English Learning Burnout and L2 Motivational Self-System among Chinese English Major Undergraduates

Abstract

This study investigates the complex relationship between English learning burnout and the L2 Motivational Self-System (L2MSS) among Chinese English major undergraduates by employing a hybrid analytical framework combining Long Short-Term Memory (LSTM) neural networks and Structural Equation Modeling (SEM). The research aims to uncover how motivational constructs—such as the Ideal L2 Self and L2 Learning Experience—influence levels of academic burnout in language learning contexts. A dataset comprising survey responses was used to extract sequential features via the LSTM model, while SEM was applied to validate theoretical relationships between latent motivational variables and observed burnout scores. The model achieved high accuracy (91.2% in training and 87.6% in testing) and demonstrated strong predictive performance across multiple metrics, including F1 score and AUC. SEM results confirmed that both the Ideal L2 Self and L2 Learning Experience are negatively correlated with burnout, indicating that enhanced motivational states can significantly mitigate emotional exhaustion and disengagement. This integrated approach highlights the importance of motivation in preventing burnout and offers actionable insights for curriculum design, pedagogical strategies, and learner support systems in Chinese higher education. The findings contribute to both theoretical advancement and practical solutions for improving student well-being and second language acquisition outcomes.