<![CDATA[The Predict–Observe–Explain (POE) model has been widely implemented in primary science education to support conceptual change and the development of science process skills. However, research on POE continues to exhibit conceptual fragmentation, reflected in the diversity of theoretical foundations, instructional designs, and learning outcome orientations. Furthermore, the conventional POE model has not fully accommodated the demands of twenty-first-century learning, which emphasize collaboration and the integration of technology. Therefore, this study aims to examine the theoretical foundations, instructional designs, and learning outcomes associated with POE, while reconceptualizing the model within the context of primary science education. This study employed a Systematic Literature Review (SLR) based on the PRISMA 2020 guidelines. Articles were retrieved from the Scopus, Web of Science (WoS), and ERIC databases, covering publications from 2010 to 2025. A total of 78 articles that met the inclusion criteria were analyzed using a thematic synthesis approach. The findings reveal that constructivism, conceptual change theory, inquiry learning, and social constructivism constitute the dominant theoretical foundations underpinning POE research. The implementation of POE has evolved toward investigation-based, collaborative, and technology-enhanced learning approaches. The review also indicates consistent positive effects on conceptual understanding, science process skills, critical thinking, communication, and collaboration. Based on these findings, this study proposes the PICOTA (Predict–Investigation–Collaboration–Technology Application) conceptual framework as a more relevant science learning model for addressing the needs of twenty-first-century education.]]>
