Sessions / Location Name: Room E308
Location not set by organizers
Location:
Building: Lecture Hall Building < Tokyo University of Science, Katsushika Campus
Personalized Vocabulary Quizzes with Google Sheets #4343
Learners should be able to study the vocabulary that they personally want or need to learn, but assessing students’ individual vocabulary can be difficult. This presentation demonstrates how the use of Apps Script with Google Sheets creates a system in which students can maintain their own personal vocabulary lists, and teachers can generate individualized vocabulary quizzes for students within minutes. The system involves students recording their individual vocabulary items in a personal Google Sheet. Items are then aggregated in a teacher's sheet, after which the teacher can click one button and generate 15-item quizzes and corresponding answer sheets. The quizzes take students three minutes to complete, and classmates can check each other’s quizzes using the answer sheets. The system will be fully demonstrated during the presentation, and there will be time at the end to discuss how session attendees might adapt the system to their own teaching contexts.
Fluid Language Pedagogy #4254
JALTCALL 2024 Keynote Speaker
Many experts in AI predict that AI will transform education, particularly through personalized learning. But what does this look like in real educational settings, especially within the structure of a classroom? How can AI-driven personalization be integrated meaningfully into language instruction without undermining the role of teachers or peer collaboration?
This paper introduces a new AI-enhanced instructional framework called Fluid Language Pedagogy (FLaP), designed to complement (not replace) classroom-based language learning. FLaP supports a hybrid learning environment in which students engage with AI chatbots and AI-generated materials tailored to their individual proficiency levels and goals, while still participating in classroom interaction, group work, and teacher-guided activities. The “fluid” aspect of FLaP refers to its adaptable structure: learners can move between self-directed AI-supported practice and collaborative classroom engagement, allowing for flexible learning pathways.
A core feature of FLaP is the use of customized large language models (LLMs) for different learner levels. We developed novice and intermediate LLMs for Japanese learners using prompt engineering—a process easily adaptable to other languages. This paper details the structure of FLaP, the development of level-specific LLMs through prompt writing, and its classroom implementation through a sample syllabus and integrated learning activities.
Developing EAP/ESAP Reading Materials with GenAI: A Boon or a Bane? #4302
With the affordances offered by GenAI (Generative AI) in nearly every aspect of language learning and teaching—including material development and assessment—both novice and experienced educators can now create tailored materials for their learners with remarkable ease. This shift has raised questions about the necessity of relying on established ELT (English Language Teaching) textbooks, which traditionally provide a well-structured curriculum and high-quality content. Drawing on his experience in developing in-house EAP (English for Academic Purposes) and ESAP (English for Specific and Academic Purposes) materials at an EMI (English as a Medium of Instruction) university, the presenter will examine the challenges of using GenAI to create reading texts and the relevant learning tasks that accompany them. These challenges encompass text readability, the frequent content overlap encountered in the question or task types, and the necessary review and revision for enhancing the quality and suitability of the generated materials. While potential solutions will be shared, one of which involves prompt writing to generate varying readability and task difficulty levels, the ultimate aim of the talk is to encourage reflection on the impact of GenAI on EAP/ESAP material development.
Leveraging ChatGPT and Grammarly for Automated Text Analysis: A Validation Study #4263
Text analysis plays a crucial role in understanding language proficiency. Evaluating the complexity, accuracy, and fluency (CAF) of texts, for example, is a common method to assess and compare language learners' proficiency under various conditions. Traditionally, such analyses have required many hours of tedious manual labor on the part of researchers and their assistants. This study investigates the validity and reliability of incorporating the free online tools ChatGPT and Grammarly to fully or partially automate these tasks. The study involved 55 students at a national university in Japan and analyzed one-paragraph summaries they wrote of a textbook listening script. Strong, significant correlations emerged between human and AI-derived measures for syntactic accuracy and syntactic complexity, as well as content comprehensiveness. These results support the efficacy of AI-based technologies for text analyses that have hitherto required human raters. In addition to the results and their implications, the presentation will detail the methods utilized for the human and AI-based measures, enabling attendees to obtain similar results. The findings indicate a widening role for AI in language research and education.
AI-Enhanced Teacher Development for Bilingual Course Design #4210
This study examines the feasibility of an AI-enhanced professional development programme designed to support EFL secondary teachers in bilingual course development in Taiwan. Employing action research methodology, the programme was implemented across six in-service teacher cohorts (N=137), with training duration varying between six hours (n=62) and two hours (n=75). The training framework incorporated lesson planning, assessment design, and evaluation measures, utilizing a Padlet platform that provided structured AI tool prompts and exemplar materials for arts and music courses. Data collection comprised group-based bilingual lesson planning products, overall survey, and researcher reflective journals. Preliminary findings suggest the programme's potential feasibility, with participants successfully completing group-based bilingual course designs despite limited prior experience with AI-enhanced pedagogy. Participants demonstrated considerable willingness to incorporate this approach in their future individualized planning. Training duration emerged as a crucial variable, with six-hour implementations generating more comprehensive outputs than two-hour ones. The study identifies several contributory conditions: online delivery mode, optimal class sizing, and structured AI prompt guidance. These insights may inform future professional development initiatives within the EFL context of approaching bilingual education. Additionally, the findings suggest promising potential for leveraging AI-enhanced course design to tackle obstacles in bilingual lesson planning.
Streamlining Learner Corpus Development with LLMs and NLP #4304
While recent studies have explored how learner corpora can help teachers develop materials that meet learners’ needs (Brezina et al., 2022), there remains a need for targeted learner corpora that provide insights into specific groups of learners (Götz & Granger, 2024). However, cleaning a corpus for analysis is time-consuming (Brezina et al., 2019). This presentation demonstrates a Large Language Model (LLM) powered corpus cleaning workflow that can address this by using advances in LLMs and Natural Language Processing (NLP) tools like spaCy and Stanza to streamline the process.
Our approach addresses this by integrating LLMs with NLP libraries to identify spelling errors, classify words (e.g., proper nouns, technical terms, foreign words), and apply structured markup. Leveraging API-based processing from modern LLMs like Claude or ChatGPT, this approach allows these LLMs to assist with the systematic analysis and cleaning of a corpus.
This presentation showcases the workflow in action. By using a subset of texts from our existing learner corpus, along with a cleaned and annotated gold-standard version of these texts, we will illustrate how LLMs facilitate preprocessing and structuring learner corpora. The results suggest that this method enhances efficiency and consistency, allowing researchers to focus on linguistic analysis rather than data cleaning.
Assessing the Accuracy and Reliability of AI-Generated Classroom Materials #4189
With increasing workloads, many educators are turning to AI to generate classroom materials quickly and efficiently. AI-created reading texts, worksheets, and listening activities offer level-appropriate content packed with target vocabulary and grammar, spanning topics from SDGs and nursing to the latest K-pop trends. But how accurate and reliable are these materials? Could they even pose risks to students' learning? Cognitive phenomena such as cryptomnesia, source confusion, and the illusory truth effect demonstrate how exposure to misinformation can lead individuals to unknowingly adopt false knowledge as fact. If AI-generated content contains errors, could ESL teachers inadvertently reinforce misconceptions in their students? This study examines the accuracy and educational value of AI-generated materials by consulting subject matter experts—including cinephiles, medical doctors, and a high court judge—who evaluated reading activities produced by three different AI models to rate the accuracy, content and educational value of the texts. Errors were identified, categorized (unimportant, bizarre, or potentially harmful), and analyzed by topic and AI model. This presentation provides essential insights into the risks and limitations of AI-generated teaching materials, helping educators make informed decisions about their use in the classroom.
Chatcasts as a Tool for Validating Custom GPTs and Classroom Task Designs #4167
This study investigates how students perceive and navigate AI-mediated discussions in role-based learning tasks. Grounded in Vygotsky’s Zone of Proximal Development (ZPD), the study explores the potential of customized GPTs as Digital Mentors (DMs) to scaffold learning. However, because Large Language Models operate as black boxes, their affordances and effectiveness require ongoing examination. To assess their impact, custom GPTs were developed for Literature Circles and Mock Trials (‘Pachi-Literature Chatbot’ and ‘Pachi-Mock Trial Chatbot’) and deployed in OpenAI’s Explore GPTs section. Students engaged in AI-supported discussions, recorded screencasts of their interactions, and added voice commentary through Stimulated Recall to reflect on their learning experiences. Post-task surveys captured further insights. Findings from Chatcasts, surveys, and field notes suggest that custom GPT configurations require iterative refinement to optimize their role in task sequencing and scaffolding. Additionally, results indicate that custom GPTs effectively supported pre-advanced students in handling complex discussion and mock trial tasks, accelerating their transition from discussion to introductory debate. These insights contribute to the ongoing refinement of AI-assisted language learning and inform best practices for integrating AI tools into structured educational contexts.
Transforming Post-Pandemic University Classrooms Through MIRO: A Workshop on Interactive, Collaborative, and Flexible Teaching Approaches #4359
This workshop integrates ICT tools, mainly the online whiteboard MIRO, into post-pandemic university classrooms to create more dynamic and engaging learning experiences. Building on teaching practices developed during the COVID-19 pandemic, the workshop demonstrates how MIRO, an online whiteboard application accessible by clicking the link that the host of the classroom provides, can serve as a central instructional hub, promoting interactivity, collaboration, and flexibility in classroom management. Participants will learn to incorporate MIRO into lesson planning, set up engaging activities, and integrate additional applications like Kahoot and Google Forms for assessment and feedback. The workshop offers hands-on experience with these digital tools, allowing participants to experiment with example tasks and collaborative exercises. Key benefits of this approach include increased student ownership of learning, reduced paper usage, enhanced class participation, accommodation of different learning styles, and flexibility for online involvement when needed. The workshop addresses potential challenges, such as student unfamiliarity with digital platforms and the need for diverse instructional formats. It concludes by discussing participant feedback and insights on effectively combining ICT tools in face-to-face contexts, drawing on student perspectives and survey results from previous classes. Participants are encouraged to bring in their laptops to participate in the workshop.
Navigating AI-Assisted Learning: The Impact of ChatGPT on Metacognitive Development in Advanced Japanese University Students #4319
Since its release in November 2022, ChatGPT has brought both opportunities and challenges to English education. While it enhances writing and engagement, concerns remain about over-reliance and ethical use. Metacognition is key in language learning, yet little is known about how ChatGPT affects this in advanced Japanese university students. This eight-week study in 2024 involved 30 students who received structured guidance on AI use, conducted research projects, recorded their interactions with ChatGPT, and reflected on their learning. It focused on three questions: (1) How does explicit guidance affect metacognitive awareness? (2) What support fosters metacognitive skills? (3) How do AI usage patterns show learner autonomy? Qualitative analysis revealed that motivated students used ChatGPT to refine their thinking, while others struggled with over-reliance. The study found that AI can support brainstorming but may hinder self-reflection without careful guidance. This presentation highlights the importance of instructional support to help students use AI intentionally. By encouraging reflective thinking and purposeful interaction with ChatGPT, teachers can promote both language development and metacognitive growth.
Practicing business English skills with AI chatbots #4323
AI is used to handle many tasks that office workers had previously done, such as writing emails and responding to customer inquiries via chat. Therefore, business English classes should prepare students to use AI tools, while also helping them improve their English skills. In this presentation, the presenter will describe and share materials from a task-based business English class in which students used Poe, a site where users can create AI chatbots for free. The class included modules on emailing, texting, telephoning, and video conferencing. The students worked with a classmate to practice each skill with an AI partner created by the teacher, then practiced the skill with a classmate before taking an individual test. In the telephoning module, the students worked together to make a hotel reservation for a customer on the telephone, with the customer played by an AI. They interacted with two additional AIs and got feedback from the teacher. Then, they practiced with their partners playing the role of the customer and received further feedback. Finally, they took an individual exam with their teacher playing the role of the customer. To sum up the semester, students created their own chatbots to perform business tasks.
A Register Analysis of Lexical Bundles in L1 Japanese English Written Essays and Spoken Monologues #4197
Learner corpora have been utilized to examine student production of multi-word units, also known as ‘lexical bundles’ (Chen & Baker, 2010; 2016, Staples et al., 2013). Lexical bundles are strings of three or more words that appear frequently in a given discourse (Biber et al., 1999). This talk presents the results of a register analysis study into the production of lexical bundles by Japanese university students in written essays and spoken monologues taken from two ICNALE corpora (Ishikawa, 2023). First, an overview will be given of key research findings into lexical bundles in learner corpora. Following this, the research study’s methodology will be outlined. The bundles were extracted through Sketch Engine and manually categorized into Biber et al.’s (2004) structural and functional taxonomies. Results show that the two registers are structurally and functionally similar to each other, suggesting that spoken monologues have more in common syntactically with writing than with speech. Also, it was found that native speakers used more noun phrases and prepositional phrases than Japanese learners, and that the latter displayed a greater reliance on verb phrases. Finally, the presenters will consider the study limitations and the possibilities for further research.
The State of CALL in Japan: A Look at the Past, Present, and Future #4206
Despite the ubiquity of technology and the expanding role of CALL in language education, studies investigating the development of CALL at the national level remain scarce (Fathali & Emadi, 2022). Accordingly, this presentation reports on a methodological review that addresses this gap in the literature by systematically reviewing CALL research conducted in Japan. The study aimed to understand the trends and methodological characteristics (e.g., research designs, research topics, settings) related to the Japanese CALL context. Five leading CALL journals (CALICO Journal, Computer Assisted Language Learning, Language Learning & Technology, ReCALL, The JALT CALL Journal) were searched for relevant research published over a 10-year period (2015-2024). A total of 67 articles were selected according to the inclusion criteria. While analysis of the data is ongoing, the initial findings suggest that Japan-based CALL research has tended to favor quantitative or mixed-method designs over purely qualitative studies. Research on the university context also appears to be dominant. The full research findings and future directions for CALL in Japan will be discussed during the presentation.