Objective This study aims to evaluate the comprehensive impact of an artificial intelligence (AI)-driven personalized learning platform based on the Coze platform on medical students’ learning outcomes, learning satisfaction, and self-directed learning abilities. It seeks to explore its practical application value in medical education and provide empirical evidence for the digital transformation of education. Methods A prospective randomized controlled trial (RCT) design was adopted, enrolling 40 full-time medical undergraduates who were stratified by baseline academic performance and then randomly assigned via computer-generated block randomization (block size = 4) into an experimental group ( n = 20, AI intervention) and a control group ( n = 20, traditional instruction). The experimental group received a 12-week personalized learning intervention through the Coze platform, with specific measures including: Dynamic learning path optimization: Weekly adjustment of learning content difficulty and sequence based on diagnostic test results; Affective sensing support: Real-time identification of learning emotions through natural language processing (NLP) with triggered motivational feedback; Intelligent resource recommendation: Integration of a 2,800-case medical database utilizing BERT models to match personalized learning resources; Clinical simulation interaction: Embedded virtual case system providing real-time operational guidance. The control group adopted the traditional lecture-based teaching model (4 class hours per week + standardized teaching materials). The following data were collected synchronously during the study period: Academic performance: 3 standardized tests before and after the intervention (Cronbach’s α = 0.89); Learning satisfaction: 5-dimensional Likert scale (Cronbach’s α = 0.84); Self-directed learning behaviors: daily average learning duration recorded in platform logs, classroom interaction frequency (transcription count of audio recordings), and literature reading volume. SPSS 26.0 was used to conduct independent samples t -tests, Pearson correlation analysis, and effect size calculations (Cohen’s d), with a preset significance level of α = 0.05. Results Academic Performance Improvement: The post-test scores of the experimental group were significantly higher than those of the control group (84.47 ± 3.48 vs. 81.72 ± 4.37, p = 0.034, effect size d = 0.72), indicating that the AI intervention yielded moderate to strong practical effects. Learning Experience Optimization, Overall learning satisfaction increased by 8.7% (17.45 ± 3.94 vs. 16.05 ± 3.69, p = 0.042, d = 0.36);Classroom participation significantly increased (16.05 ± 3.36 times/session vs. 7.40 ± 3.57 times/session, p = 0.026, d = 0.83), reflecting the effectiveness of emotional support and interaction design. Enhanced Self-Directed Learning Ability, Daily average learning duration extended by 41.5% (49.25 ± 18.59 vs. 34.80 ± 18.32 min, p = 0.048, d = 0.49); Literature reading volume increased by 48.3% (25.95 ± 7.01 articles vs. 17.50 ± 7.64 articles, p = 0.008, d = 1.14). Correlation Analysis: In the experimental group, self-directed learning duration ( r = 0.261, p = 0.045) and reading volume ( r = 0.409, p = 0.008) showed significant positive correlations with academic performance, validating the platform’s mechanism of promoting deep learning through behavioral intervention. Conclusion AI-driven personalized learning platforms (AI-PLPs) significantly enhance medical students’ learning outcomes, classroom engagement, and self-directed learning abilities through dynamic resource adaptation, affective computing, and behavioral data analysis. The study confirms artificial intelligence’s potential in medical education to balance knowledge delivery and competency cultivation, though its long-term effects and ethical risks require further validation. Future directions include multicenter large-sample studies, longitudinal tracking, and interdisciplinary applications to advance the intelligent transformation of educational models.
Yajun Chen (2025) studied this question.