Generative AI-supported evidence-informed use of L1 in the L2 classroom

It is well known that L1, when used purposefully and strategically, can help with L2 learning (Shin et al., 2020). However, in many contexts, deep-rooted monolingual immersion approaches, ‘e.g. English only’ still dominate (Lin, 2015). These policies can lead teachers to refrain from using L1 or use it deceptively. It can also mean that teachers are not prepared with the knowledge or practices that can lead to evidence-informed uses of L1 (Cummins, 2007). It has been suggested that generative AI advances (e.g. ChatGPT) may aid the development of teachers’ professional knowledge development and evidence-informed practices (Kohnke et al., 2023).

In this presentation, we share the findings of an intervention study that aimed to raise preservice English language teachers’ awareness of the purposeful use of L1 while exploring the potential generative AI tools can have in aiding teachers in their development of the knowledge and skills needed to utilize L1 in L2 classrooms. The study involved data collected from 41 preservice teachers studying a full-time Master of Education in English Language Teaching in Hong Kong.

Data were collected through a pre- and post-intervention mixed-method survey and follow-up group interviews. The findings show that explicit instruction on the use of L1 in L2 classrooms can increase pre-service teachers' willingness to use L1, knowledge about evidence-informed teaching practices, and how AI can support their planning and implementation of L1. Practical implications for teacher education and development are provided.