Are there specific neuropsychological deficits underlying poor insight in first episode psychosis?

Insight in psychosis is a multi-dimensional phenomenon, and has been hypothesised to have some sort of neuropsychological basis. It is unclear to what extent specific neuropsychological abilities are able to predict insight beyond the effect of generalised cognitive ability. We aimed to test this association, alongside the relationship of insight with illness duration and diagnosis, in a sample of first episode psychosis patients. 110 first episode psychosis patients were recruited and a comprehensive assessment was administered, including insight, symptoms, diagnosis and neuropsychological function. Low insight was related to worse performance in a variety of neuropsychological tasks. Regression analysis tested whether any specific tasks were related to insight (or dimensions of insight) beyond the effect of IQ. Verbal memory had an effect on total insight and all dimensions of insight (except compliance) beyond the effect of IQ. Insight appeared to vary with diagnosis, with those diagnosed with depressive affective psychoses having better insight than those with manic affective psychoses. There was no relationship between insight and DUP, but there was a relationship between time spent in treatment before assessment and insight, even after controlling for severity of symptoms. These results suggest a model of insight in early psychosis with a significant neuropsychological component, particularly with verbal memory but also with generalised cognitive ability. There is likely to be a social component to insight affected by initial time spent in contact with treatment, helping patients to understand and come to terms with their illness. © 2011 Elsevier B.V.