The ability to properly evaluate one’s own academic progress has long been considered a predictor of academic success. However, its distinctive role in the context of computational mathematics remains underexplored. Grounded in social cognitive theory, this study investigates the critical role of self-regulated learning (SRL) strategies in enhancing mathematics learning, particularly in programming-based contexts. Focusing on two components of SRL, self-awareness and reflection, the study provides empirical evidence on the psychological effectiveness of SRL in academic outcomes through the implementation of an e-portfolio-based intervention. Using Bayesian inference, the study models individual learning processes, offering personalized insights for effective educational interventions. The analysis reveals that the use of e-portfolios significantly fosters self-awareness and enhances learning among students. Nevertheless, the study also addresses psychological challenges in programming-based mathematical education, such as complex problem-solving and abstract thinking. The findings highlight the need for interactive, technology-enhanced teaching approaches to keep university-level students engaged and motivated. Key psychological implications are discussed for relevant measures in mathematics education.