The current review of the role of executive function (EF) in reading provides a brief summary of analyses with a large-scale longitudinal dataset and a meta-analysis, along with proposing a framework for designing EF training studies. The 1st study, based on latent growth models with structured residuals, demonstrated a longitudinal reciprocal relation between reading and EF in Grades 2-5, but only for high-performing students and not for the general population sample or those with reading difficulties. The 2nd study, based on meta-profiling analysis and meta-analytical structural equation modeling of 378 studies, showed unique contributions of EF to reading difficulties after controlling for language skills and suggested a reading difficulty-EF deficit vicious cycle with development. The 3rd component of this review describes a framework for a domain-specific approach to intervention. Based on the findings of these studies, I propose that working memory training tasks should closely link the central executive (attentional control) with the use of long-term memory through retrieval practice in a specific academic domain, and training tasks should promote strategy use that can be effectively applied (i.e., transferred) to different academic tasks.