Learner Agency and the Negotiation of EFL Motivation in GenAI ‐Mediated Drama‐Based Learning: A Qualitative Case Study
Abstract The rapid integration of generative artificial intelligence (GenAI) is reshaping language learning environments, yet little is known about how it becomes embedded within learners' motivational and agentive experiences in collaborative, informal settings like English drama‐based learning. This qualitative case study examines EFL learners' motivation during a 12‐week GenAI‐mediated English drama production cycle in a Chinese university drama club. Guided by Dörnyei's (2009) L2 Motivational Self System (L2MSS), the analysis examines learners' accounts of the Ideal L2 Self, Ought‐to L2 Self, and L2 Learning Experience alongside three analytic patterns of agency identified in this case: authorship, judgment, and collaboration. Data were drawn from focus groups, individual interviews, drama scripts, GenAI interaction artifacts, and rehearsal observations. Findings indicate that participants experienced GenAI as part of their motivational ecology, particularly in relation to access to advanced language, emerging reference points for “good English,” and the ways collaborative decisions were negotiated. Participants also described different ways of maintaining ownership, evaluating GenAI‐generated language, and coordinating decisions with others. The study extends L2 motivational research by revealing how L2MSS dimensions acquire context‐specific meanings in GenAI‐mediated drama‐based language learning and underscores the importance of fostering critical judgment and sustained human collaboration in GenAI‐supported language learning.
Authors
- Timothy P. Teo (ORCID: https://orcid.org/0000-0002-7552-8497)
- Meng Xiong (ORCID: https://orcid.org/0009-0007-2602-769X)
Institutions
- Chinese University of Hong Kong (HK)
- Hubei University (CN)
Publication Details
- Journal
- TESOL Quarterly
- Published
- 2026-09-30
- DOI
- https://doi.org/10.1002/tesq.70235
- Primary Topic
- EFL/ESL Teaching and Learning
- Type
- article
- Field-Weighted Citation Impact
- 0.00