Research Article

A Conceptual Framework for LLM-Based Digital Scaffolding in Academic Writing: Adaptive Support for EFL Learners

Authors

  • EBTESAM GHALI T. ALANEZI English Language and Translation Department, College of Arts, University of Hafr Al Batin, Saudi Arabia

Abstract

Large language models (LLMs) are currently having an impact on English as a Foreign Language (EFL) learners’ academic writing development and performance across planning, drafting, revision, feedback incorporation, and self-assessment processes. Although several recent studies have provided empirical findings on learners’ perceptions of AI-assisted writing, the nature of LLM feedback, AI’s impact on learners’ academic honesty, etc., scant research has focused on explaining how LLM-based support can be sequenced and structured as digital scaffolding for learners’ independent writing versus open-ended LLM text generation. To address this gap, the current article provides an explanatory framework for designing LLM-based digital scaffolding support specifically for EFL learners’ academic writing development. Employing a conceptual–methodological approach to analysis, this study conceptually reviews and synthesizes relevant recent scholarship on digital scaffolding, sociocultural learning, cognitive load theory, self-regulated learning, prompt engineering, adaptive feedback design, teacher-mediated LLMs, and learners’ ethical and effective use of AI. Due to the nature of the study’s focus on conceptual framework development, no empirical sample of participants was surveyed or observed. The final proposed framework was built through a synthesis of the previous literature and a construction of novel pedagogical models. Specifically, the framework suggests integrating learner–task–stage profiling with three-layer scaffolding (pre-writing, during-writing, and post-writing), calibrated prompting, adaptive feedback, reflective revising, teacher mediation, and ethical safeguards. The proposed explanatory framework provides researchers and teachers with a theoretically based pedagogical model to use LLMs to empower learner agency, metacognitive monitoring, academic integrity, and gradual release of responsibility in EFL academic writing.

Article information

Journal

International Journal of Linguistics, Literature and Translation

Volume (Issue)

9 (11)

Pages

229-245

Published

2026-11-14

How to Cite

EBTESAM GHALI T. ALANEZI. (2026). A Conceptual Framework for LLM-Based Digital Scaffolding in Academic Writing: Adaptive Support for EFL Learners. International Journal of Linguistics, Literature and Translation, 9(11), 229-245. https://doi.org/10.32996/ijllt.2026.9.11.21

Downloads

Views

27

Downloads

10

Keywords:

Large language models, digital scaffolding, EFL academic writing, adaptive feedback, prompt engineering, learner autonomy