BackgroundReconsolidation-based interventions have been suggested to be a promising treatment strategy for substance use disorders. In this study, we aimed to investigate the effectiveness of a working memory intervention to interfere with the reconsolidation of alcohol-related memories in a sample of non-treatment seeking heavy drinkers.MethodsParticipants were randomized to one of the two conditions that underwent a 3-day intervention: in the experimental condition, a 30-min working memory training was performed immediately after a 15-min memory retrieval session (i.e., within the memory reconsolidation time-window), whereas in the control condition, the working memory training was performed prior to a memory retrieval session.ResultsIn contrast to our original hypothesis, a high working memory load after memory retrieval did not interfere with the reconsolidation of those memories while a high working memory load prior to memory retrieval (the original control condition) strongly reduced retrieval-induced craving and craving for alcohol at follow-up.ConclusionWhereas the neurocognitive mechanism behind this effect needs to be further investigated, the current findings suggest that, if replicated, working memory training prior to addiction-related memory retrieval has the potential to become an effective (adjunctive) intervention in the treatment of substance use disorders.Electronic supplementary materialThe online version of this article (10.1007/s00213-017-4785-4) contains supplementary material, which is available to authorized users.