What’s Hot in Mathematical Philosophy Dunja ˇ Seˇ selja May 11, 2019 Computational modeling in the form of agent-based models (ABMs), with a long tradition in biomedical and social sciences, has become increas- ingly popular in philosophy of science and social epistemology. In particular, simulations of scientific inquiry have been used for tackling a variety of ques- tions concerning social aspects of science: from the impact of different social networks on the efficiency of knowledge acquisition, to the division of cogni- tive labor, to the study of norms that guide scientists facing disagreements etc. At the same time, the proposed models tend to be highly idealized, raising the question: What can we learn from them? As a result, discus- sions on the epistemology and methodology of models and idealizations have intensified in recent years, marking a new phase in the literature on ABMs of science. We can roughly distinguish three phases in the research on ABMs of scientific inquiry (inspired by Thiele’s et al. (JASSS, 17(3)11, 2014) two- phase distinction of the research on ABMs in general). The first phase is marked by the introduction of agent-based modeling to philosophy of science as a method that can be fruitfully applied to some important problems (some of which are mentioned above). The pioneering works of Zollman, Weisberg and Muldoon, Grim and Singer, Douven (building on the famous Hegselmann and Krause’s model), De Langhe—among others—kick-started this line of research around 2010. Most work done in this phase aimed at showing how to simulate scientific inquiry and how to use the simulations to fruitfully tackle philosophical problems. The emphasis here was on the fertility of the method, rather than on the reliability of the models or on the specification of their explanatory features. In the second phase, previously proposed ABMs have been put to robust- ness tests and extended to novel application contexts. The robustness analy- sis includes the examination of the results of a model with respect to changes in parameter values (sensitivity analysis) and with respect to changes in 1