Theoretical and Psychological Mechanisms of Perceptual-Motor Learning in AI Bots-Assisted Art Education

The integration of Artificial Intelligence (AI) into art education presents a complex shift in perceptual-motor learning, yet the psychological mechanisms associated with learners' responses to these digital tools remain insufficiently mapped. This study investigates the structural relationships among self-efficacy, cognitive load, attention allocation, learning strategy preference, task motivation, and perceived difficulty within the context of AI-assisted art education. Drawing on a sample of 637 Chinese art learners, the research employs structural equation modeling (SEM) to examine correlational patterns and structural associations among these self-reported psychological constructs. The results suggest that self-efficacy was negatively associated with cognitive load and positively associated with self-reported attention allocation. Perceived difficulty was associated with higher cognitive load and lower self-efficacy, indicating that difficulty may operate as a double-edged factor in AI-assisted art learning. Attention allocation was positively associated with learning strategy preference, and learning strategy preference was positively associated with task motivation; however, these paths should be interpreted as structural associations rather than evidence of mediation or causal effects. These findings complicate the assumption that technological affordances alone enhance learning, highlighting instead that learners' cognitive and self-regulatory resources are closely linked to their experiences of AI-supported art practice. The study concludes by arguing for pedagogical and design approaches that prioritize cognitive economy, adjustable feedback, and the cultivation of adaptive mastery beliefs in technologically enriched artistic environments.