Speaker
Abstract section 5: References
Mishra, P., & Koehler, M. J. (2006). Technological Pedagogical Content Knowledge: A Framework for Teacher Knowledge. Teachers College Record, 108(6), 1017-1054. https://doi.org/10.1111/j.1467-9620.2006.00684.x
Sweller, J. (1988). Cognitive load during problem solving: Effects on learning. Cognitive Science, 12(2), 257-285.
Abstract section 3: Content/method
The presenter will share specific prompt templates that successfully isolated individual skills for targeted practice discussing successes, such as feedback delivered in digestible formats that help students self-identify weaknesses. It will also address challenges, including the tendency for AI feedback to become too long or complex at various assignment stages. Furthermore, survey results regarding student reactions, specifically their hesitation to trust AI feedback and the resulting importance of instructor guidance, will be discussed.
Abstract section 2: Contribution/research questions
This session explains how to use GenAI to bridge the gap between generic suggestions and individualized instruction. It demonstrates practical methods for generating tailored feedback on writing structure (clarity, organization, and word choice) and pronunciation (suprasegmentals like stress, rhythm, and intonation). The goal is to share successes and challenges from an EAP classroom to help teachers adopt immediate, practical adjustments.
Abstract section 1: Relevance
Providing individualized feedback is essential for EAL learners, yet the time-intensive nature of manual correction often limits scalability. While GenAI offers a solution for generating personalized examples and practice, instructors must navigate challenges such as feedback length and student trust. Applying the Technological Pedagogical Content Knowledge (TPACK) framework (Mishra & Koehler, 2006) and Cognitive Load Theory (Sweller, 1988) ensure AI tools support the integration of GenAI tools into feedback rather than hinder the learning process. By focusing on practical instructor-guided enhancements, feedback for writing and speaking tasks can be specific timely without overwhelming the learner or the instructor.
Abstract section 4: Outcomes/results
Participants will learn how to design GenAI prompts that produce scalable, skill-focused feedback for writing and pronunciation. They will gain insights into student preferences and the necessity of instructor oversight when delivering AI-generated suggestions. Attendees will leave with a set of prompt templates designed to provide individualized feedback while avoiding common pitfalls, such as feedback fatigue or over-automation.
| Title | Student-centered AI Feedback for Pronunciation & Writing |
|---|---|
| Teaching Context | College and university education |