Formative assessment as a self-regulation engine in blended language learning: A case study with top-tier engineering talents

This study addresses the gap between result-oriented evaluation and process-oriented learning by examining how formative assessment operates as a self-regulation (SRL) engine within a learning-centered blended Japanese course for top-tier engineering students at Xi'an Jiaotong University. Using a mixed-methods design, the study involved 40 first-year engineering students enrolled in the course (quantitative questionnaire and performance data, N = 40), complemented by two rounds of reflective learning journals ( n = 87 entries) and semi-structured interviews with four representative participants. Formative mechanisms were embedded across online–offline activities (SPOC analytics, periodic checkpoints, reflective journals, peer interaction) to create a sustainable, dialogic feedback ecology. A mixed-methods design shows that continuous, multi-agent feedback strengthens learners' motivation, strategy use, and reflective regulation, reduces anxiety, and supports the internalization of learning goals. The model turns evaluation into a cyclical SRL process that sustains engagement and improves language learning quality. The study offers practical guidance for curriculum reform in foreign-language education and contributes a transferable blueprint for engineering contexts where autonomous, lifelong learning capacities are essential. • Formative assessment functions as an SRL engine in blended Japanese courses • A mixed-methods design combines performance data, journals, and interviews • Continuous feedback enhances motivation, self-regulation, and engagement • Findings inform sustainable formative assessment in engineering education