Effective Human-AI Interaction as Observational Learning for Multisensory Creative Fashion Design: Sound(music)-Text-Image Processing

The rapid advancement of artificial intelligence (AI) increasingly challenges the traditional view of creativity as a uniquely human domain, raising concerns about the erosion of human creative capacities. This study investigates whether incorporating auditory stimuli into visually-driven design activities can enhance creativity, and further explores how multisensory training involving images, language, and music (modern classical) can support human-AI collaboration. An exploratory three-month study was conducted with 23 third-year fashion design students, who engaged with the modern classical piece Cello Sonata No. 1 ‘Gyeongya’ while utilising Generative AI tools (Midjourney, DALL-E 3, and Bing Image Creator) throughout the concept development and design stages. Participants were guided to use linguistic thinking activities to bridge auditory and visual perceptions. The findings demonstrate that multisensory stimulation, particularly when coupled with structured linguistic reflection, facilitated cognitive expansion and enabled the generation of a wider range of visual imagery. These outcomes validate the potential of human-AI interaction in enhancing synergistic creativity by integrating auditory and visual elements. This study contributes to the emerging field of interdisciplinary creative education, offering new insights into how AI technologies and multisensory approaches can be strategically applied to strengthen creative thinking in fashion design pedagogy.