Centralized Funding and Epistemic Exploration

Abstract

Computer simulation of an epistemic landscape model, modified to include explicit representation of a centralized funding body, show the method of funding allocation has significant effects on communal trade-off between exploration and exploitation, with consequences for the community’s ability to generate significant truths. The results show this effect is contextual, and depends on the size of the landscape being explored, with funding that includes explicit random allocation performing significantly better than peer review on large landscapes. The article proposes a way of incorporating external institutional factors in formal social epistemology, and offers a way of bringing such investigations to bear on current research policy questions. 1Introduction2Theoretical Background3Model Description4Simulation Details 4.1Simulating the epistemic landscape4.2Simulating agents4.3Simulating communal knowledge4.4Simulating funding strategies4.5Simulating merit dynamics5Results and Discussion 5.1Experiment 1: The winner-takes-it-all mechanism only5.2Experiment 2: All dynamic mechanisms5.3Experiment 3: Adding a new funding mechanism (triage)5.4Experiment 4: Varying the degree of myopia5.5Experiment 5: Variability of individual epistemic gain5.6Experiment 6: Likelihood of renewal6Discussion7Conclusion

A mid-level approach to modeling scientific communities

“…First, detailing a process for generating mid-level models by incorporating empirical information into high-level theoretical models. Second, providing an exemplar of a mid-level model by extending high-level models of science funding allocation (Avin 2017). The results of my model show that allocating funding by modified lottery systems generates comparable results to other funding methods.…”

“…The original landscape used by Weisberg and Muldoon (2009) consists of a flat landscape with two peaks, one higher than the other (Figure 1). Many modelers of scientific communities have drawn on the concept of an epistemic landscape, and different models incorporate new features or more complicated landscapes (Grim 2009, Grim et al 2013, Alexander et al 2015, Thoma 2015, Muldoon 2013, Avin 2017, Pöyhönen 2016. In section 3 I discuss my model, which uses a similarly structured epistemic landscape.…”

“…In a how-possibly model there is no need to show that the modeling results are likely, or that they do in fact obtain in the world; the results are merely possible. Other authors have advocated the use of agent-based models as formalized thought experiments (Avin 2014, Currie andAvin, 2017). While these are productive uses for such models, my project takes a different approach.…”

“…I start with a high-level agent-based model of science funding developed by Avin (2017). Avin's model provides the structure for two essential features of science funding systems: 1) a distribution of scientific projects 2) mechanisms for selecting which projects to fund.…”

Argumentative landscapes: the function of models in social epistemology

“…For this reason, it is important to pay close attention to their details and argumentative goals. For example, Shahar Avin's (2019) main aim is to examine the role of funding-allocation mechanisms. Deviating from the original goals of the Weisberg-Muldoon model, Avin primarily attempts to direct the attention of social epistemologists from individual motives and learning strategies to institutional arrangements.…”