Abstract Introduction: Methamphetamine abuse has been growing during the past decade. Recent studies show that computerized training intervention can be useful. The present study compared a variety of cognitive bias modification methods (i.e., attention bias modification, approach bias modification and their combination) on attention bias, approach bias and relapse among methamphetamine abusers. Method: A total of 108 patients were recruited as the primary sample and the obtained data from 97 patients were analyzed in the final analysis. After being matched according to demographical characteristics, participants were assessed for cognitive bias and relapse to measure the baseline. Afterwards, they were randomly assigned to three different training conditions receiving eight training sessions and a control condition. Finally, all participants underwent a post-assessment similar to the pre-assessment. Results: Analysis of covariance (ANCOVA) indicated significant reduction both in cognitive bias and relapse in the experimental conditions compared to the control condition. Post-hoc tests suggested no significant differences in relapse in the experimental groups. Furthermore, attention bias decreased in groups after Attention Bias Modification (ABM) and approach bias decreased in the groups which received Approach Avoidance Task (AAT-T). Conclusion: It can be suggested that different treatments of cognitive bias modification, by reducing cognitive bias and relapse, could be beneficial for methamphetamine abusers.