Fostering Adolescent Engagement in Generative AI Art Therapy: A Dual SEM-ANN Analysis of Emotional

XIAOJIE PENGYANRAN QIANQIAN BAO^*

• Hanyang University, Seoul, Republic of Korea

This study is centered on the application of generative artificial intelligence (AI) art in digital art therapy and systematically investigates the mechanisms by which it influences adolescents’ interest-driven participation. Drawing on a cross-sectional survey of 444 junior and senior high school students in Hubei Province, a predictive model was constructed and validated by integrating Emotional Design Theory with the Technology Acceptance Model (TAM), using both structural equation modeling (SEM) and artificial neural network (ANN) analysis.The results revealed that perceived usefulness (PU), perceived ease of use (PEOU), perceived fun (PF), and perceived trust (PT) significantly influenced users’ attitudes toward use (ATT) (p < 0.001), whereas ATT, PF, and PT were identified as strong predictors of interest-driven participation. By contrast, the behavioral level was found to exert no direct effect on perceived enjoyment (PE), indicating that emotional resonance and technological friendliness may play a more decisive role.These findings were corroborated by ANN analysis, in which attitude toward use was identified as the most important predictor (100% normalized importance), substantially surpassing perceived enjoyment (19.3%). The results underscore the importance of integrating intuitive design, seamless interaction, and trust-building mechanisms to sustain adolescents’ engagement in AI-assisted art therapy. This study contributes a theoretical foundation for understanding adolescent interest formation mechanisms and offers practical insights for optimizing digital interventions, emotional design, and mental health support strategies.