The emergent integrated network structure of scientific research

TL;DR

This paper uses network science to map and analyze the interconnected structure of scientific research topics in PNAS, revealing increasing integration and interdisciplinarity over time, which correlates with higher impact factors.

Contribution

It introduces a novel network-based approach to visualize and quantify the structure and interdisciplinarity of scientific research, highlighting the importance of integrated knowledge.

Findings

Networks show small-world architecture with dense within-cluster and sparse between-cluster links.

Clusters often span multiple disciplines, not aligning with article classifications.

Interdisciplinarity correlates positively with journal impact factor.

Abstract

The practice of scientific research is often thought of as individuals and small teams striving for disciplinary advances. Yet as a whole, this endeavor more closely resembles a complex system of natural computation, in which information is obtained, generated, and disseminated more effectively than would be possible by individuals acting in isolation. Currently, the structure of this integrated and innovative landscape of scientific ideas is not well understood. Here we use tools from network science to map the landscape of interconnected research topics covered in the multidisciplinary journal PNAS since 2000. We construct networks in which nodes represent topics of study and edges give the degree to which topics occur in the same papers. The network displays small-world architecture, with dense connectivity within scientific clusters and sparse connectivity between clusters. Notably, clusters tend not to align with assigned article classifications, but instead contain topics from various disciplines. Using a temporal graph, we find that small-worldness has increased over time, suggesting growing efficiency and integration of ideas. Finally, we define a novel measure of interdisciplinarity, which is positively associated with PNAS's impact factor. Broadly, this work suggests that complex and dynamic patterns of knowledge emerge from scientific research, and that structures reflecting intellectual integration may be beneficial for obtaining scientific insight.