Online self-regulated English learning and English learning anxiety in digital contexts: the mediating role of intrinsic motivation

Background: With the rapid digital transformation of higher education, online self-regulated English learning (OSRL) has become increasingly important; however, its emotional consequences, particularly English learning anxiety, remain insufficiently understood. In these digital learning environments, students are expected to regulate their learning processes while also coping with emotional challenges such as English learning anxiety. Nevertheless, little is known about how different dimensions of OSRL relate to anxiety and whether intrinsic motivation may explain these associations. Methods: A cross-sectional design was employed, and 1,361 valid responses were collected from university students in China using the Maike Online Questionnaire Platform. Three validated instruments were used: the Online Self-Regulated Learning Questionnaire (OSLQ) to assess six dimensions of learners' self-regulatory behavior, the English Class Performance Anxiety Questionnaire (ECPAQ) to measure anxiety in English learning contexts, and the Intrinsic Motivation subscale of the Language Learning Orientations Scale (LLOS-IM) to assess intrinsic motivation. Data were analyzed using Spearman correlation and bootstrap-based multiple mediation analysis (5,000 resamples) to examine the direct and indirect associations between online self-regulated English learning dimensions and English learning anxiety. Results: = 0.006), suggesting that social support may serve as a protective factor in digital English learning contexts. Self-evaluation showed a suppression-like indirect pattern (i.e., indirect and direct effects operating in opposite directions) (indirect a × b = 0.116, accounting for 22.65% of the total effect), indicating that while self-monitoring enhances learning control, it may also amplify performance pressure and emotional strain. In contrast, goal setting exerted a fully mediated effect through intrinsic motivation (a × b = 0.234, accounting for 100% of its total effect), showing that motivational quality is a key determinant of whether self-regulatory behaviors reduce or exacerbate anxiety. Conclusion: This study suggests that online self-regulated learning does not uniformly alleviate English learning anxiety; rather, its emotional associations depend on the interaction between intrinsic motivation and social-cognitive factors. While excessive self-evaluation may trigger anxiety through heightened self-expectation, adaptive goal setting can enhance intrinsic motivation and indirectly mitigate anxiety. These findings highlight the importance of considering motivational quality and contextual factors when interpreting the emotional outcomes of self-regulated learning, and offer empirical guidance for designing low-anxiety, motivation-enhancing digital learning environments for English language instruction.