Original Research
Optimizing EFL Article Accuracy with Hybrid AI-Teacher Feedback
- Abstract
- Full text
- Metrics
This study examines the effectiveness of various feedback models in enhancing grammatical accuracy in English article usage among 140 Iranian EFL medical students. Using a randomized experimental design, participants were assigned to five groups: AI feedback alone (n=28), AI with immediate oral teacher feedback (n=30), AI with delayed written teacher feedback (n=35), a hybrid model combining AI with both teacher feedback types (n=24), and a control group receiving traditional feedback (n=23). To evaluate performance, we administered pre-tests, post-tests, and delayed post-tests, while a post-study survey gauged learner attitudes. The data revealed that the hybrid model yielded the most substantial improvements in accuracy and long-term retention (post-test M=21.83; delayed post-test M=21.29). The next most effective condition was AI coupled with immediate oral feedback. Notably, AI feedback used in isolation surpassed AI with delayed written feedback, underscoring the crucial role of immediacy in learning. Furthermore, the survey showed a strong positive correlation between learner attitudes and performance (r=.32, p<.001), with the hybrid group expressing the most positive perceptions. These findings highlight the pedagogical value of integrating AI tools with timely, multi-modal teacher feedback to foster linguistic development and learner engagement in EFL contexts.
Page 1 of
Download Count : 16
Visit Count : 135
Keywords
AI Feedback; EFL learners; Hybrid Feedback; Grammatical Accuracy; Learner Attitudes
Author(s) Information
Publisher’s Note
The claims, arguments, and counter-arguments made in this article are exclusively those of the contributing authors. Hence, they do not necessarily represent the viewpoints of the authors’ affiliated institutions, or EUROKD as the publisher, the editors and the reviewers of the article.
Acknowledgements
The authors would like to express their sincere gratitude to the participants who took part in this study and to the instructors who supported the implementation of the research procedures. We also acknowledge the support of Central Asian University for providing ethical approval and facilitating the research process.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
CRediT Authorship Contribution Statement
Afshin Soori: Conceptualization, Methodology, Investigation, Formal Analysis, Writing – Original Draft
Laleh Khojasteh: Data curation, Methodology, Supervision, Writing – Review & Editing, Project Administration
Sedigheh Shakib Kotamjani: Writing – Review & Editing, supervision- Funding Acquisition
Sanaz Mojarrad: Data Curation, Investigation, Visualization, Writing
Generative AI Use Disclosure Statement
Generative AI tools (Grok-4 by xAI) were used solely for basic text editing, minor copyediting, and formatting assistance consistent with standard word processing functions (e.g., Microsoft Word grammar/style checks). No AI was involved in generating content, analyzing data, creating code, drafting text, refining research design, or producing figures/tables/images. All content reflects the original authors’ work.
Ethics Declarations
World Medical Association (WMA) Declaration of Helsinki–Ethical Principles for Medical Research Involving Human Participants
This study was conducted in accordance with the ethical standards outlined in the World Medical Association Declaration of Helsinki (2013). Ethical approval was obtained from the Institutional Review Board (IRB) of Central Asian University (Ethical Review Code: CAU-IRB-2025-032). All participants were informed about the purpose of the study, and informed consent was obtained before participation. Participation was voluntary, and all data were anonymized to ensure confidentiality. The study involved no physical or psychological risk to participants.
Competing Interests
The authors declare that there are no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability
The datasets generated and/or analyzed during the current study are not publicly available due to ethical and privacy considerations but are available from the corresponding author on reasonable request. All data have been anonymized to protect participants’ confidentiality.