AI versus humans: Impact of IPA, AI apps, and peer interactions on the English skills of EFL adolescents

초록

This study examines the comparative impacts of conversational artificial intelligence (AI), structured AI applications, and peer interaction on English language development among students learning English as a foreign language (EFL). While prior research has distinguished between intelligent personal assistants (IPAs), such as Google Assistant, which enable conversational AI-human interaction, and AI language learning apps that provide automated speech evaluation and structured dialogues, few studies have compared these approaches with human interaction. We recruited 207 seventh-grade Korean EFL students from middle schools and divided them into three experimental groups (Google assistant, AI app, and peer interaction) and a control group (teacher-centered instruction). Over 12 weeks, students participated in weekly interventions. Pre- and post-tests assessed their English speaking, listening, and vocabulary skills, while teacher interviews provided qualitative insights. The findings indicated that all experimental groups significantly outperformed the control group in speaking skills, while no significant differences were found among the groups in listening and vocabulary gains. Teacher interviews revealed that conversational AI (Google Assistant) promoted dynamic and engaging exchanges that sustained student interest and enjoyment. The structured AI app offered gamified and scaffolded practice with immediate feedback, enhancing basic speaking skills and confidence, particularly for shy or lower-performing students. Peer interactions fostered authentic communication and peer collaboration opportunities that AI could not fully replicate. These findings underscore the complementary benefits of AI tools and peer interactions in EFL learning, highlighting the need for a balanced integration of both approaches to effectively enhance EFL students’ speaking skills. © 2026 Elsevier Ltd.

키워드