Enhancing EFL writing with visualised GenAI feedback: A cognitive affective theory of learning perspective on revision quality, emotional response, and human-computer interaction

With the rapid development of large language models and natural language processing technologies, Generative AI (GenAI) chatbots have offered new opportunities for supporting EFL learners in writing and revision. However, existing GenAI chatbots’ output is predominantly presented in plain, text-only formats, which may impose a high cognitive load and trigger negative emotional responses. Despite growing interest in GenAI-assisted writing instruction, limited research has examined how visual enhancements to GenAI chatbot output might improve revision outcomes and learners’ emotional experiences. Grounded in the Cognitive-Affective Theory of Learning with Media, this study employed a self-developed GenAI-powered writing chatbot to investigate the effects of visualised feedback on EFL learners’ writing performance and emotional responses during revision. A 2 (time: pre-test and post-test) × 2 (feedback mode: visualised vs. non-visualised) quasi-experimental design was employed. Group A (N = 30) received standard text-only feedback, while Group B (N = 30) received visualised feedback incorporating colour variation, tabular formatting, and bolded text. Results from pre-and post-tests and questionnaires indicated that visualised feedback significantly improved coherence and cohesion in learners’ writing, reduced negative emotions, and resulted in lower cognitive load. These findings offer practical implications for language educators, learners, and developers, highlighting the critical role of how AI-generated content is presented in building emotionally supportive and cognitively effective GenAI-assisted learning environments.