Supporting Young Foreign Language Learners With Image Generative AI in Digital Multimodal Composing: Benefits and Challenges

Abstract

Integrating generative artificial intelligence (GenAI) into digital multimodal composing (DMC) offers promising avenues for young Foreign Language (FL) learners' learning experiences. Image‐based GenAI tools enable students to generate personalized images from written prompts to help convey meanings, and DMC combines text, images, and sound to create meaning. However, most existing research on GenAI and DMC focuses on text‐based GenAI use and older learners, leaving image GenAI usage and younger primary school students underexplored. This action research addressed this gap by designing a DMC task for primary school FL students ( N  = 33, aged 8–9) using image‐based GenAI. Students engaged in structured pre‐writing tasks applied prompt‐engineering strategies targeting key language features and AI‐generated images to support idea visualization and creative expression. Peer evaluation further fostered collaborative feedback. Classroom video recordings, group interviews, and AI‐generated images revealed positive reactions, including sustained engagement, enhanced agency, and spontaneous collaboration, even though challenges with prompting GenAI for specific outputs were noted. These findings suggested that the integration of image‐based GenAI within scaffolded DMC tasks and processes was mostly perceived positively by young FL learners. Educators are encouraged to adopt image GenAI‐supported DMC with explicit prompt‐engineering instruction and peer interaction to support their needs in language development.