Abstract
As an emerging pedagogical approach in art education, artificial intelligence-assisted drawing holds significant value in enhancing students’ creative abilities and digital literacy. However, students’ continuance learning intention and its underlying mechanisms in AI-assisted drawing courses remain underexplored. This study aims to systematically investigate the key factors influencing students’ continuance learning intention in AI-assisted drawing courses, guided by coolness theory and expectancy-value theory as theoretical frameworks. Using a mixed-methods approach combining structural equation modeling (SEM) and fuzzy-set qualitative comparative analysis (fsQCA), this research examined the predictive effects of performance expectancy, effort expectancy, perceived coolness, attitude, and anxiety on college students’ Behavioral Intention among 365 university students. The SEM analysis revealed that attitude exerted the most significant influence on students’ Behavioral Intention, while performance expectancy and perceived coolness also demonstrated significant positive effects, and anxiety showed a significant negative impact. Additionally, self-efficacy had a significant positive effect on students’ continuance learning intention. These factors collectively explained 77.1% of the variance in students’ continuance learning intention (R²), while explaining 52.4% of the variance in behavioral intention. Furthermore, gender moderated the path from perceived usefulness to continuance intention, and grade level moderated the path from self-efficacy to continuance intention. The fsQCA results revealed that students’ continuance learning intention is not determined by a single factor but achieved through various combinations of multiple factors, with five configurational solutions identified as leading to similar outcomes. At the theoretical level, this study successfully applied coolness theory and expectancy-value theory to the context of AI-assisted art education, expanding the applicability of these theories in creative technology learning and revealing the interactive mechanisms among cognitive appraisal, technology perception, and social influence. At the practical level, the findings provide empirical evidence and diversified intervention strategies for teachers, schools, and curriculum designers to optimize AI-assisted drawing instruction, cultivate students’ positive attitudes, and enhance their technological experience.