence accumulation occurred in SZ only during eye contact detection. In SZ, less sensitive evidence accumulation during eye contact detection and gender identification—not visual integration—was associated with worse paranoia, even after controlling for general processing difficulties. Conclusions: Aberrant self-referential social processing in SZ may be shaped by mechanisms supporting social and self-referential—but not basic visual—information processing. Paranoia was associated with disrupted processing of social—not non-social—information. Therefore, disrupted evidence accumulation in SZ may have differential symptom associations across social and non-social contexts.