AI로 재구성된 수준별 텍스트를 활용한 고등학교 영문학 수업 효과 연구

Utilizing AI-Generated Leveled Texts in EFL Literature Class

This study aims to explore the use of AI to adapt English literature for easier comprehension in EFL classes, addressing the challenge learners face with the complexity of literary texts. By utilizing AI-assisted learning tools, specifically automatic text generation capabilities of ChatGPT, complex literary works are paraphrased into simplified versions. This study tested the effectiveness of AI-generated leveled texts through a case study involving twelve high school students reading various levels (1, 2, 3) of the adapted novel Wonder over three weeks. Students engaged in pleasure reading and evaluated the texts on aspects such as flow, coherence, difficulty, and engagement. The experiment’s findings indicate that the level-adapted texts were highly beneficial, offering more advantages than drawbacks. Students reported increased immersion and enjoyment, and even attempted more challenging texts. The adaptation’s effectiveness varied by proficiency; upper-level learners used simplified texts as supplementary materials, middle-level students were more engaged, and lower-level students found even the simplest texts challenging. Overall, students reported increased self-efficacy and motivation, suggesting that AI-modified texts enhance reading speed and comprehension. Despite its benefits, limitations in AI-generated leveled texts were recognized, pointing to potential areas for further refinement. This approach offers a promising avenue for making English literature more accessible to EFL learners, expanding educational possibilities for those intimidated by original literary works.