The influence of AI-driven personalized foreign language learning on college students' mental health: a dynamic interaction among pleasure, anxiety, and self-efficacy.

Introduction: This study examines the effects of AI-driven personalized foreign language learning on college students' mental health, with a focus on the dynamic interaction among pleasure, anxiety, and self-efficacy. Drawing on cognitive load theory, self-determination theory, and dynamic systems theory, the research constructs a mental variable influence model to explore how emotional and cognitive factors shape learners' experiences. Methods: A mixed-methods design was adopted, integrating questionnaire surveys, experimental research, and time series analysis. College students were randomly assigned to an experimental group receiving AI-driven personalized learning or a control group using traditional learning methods. The study tested hypotheses regarding the relationships among pleasure, anxiety, and self-efficacy and tracked their temporal evolution during the learning process. Results: Comparative analysis revealed that AI-driven personalized learning significantly enhanced pleasure, reduced anxiety, and strengthened self-efficacy compared with traditional methods. Pleasure and self-efficacy exerted a mitigating effect on anxiety, while heightened anxiety negatively influenced self-efficacy. Time series analysis further showed a phased pattern: after an adaptation period, pleasure and self-efficacy progressively increased, while anxiety levels demonstrated a sustained decline over time. Discussion: The findings provide empirical insights into the interaction mechanisms among mental variables in AI-supported learning. They suggest that AI-driven systems should integrate emotional regulation mechanisms-such as adaptive feedback, personalized emotional support, and social interaction functionalities-to enhance learners' mental experiences and sustain learning persistence. This study contributes to the optimization of AI-driven personalized learning models, supports the design of intelligent educational technologies, and strengthens mental health protection for foreign language learners.