AI-driven mixed-methods analysis of technology dependence: Personality-moderated pathways to Oral English anxiety in...

Intelligent education systems are reshaping the language learning ecosystem; however, the psychological mechanisms linking technology dependency (TD) and oral English anxiety (OA) remain under-explored. This study employs a mixed-methods approach grounded in Cognitive Load Theory (Sweller, 1988) and Social Cognitive Theory (Bandura, 1986), supplemented by Trait Activation Theory to examine personality moderation associations.Surveying 716 university students in Beijing, Shanghai, Guangzhou, and Chongqing, it investigates the mediating effects of psychological burden (PB) and digital self-efficacy (DSE) on the TD-OA relationship, alongside the moderating associations of personality traits. Mplus structural equation modelling was combined with Python XGBoost-SHAP machine learning. SEM revealed an asymmetric dual-path mechanism: PB acted as a complementary mediator, amplifying the positive association between TD and OA, while DSE functioned as a competitive mediator, marginally buffering the association with a path ratio of 7.83:1. Personality traits significantly modulated these relationships: Extraversion (EX) enhanced resilience to technology-induced anxiety, while Neuroticism (NEU) amplified vulnerability. Mixed-methods enhanced predictive accuracy by detecting complex nonlinear interactions. The study contributes in two key areas: 1. A dual-process model explaining technology dependence by highlighting differentiated associations through cognitive load and efficacy pathways; 2.A trait adaptation framework that clarifies how personality traits are associated with psychological adaptation in technology learning environments. Practical implications inform the development of "anxiety-sensitive" intelligent tutoring systems. This research advocates for adaptive adjustments based on psychological burden and personality-oriented personalised intervention strategies, emphasizing the importance of protecting learners' mental health while pursuing technological efficiency.