What are behavioral sciences? Epistemic boundaries and paradigms from a Delphi study

What are the Behavioral Sciences? Despite their widespread use in governments, international organizations, and policy innovation contexts, the term circulates with conceptual ambiguity and permeable boundaries. From a Science and Technology Studies (STS) perspective, this ambiguity is not an anomaly but evidence of an evolving and heterogeneous epistemic configuration, marked by competing epistemologies, practices, and regimes of expertise. Yet few studies have examined how specialists themselves define and demarcate this field. This study seeks to answer this question by examining the current understanding of the Behavioral Sciences using a two-round Delphi design informed by scientometric mapping, interviews, and a survey. The scientometric stage analyzed 4429 open-access publications (1956–2023) indexed under the term behavioral sciences, identifying 47 structuring references through co-citation network analysis. Subsequently, semi-structured interviews were conducted with 27 experts from universities, governments, and international organizations (Round 1), followed by an online questionnaire with 23 experts (Round 2) to validate areas of consensus and dissent regarding definitions, paradigms, and methods. The results suggest the presence of three main paradigmatic configurations, namely decision-making processes (95.6%), individual behavior planning (91.3%), and behavior change according to context (69.6%). In terms of methods, the findings indicate the centrality of an empirical-experimental axis, particularly advanced statistical analyses and randomized controlled trials. A high level of consensus was reached regarding the proposed general definition of the Behavioral Sciences (91.3% agreement among experts), whereas consensus was low regarding the object of study (50%). In summary, a key contribution of this study is demonstrating that Behavioral Sciences cannot be reduced to a set of techniques aimed at inducing desirable behaviors but should instead be understood as a sociotechnical project of knowledge production and social intervention.