<![CDATA[The rapid advancement of digital technology and artificial intelligence has reshaped the landscape of physics education, emphasizing the need for instructional models that effectively promote students’ critical thinking skills. This study aims to synthesize and map the development of Hybrid STEM–Project-Based Learning (STEM–PjBL) models supported by web-interactive learning environments and deep learning approaches in physics material education. A hybrid review method was employed, integrating a Systematic Literature Review (SLR) and bibliometric analysis of 30 peer-reviewed articles indexed in Scopus and SINTA from 2020 to 2026. The SLR results indicate that STEM-oriented PjBL consistently enhances students’ critical thinking, problem-solving, and conceptual understanding, particularly when implemented through authentic and interdisciplinary projects. Bibliometric findings reveal a growing research trend toward web-based learning and artificial intelligence, with deep learning emerging as a promising yet underexplored component within STEM–PjBL frameworks. However, empirical integration of adaptive deep learning mechanisms in physics material learning remains limited. This study contributes by highlighting critical research gaps and proposing a conceptual Hybrid STEM–PjBL Web-Interactive Model with Deep Learning, which integrates pedagogical design, digital interactivity, and intelligent adaptability to optimize critical thinking development. The findings provide theoretical, practical, and methodological insights for advancing AI-enhanced STEM education.]]>
Copyrights © 2025
