How Science Can Reward Cooperation, Not Just Individual Achievement

Accounting for indirect contributions • Key questions • Conclusion

Science has a long history of emphasizing individual achievement. Consider eponymy—the convention of naming a scientific discovery after the individual who discovered it. Eponymy leads us to speak of Newtonian physics, Darwinian evolution, and Pavlovian conditioning. 

There are numerous scientific prizes awarded to either individuals (e.g., the APS William James or James McKeen Cattell fellow awards) or a small number of individuals (e.g., the Nobel prize, awarded to a maximum of three individuals). 

Further, scientists earn grants and promotions largely on the basis of individual achievements, such as first-authored publications, keynote talks at conferences, and past grants as a principal investigators.  

And yet the reality of scientific discovery is different. Knowledge production is often not the result of individual efforts, but instead stems from cooperation and social interaction, including discussions, assistance, criticism, and education.  

The ways that scholars contribute to science via their interactions with others can be thought of as indirect contributions to knowledge production. Indirect contributions can be either positive or negative. For example, positive indirect contributions include being a diligent peer reviewer, serving as a dedicated mentor, improving the work climate in scientific communities, and detecting fraud and statistical errors in published articles. Negative indirect contributions include being an exploitative mentor, engaging in fraudulent or questionable research practices, and overselling research.  

Scientists should be rewarded or punished for their entire range of contributions to science, not just their individual achievements.  

Accounting for indirect contributions 

In our recent article “Shifting the level of selection in science”, coauthored with Karthik Panchanathan and Paul Smaldino and published in Perspectives on Psychological Science, we propose a different approach to scientific recognition and rewards: shifting the focus away from individual scientists and toward the larger groups in which scientists are embedded. We review evidence from diverse fields that have a long history of shifting the level at which rewards and recognition occur. We also elaborate on the implications of such a shift for science policy.  

Why should recognition and rewards shift toward scientific groups?  

For a simple analogy, consider the challenges faced by poultry farmers. Poultry farmers aim to breed hens in ways that maximize lifetime egg production. You might expect that breeders simply select the most productive hens to reproduce. However, reality is not so simple, because each hen’s productivity is affected by the behavior of other hens in their social environment. It turns out that the most productive hens are also the nastiest, feather-pecking and cannibalizing other hens in their coop. As a result, selectively breeding the most productive hens can actually lead to lower overall egg production. 

How do breeders address this problem? Instead of selecting the most productive individuals, breeders can select the most productive coops. By choosing the most productive groups to reproduce, breeders implicitly account for hens’ indirect effects on group members. In one case, using this approach caused mortality to drop from 68% to 9% in just a few generations and laying to increase from 91 to 237 eggs per hen.  

Improving scientific research is a larger challenge than maximizing egg yield. Yet, the underlying principle used—shifting selection away from individuals and toward the groups in which individuals work—has been successfully applied in many fields, including team sports and professional organizations.  

Why is shifting to the group level so powerful? 

Group-level rewards create a shared fate among group members, where each individual’s success becomes tied to the success of the group. This promotes cooperation with fellow group members, while also disincentivizing competition within groups. Shifting selection toward groups thus offers an intriguing, theoretically rigorous, evidence-based approach to recognizing scientists’ indirect contributions, promoting collaboration, and even improving the scientific working climate. 

Key questions 

How can increased recognition and rewards for scientific groups be accomplished in practice? As a starting point, policymakers and professional organizations should consider the following: 

• Scientific research is conducted at multiple levels of hierarchical organization (research labs, departments within universities, entire scientific areas, etc.), with complex interactions between them. At which levels of the organizational hierarchy will recognition and rewards be implemented and why? 
• Scientists operate in a complex environment with diverse incentives, ranging from grant funders’ criteria to standards for obtaining a promotion at their universities. What mechanism of recognition and rewards will be used for implementing the new incentive structure? 
• Scientific contributions are diverse, and include making novel discoveries, improving the reliability of published research, developing theory, and enhancing the happiness and well-being of other scientists. Which outcomes will be the target of the group-level incentives and how will the various outcomes be used to assess whether the new incentives have improved science?  

Group-level incentives have a long history, but few studies examine the consequences of implementing them in the scientific world. As scientists increasingly embrace the importance of collaborative research, now is an ideal time to test alternative mechanisms of recognition and rewards. With this in mind, we propose five practical approaches for shifting the level of selection in science.   

1. Shift scientific evaluation to more strongly weigh group outcomes. 

Just as corporations base employee bonuses on company performance, scientific rewards can hinge on the performance of groups such as labs, departments, or scientific societies.  

2. Provide more funding for supportive positions. 

Scientific reforms should fund permanent positions for individuals whose primary role is not to produce research but rather to help others improve their research output. Such positions could include departmental statisticians, technicians, data managers, mentors and instructors, and individuals who facilitate communication among scientists or between fields.  

3. Create group-level competitions. 

Funders can create competitions that reward scientists for making progress on critical scientific or societal problems. One example is the $1 million prize that the nonprofit Clay Mathematics Institute offers for solutions to unsolved mathematics problems. Funders can also encourage antagonistic collaborations between teams, encouraging scientists with opposing views to jointly advance a disputed knowledge area. The Templeton Foundation has adopted this approach with its Structured Adversarial Collaboration Initiative for research on consciousness.  

4. Provide more funding for scientific “policing”. 

Policing helps suppress harmful behaviors within groups, where individuals gain personal benefits at others’ expense. One approach is to fund independent entities that evaluate research quality or conduct audits of researchers, such as the Strategic Council for Research Excellence, Integrity, and Trust at the National Academies of Sciences, Engineering, and Medicine in the United States. Additionally, grants and permanent positions could be awarded to individuals who engage in scientific criticism, including rigorous peer review, fraud detection, or “red teaming”.  

5. Develop advanced techniques for measuring indirect contributions. 

Current approaches to measuring indirect contributions (such as letters of recommendation to assess collegiality) are relatively unsophisticated, and there is substantial room for innovation. Existing ideas include the Contributor Roles Taxonomy and citation-based algorithms to determine collective credit allocation in multiauthor articles. Methods could also be developed to identify ways in which scientists harm scientific progress or the well-being of their colleagues.   

Conclusion  

Scientists’ behavior is dramatically affected by the system of recognition and rewards. By focusing on individual achievements, current reward structures exacerbate the disconnect between what is best for scientists’ careers and what is best for science. To promote a healthier, fairer, and more cooperative science, policymakers must consider shifting the level at which scientific rewards occur, away from individuals and toward the larger groups in which individuals are embedded. Understanding how to best accomplish this shift should be a top priority for science policy in the upcoming decade.

Back to top

Feedback on this article? Email [email protected] or login to comment.

References 

For readability, we have chosen to exclude references. All references can be found in 

Tiokhin, L., Panchanathan, K., Smaldino, P. E., & Lakens, D. (2023). Shifting the level of selection in science. Perspectives on Psychological Science. https://doi.org/10.1177/17456916231182568