When Stories Get a Visual Voice: AI-Generated Images, a Visual Pedagogy for Enhancing Student Engagement and Critical Analysis in World Literature

Background/Purpose: This study investigates the pedagogical potential of AI-generated images to enhance student engagement and critical analysis in world literature curricula. Grounded in Reader-Response Theory, it explores how algorithmic visuals impact student interpretation, addressing a gap in understanding technology's role in fostering aesthetic literacy. Materials/Methods: A mixed-method approach was employed in an undergraduate literature course using surveys (N=30) and educator interviews. Students engaged in short stories (e.g., Kafka, Hemingway) accompanied by AI-generated images (Midjourney v5.2). Questionnaires measured emotional/intellectual engagement, and open-ended responses captured interpretive reasoning. Data were analyzed via ANOVA and thematic coding (NVivo14). Results: AI images significantly boosted student emotional engagement (+22%, *p*<0.01), effectively hooking reluctant readers. However, interpretive diversity decreased (3 vs. 9 thematic variants in control groups), potentially narrowing critical thinking. Disparities emerged: women engaged more with symbolic visuals (77.4% emotional connection), while men preferred literal depictions (69.8% analytical focus). AI's cultural biases were identified as a teachable moment for digital literacy by 28% of participants. Conclusion: AI visuals deepen student-text interactions but risk homogenizing interpretation. The study highlights AI’s value as a scaffolded pedagogical tool for differentiating instruction, advocating for ethical prompt-crafting exercises to teach algorithmic bias and preserve hermeneutic openness. Findings urge interdisciplinary collaboration to harness AI's capacity for enriching literary learning.