This study investigated the differential effects of implicit and explicit instruction of refusal strategies in English and whether and how the impacts of instruction methods interact with learners' working memory capacity (WMC). 78 learners of English were assigned to three groups (explicit, implicit, and control). Implicit instruction was operationalized through input enhancement and provision of recast. In the explicit instruction group, participants received description and exemplification of refusal strategies and were provided with explicit corrective feedback. Prior to the treatment, all participants took WMC test, Discourse Completion Test (DCT) and completed a pragmatics comprehension questionnaire (CQ). Results revealed that explicit instruction was more effective than implicit instruction for both production and comprehension of refusals and that both implicit and explicit groups maintained the improvement in the delayed post-test administered two months later. In addition, whilst WMC scores were positively and strongly correlated with gains in the immediate and delayed post-test for both DCT and CQ in the implicit group, no meaningful relationship was found for explicit and control groups. The unique feature of this research is demonstrating that explicit instruction of refusal strategies equalizes learning opportunities for all learners with differential levels of WMC.