The issues of metacognitive scaffolding in science education (MSiSE) have become increasingly popular and important. Differing from previous content reviews, this study proposes a series of quantitative computer-based analyses by integrating document co-citation analysis, social network analysis, and exploratory factor analysis to explore the intellectual structure of the MSiSE literature (i.e. the relationships within and between subfields of MSiSE). Co-citation refers to any two articles that are jointly referenced in other articles. After the computation of co-citation analysis, 27 articles that have been co-cited at least once by follow-up studies as references were identified as the final set of core articles. The whole co-citation profile of 27 cores with the 434 links was then visualized in a network through social network analysis, representing an overview for the intellectual structure of core MSiSE studies. The most cross-referenced underpinnings in the network focused on adaptive scaffolding for self-regulated learning to enhance students' conceptual understanding and on younger students' metacognition in online science inquiry learning environments. Furthermore, two emerging topics in the network were identified through an exploratory factor analysis as "non-technological metacognitive scaffolding media," and "behavior patterns & task analysis in technology-infused environments." Overall, the study provides an innovative review method of scholarly communication in the MSiSE literature.