A multimodal study of decoding heritage tourists’ perception and emotion in historic urban areas of Ganzhou

Abstract In the context of global urban renewal, historic urban areas (HUAs) face the challenge of balancing heritage preservation with sustainable development. Heritage tourism is a key pathway to achieve this balance. The emotional responses and behaviors of heritage tourists, shaped by perception, influence the directions of renewal. Related studies rely on single-modal data and lack an in-depth exploration of tourists’ perceptual dimensions and intuitive emotions. This study examines historic urban areas of Ganzhou and develops a multimodal framework comprising “text topic mining, visual element extraction, and emotional mechanism modeling” to reveal tourists’ perceptual dimensions and how spatial elements influence emotion. Findings revealed that tourists focus on three thematic areas: history and culture, historic cityscape, and leisure experience. Natural and facility-related elements positively impact tourist emotion, whereas architecture, road, and vegetation show a negative correlation. This study offers theoretical support for the transformation of HUAs from static heritage preservation to living heritage.