Divergent outcomes of AI anxiety: A dual-pathway model of cognitive appraisal on learning behaviors among university students

The rapid integration of artificial intelligence (AI) into higher education is producing divergent learning behaviors, as student AI anxiety appears to both hinder and motivate learning. However, the psychological mechanisms that explain why these divergent responses occur remain underexplored. To address this gap, this study investigates how AI anxiety is associated with university students' motivated and avoidance learning by examining challenge and hindrance appraisals as key mediating mechanisms. The study employed a cross-sectional questionnaire design using established scales adapted to the educational AI context. An online survey was administered to students from three universities in China, yielding 591 valid responses after data screening. Results show that AI learning anxiety is primarily associated with hindrance appraisal, while AI job replacement anxiety is associated with both challenge and hindrance appraisals. Challenge appraisal is positively associated with motivated learning and negatively associated with avoidance learning, whereas hindrance appraisal shows the opposite pattern. AI learning anxiety exhibits consistent negative effects through both direct and indirect pathways, while AI job replacement anxiety exerts entirely indirect effects mediated by appraisal processes. These findings highlight cognitive appraisal as a crucial mechanism explaining the divergent behavioral associations of AI anxiety and offer valuable insights for educational intervention.