Cognitive functioning in people with psychotic experiences: a systematic review and meta-analysis study

Abstract

Earlier research suggested that psychotic experiences (PEs), the extended-psychosis phenotype, are associated with cognitive impairment. Recent studies, however, revealed more mixed findings, and patterns and magnitude of cognitive deficits in PEs remain uncertain. We aimed to systematically review and quantitatively synthesize estimates of cognitive functioning covering a wide array of domains in individuals with versus without PEs. We systematically searched four databases from inception to 6 July 2023. We generated pooled effect size (Hedges'g) using random-effects models. Subgroup analyses and meta-regression examining the moderating effect of sex, age at PE assessment, study design, cognitive task, and PE assessment instrument on cognitive functioning were performed when applicable. The study was registered with PROSPERO (CRD42023442528). Twenty-seven and six studies were included for meta-analysis of cognitive functioning comparing individuals with versus without PEs (n = 82,561; 10,251 individuals with PEs) and individuals with high-level versus low-level PEs (n = 8062; 813 individuals with high-level PEs), respectively. Individuals with PEs exhibited worse cognitive performance in general cognition (Hedges'g = -0.10 [95%CI = -0.18 to -0.02]), verbal fluency (Hedges'g = -0.05 [95%CI = -0.10 to -0.00]), visual memory (Hedges'g = -0.21 [95%CI = -0.38 to -0.03]), and working memory (Hedges'g = -0.16 [95%CI = -0.28 to -0.04]). Meta-regression revealed that general cognition associated with PEs was related to younger age (z = 3.37, p = 0.001), male sex (z = -2.59, p = 0.010), and cognitive assessment before PE assessment (z = -2.15, p = 0.031), whereas working memory in individuals with PEs was associated with concurrent cognitive and PE assessment (z = 6.19, p < 0.001). We failed to find moderating effect of the choice of PE assessment instrument or cognitive task on cognitive functioning in PEs. Additional analysis showed no significant difference in the performance of any cognitive domains between individuals with high-level versus low-level PEs. Limitations included studies primarily derived from Western countries, no social-cognitive domains, and varied PE measurement. In sum, PEs are associated with milder and more circumscribed cognitive impairment relative to psychotic disorders. Future research is required to clarify differential cognitive trajectories between individuals with transient PEs and persistent/recurrent-PEs.