ScholarWorks@숙명여자대학교

초록

This study reviews domestic and international research on AI-based and multimodal feedback in English writing education, focusing on their key characteristics, learning effects, and pedagogical value. It explores how these feedback types help overcome the limitations of traditional feedback, such as time demands and lack of immediacy, while complementing each other to promote learner engagement. AI-based feedback, powered by natural language processing and machine learning, offers immediate, personalized, and repeatable responses that promote grammatical accuracy, language awareness, and self-regulated revision. Multimodal feedback, provided through audio, video, or screencast, combines verbal and visual cues to clarify meaning, increase motivation, and foster a stronger sense of connection between teachers and students. Integrating these two types of feedback can create a balanced feedback ecology that supports both cognitive precision and emotional engagement. Ultimately, their success depends on feedback literacy, the capacity of teachers and learners to interpret and apply feedback critically. By reframing feedback as an interactive and learner-centered process of reflection and knowledge construction rather than a corrective act, this study highlights the pedagogical importance of cultivating feedback literacy. It also calls for further research on hybrid feedback practices considering learner factors such as self-efficacy and digital competence across diverse contexts.

키워드