Visualizing Migration Narratives with Generative AI: A Multi-Modal Approach to Cultural and Emotional Storytelling

Abstract

Although there’s been a worldwide increase in the number of people migrating due to different social factors, their stories and voices have been largely left unheard by the general public. Existing visualizations often generalize, overlooking emotion, metaphor, and context at the individual level. This study tackles the challenges of personal stories and public outreach by utilizing Large Language Models (LLMs)—AI systems designed to understand and generate human-like text. By combining LLMs, sentiment analysis, and image generation to analyze migration stories, the study uncovers social shifts, emotional trends, and cultural themes. This work contributes a multimodal language-to-visual pipeline and a replicable reporting template, focusing on emotional resonance in migration narratives. RetrievalAugmented Generation (RAG) allows the LLM to access specific information before responding. Through prompt engineering and text-to-image synthesis, the study creates symbolic visualizations and develops an evaluation system in which the LLM assesses whether prompts effectively convey emotion, culture, and meaning. A survey was conducted to evaluate the system’s effectiveness. Over 95.9% of respondents (n > 120) reported that at least one AI-generated image clearly evoked feelings related to migration, with an average of 73.2% per image feeling it effectively conveyed immigration concepts. This project increases public awareness of identity, belonging, loss, adaptation, and cultural survival, and explores multimodal generative AI as a tool for cultural interpretation and inclusive social development.

Keywords

• migration narratives
• multimodal generative AI
• Large Language Models
• texttoimage synthesis
• symbolic visualization

References

• [1] Agüera y Arcas, B. (2022). Do Large Language Models Understand Us? Daedalus, 151(2), 183–197.
• [2] Dahlan, D. (2022). Social media and the reading culture revolution among digital natives. COMMICAST, 3(1), 105-112.
• [3] Farrar, J., Arizpe, E., & Lees, R. (2024). Thinking and learning through images: A review of research related to visual literacy, children's reading, and children's literature. Education 3–13, 52(7), 993–1005.
• [4] Fregly, C., Barth, A., & Eigenbrode, S. (2023). Generative AI on AWS: Building context-aware multimodal reasoning applications. O'Reilly Media, Inc.
• [5] Guo, Y., Guo, M., Su, J., Yang, Z., Zhu, M., Li, H., ... & Liu, S. S. (2024). Bias in large language models: Origin, evaluation, and mitigation. arXiv preprint arXiv:2411.10915.
• [6] Jones, N. & Walton, R. (2018). 10 Using Narratives to Foster Critical Thinking about Diversity and Social Justice. In A. Haas & M. Eble (Ed.), Key Theoretical Frameworks: Teaching Technical Communication in the Twenty-First Century (pp. 241-267). Utah State University Press.
• [7] Nakamura, K. (2013). Making sense of sensory ethnography: The sensual and the multisensory. American Anthropologist, 115(1), 132–144.
• [8] Levi, E., Bayerlein, M., Grimalda, G., & Reggiani, T. (2025). Narratives of Migration and Political Polarization: Private Preferences, Public Preferences, and Social Media. IZA-Institute of Labor Economics.
• [9] Li, H., Wang, Y., & Qu, H. (2024). Where are we so far? Understanding data storytelling tools from the perspective of human-AI collaboration. In Proceedings of the CHI Conference on Human Factors in Computing Systems (CHI '24) (pp. 1–19). ACM.
• [10] Mead, M. (1952). The training of the cultural anthropologist. American Anthropologist, 54(3), 343-346.
• [11] Olier, J. S., & Spadavecchia, C. (2022). Stereotypes, disproportions, and power asymmetries in the visual portrayal of migrants in ten countries: An interdisciplinary AI-based approach. Humanities and Social Sciences Communications, 9(410).
• [12] Oppenlaender, J. (2022). The creativity of text-to-image generation. In Proceedings of the 25th International Academic Mindtrek Conference (Academic Mindtrek 2022) (pp. 1–11). ACM.
• [13] Paul, S. (2024, April 26). The picture-superiority effect: Harness the power of visuals. Nielsen Norman Group.
• [14] Bommasani, Rishi, et al. 2021. On the Opportunities and Risks of Foundation Models. CoRR abs/2108.07258 (2021), 212 pages. arXiv:2108.07258.
• [15] Stoller, Paul. 1989. The Taste of Ethnographic Things: The Senses in Anthropology. Philadelphia: University of Pennsylvania Press.
• [16] Westmoreland, M. R. (2022). Multimodality: Reshaping anthropology. Annual Review of Anthropology, 51, 173–194.