Exploring the Synergistic Effects of LLM-Based Conversational AI and Teacher Intervention on L2 Learners' Interactional Competence

Abstract

This study investigates how Large Language Model (LLM)-based conversational AI and human teacher intervention affect the development of Second Language (L2) Interactional Competence (IC). Employing a two-phase qualitative research design, the study first analyzes interaction texts and reflective journals from five college volunteers. This initial phase isolates the AI's effect on verbal IC dimensions, specifically topic management, turn management, interactive listening, and breakdown repair, until theoretical saturation is reached. Building on this established baseline, the investigation proceeds to a high school oral workshop setting, where classroom observations and an in-depth teacher interview evaluate a human-AI collaborative teaching model. The data shows that AI provides a low-pressure practice environment that effectively improves basic turn-taking and topic initiation. However, AI fundamentally lacks the emotional resonance and non-verbal behavior guidance required for authentic human communication. Teacher intervention is absolutely necessary to scaffold non-verbal behaviors like eye contact and physical gestures. Teachers also play a vital role in adjusting conversational prompts when the AI misunderstands the student. The study concludes that combining AI's capacity for high-frequency linguistic practice with a human teacher's pragmatic and emotional guidance offers a highly useful approach to cultivating L2 learners' interactional competence.