This research compares how second language learners process two types of written feedback: reformulation and direct correction. On a two-stage composition-and-comparison task, 22 adult learners of English as a second language taking an academic writing course at a large midwestern U.S. university participated in a repeated-measures study in which they wrote two argumentative essays and received feedback in the form of reformulation on one and direct correction on the other in counterbalanced fashion. During the comparison stage for each essay, learners completed think-alouds, which were used to gauge how they processed the two types of feedback. The findings reveal that learners processed sentential and paragraph-level errors more deeply but overlooked surface-level errors when they received reformulation as feedback; the reverse was the case when they received direct correction. Results therefore suggest that there may not be a one-size-fits-all answer for written corrective feedback but that different errors respond to feedback differently.