Evolution of collaboration and optimization of impact: self-organization in multinational research

The expanding presence of multinational research teams highlights the importance of characterizing the outcomes of international collaboration. Herein, we characterize the organization of international collaborations over four decades and report several advances which may be instructive of strategies to promote effective multinational teams. First, we uncover immutable and scale-dependent collaborative behaviors of countries governing the degree of foreign interactions and distribution of ties in multinational research. Second, we show an increased likelihood over time for countries to collaborate with nations of unequal scientific, economic, and social stature. Third, we demonstrate that temporal changes in foreign interactions result in the optimization of citation gains globally despite a lack of an association between the incidence of ties and impact in scientific relationships. These findings suggest that the assembly of multinational teams is naturally bounded and a self-organizing process. We propose that competing altruistic and egocentric citation priorities may influence partner selection resulting in modern multinational teams engendering greater absolute citation measures. Central to self-organizing systems such as market economies is the preservation of autonomy and easy diffusion of information, suggesting that enforcing these elements in science may enhance multinational teams.