Investigating the Impact of AI Generated Mnemonic Illustrations on Learning Japanese Kanji

Abstract

Learning Japanese Kanji is challenging for novice learners due to character volume, visual complexity, and limited personalized resources. While various strategies exist, the use of illustrations in Kanji learning, despite their proven benefits, remains underexplored. This study investigated the impact of AI-generated mnemonic illustrations on Japanese as a Foreign Language (JFL) learners' Kanji attainment, cognitive workload, and attitudes. Using a quasi-experimental design, 84 undergraduate engineering students with no prior Japanese knowledge were divided into three groups. Group M learned Kanji using only mnemonics. Group MI used mnemonics alongside AI-generated mnemonic illustrations. Group MID used mnemonics, AI-generated mnemonic illustrations, and also drew the Kanji characters. Quantitative analysis revealed that using AI-generated mnemonic illustrations (Group MI) did not significantly improve overall Kanji learning attainment compared to using only mnemonics (Group M) or mnemonics with illustrations and drawing (Group MID). However, Group MI experienced significantly lower perceived temporal demand (a component of cognitive workload) than Group M. Furthermore, the active engagement of drawing Kanji (Group MID) significantly enhanced participants' perception of the usefulness of learning Kanji compared to Group MI. Qualitative analysis also highlighted that mnemonic illustrations helped learners correlate Kanji characters with their meanings. These findings suggest that while AI-generated mnemonic illustrations may not directly boost short-term Kanji attainment, they can positively influence the learning experience by reducing perceived temporal demand and enhancing attitude(usefulness) when combined with active learning. The ability of AI to create visual aids shows promise for improving Kanji pedagogy.