Data from 75 participants in a longitudinal study of methamphetamine (MA) abuse were used to differentiate the cognitive performance of those who remained abstinent, relapsed, or continued to use during treatment. Participants were divided into three groups: continuous abstinence, initial abstinence but relapse, and continuous use. Groups did not differ on age, education, gender or ethnicity. Participants in the longitudinal study completed a battery of cognitive tests within 7 days of their last use of MA, then were re-tested monthly for up to 6 months (average time for this analysis was 92 days). For episodic memory, the relapse groups performance was worse than the abstinent and significantly worse than that of the continued use group who had the best performance on all measures. Relapse to methamphetamine use may affect episodic memory differently than it affects the other cognitive functions measured.