Investigating Robot-Assisted Second Language Teaching Under Social Interactive Education: A Systematic Review

Abstract

Artificial Intelligence (AI) supported educational robots hold great promise for second language teaching. Robots using advanced AI capabilities such as emotion recognition and real-time feedback can create dynamic and interactive environments that improve engagement and educational outcomes. Additionally, they have the ground-breaking potential to enable personalised teaching adapted to the unique needs of different learners. This paper aims to systematically explore the recent developments of educational robots in second language learning and teaching through conducting a systematic literature review to investigate the research gaps and propose a novel framework for AI-supported personalised teaching with educational robots. This study utilised Scopus to search for recent publications (2017-2024), applying specific criteria such as keywords, publication dates, and subject areas to ensure the relevance of the results. These criteria have been carefully designed to identify high-quality research that is consistent with the scope of the study. Based on these search criteria, 25 studies were identified using the PRISMA method, which systematically screened and selected studies for inclusion criteria. The results highlight that the novelty and social interaction capabilities of educational robot design, coupled with the support of AI technology, provide language learners with engaging new learning opportunities. These robots, by incorporating advanced AI features, can simulate real-life social interactions, making language practice more immersive and motivating for learners. These opportunities enhance the interactivity of teaching sessions by instantly recognising students' questions and providing pertinent feedback. Therefore, this study suggests the benefits that AI and robotics offer for second language teaching, particularly in offering greater potential for personalised learning and interactive teaching. However, despite these advantages, this remains a key implementation challenge in integrating educational robotics with traditional teaching. The willingness of teachers to embrace educational robotics, the need to coordinate with current lessons, and the high cost of robotics are all still barriers that must be addressed. These limitations highlight the importance of further research and development to adapt different teaching environments, grounded on the initial design of the work. Future work includes a planned validation study in a second language school in Wales, scheduled for autumn 2025.