The purpose of this dissertation was to compare two models of the relationships among neurocognition, social cognition, and functional outcome in schizophrenia and high-functioning autism (HFA). Forty-five participants with schizophrenia and thirty-four participants with HFA completed a battery of neurocognitive and social cognitive tasks, and the Social Functioning Scale as a measure of functional outcome. Composite variables were created for all three constructs. Within the schizophrenia sample, path analyses revealed a significant and negative relationship between neurocognition and functional outcome, a nonsignificant, positive relationship between social cognition and functional outcome, and modest support for social cognition serving as a mediator between neurocognition and functional outcome. Within the HFA sample, neither neurocognition nor social cognition significantly predicted functional outcome. Consistent with previous research, a strong, positive relationship between neurocognition and social cognition emerged in both samples, although this appears to be more robust for individuals with schizophrenia. In contrast, only a small proportion of the variance in functional outcome was accounted for by the models. It is suggested that use of the Social Functioning Scale may have contributed to these findings.