This is a case study dealing with the relationship between consciousness-raising (CR) grammar tasks and noticing, replicating the work of Fotos (1993). Using only a reading and underlining task as a measure of noticing in order to successfully capture noticing as a complex cognitive process is inadequate because the way in which learners process target grammar structures can vary greatly depending on their level of awareness. Therefore, post-experimental stimulus-cued retrospective interviews were added to obtain both quantitative and qualitative detailed information on processing input. Nine post-secondary ESL learners who studied in a university preparation course participated in this study. Present and past counterfactual conditionals were chosen as target grammar structures. The participants were randomly divided into two groups: a direct CR group given a teacher-centered deductive lesson and an indirect CR group where the students learned the target structure inductively in a dyad style. Pre- and post-Grammatical Judgment Tests (GJTs) were also administered to assess the effectiveness of these two methods of instruction. One week after the treatment, each student engaged in a reading and underlining activity immediately followed by a retrospective interview. Although no quantitative difference was found between the two groups in noticing and GJT scores, qualitative data revealed that the learners processed the underlined target structures very differently.