16–18 May 2025
Kanda University of International Studies (神田外語大学)
Asia/Tokyo timezone

Generative AI as a Tool for Teaching Discourse Competence in EFL

18 May 2025, 13:05
25m
BLDG 8/1F-105 (Kanda University of International Studies (神田外語大学))

BLDG 8/1F-105

Kanda University of International Studies (神田外語大学)

30
Practice-oriented Oral Face-to-face presentation Computer Assisted Language Learning B8-109 SAT: Critical Thinking, Study Abroad & CALL; SUN: CALL

Speaker

Javier Salazar (ICLE SIG - University of Tsukuba)

KEYWORDS

Generative AI
Prompt Engineering
Discourse Competence
Communicative Competence

ABSTRACT

The ability to produce coherent and cohesive utterances during a conversation, commonly referred to as discourse competence (Celce-Murcia et al., 1995) is one of the many challenges of developing “natural” conversational skills in EFL (Bachman, 1991). In the Japan context, there is a number of difficulties Japanese learners face in this regard, e.g. structural discrepancies between Japanese and English in terms of how thoughts tend to be chained during speech as well as the inherent “unnaturalness” of the conversations that appear in some English textbooks (Ellis, 1991; Haugh, 2019). In this practice-oriented presentation, a Generative AI-based strategy for overcoming some of these challenges will be presented. Specifically, a chatbot was prompt-engineered for conversing with students; whilst coaching them on how to organize sentences coherently and use linking words and transitional phrases to create cohesive discourse. The classroom application of this tool will be demonstrated through student outputs and testimonials. Finally, the limitations of using this tool and the general transferability of skills learned through chatbot-based conversations to non-AI mediated, real-world interactions will be discussed.

TITLE Generative AI as a Tool for Teaching Discourse Competence in EFL
RELEVANT SIG Computer Assisted Language Learning (CALL)
FORMAT Practice-oriented Oral Face-to-face presentation (25 minutes, including Q&A)

Author

Javier Salazar (ICLE SIG - University of Tsukuba)

Presentation materials

There are no materials yet.