AI-MEDIATED METACOGNITIVE SCAFFOLDING IN EFL WRITING INSTRUCTION: EVALUATING LLM FEEDBACK ON SYNTACTIC COMPLEXITY AND COHESION
Keywords:
applied linguistics, generative AI, metacognitive scaffolding, EFL writing, syntactic complexity, text cohesion, large language models.Abstract
This study explores how feedback from Large Language Models (LLMs) supports metacognitive processes in english as a Foreign Language (EFL) writing instruction. We evaluated the impact of AI feedback on sentence structure complexity and overall text cohesion in student essays. The findings indicate that AI-guided prompts help learners self-correct their errors and improve sentence flow. This leads to higher structural clarity and better logical flow in academic writing.
References
1. Crossley, S. A., & McNamara, D. S. (2014). Does second language writing quality depend on linguistic features? A meta-analysis. Journal of Second Language Writing, 24, 1-16.
2. Hyland, K. (2019). Second Language Writing. Cambridge University Press.
3. Warschauer, M., & Grimes, D. (2008). Automated writing evaluation: Defining the classroom research agenda. Language Teaching Research, 12(1), 22-49.


