Dynamic Transmission and Innovative Transformation of Cultural Heritage: Generative Artificial Intelligence Practices Based on Cultural Cognitive Models

The rapid advancement of digital technologies is compelling the field of cultural heritage preservation to shift from static conservation toward dynamic transmission and innovative transformation. In response, this study proposes a Generative AI (GenAI) application approach based on a cultural cognitive model. First, a cognitive structure of cultural symbols is constructed based on symbolic interactionism, and grounded theory is applied to analyze how specific user groups interpret and internalize these symbols, thereby establishing a cultural cognition system. An enhanced Delphi method is then employed to synthesize expert judgments and develop a multi-level cultural-symbol dataset. The dataset is integrated into generated models through stable diffusion models and Low-Rank Adaptation (LoRA) to strengthen their capacity for recognizing and generating culturally significant features. The feasibility and effectiveness of the proposed model are evaluated through expert-based assessments. To further examine its generalizability, the study conducts a case application using Shanghai-style furniture design. The results demonstrate substantial improvements in output quality and alignment with design requirements. This research provides a reproducible methodology for the digital safeguarding and innovative development of cultural heritage, while expanding the application scenarios of AI technologies in the protection of intangible cultural heritage.