Rationale: A cognitive model of functional outcome in schizophrenia posits that neurocognitive deficits can contribute to failure experiences and difficulties in daily living that may lead to the development of dysfunctional attitudes and negative appraisals about one's self and one's ability to perform goal-directed tasks. These cognitions and negative appraisals could increase the likelihood of negative symptoms, such as amotivation, apathy, anhedonia, and contribute to a lack of engagement in goal-directed tasks. Both defeatist performance beliefs and related social disinterest attitudes have emerged as promising targets for cognitive therapy interventions. Defeatist performance beliefs refer to an individual's tendency to overgeneralize from past failures to expected future performance (e.g. "If I fail partly, it is as bad as being a complete failure", "If I ask a question, it makes me look inferior"). Social disinterest attitudes refer to an individual's tendency to show less interest in social relationships and difficulty in anticipating pleasure from social-interpersonal relationships. These cognitions have emerged as potential treatment targets, but research is still lacking concerning whether cognitive therapy can improve negative symptoms and functioning in schizophrenia by changing dysfunctional attitudes. Design: The current study examined whether dysfunctional attitudes were related to outcome in psychosocial interventions for negative symptoms and social functioning. Data were examined from two completed clinical trials of persons diagnosed with either schizophrenia or schizoaffective disorder (N=165) who were assigned to either Cognitive Behavioral Social Skills Training (CBSST) or an active goal-setting intervention called goal-focused supportive contact (GFSC). The primary aim was to evaluate whether mediational models that were supported by our previous cross-sectional research are confirmed in this longitudinal design. The stability of these models across time can help establish temporal-precedence, which is not possible in cross-sectional tests of mediation. The secondary aim of this study was to evaluate whether baseline levels of dysfunctional attitudes moderated negative symptom and functioning outcomes in CBSST. Results: The longitudinal mediation models fit well and all proposed relationships were in the correct direction and generally significant. While support for complete mediation across all time points was not found, there was evidence for partial mediation and the stability of the relationships across time. With regard to moderation, a series of linear regressions showed no significant moderation of negative symptom and functional outcomes based on baseline level of dysfunctional attitudes. Thus, while we did not find support for dysfunctional attitudes as a moderator of change in CBSST versus GSFC, data did support a theoretical model in which changes in dysfunctional cognitions predicted improvement in negative symptoms and social functioning. Conclusion: Future studies should continue to evaluate relationships between dysfunctional cognitions and functional outcomes in schizophrenia, as there was support for these constructs being related. In addition, the impact of psychosocial interventions on negative symptoms and functional outcome in schizophrenia might be improved by strengthening their focus on reducing the severity of dysfunctional attitudes.