Mapping research on ChatGPT and LLMs in EFL writing: A scoping review with bibliometric analysis
TL;DR
The synthesis of the reviewed studies indicates that AI-assisted feedback is frequently associated with improvements in grammar, vocabulary, coherence, and learner engagement, and several studies also highlight challenges, including concerns about academic integrity, students’ overreliance on AI, and the necessity for teacher guidance.
Abstract
The present paper investigates the current trends of second language learning and teaching mediated by generative artificial intelligence (GenAI) tools, like ChatGPT and other large language models within the English as a foreign language (EFL) writing context. Scopus, which is a major database, was carefully researched using targeted keywords. All the inclusion and exclusion criteria were implemented consistently, and a total set of 43 articles were finally analyzed to synthesize the present review. The PRISMA method was implemented to ensure transparency and reliability, whereas the bibliometrix R-tool was utilized to generate clear and vivid visual depictions of the bibliometric data. This study adopts a complementary dual-method design, combining a scoping review approach with a bibliometric analysis to investigate the use of GenAI in EFL writing instruction. The scoping review serves as the primary analytical component of the study, identifying major pedagogical themes, applications, and challenges related to artificial intelligence-supported writing enhancement. The bibliometric analysis is used to contextualize the structure and evolution of the field by mapping its intellectual structure, publication evolution, influential publications, productive authors, institutional affiliations and keyword trends. The synthesis of the reviewed studies indicates that AI-assisted feedback is frequently associated with improvements in grammar, vocabulary, coherence, and learner engagement. Several studies also highlight challenges, including concerns about academic integrity, students’ overreliance on AI, and the necessity for teacher guidance. By systematically synthesizing these results, this review offers a clear picture of current practices, emerging trends, and research gaps, providing insights for researchers, educators and policymakers in the EFL domain. This study has certain limitations such as filtering constraints and strict time limits, but there are some useful guidelines that researchers can utilize in their future research.