This paper reports on an investigation into the effect of highlighting metadiscourse markers on Iranian L2 learners’ reading comprehension of expository and argumentative texts. To this end, the researchersdivided 45 higher L2 proficiency learners and 45 lower L2 proficiency learners randomly into 3 groups, and gave 3 different versions of a reading comprehension test—metadiscourse-free, metadiscourse-rich, and metadiscourse-highlighted—to each group. Then, all the participants read the expository and argumentative texts and then answered the related questions. The results of a two-way ANOVA and an MANOVA indicated that metadiscourse markers and the participants’ level of L2 proficiency significantly influenced their reading comprehension, whereas genre, or the type of reading text, was not a significantly effective variable. Specifically, both the higher and lower L2 proficiency participants comprehended the metadiscourse-rich better than the metadiscourse-free versions and also comprehended the metadiscourse-highlighted versions the best. The results could have implications for L2 research and pedagogy.