ARCS-iLearn : An AI Agent-Based Vocabulary Learning Framework Grounded in ARCS Motivation Theory

• In Event: Virtual Poster Session
  In Virtual iPresentation Only: SIG-Bilingual Education Research Virtual Poster Session

Abstract

With the rapid development of generative AI, large language model-based agents have shown strong potential in language education. This study investigates the practical application of the Coze agent in English vocabulary teaching, integrating the ARCS motivational model to design an agent-supported learning pathway. After reviewing the ARCS model’s role in language instruction and recent AI advancements, a pathway focusing on “Attention–Relevance–Confidence–Satisfaction” was constructed. Leveraging Coze’s context generation and adaptive feedback, a six-week experiment at T Middle School found that the agent-based pathway significantly improved vocabulary acquisition and learner motivation compared to traditional instruction. These findings confirm that large language model agents can effectively sustain learning motivation and optimize vocabulary learning, offering a promising paradigm for intelligent language education.

Authors

• Yuxi Zhang, central china normal university

• jianping huang, Beijing Normal University