A Computationally Supported Spiral Symbiosis Framework for Art-and-Technology Education: Generative AI, Parametric Modelling and OBE-based Learning Analytics

Abstract

We report on a three-cycle action research project that restructured Cultural Inheritance and Innovation, a third-year core studio course in art-and-technology education delivered at two universities in the Guangdong–Hong Kong–Macao Greater Bay Area during the 2023–2024 academic year. The redesign was prompted by a persistent problem: students could produce visually competent cultural prototypes yet struggled to articulate why the work mattered beyond stylistic citation. We introduced generative AI, Grasshopper-supported parametric modelling and a rubric-based analytics matrix sequentially across the cycles, not as simultaneous add-ons but as a digitally mediated workflow in which each tool entered only after the previous stage had proved stable. Drawing on classroom observations, interviews and 137 student projects generated across the cycles, we established a workflow that requires cultural feature extraction to precede prompt-driven image generation, which in turn feeds parametric transformation and OBE-aligned evaluation. The insistence on this sequencing—particularly the requirement that students defend their AI prompts against the cultural feature matrix before generating images—was the most consequential change we made. The resulting prototypes showed greater traceability between cultural inquiry and digital production, though the improvement was uneven: students with prior scripting experience benefited significantly more from the parametric stage than those without.