Background. Results of previous research examining long-term residual effects of marijuana use on cognition are conflicting. A major methodological limitation of prior studies is the inability to determine whether differences between users and non-users are due to differences in genetic vulnerability preceding drug use or due to the effects of the drug. Method. Fifty-four monozygotic male twin pairs, discordant for regular marijuana use in which neither twin used any other illicit drug regularly, were recruited from the Vietnam Era Twin Registry. A minimum of 1 year had passed since the marijuana-using twins had last used the drug, and a mean of almost 20 years had passed since the last time marijuana had been used regularly. Twins were administered a comprehensive neuropsychological test battery to assess general intelligence, executive functioning, attention, memory and motor skills. Differences in performance between marijuana-using twins and their non-using co-twins were compared using a multivariate analysis of specific cognitive domains and univariate analyses of individual test scores. Dose-response relationships were explored within the marijuana-using group. Results. Marijuana-using twins significantly differed from their non-using co-twins on the general intelligence domain; however, within that domain only the performance of the block design subtest of the Wechsler Adult Intelligence Scale - Revised reached a level of statistical significance. Conclusions. Out of the numerous measures that were administered, only one significant difference was noted between marijuana-using twins and their non-using co-twins on cognitive functioning. The results indicate an absence of marked long-term residual effects of marijuana use on cognitive abilities.