Scientists often base their reasoning on flawed data, overlook confounding factors, or employ incorrect diagnostic methods. Yet, current simulation models of scientific inquiry typically assume that the data underlying scientific assessments is gathered through adequate methods, such as appropriate experimental procedures. In this paper, we introduce a computational model in which scientists can produce and update their beliefs on spurious evidence and ignore properly gathered evidence. In this way, we investigate how misinterpreting evidence impacts the success of the collective inquiry. We find that misinterpretation has two central effects: diversification and distortion. Diversification fosters diverse evidential bases, promoting extensive exploration, while distortion leads to incorrect beliefs, swaying agents towards sub-optimal actions. Using this insight, we show under which conditions misinterpretation benefits a community and what mechanisms allow scientists to profit from it. Moreover, we challenge the predominantly positive view of transient diversity—where scientific communities engage in parallel exploration of rival theories. We demonstrate that under severe distortion, transient diversity intrinsically increases the chance of the community converging on a wrong theory.

When, why and how are scientific ideas worthy of pursuit has long been a subject of philosophical debates. This paper serves as an introduction to the Topical Collection "Pursuitworthiness in Science" by providing an overview of the literature on this topic, explicating the notion of pursuitworthiness judgments and by clarifying its applications to various contemporary discussions in philosophy of science and beyond.

Communication plays a pivotal role in social phenomena such as belief polarization, scientific inquiry, and collective problem-solving. Agent-Based Models (ABMs) are computational tools that simulate the emergence of macro-level phenomena from micro-level interactions among agents. This paper focuses on Argumentative Agent-Based Models (AABMs), a specialized subset of ABMs that study argumentative communication, where agents provide reasons to support or counter opinions. We present a systematic overview of AABMs, detailing their design, methodologies, and applications across disciplines. Key research questions include understanding the dynamics of consensus versus polarization, the conditions for epistemic reliability in collective decision-making, and the mechanisms that foster efficient collaboration within diverse groups through argumentative exchanges. By synthesizing contributions from computer science, social science, and philosophy, this paper serves as both an entry point for newcomers and a comprehensive resource for researchers advancing the study of AABMs.

. This chapter is devoted to robustness analysis, a common practice in modelling, where researchers vary features of a model and study the impact of changes on its behavior. After presenting the three most prominent types discussed in the philosophical literature, the chapter reviews the debate surrounding the epistemic role of this practice, focusing on the contested issue of its evidential import. The discussion highlights the multiple roles of robustness analysis, including the value of not establishing the robustness of a particular modeling result.

In this paper, we present an agent-based model for studying the impact of 'myside bias' on the argumentative dynamics in scientific communities. Recent insights in cognitive science suggest that scientific reasoning is influenced by 'myside bias'. This bias manifests as a tendency to prioritize the search and generation of arguments that support one's views rather than arguments that undermine them. Additionally, individuals tend to apply more critical scrutiny to opposing stances than to their own. Although myside bias may pull individual scientists away from the truth, its effects on communities of reasoners remain unclear. The aim of our model is two-fold: first, to study the argumentative dynamics generated by myside bias, and second, to explore which mechanisms may act as a mitigating factor against its pernicious effects. Our results indicate that biased communities are epistemically less successful than non-biased ones, and that they also tend to be less polarized than non-biased ones. Moreover, we find that two socio-epistemic mechanisms help communities to mitigate the effect of the bias: the presence of a common filter on weak arguments, which can be interpreted as shared beliefs, and an equal distribution of agents for each alternative at the start of the scientific debate.

AbstractScientific disagreements are an important catalyst for scientific progress. But what happens when scientists disagree amid times of crisis, when we need quick yet reliable policy guidance? In this article, we provide a normative account for how scientists facing disagreement in the context of “fast science” should respond and how policy makers should evaluate such disagreement. Starting from an argumentative, pragma-dialectic account of scientific controversies, we argue for the importance of higher-order evidence (HOE), and we specify desiderata for scientifically relevant HOE. We use our account to analyze the controversy about the aerosol transmission of COVID-19.

Scientific disagreements sometimes persist even if scientists fully share results of their research.In this paper we develop an agent-based model to study the impact of diverging diagnostic values scientists may assign to the evidence, given their different background assumptions, on the emergence of polarization in the scientific community.Scientists are represented as Bayesian updaters for whom the diagnosticity of evidence is given by the Bayes factor.Our results suggest that an initial disagreement on the diagnostic value of evidence can, but does not necessarily, lead to polarization, depending on the sample size of the performed studies and the confidence interval within which scientists share their opinions.In particular, the more data scientists share, the more likely it is that the community will end up polarized.

Discussion of epistemic responsibility typically focuses on belief formation and actions leading to it. Similarly, accounts of collective epistemic responsibility have addressed the issue of collective belief formation and associated actions. However, there has been little discussion of collective responsibility for preventing epistemic harms, particularly those preventable only by the action of an unorganized group. We propose an account of collective epistemic responsibility that fills this gap. Building on Hindriks's (2019) account of collective moral responsibility, we introduce the epistemic duty to join forces. Our theory provides an account of the responsibilities of scientists to prevent epistemic harms during inquiry.

Agent-based modelling has become a well-established method in social epistemology and philosophy of science but the question of what kind of explanations these models provide remains largely open. This paper is dedicated to this issue. It starts by distinguishing between real-world phenomena, real-world possibilities, and logical possibilities as different kinds of targets which agent-based models (ABMs) can represent. I argue that models representing the former two kinds provide how-actually explanations or causal how-possibly explanations. In contrast, models that represent logical possibilities provide epistemically opaque how-possibly explanations (Šešelja et al., 2022). While highly idealised ABMs in the form in which they are initially proposed typically fall into the last category, the epistemic opaqueness of explanations they provide can be reduced by validation procedures. To this purpose, an examination of results of simulations in terms of classes of models can be particularly helpful. I illustrate this point by discussing a class of ABMs of scientific interaction and the claim that a high degree of interaction can impede scientific inquiry.

The paper examines recent developments in agent-based modeling of scientific inquiry with a special focus on network epistemology. It provides a survey of different types of ABMs studying network effects in scientific inquiry: ABMs based on bandit problems, ABMs based on epistemic landscapes and ABMs based on argumentative dynamics. It further presents models that study the impact of biased and deceptive researchers on the success of collective inquiry. The paper concludes with a discussion on the contribution of ABMs to the broader field of philosophy of science given their highly idealized nature.

Abstract In this paper I examine the epistemic function of agent-based models (ABMs) of scientific inquiry, proposed in the recent philosophical literature. In view of Boero and Squazzoni’s (2005) classification of ABMs into case-based models, typifications and theoretical abstractions, I argue that proposed ABMs of scientific inquiry largely belong to the last category. While this means that their function is primarily exploratory, I suggest that they are epistemically valuable not only as a temporary stage in the development of ABMs of science, but by providing insights into theoretical aspects of scientific rationality. I illustrate my point with two examples of highly idealized ABMs of science, which perform two exploratory functions: Zollman’s (2010) ABM which provides a proof-of-possibility in the realm of theoretical discussions on scientific rationality, and an argumentation-based ABM (Borg et al. 2019, 2017b, 2018), which provides insights into potential mechanisms underlying the efficiency of scientific inquiry.

The history of the research on peptic ulcer disease (PUD) is characterized by a premature abandonment of the bacterial hypothesis, which subsequently had its comeback, leading to the discovery of Helicobacter pylori the major cause of the disease. In this paper we examine the received view on this case, according to which the primary reason for the abandonment of the bacterial hypothesis of PUD in the mid-twentieth century was a large-scale study by a prominent gastroenterologist Palmer, which suggested no bacteria could be found in the human stomach. To this end, we employ the method of digital textual analysis and study the literature on the etiology of PUD published in the decade prior to Palmers article. Our findings suggest that the bacterial hypothesis of PUD had already been abandoned before the publication of Palmers paper, which challenges the widely held view that his study played a crucial role in the development of this episode. The paper makes two main contributions to the literature in integrated history and philosophy of science. First, we suggest that the received narrative on this historical episode, commonly used by philosophers, needs to be revised. Second, we introduce the notion of a declining research program and argue for its importance as a unit of socio-epistemic analysis, especially in combination with normative assessments, such as pursuitworthiness of scientific theories.

The history of the research on peptic ulcer disease (PUD) is characterized by a premature abandonment of the bacterial hypothesis, which subsequently had its comeback, leading to the discovery of Helicobacter pylori-the major cause of the disease. In this paper we examine the received view on this case, according to which the primary reason for the abandonment of the bacterial hypothesis in the mid-twentieth century was a large-scale study by a prominent gastroenterologist Palmer, which suggested no bacteria could be found in the human stomach. To this end, we employ the method of digital textual analysis and study the literature on the etiology of PUD published in the decade prior to Palmer's article. Our findings suggest that the bacterial hypothesis had already been abandoned before the publication of Palmer's paper, which challenges the widely held view that his study played a crucial role in the development of this episode. In view of this result, we argue that the PUD case does not illustrate harmful effects of a high degree of information flow, as it has frequently been claimed in the literature on network epistemology. Moreover, we argue that alternative examples of harmful effects of a high degree of information flow may be hard to find in the history of science.

This paper is a reaction to 'Styles of Thought on the Continental Drift Debate' by Pablo Pellegrini, published in this journal. The author argues that rationalist accounts of the continental drift debate fail because they overlook important issues. In this discussion we distinguish various forms of rationalism. Then we present a sophisticated rationalist account of the continental drift debate and argue that it is satisfactory because it explains all the central developments in that debate. Finally, we point to a problematic tension in Pellegrini's paper and unravel an underlying ambiguity.

Formal models of scientific inquiry, aimed at capturing socio-epistemic aspects underlying the process of scientific research, have become an important method in formal social epistemology and philosophy of science. In this introduction to the special issue we provide a historical overview of the development of formal models of this kind and analyze their methodological contributions to discussions in philosophy of science. In particular, we show that their significance consists in different forms of ‘methodological iteration’ (Elliott 2012) whereby the models initiate new lines of inquiry, isolate and clarify problems with existing knowledge claims, and stimulate further research.

The article presents an agent-based model (ABM) of scientific interaction aimed at examining how different degrees of connectedness of scientists impact their efficiency in knowledge acquisition. The model is built on the basis of Zollman's ([2010]) ABM by changing some of its idealizing assumptions that concern the representation of the central notions underlying the model: epistemic success of the rivalling scientific theories, scientific interaction and the assessment in view of which scientists choose theories to work on. Our results suggest that whether and to what extent the degree of connectedness of a scientific community impacts its efficiency is a highly context-dependent matter since different conditions deem strikingly different results. More generally, we argue that simplicity of ABMs may come at a price: the requirement to run extensive robustness analysis before we can specify the adequate target phenomenon of the model.(1)

If a given scientific community faces an epistemic harm that could be prevented only by a collective action, what kind of epistemic duties fall on each of the given scientists? In this paper we propose an account of collective epistemic responsibility, which addresses this and related questions. Building on Hindriks’ (2018) account of collective moral responsibility, we introduce the Epistemic Duty to Join Forces as a norm consisting of two sub-norms: first, a duty of individuals to approach other relevant agents raising awareness about the epistemic harm, expressing willingness to prevent it, and encouraging others to do the same; and second, a duty of those who have expressed their commitment to join forces, to prevent the given epistemic harm. We argue that our account has a distinctly epistemic character, irreducible to the accounts of collective moral responsibility. As such, it fills an important gap in the literature on epistemic responsibility. In contrast to previous accounts of epistemic responsibility, which have been concerned with the conditions of responsible belief formation and holding, our approach concerns responsibility for other kinds of performances, specifically those aimed at preventing epistemic harms.