07.02 The merits and defects of competition

Aim of the project

Scientists are naturally incentivized to adopt a competitive mode. This makes scientific work fast and exciting. Notably, competition has a built-in connotation of shared intention: all parties involved partake in the same race. However, the competitive character of the shared activity also hinders further collaboration. And as current affairs show, science needs functional collaboration schemes to respond to pressing societal demands. The collective search of treatments and vaccines for novel viruses is a case in point. Expectedly, universities and research institutes try to promote collaborations between scientists and disciplines, especially when they don’t arise naturally. How can we motivate competitive scientists to collaborate more naturally? And/or How can we make the adversarial attitude of research epistemically beneficial How can we implement Open Science practices when competitive scientists are prone to secrecy? Our research projects will produce insights that can help to reconcile the merits of collaboration and competition.

Theoretical background

Following a broadly competitive idea of social exchange inherited from economics, many of our social arrangements and institutions are built on the principle that individuals and groups will fight for the same resources and attempt to claim these resources at the expense of others. This conception of human behaviour, dominant in explaining economic exchange, has not only determined our thinking about social exchange, it has influenced the design of our social institutions and thereby shaped our world.

The institutions of science are a good example of this general pattern. Several social processes are at the heart of how science functions: the distribution of scarce research funds, the allocation of attention and reputation, and the distribution and investment of human capital, all these processes are guided by competition [2,3,5]. Researchers and research groups compete for prestigious grants, for the scarce places in high-ranking journals, for talent and for experimental facilities and data.

All the while the goals of the scientists engaged in fierce competition may be perfectly aligned. Medical researchers might all be concerned with finding a cure for some disease, astronomers with discovering the deep structure of our galaxy, ecologists with determining good strategies for conserving species, and so on. Moreover, the competition within science is in itself a shared activity, and recognizably so through formal and informal rules of engagement. Both the shared goals of science and the shared social structure in which competition takes place have led to cooperative arrangements in science, such as data sharing and collaboration on data collection and analysis, and most importantly the sharing of research findings by publication. But competitive elements remain, e.g., the priority rule in journal publishing [1,4]. This raises numerous questions. Are the competitive arrangements conducive to achieving the shared goals? Do they primarily stimulate researchers into greater productivity, or do they rather hamper the progress of scientific understanding?

Research design

Broadly speaking we are interested in the impact of competition on social exchange at three levels of description: the individual, the organisation, and the system. These levels relate loosely to the following questions:

• How are the motivations of scientists impacted by competitive institutions?
• What impacts do competitive arrangements have on epistemic diversity?
• How is information sharing affected by a competitive environment?

These are to some extent empirical questions, which have to be addressed by social psychologists and sociologists in experimental and in real-life settings. But the questions also have more conceptual and normative aspects. Assuming a basic mathematical or computational model of deliberation and information exchange, we can investigate systematically what the consequences are of imposing particular norms, incentive structures, and social arrangements. In addressing these systematic investigations we connect to a thriving field of computational and formal social epistemology, in which the epistemic characteristics of various social arrangements are being studied (e.g., [5,6]).

In our project we aim to investigate a number of specific hypotheses about the impact of competition on cooperation. On the one hand there are reasons to suspect that competition negatively impacts cooperation at all levels and in all areas. Competition is easily seen as setting in motion a vicious cycle, i.e., a negative feedback loop. Selective information sharing will fragment a research community and reduce access to epistemic diversity, and this will feed into more selective information sharing (hypothesis 1). Scientists who have more cooperative inclinations will be less motivated to contribute, and this will exacerbate the already existing tendency towards competitiveness (hypothesis 2).

On the other hand, competition is robustly present in our social institutions, in science as much as anywhere else. So: what could be its value or benefit? One option is that it has positive effects in a particular phase in the development of an epistemic community (hypothesis 3), namely as a way to diversify the set of approaches entertained in the community. Another option is that it is useful for reaching specific types of sub-goals, as part of a larger collaborative effort (hypothesis 4), for instance when the sub-goal is clearly delineated and mostly to do with optimization. Our hope is that systematic and conceptual studies will help us to a better understanding of why so-called weak solidarity, in which competitive and cooperative aspects are brought together, is effective. Ultimately this may contribute to better guidelines for how social institutions, in science and elsewhere, can benefit from competitive elements, without falling prey to their drawbacks.

Methodologically, the projects will use a variety of tools from philosophy, social sciences, and social studies of science (STS). On the one hand, in addition to traditional conceptual analysis, we will adopt technical tools used today to do theoretical work in social epistemology. These include formal modelling with game-theoretic tools and computational simulations of social systems. With these tools, we will be able to explore the different hypotheses in a principled way. Importantly, we will do this modelling work in an empirically-informed fashion, calibrating the models with close attention to relevant empirical literature from the social and behavioural sciences. This will allow us to develop the projects accounting for the limitations of purely formal models and produce insights that can be informative about real-case scenarios. In parallel, given that the projects intend to be sensitive to contextual and historical aspects, we will develop case studies that account for the particularities of the different institutions (e.g., scientific institutions) in question.

Literature

• Romero, F. (2017). Novelty versus replicability: Virtues and vices in the reward system of science. Philosophy of Science, 84(5), 1031-1043.
• Stephan, P. E. (2012). How economics shapes science. Harvard University Press.
• Strevens, M. (2003). The role of the priority rule in science. Journal of Philosophy, 100(2), 55–79.
• Tiokhin, L., & Derex, M. (2019). Competition for novelty reduces information sampling in a research game - a registered report. Royal Society Open Science, 6(5), 180934.
• Zollman, K. J. S. (2018). The credit economy and the economic rationality of science. Journal of Philosophy, 115(1), 5–33.
• Heesen, R. & Romeijn, J.W. (2019). Epistemic Diversity and Editor Decisions: a Statistical Matthew Effect. Philosophers’ Imprint, Vol. 19, No. 39.

Project stakeholders

dr. Felipe Romero (Philosophy)