Speaker
Description
Cross-examination is one of the most difficult skills for second language learners in debate because students must produce strategic questions under real-time interactional pressure. This session introduces a practical approach to teaching cross-examination using a custom AI mentor that allows students to rehearse questioning strategies before mock trials and debates. Participants will see how text- and voice-based AI rehearsal can scaffold questioning skills and help learners participate more effectively in argumentative discussions.
Abstract section 2: Contribution/research questions
The purpose of this presentation is to demonstrate a practical classroom approach for teaching debate cross-examination through AI-supported rehearsal. Participants will learn how a custom GPT mentor (“Pachi”) can simulate witnesses and opponents so students practice questioning strategies before participating in mock trials and debates. Drawing on research on institutional questioning and courtroom discourse, the session illustrates how AI rehearsal helps learners develop greater awareness of question design and interactional strategy during argumentative discussions.
Abstract section 3: Content/method
The session introduces a sequence of preparation activities using a custom AI mentor. Students first practice designing questions in text-mode chat, focusing on morphosyntactic forms (e.g., WH and polar questions) and basic discourse functions that expand or constrain response opportunities. Voice-mode rehearsal is then used to practice intonation and delivery while developing lines of evidential questioning. These preparation activities lead into classroom mock trials and debates where students apply questioning strategies during cross-examination.
Abstract section 1: Relevance
Classroom debates promote critical thinking and L2 learning, yet students often struggle with cross-examination because they lack experience designing strategic questions. Task preparation research shows that learners benefit from rehearsing language prior to performance (Ellis, 2003). Studies of institutional questioning demonstrate how morphosyntactic question design shapes response possibilities in courtroom examinations (Mortensen, 2017; Seuren, 2019). Classroom simulations such as mock trials support the development of L2 argumentative skills (Luchini & Cresci, 2026). Recent work on GenAI suggests that dialogic interaction with AI systems can scaffold language development through guided interaction within learners’ zones of proximal development (Özturan & Shrestha, 2025).
Abstract section 4: Outcomes/results
Participants will examine classroom materials used to prepare students for cross-examination, including example prompts, student-generated questions, and short interaction excerpts from AI-supported rehearsal sessions. The presenter will demonstrate how the custom GPT mentor guides learners in refining questions from open interpretive prompts to strategically constraining forms used in argumentative exchanges. Attendees will discuss how these activities can be adapted for their own classes and explore ways to integrate AI rehearsal into discussion-based tasks such as literature circles, mock trials, and debates. Participants will leave with adaptable preparation activities and practical strategies for using AI tools to support argumentative discussion skills.
Abstract section 5: References
Ellis, R. (2003). Task-based language learning and teaching. Oxford University Press.
Luchini, P. L., & Cresci, K. L. (2026). From the courtroom to the classroom: Enhancing L2 learners’ argumentative skills through a mock trial simulation. Simulation & Gaming. Advance online publication. https://doi.org/10.1177/10468781261416273
Mortensen, K. (2017). The interactional organization of questioning in institutional interaction. Journal of Pragmatics, 108, 145–160.
Özturan, T., & Shrestha, P. N. (2025). ChatGPT’s potential to enhance language learners’ proximal development through dialogic interaction. In C. A. Chapelle, G. H. Beckett, & B. E. Gray (Eds.), Researching generative AI in applied linguistics (pp. 11–39). Iowa State University Digital Press. https://doi.org/10.31274/isudp.2025.211.02
Seuren, L. M. (2019). Questioning in court: The construction of direct examinations. Discourse Studies, 21(3), 340–357. https://doi.org/10.1177/1461445618770483
| Title | Teaching Debate Cross-Examination with AI Mentors |
|---|---|
| Teaching Context | College and university education |