The aims of the study were: first to examine, using clinical symptoms of patients as a template, whether the correlated but independent dimensions of positive, negative and depressive symptoms that have been identified in clinical psychosis, also have a distribution as non-clinical experiences in the general population; and second, to establish to what degree population variation in experience of positive and negative features of psychosis is actually independent of experience of depression. In a representative population sample of 932 young men, we measured experiences of positive, negative and depressive features of psychosis, using a 40-item self-report instrument. Confirmatory factor analysis was used to compare the fit of hypothesized one-, two- and three-factor solutions. A three-factor model of separate depressive, positive and negative dimensions provided a better fit to the data than either a two-factor or unidimensional model. All three dimensions were correlated with each other, but also showed good discriminant validity in relation to established scales, confirming their relative independence. The data suggest that the correlated dimensions of clinical psychosis also have a distribution in the general population, and that depressive symptoms may form an integral part of psychosis-like experiences in the general population.