Evaluating the Longitudinal Effects of AI-Enhanced Collaborative Dialogue Modes on Computational Thinking and Language Proficiency in EFL Learners: A Mixed-Methods Approach.

This study investigates the comparative effectiveness of three artificial intelligence (AI)-powered collaborative dialogue approaches—guided feedback (GF), interactive Q&A (IQ), and inquiry-driven scaffolding (IS)—on computational thinking (CT) and English language proficiency (LP) among 448 Chinese tertiary-level English as a Foreign Language (EFL) learners. Conducted as a triple-blind, randomized controlled trial over 18 sessions, the research found that the guided feedback group achieved the largest and most durable gains in both CT and LP, significantly outperforming the other AI modes and a traditional control group, with superior skill retention at delayed posttests. Qualitative analyses revealed that GF facilitated metacognitive monitoring and error correction, whereas IQ and IS approaches imposed higher cognitive load and risked learner overreliance on AI, potentially constraining creativity and self-regulation. The findings emphasize that AI’s pedagogical design—particularly structured, adaptive feedback aligned with learner readiness—is critical for sustainable dual-skill development, while cautioning that these results may be culturally specific and recommending cross-cultural replications to validate applicability in diverse educational contexts.