A groundbreaking international study involving Kobe University has recently challenged longstanding assumptions about human cooperation in social networks. By redesigning classic economic games to grant individuals the freedom to tailor their behavior toward different social partners, researchers found dramatic increases in cooperative, trusting, and fair behavior. This research not only calls into question traditional experimental methods that restrict behavioral options but also offers profound insights into the dynamics of social interaction and human prosocial potential.
Economic and social scientists have long relied on game theoretic models such as the prisoner’s dilemma, trust games, and ultimatum games to capture decision-making processes within networks. These frameworks have consistently suggested that true cooperation is relatively rare, with only about one in seven players maintaining cooperative behavior over repeated interactions. However, previous designs often imposed uniform behavior constraints on participants, requiring them to treat all members of their network identically. This simplification overlooked a fundamental human skill: managing differentiated social relationships and adapting behaviors accordingly.
Motivated by this limitation, a collaborative effort between Kobe University’s computational social science department and researchers from Northwestern Polytechnic University in Xi’an introduced a novel experimental setup. In this new paradigm, over 2,000 university students across China participated in iterations of classical economic games, but crucially, some were allowed to exercise ‘networking agency’—the ability to customize their strategic decisions toward different peers. This feature more realistically mimics real-world social environments where individuals modulate trust, cooperation, or fairness depending on complex social cues and histories.
The results were nothing short of remarkable. As published in Nature Human Behaviour, unrestricted participants exhibited cooperation rates soaring from a meager 14% under constrained scenarios to an astonishing 80% under full agency. This transformative leap was echoed in enhanced displays of trust and equitable behavior across the trust and ultimatum games. Even when only subsets of players enjoyed behavioral freedom, the ripple effects elevated prosocial tendencies throughout the network, albeit temporarily amplifying inequality as adept actors leveraged their agency advantages. A key observation was that when everyone had full agency, inequality diminished as collective wealth rose, indicating a more balanced and productive social fabric.
Beyond aggregate behavior, the temporal analysis revealed that agency influenced player strategies from the very first round. Coauthor Danyang Jia emphasized that this was less about learning curves and more about the intrinsic ability to initiate differential interactions. Populations with agency evolved toward conditional cooperation—embodied in strategies such as tit-for-tat—and generous trust, while constrained groups gravitated toward antisocial default strategies, not due to inherent selfishness but because of environmental limitations on their choices.
This nuanced understanding suggests that prior experimental frameworks have systematically underestimated the prosocial potential embedded within human sociality by artificially restricting interaction heterogeneity. By failing to incorporate the capacity for individualized social calibration, traditional models gave a distorted picture of how cooperation, trust, and fairness naturally emerge and stabilize within groups.
The implications of these findings ripple across disciplines. In sociology, anthropology, psychology, and economics, accurately modeling social cooperation requires systems that reflect real-world interaction diversity. More succinctly, behavioral experiments must allow for differentiated social agency if they are to capture authentic human dynamics. Ivan Romić of Kobe University stresses the ethical and practical value of equalizing individuals’ opportunities to personalize their social engagements, positing that such equality fosters enhanced prosociality and, consequently, healthier social ecosystems.
This study represents a significant methodological advancement in experimental social science, coupling large-scale participant pools with intricate behavioral freedom manipulations. The research was supported by prestigious funding sources including the National Science Fund for Distinguished Young Scholars, the Japan Society for the Promotion of Science, and leading Chinese science foundations. Collaborative ties stretched across influential institutions such as the Northwestern Polytechnical University, Yunnan University of Finance and Economics, Shanghai Jiao Tong University, Aalto University, and China Telecom Corp Ltd, underscoring the multidisciplinary and international commitment to advancing understanding of human social behavior.
Kobe University, which spearheaded the research alongside these global partners, is anchored in a rich academic tradition since 1902 and currently serves as one of Japan’s prominent comprehensive research universities. Its dedication to integrating natural and social sciences to cultivate interdisciplinary leaders has propelled its contributions to unveiling the complexities of social cooperation and behavioral economics.
By inviting individuals to act with nuanced discretion toward different network members, this research illuminates how humans employ social intelligence to construct trust, cooperation, and fairness much more robustly than previously recognized. The findings advocate for a paradigm shift in economic game design and social experimentation, moving toward frameworks that facilitate tailored social interactions. Such evolution in research practice will better capture the depth of human prosociality and offer strategic pathways for cultivating cooperation in diverse social systems.
This new understanding resonates deeply in contemporary social challenges where mistrust and inequality hinder collective well-being. It suggests that allowing or even encouraging individuals to calibrate their social behaviors individually rather than assume uniform rules can unleash latent cooperative potential. From organizational behavior and digital platforms to policy design and community building, the principles revealed here could guide more effective interventions for fostering trust and equity in complex societal networks.
At its core, the study is a testament to the intricate interplay between freedom of choice and prosocial behavior. By bridging computational social science with experimental economics, it paves the way for richer, more accurate models of human sociality. As behavioral experiments evolve to incorporate these dynamic elements, our understanding of cooperation, trust, and fairness will expand—potentially transforming how societies conceptualize and nurture collaboration in an increasingly interconnected world.
Subject of Research: People
Article Title: Social networking agency and prosociality are inextricably linked in economic games
News Publication Date: 24-Sep-2025
Web References: https://doi.org/10.1038/s41562-025-02289-0
References: Published study in Nature Human Behaviour
Image Credits: Kobe University, created with material from Alexander Kaufmann, Amanda Jones, Heather Gill, Max Harlynking, Paola Aguilar, Rita Chou, Style Studio and Tim Mossholder via Unsplash
Keywords: social cooperation, prosocial behavior, economic games, prisoner’s dilemma, trust game, ultimatum game, social networks, behavioral experiments, networking agency, fairness, human sociality, interdisciplinary research
