Social and non-social measures of cognition for predicting self-reported and informant-reported functional outcomes in early psychosis

The main aim of this study was to investigate the individual contributions of neurocognitive and social-cognitive domains to self-reported and informant-reported functional outcome in early psychosis. We also sought to further characterize the nature of cognitive impairments in this sample and explore the interrelationships between the social-cognitive measures and how they correlate with measures of neurocognition and clinical symptoms. In this study, 70 patients (mean age: 24.1; 87.1% males) with primary psychotic disorder diagnosed in the previous 5 years were assessed on multiple neurocognitive (processing speed, attention, working memory, immediate verbal memory, delayed recall, visual reasoning, inhibition, planning, cognitive flexibility), and social-cognitive domains (theory of mind (ToM), emotion recognition, attributional style, metacognitive overconfidence) as well as measures of clinical symptoms. Functional outcome was assessed with three self-reports and two informant-reports. On average, patients performed one or more SD below healthy controls on measures of delayed recall, ToM and metacognitive overconfidence. Emotion recognition and ToM were intercorrelated and correlated with multiple neurocognitive domains and negative symptoms. Attributional style correlated with positive symptoms. In the context of multiple variables, self-reported functional outcomes were predicted by attributional style, whereas emotion recognition and immediate verbal memory predicted variance in informant-reported community functioning. These results support the suggestion of a likely distinction between the predictive factors for self-reported and informant-reported functional outcome in early psychosis and suggest that consideration of self-assessment of functional outcome is critical when attempting to evaluate the effects attributional style has on functional disability.