중급 한국어 학습자의 듣기 이해 처리 유형 연구 ― 상향식·하향식 처리 양상에 따른 집단 비교 ―

초록

This study classifies learner types based on differences in bottom-up and top-down processing patterns among intermediate Korean-language learners and examines the characteristics and educational needs of each type. It is intended to demonstrate that learners at the same proficiency level can exhibit different listening comprehension processing patterns, thereby providing foundational data for learner-centered Korean listening instruction. To this end, a survey was administered to 125 Level-3 Korean learners. Listening difficulties were categorized into bottom-up and top-down processing domains. Learners were classified into four groups based on the median values of each domain: low difficulty in both domains, high difficulty in bottom-up processing only, high difficulty in top-down processing only, and high difficulty in both domains. The study then analyzed learners’ perceptions of listening class effectiveness and post-listening activities helpful for improving listening ability. The study determined that the four groups perceived class effectiveness and post-listening activities differently. These differences primarily manifested themselves in ways that compensated for the processing areas in which the learners experienced relative difficulty. The findings suggest that even learners at the same intermediate level may perceive different instructional activities and support strategies as effective depending on their listening comprehension processing patterns. Therefore, Korean listening education should consider these individual differences to provide differentiated instructions and scaffolding. Furthermore, generative artificial intelligence may expand the potential of personalized listening instruction by providing tailored activities and feedback according to learners’ processing patterns. This study contributes to the field by treating listening difficulties not simply as differences in proficiency but as variations in listening comprehension processing patterns. However, because learner types were identified based on self-report questionnaires, directly reflecting actual listening performance processes was one area of limitation for this study.

키워드