In schizophrenia, neurocognition, social cognition and functional outcome are all inter-related, with social cognition mediating the impact that impaired neurocognition has on functional outcome. Less clear is the nature of the relationship between neurocognition, social cognition and functional outcome in individuals at clinical high risk (CHR) for psychosis. 137 CHR participants completed a neurocognitive test battery, a battery of social cognition tasks and the Social Functioning Scale. Confirmatory factor analysis showed that all social cognition tasks were reliable and valid measures of the latent variable. The path from neurocognition to functioning was statistically significant (standardized coefficient β=0.22, p<0.01). The path from social cognition to functioning was also statistically significant (β=0.27, p<0.05). In the mediation model the bootstrapping estimate revealed a nonsignificant indirect effect that was the association of social cognition with neurocognition and with functional outcome (β=0.20, 95% CI=-0.07 to 0.52, p=0.11). However, social cognition was significantly associated with neurocognition (β=0.80, p<0.001) and the path from neurocognition to functioning was no longer significant as soon as the mediator (social cognition) was entered into the mediation model (β=0.02, p=0.92). All of the model fit indices were very good. Unlike what has been observed with psychotic patients, social cognition does not seem to mediate the pathway from neurocognition to functional outcome when assessed with a measure of social attainment in individuals at CHR for psychosis.