Thirty-Second International Conference on Learning

Abstract

Prosodic competence—mastery of stress, rhythm, intonation, and pitch—plays a crucial role in effective communication, yet many adult EFL learners struggle to develop these skills. While metacognitive strategies such as goal setting, planning, self-reflection, and self-regulation have been shown to enhance learning outcomes, their impact on prosodic competence remains underexplored in EFL contexts. This study addresses this gap by examining the link between metacognitive strategies and prosodic competence, focusing on the use of AI-generated exercises to support learning. Twenty-five sophomore TESOL students from a Costa Rican public university participated in a 15-week intervention during an English Phonology class. An explanatory sequential design (QUAN-Qual) was used. The quantitative phase included correlational analysis to assess the relationship between metacognitive strategies and prosodic gains, keeping in mind confounding variables such as teaching styles. The qualitative phase involved thematic analysis of students’ reflective journals, peer feedback, and survey responses, triangulated with activity logs to explore their experiences with AI tools and metacognitive activities. While no significant correlation emerged in the quantitative analysis, the qualitative phase highlights areas where students reported growth, suggesting that AI-driven, metacognitive strategies foster broader learning benefits beyond test performance. This research offers insights into the role of AI in enhancing language learning by promoting self-regulation and provides valuable implications for integrating metacognitive practices into EFL instruction. By combining human learning with machine-generated support, the study contributes to understanding the challenges and opportunities presented by AI in education.