Although Generative AI (GenAI) has the potential to create materials for EAP students to learn academic vocabulary, this potential has received little empirical investigation. To address this gap, this study developed three ChatGPT corpora of extensive reading materials targeting EAP learners at different vocabulary levels. Each corpus was evaluated against key conditions needed for vocabulary through extensive reading (comprehensibility, repetition, spaced retrieval, varied encounters, and elaboration). The results indicate that ChatGPT can be a useful resource for generating extensive reading materials to support academic vocabulary learning among EAP learners. However, to optimise learning outcomes, these materials need further revision and pedagogical mediation from teachers.
In this research-into-practice paper, we examined how 14 postgraduate EAP student teachers in Hong Kong engaged with generative artificial intelligence (GenAI) tools to develop teaching materials during a professional practice course. Drawing on reflective journals, other participant-generated materials, and stimulated recall interviews, we traced the participants' developmental trajectory from tentative experimentation to context-sensitive integration. The findings were interpreted through the technological pedagogical content knowledge (TPACK) and substitution augmentation modification redefinition (SAMR) frameworks. Three interconnected themes emerged: developing GenAI confidence and technical competence, renegotiating the EAP teacher's role, and contextualising AI outputs for Chinese EAP classrooms. The findings suggested that EAP teacher education programmes should integrate GenAI training into pedagogical and content knowledge development. They should provide structured opportunities for experimentation, peer collaboration, and reflection, and address contextual adaptation and ethical transparency.
Narrative structure, also known as the shape of stories, refers to the way a story is organized and presented from a semantic progression perspective. In academic writing, high-quality research articles are often metaphorically described as well-told academic stories, as narrative devices can present complex ideas in a clear and engaging manner. However, the role of the narrative structure in academic writing remains underexplored. To address the issue, we examined the narrative structure in research articles and its relationship with citation counts from three dimensions (i.e., speed, volume, and circuitousness) and in different disciplines. The results revealed significant cross-disciplinary differences in the narrative structure of research articles. Specifically, articles in medical and life sciences exhibited the highest levels across all three dimensions, while those in social sciences demonstrated the lowest speed and volume, and articles in applied sciences showed the lowest circuitousness. Furthermore, speed was negatively related to citation counts, whereas both volume and circuitousness were positively related to citation counts, except that volume showed a negative relationship in medical and life sciences. To the best of our knowledge, this study is probably the first cross-disciplinary investigation of the narrative structure in academic writing. Possible explanations for the findings were discussed. Implications for researchers and instructors of academic writing were also provided.
The challenges of English for Academic Purposes (EAP) lie not only in mastering vocabulary and grammatical rules, but also in acquiring discipline-specific patterned language. While causality is a core element of academic reasoning, the recurrent lexico-grammatical patterns expressing causality remain inadequately addressed. This study adopts local grammar and construction grammar approaches to explore causal expressions in academic discourse. By integrating the descriptive granularity of local grammar with the structural systematicity of construction grammar, the research provides a bottom-up method for identifying causal constructions. Using a self-constructed corpus of research articles in applied linguistics, the study systematically identifies 24 local grammar patterns that realize causal functions. These micro-level patterns are then abstracted into five meso-level causal constructions, which together form a hierarchically organized and interconnected constructional network. The findings reveal that causality in academic writing is instantiated through highly conventionalized pattern-function pairings and show how local grammar patterns can provide a data-driven basis for identifying causal constructions. Pedagogically, the resulting inventory supports the development of a more systematic and function-oriented approach to teaching causal language in EAP.
The rapid uptake of Generative AI (GenAI) in higher education is reshaping how international students engage with academic work in L2 English, including how they navigate academic challenges. However, the implications of GenAI for academic literacies remains underexplored. This study investigates how international MA TESOL students in the UK engage with GenAI as a mediational tool in response to academic English challenges. Grounded in sociocultural theory and a practice-oriented conceptual framework which combines academic literacies and tool-mediated activity, the study employs a mixed-methods design. Survey data were collected from 82 students, and semi-structured interviews were conducted with 8 students and 5 lecturers from MA TESOL programmes. Findings indicate that MA TESOL students' academic challenges extend beyond L2 English proficiency to encompass a range of academic discourse practices, including academic writing, critical thinking, and understanding and meeting assessment requirements. Furthermore, findings provide rich descriptions of how students use GenAI tools to mediate critical thinking and academic engagement. However, lecturers raised concerns that students' GenAI use was masking academic challenges, thus highlighting a potential tension in the effectiveness with which students are mobilising GenAI as a mediational tool. By situating students’ challenges within an academic literacies framework, this study shows that GenAI use is not merely a coping strategy but a tool-mediated activity that has the potential to reshape how academic literacies are manifested in L2 English higher education.
While the potential of generative AI (GenAI) in supporting EAP reading assessment materials development has been widely recognized, its application is accompanied by notable challenges due to the iterative nature of materials development and the inherent complexity of GenAI. This highlights the importance of teacher resilience, a capacity to manage challenges and adapt to evolving conditions. Yet, how teachers demonstrate resilience in addressing such GenAI-related challenges and how their resilience develops over time remain underexplored. This semester-long case study examined the development of resilience among four teachers who used GenAI to design EAP reading assessment materials, and investigated the factors that influenced the development. Data were collected from multiple sources, including human-GenAI interaction logs, drafts and final versions of assessment materials, and semi-structured interviews. Three key phases of resilience development were identified: (1) reactive resilience for instrument adaptation, (2) proactive resilience for pedagogy optimization, and (3) integrative resilience for mutual empowerment. In each phase, the teachers mobilized various dimensions of resilience (i.e. motivational, professional, emotional and social). The protective and risk factors influencing teachers’ resilience development were also discovered across individual (e.g. intrinsic motivation, sense of responsibility, and low prompt literacy), interpersonal (e.g. feedback from students, support from colleagues, and unwillingness to seek help), and organizational (e.g. professional training programmes, lack of sustained institutional communication mechanisms, and non-teaching duties) levels. These findings have implications for cultivating teacher resilience and contextualizing the use of GenAI for effective EAP reading assessment materials development.
Explicit instruction in characteristics of academic language can be helpful for students in preparing for reading academic texts in English as a second language (L2) and familiarity with academic language is an explicitly stated goal in the curriculum for English in the study preparatory program in upper secondary schools in Norway. English-language texts constitute a large proportion of the course materials used in Norwegian universities, but research suggests that many students struggle to read academic texts in English. This study explores how academic L2 English is taught to study preparatory students, based on findings from an online survey of 73 upper secondary English teachers across Norway. We focused on teachers’ conceptualizations of academic language and academic vocabulary and how these concepts are reported to be implemented in teaching practices. While 96% of teachers reported confidence in their understanding of academic language, most defined it broadly as ”formal language” or focused on salient text features, like lack of contracted forms or personal pronouns, which suggests limited understanding. The teachers acknowledged the challenges academic texts pose for students and their own role in preparing students for university reading, but few reported having received training or further development in teaching academic language. The findings suggest that providing more training for teachers could promote a clearer focus on academic language, helping study preparatory students to build on their academic English skills to better prepare for university reading.
The rapid advancement of generative artificial intelligence (GenAI) is reshaping the ways in which English for Academic Purposes (EAP) course materials are developed, adapted and evaluated. Appropriate and effective collaboration with GenAI in EAP course material design has been receiving significant scholarly and practical attention. This case study proposes a design thinking-informed pedagogical design framework for EAP material development with GenAI (DT-GenEAP), mapping human-AI collaboration within an iterative instructional design process for EAP materials. This study also demonstrates the implementation of DT-GenEAP in the design of course materials for English Academic Writing in the Digital Environment. A subset of the course materials was piloted in the Test stage, and students' feedback was collected through student journals and classroom observations to inform revisions to the outputs of the other design stages. The facilitative impacts of human–AI collaboration within the DT-GenEAP framework on EAP instructional design and material development are discussed, along with the importance of maintaining teachers' oversight of their collaboration with GenAI. Implications for teachers’ professional development for effective and ethical collaboration with GenAI are offered.
Graduate L2 students frequently experience difficulty initiating and sustaining research proposal writing, often manifesting as observable drafting disruption. This Research into Practice paper reports on a staged instructional workflow implemented in a graduate EAP writing course in Taiwan, where selected NotebookLM features were integrated as pedagogical scaffolds to support proposal development. The workflow was organized around four writing stages: planning, monitoring, reflecting, and revising, and incorporated explicit guardrails for responsible AI use. Data documenting the practice included proposal drafts, weekly learning logs, reflective essays, and an end-of-course survey. Drawing on classroom evidence, the paper illustrates how instructor-designed drafting prompts supported task initiation, Audio Overviews (auto-generated podcasts) facilitated comprehension and re-engagement, and study guides, FAQs, and briefing documents aided synthesis and organization. The practice illustrates how AI-supported scaffolding helped students sustain writing momentum and develop clearer proposal structures. Implications are discussed for EAP teachers seeking to integrate generative AI tools into graduate writing instruction in pedagogically purposeful ways.
Plain Language Summaries of Publications (PLSPs) are articles with a visually enhanced format that provide an extended summary of a published research article. The purpose of this study was to examine how medical research articles are recontextualized into PLSPs to make recent research findings accessible to non-specialist audiences. Although there has been research on text-only Plain Language Summaries (PLSs), the orchestration of semiotic resources in PLSPs to convey meaning has not been explored. In this study, a multimodal approach to genre analysis was adopted to analyze the move structure of a corpus of 50 medical PLSPs and examine the multimodal realization of these moves. The move analysis reveals a relatively stable rhetorical structure across PLSPs, with moves that prioritize information relevant to the non-specialist reader. The multimodal analysis shows how semiotic resources, including written language, typography, color, and visuals, are orchestrated to facilitate comprehension. Visuals, ranging from diagrams to data displays and micro-visuals, serve a key role in conveying meaning in PLSPs. They are combined with the verbal text in relations of concurrence and complementarity to enhance accessibility and support narrative flow. These findings suggest that EAP instruction must expand beyond traditional academic writing and include training in plain language translation and multimodal literacy.
Despite growing interest in L2 student feedback agency in higher education, how it is constructed through feedback discourses remains under-researched. To address this gap, this qualitative study drew on a discursive agency perspective to examine how three feedback discourses constructed the feedback agency of three Chinese taught master's students in disciplinary writing at a New Zealand university. Data were collected from teacher feedback on student writing, feedback-related documents (e.g., assignment guidelines, exemplars, and emails), and student-generated data, including informal meetings, reflective diaries, and semi-structured interviews. The findings reveal that (1) L2 student feedback agency was constructed through feedback discourse, as reflected in the structure of feedback practices, opportunities for agency, and student positioning; (2) although the three discourses were all teacher-led, they shaped L2 student feedback agency differently, with opportunities for agency ranging from largely foreclosed to only implicitly available to more openly afforded; and (3) participants' uptake of these positionings varied across the three cases, taking the form of compliance, partial negotiation, and partial uptake. The study also identifies six feedback opportunities that may support L2 student agency: topic choice, peer review, varied tasks, open assessment criteria, chances for clarification, and mitigated feedback. The study concludes with implications for feedback design and student agentic engagement with feedback in disciplinary writing in higher education.
English for Academic Purposes (EAP) aims to develop the language skills necessary for academic achievement, frequently incorporating instruction in both academic and discipline-specific vocabulary. Additionally, EAP facilitates the acquisition of academic competencies such as writing, with particular emphasis on morphosyntactic structures. In the context of English Language Teaching (ELT), gamification has been employed to enhance learner motivation and promote communicative learning. Informed by Self-Determination Theory (Deci & Ryan, 1985), gamification aims to foster intrinsic motivation by addressing learners' needs for competence, autonomy, and relatedness. Empirical studies suggest that gamified activities can increase engagement and improve educational outcomes and may also facilitate vocabulary acquisition when implemented with appropriate digital tools and methodologies (Rabea & Abdalgane, 2022; Chen et al., 2019; He et al., 2023). Digital Educational Escape Rooms (DEERs) constitute a form of gamified learning facilitated by specialised software and, in certain instances, immersive technologies. Nevertheless, such technologies can introduce financial and technical challenges. The present study investigates a DEER delivered exclusively via Google Forms, a cost-effective platform that supports the integration of text, images, audio, and video. Employing an action research design, this study explored the relationship between participation in an escape-room activity and students' performance on morphosyntactic assessments. Final morphosyntax test results were compared between participants and non-participants using both descriptive and statistical analyses.
This study investigates the use of interactional metadiscourse in academic writing based on a corpus of 480 research articles drawn from four disciplines. Using a clustering approach, the analysis identifies five distinct uses of interactional metadiscourse characterized by different distributions and combinations. The resulting clusters show partial disciplinary concentration alongside substantial cross-disciplinary overlap, suggesting that disciplinary convention remains important but does not fully account for the distribution of interactional metadiscourse across texts. Theoretically, our results complement previous discipline-based research by revealing recurring text-level distributional patterns and within-discipline heterogeneity. Pedagogically, our findings highlight the need for a repertoire-based approach to academic writing instruction, which encourages novice writers to identify and strategically deploy diverse writing strategies. Methodologically, our results illustrate the value of cluster analysis in revealing meaningful patterns of variation that remain obscured under traditional frequency-based comparisons.
