The DOI matches its Crossref record; 66 of the 68 references it cites were found in a registry or an index, and 2 are not checked yet; none of the 47 sources its claims cite is checked for support yet.
Exploring English for academic purposes learners’ engagement with ChatGPT-4 in academic writing revision: a case study from a private language institute in China
Open accessAbstract
While a growing body of research has examined the affordances and effects of generative artificial intelligence (GenAI) in language learning, less attention has been paid to how learners engage with these tools during the writing revision process across behavioural, cognitive, and affective dimensions. This qualitative multiple-case study investigates how eight English for Academic Purposes (EAP) learners engage with ChatGPT-4.0 while revising academic writing. Data included screen recordings of learner–AI interactions, interaction transcripts, stimulated recall sessions, and semi-structured interviews. The findings reveal that learner engagement is heterogeneous and multidimensional. Behavioural and cognitive engagement ranged from deliberate, evaluative processing to more mechanical uptake, including instances of overreliance on AI-generated feedback. Affective engagement was similarly mixed: while participants valued the immediacy and accessibility of feedback, some reported a lack of social and contextual support compared to teacher guidance. These findings highlight the importance of teacher mediation and critical AI literacy in supporting meaningful engagement with GenAI. The study contributes to emerging research on AI-assisted language learning by specifying the conditions under which GenAI feedback may foster, rather than undermine, reflective and critical engagement in EAP writing.
Checked 2026-11-18 · 2 references not checked yet
Overview
CoChat Summary
TL;DR
This qualitative multiple-case study of eight Chinese EAP learners found heterogeneous behavioural, cognitive, and affective engagement with ChatGPT-4 during academic writing revision, ranging from deliberate evaluation to mechanical uptake and overreliance. It highlights teacher mediation and critical AI literacy as important for meaningful GenAI engagement.
Key findings
- Across the eight participants, 246 ChatGPT suggestions were incorporated into revised drafts and 70 were rejected; all participants accepted more suggestions overall than they rejected, and lower-order feedback was taken up more frequently than higher-order feedback (Table 4).
- Behavioural engagement varied: students spent an average of 57.8 min per revision session, with some reviewing and comparing suggestions before revising, while others accepted rapidly or mechanically—e.g., P6 described clicking “accept, accept” for everything and called the process “meaningless.”
- Cognitive engagement was mixed: some learners compared suggestions with prior knowledge or external sources and rejected them (e.g., P7 rejected an em dash-to-colon suggestion by citing EAP writing samples), with such responses observed in five of eight learners; others used ChatGPT for idea development, as when P4 used generated content to form a “white lie” example.
- Affective engagement was also mixed: P1, P3, P4, P5, and P7 reported reduced anxiety, confidence, or emotional support, while P1, P4, P5, P6, and P7 expressed concerns about overreliance or dependency; several participants (P1, P2, P4, P7, P8) referred to teacher guidance and valued step-by-step scaffolding.
Methods
- Qualitative multiple-case study with eight EAP learners at a private language institute in China; participants were aged 19–29, undergraduate/postgraduate, upper-intermediate to advanced in English, from Business, STEM, Health, and Humanities/Social Sciences, and recruited purposively after a one-hour ChatGPT seminar.
- Data collection had three phases: Phase 1 screen recordings of ChatGPT-supported revision plus drafts, interaction logs, and final essays (May 2025); Phase 2 delayed stimulated recall within 48 h and semi-structured interviews (May–July 2025); Phase 3 follow-up semi-structured interviews two weeks after main data collection (August–September 2025).
- Iterative thematic analysis used the triadic engagement framework; units were revision episodes for screen recordings (prompt, ChatGPT feedback, learner response) and meaning segments for stimulated recall/interviews, with two coding cycles, cross-data triangulation, and collaborative review of coding decisions.
Limitations
- Stated: the one-hour training seminar created a lightly scaffolded context, so observed engagement may partly reflect pedagogical framing rather than unprompted use; the study also had no baseline or comparison condition, so it cannot support definitive causal claims about anxiety reduction, writing improvement, or relative pedagogical value.
- Stated: the small, context-specific sample of eight learners from a single private institute limits generalizability, and cognitive/affective findings rely partly on retrospective stimulated recall and interviews subject to recall or social desirability biases.
- Evident from the text and acknowledged by the authors: uptake frequency is descriptive and does not indicate underlying cognitive or evaluative processing, so high observable acceptance can coexist with mechanical uptake.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures
Verification
Claims the paper makes
0 Valid · 4 Not valid · 34 Not checked · 95 Own results and interpretations
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For example, in P7’s revision of “nice birthday gifts” to “thoughtful birthday gifts,” the change followed a system-generated comment that the original wording was vague in an academic context.”
§ Exploring English for academic purposes learners’ engagement with ChatGPT-4 in academic writing revision: a case study from a private language institute in China > Results > Behavioural engagement
plainstates a causeSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In addition to these responses, the data also included expressions of frustration and concern. These responses were reported in situations where participants described a mismatch between their intended prompts and the system’s output.”
§ Exploring English for academic purposes learners’ engagement with ChatGPT-4 in academic writing revision: a case study from a private language institute in China > Results > Affective engagement
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on an assumption not shown here.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Analysis of stimulated recall and interview data showed that learners’ affective engagement with ChatGPT was complex rather than uniformly positive or negative.”
§ Exploring English for academic purposes learners’ engagement with ChatGPT-4 in academic writing revision: a case study from a private language institute in China > Results > Affective engagement
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on an assumption not shown here.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The analysis of screen recordings, interaction logs, and revised drafts showed that behavioural engagement was evident across all eight learners, although its form and intensity varied.”
§ Exploring English for academic purposes learners’ engagement with ChatGPT-4 in academic writing revision: a case study from a private language institute in China > Results > Behavioural engagement
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Analysis of stimulated recall (SR) and interview data showed variation in how learners processed, evaluated, and used ChatGPT-generated input during writing revision.”
§ Exploring English for academic purposes learners’ engagement with ChatGPT-4 in academic writing revision: a case study from a private language institute in China > Results > Cognitive engagement
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on an assumption not shown here.
64 more claims
Create a free account to see all 133 claims and to verify the 34 references still open.
- Probably not a substantial claim. 42
- It may state two separate claims as one. 37
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ Exploring English for academic purposes learners’ engagement with ChatGPT-4 in academic writing revision: a case study from a private language institute in China > Discussion > Pedagogical implications
“ChatGPT was used exclusively by the participants for revising their work, rather than as a tool for generating text.”
§ Exploring English for academic purposes learners’ engagement with ChatGPT-4 in academic writing revision: a case study from a private language institute in China > Results > Affective engagement
“ChatGPT immediately generated a full essay. Structurally, it was almost perfect—or so it seemed to me.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Exploring English for academic purposes learners’ engagement with ChatGPT-4 in academic writing revision: a case study from a private language institute in China > Results > Affective engagement
“In addition, several participants (P1, P2, P4, P7, P8) referred to their experiences with teacher guidance when describing their use of ChatGPT.”
§ Exploring English for academic purposes learners’ engagement with ChatGPT-4 in academic writing revision: a case study from a private language institute in China > Results > Affective engagement
“For example, P6 stated:”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Exploring English for academic purposes learners’ engagement with ChatGPT-4 in academic writing revision: a case study from a private language institute in China > Results > Behavioural engagement
“As shown in Table 4, all participants accepted more suggestions than they rejected across both higher-order concerns (HOCs) and lower-order concerns (LOCs).”
§ Exploring English for academic purposes learners’ engagement with ChatGPT-4 in academic writing revision: a case study from a private language institute in China > Results > Behavioural engagement
“| P1 | 25 | 0 | 6 | 14 | 31 | 14 | 45 |”
Citations
Cites
School Engagement: Potential of the Concept, State of the Evidence
Fredricks, J. A., Blumenfeld, P. C., & Paris, A. H. (2004) · Review of Educational Research · doi:10.3102/00346543074001059 · Cited by 12,581 · cited 4×
“Much of the literature conceptualises cognitive engagement as a multidimensional construct encompassing learners’ investment in deep processing, strategic regulation, and sustained attention during learning tasks (Fredricks et al., 2004; Greene, 2015).”
InfluentialChatGPT for good? On opportunities and challenges of large language models for education
Kasneci, E., Sessler, K., Küchemann, S., Bannert, M., Dementieva, D., Fischer, F., Gasser, U., Groh, G., Günnemann, S., Hüllermeier, E., et al. (2023) · Learning and Individual Differences · doi:10.1016/j.lindif.2023.102274 · Cited by 7,326
Purposive sampling: complex or simple? Research case examples
Campbell, S., Greenwood, M., Prior, S., Shearer, T., Walkem, K., Young, S., Bywaters, D., & Walker, K. (2020) · Journal of Research in Nursing · doi:10.1177/1744987120927206 · Cited by 3,772
Chatting and cheating: Ensuring academic integrity in the era of ChatGPT
Cotton, D. R. E., Cotton, P. A., & Shipway, J. R. (2024) · Innovations in Education and Teaching International · doi:10.1080/14703297.2023.2190148 · Cited by 3,087
Ray, P. P. (2023) · Internet of Things and Cyber-Physical Systems · doi:10.1016/j.iotcps.2023.04.003 · Cited by 2,463
Students’ voices on generative AI: perceptions, benefits, and challenges in higher education
Chan, C. K. Y., & Hu, W. (2023) · International Journal of Educational Technology in Higher Education · doi:10.1186/s41239-023-00411-8 · Cited by 2,301
Multiple goals, multiple pathways: The role of goal orientation in learning and achievement
Pintrich, P. R. (2000) · Journal of Educational Psychology · doi:10.1037/0022-0663.92.3.544 · Cited by 2,254
ChatGPT for Language Teaching and Learning
Kohnke, L., Moorhouse, B. L., & Zou, D. (2023) · RELC Journal · doi:10.1177/00336882231162868 · Cited by 1,371
Feedback on second language students' writing
Hyland, K., & Hyland, F. (2006) · Language Teaching · doi:10.1017/s0261444806003399 · Cited by 1,016
Does Training on Self-Regulated Learning Facilitate Students' Learning With Hypermedia?
Azevedo, R., & Cromley, J. G. (2004) · Journal of Educational Psychology · doi:10.1037/0022-0663.96.3.523 · Cited by 859
Research works printed but not linked: 7
- Ashford, 2003, Reflections on the looking glass: a review of research on feedback-seeking behaviour in organizations, J. Manage., 29, 773, 10.1016/s0149-2063 · doi:10.1016/s0149-2063
- Bond, 2022, Handbook of Open, Distance and Digital Education, 1
- Du, 2025, 2025, the 7th International Conference on Computer Science and Technologies in Education (CSTE), 856
- Gass, 2000, Stimulated Recall Methodology in second Language Research
- Schunk, 2018, Handbook of Self-Regulation of Learning and Performance, 1
- Zhang, 2018, Student engagement with the teacher and automated feedback on L2 writing, Assess. Writing, 36, 90, 10.1016/jasw.2018.02.004 · doi:10.1016/jasw.2018.02.004
- Zimmerman, 2002, Becoming a self-regulated learner: an overview, Theory Pract., 41, 64, 10.1207/s15430421tip4102 · doi:10.1207/s15430421tip4102
Cited by
Du, J., Liu, Z., & Liao, N. (2026) · Acta Psychologica · doi:10.1016/j.actpsy.2026.107471 · Cited by 3
Self-citationLi, M. (2026) · Journal of English Language Teaching and Applied Linguistics · doi:10.32996/jeltal.2026.8.9.1 · Cited by 2
Educational Chatbot-Supported Academic Writing in Higher Education: A Rapid Systematic Review
Bommanaboina, L. D., & Bommanaboina, R. D. (2026) · Revista de educación y derecho · doi:10.1344/reyd2026.34.56975 · Cited by 1
Sheng, N. Y., & Yaccob, N. S. (2026) · International Journal of Research and Innovation in Social Science · doi:10.47772/ijriss.2026.100601289
Tao, Y., Meyer, J., & Kizilcec, R. F. (2026) · arXiv · doi:10.48550/arxiv.2609.23936
Metadata
- Authors
- Du, J. L., & Chen, Y.
- Affiliations
- University of Otago (NZ)
- Chongqing University of Engineering
- Corresponding author
- Chen, Yiran
- Venue
- Frontiers in Artificial Intelligence
- Publisher
- Frontiers Media SA
- Published
- 2026-06-01
- Volume
- 9
- Pages
- 1802007-1802007
- ISSN
- 2624-8212 (ISSN-L)
- PMC
- PMC13267182
- OpenAlex
- W7163021995
- Cited by
- 6
- References
- 66
- Citations by year
- 2026: 6
- FWCI
- 41.76
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Frontiers in Artificial Intelligence (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- Chongqing Municipal Education Commission (KJZD-M202501901)
- Topics
- Social Sciences › Arts and Humanities › Language and Linguistics › EFL/ESL Teaching and Learning
- Social Sciences › Education › Artificial Intelligence in Education
- Social Sciences › Education › English Language Learning and Teaching
- Keywords
- ChatGPT-4, generative AI, English for academic purposes, student engagement, AI-assisted language learning, AI-generated feedback, teacher mediation, critical AI literacy, overreliance on AI, emotional engagement, cognitive engagement, behavioral engagement, private language schools
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-11-18, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-11-18 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Exploring English for academic purposes learners’ engagement with ChatGPT-4 in academic writing revision: a case study from a private language institute in China. https://app.cochat.ai/papers/10.3389/frai.2026.1802007
- The paper
- Du, J. L., & Chen, Y. (2026). Exploring English for academic purposes learners’ engagement with ChatGPT-4 in academic writing revision: a case study from a private language institute in China. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1802007
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.

