Belief Polarization

Political polarization is one of the most discussed challenges facing contemporary democracies and is often associated with a broader epistemic crisis. While inspiring a large literature in political science, polarization’s epistemic problems also have significance for normative democratic theory, and this study develops a new approach aimed at understanding them. In contrast to prominent accounts from political psychology—group polarization theory and cultural cognition theory—which argue that polarization leads individuals to form unreliable political beliefs, this study focuses on system-level diversity. It (...) argues that polarization’s epistemic harms are best located in its tendency to reduce the diversity of perspectives utilized in a democratic system and in how this weakens the system’s ability to identify and address problems of public concern. Understanding such harms is also argued to require a greater consideration of the political dynamics of polarization and issues of elite discourse, alongside political psychology. (shrink)

Handbook Abstract: Beliefs play a central role in how we understand ourselves, each other, and the world. So it’s no surprise that decades of research across the social sciences have been dedicated to documenting the particular beliefs that people hold. But what makes something a belief, rather than a different kind of representation or commitment? What does it mean to believe? These questions have received a great deal of attention within philosophy, but without the benefit of systematic, empirical investigation. That’s (...) beginning to change as psychologists, anthropologists, and other cognitive scientists broaden the scope of empirical and theoretical work on belief. How does belief relate to perception and memory? Do we have control over what we believe? How did the capacity for believing evolve? Can nonhuman animals have beliefs? Does believing come in degrees? And do we have ethical obligations to believe certain things and not others? All these questions and more are addressed in this Handbook, which brings together state-of-the-art work in philosophy and the cognitive sciences—including, anthropology, cognitive psychology, computational modeling, political science, and more. The result is a resource for novices and experts alike, not only detailing recent progress, but also charting a course for further developments in the cognitive science of belief. (shrink)

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. (shrink)

Conciliationism holds that it is rational to modify one's beliefs in the face of disagreement. But extant conciliatory norms yield incorrect results in cases involving excessive confidence--cases in which one's interlocutors are sure, or almost sure, that their opinions are correct. After explaining the problem of excessive confidence, I show that a Bayesian approach to Conciliationism handles the problem elegantly and effectively. Further, it has desirable--indeed essential--features, including the ability to (i) contend with multiple interlocutors, (ii) deal with gradations in (...) competence, (iii) incorporate ubiquitous interdependence, and (iv) account for critical contextual features of cases of disagreement. (shrink)

Standardly, echo chambers are thought to be structures that we should avoid. Agents should keep away from them, to be able to assess a fuller range of evidence and avoid having their confidence in that information manipulated. This paper argues against that standard view. Not only can echo chambers be neutral or good for us, but the existing definitions apply so widely that such chambers are unavoidable. We are all in large numbers of echo chambers at any time – they (...) can be found not just on social media or in political groups, but in almost every social or epistemic group we could categorise ourselves into. Because we are finite and fallible, we cannot escape them and need to exist in them just to get by. The concept, then, does not actually capture something as structurally problematic as the paradigmatic cases would suggest. Our way of using the term in social epistemology needs to change. (shrink)

In recent years, the term “Christian nationalism” has been popularized widely in mass media and academia. But the concept is still not well understood, and its connection to other MAGA-related phenomena like QAnon or “Christofascism” warrant examination. In what follows, I cover each in turn as they relate to President Trump’s MAGA base of support.

Contemporary discourse around belief carries an unstated assumption that, once made explicit, almost nothing in it survives. Our assumption is that what people think matters in itself. Epistemology treats belief-coherence as intrinsically valuable. Media criticism worries about misinformation as if false beliefs were inherently corrosive. AI alignment frets over models producing untruths as if untruths were a load-bearing concern. Education optimizes for the transmission of correct propositions. Across these domains, the working premise is that the coherence and accuracy of a (...) person's worldview is itself the thing at stake, and that getting people to hold better beliefs is therefore a high-value operation. That premise is undefended, and on examination, it does not survive. What matters is not belief but action, and what selects against incoherence is not the symbolic operators correcting wrong views but reality vetoing failed behavior. Once this is seen clearly, almost everything that gets discussed under the heading of belief turns out to be either irrelevant to persistence or load-bearing through a mechanism the discourse misidentifies. The derivation that follows works through the structural argument, names what actually matters at the individual scale, and identifies why the discourse keeps drifting back to belief-content even when its own framework would direct attention elsewhere. (shrink)

Polarisation is a widespread societal issue, dividing populations into opposing groups whose beliefs may span multiple, seemingly unrelated, topics. This phenomenon creates distinct cultural groups with internally consistent yet mutually opposing beliefs, strengthening group identities and deepening societal divides. We develop a mathematical model to investigate how polarisation and emergent associations between unrelated beliefs occur. Unlike prior models assuming irrationality, informational segregation, or similarity biases, our model employs an epistemic distinction between two filtering mechanisms: content and source filtering. In content (...) filtering, agents evaluate beliefs based on intrinsic coherence, rejecting information opposing their current beliefs. Source filtering, in contrast, involves adopting beliefs based on the perceived credibility of information sources, where credibility corresponds to mutual consistency between individuals’ belief systems. We further incorporate external signals through varying innovation and loss rates to reflect empirical biases from real-world data. Simulations show that source filtering rapidly clusters beliefs into two dominant, highly polarised groups. Within each group, beliefs become tightly correlated, creating strong opposition between the groups across all cultural traits. Thus, new associations emerge spontaneously between previously unrelated beliefs. Arbitrary and initially independent beliefs transform into ideological bundles serving as clear group signals. External signals further reinforce these associations, allowing advantages of particular beliefs (e.g. empirical evidence) to transfer to associated beliefs. Content filtering can also produce polarised groups, but beliefs remain aligned with preexisting logical or factual structures. Source filtering mirrors algorithmic biases such as collaborative filtering prevalent on digital platforms, where content exposure depends heavily on shared user preferences rather than intrinsic coherence. Our findings illustrate how these biases amplify polarisation by reinforcing homogeneous belief packages, intensifying ideological divides within cultural systems. (shrink)

Strategic science skeptics criticize scientific claims solely to promote non-epistemic goals. I will analyze and debunk a philosophically neglected argument exploited by strategic science skeptics: the argument from disagreement. The core of this argument is that one should lower one’s confidence in a scientific claim when having learned that there is a scientific disagreement about this claim. I will develop a (Bayesian) Justificatory Account of Multiple Testimony to provide a normative characterization of how learning about agreements and disagreements is connected (...) to confirming and disconfirming scientific claims. I will use this account to debunk the argument from disagreement. (shrink)

The main purpose of my paper is to work out an experiential notion of religious diversity. This means characterising religious diversity in terms of relational and doxastic features. Such a proposal differs from mainstream approaches to religious beliefs (at least from a philosophical viewpoint) because these handle the epistemic dimension of faith as a purely epistemic matter. On the contrary, my idea consists of highlighting how the epistemic evaluation of opposing religious propositions is the outcome of an interpersonal process of (...) evidence sharing wherein the particularities of the involved individuals matter. In the introductory section, I will define the topic of my paper. In the subsequent one, I will characterise how the mainstream approach to religious diversity and my own contrast. In the third section, I will develop in a few details the main reasons for why the epistemic approach is unsatisfactory. In the fourth one, I will provide a thought experiment for religious diversity, and I will set forth my considerations about what the thought experiment shows, with a focus on issues about the epistemology of religious disagreements. I will conclude my paper with a brief overview of the main consequences of my proposal. (shrink)

Sami Pihlström argues that, for principled reasons, we have a duty not to listen to racists. Although this stance can seem admirable, I worry that by cutting itself off from evidence, a refusal to listen leaves wrongfully accused persons no means of exonerating themselves. Moreover, given that concepts like racism now encompass beliefs and acts that many rightly consider sensible, a policy of silence risks dismissing implausibly large numbers of people as immoral. Stressing that listening is not acquiescing, I urge (...) Pihlström to think more carefully about the consequences of his stance, especially since it would increase the likelihood of conflict. (shrink)

Much work in feminist social epistemology shares the basic assumption that ignorance generated by silencing (and related phenomena) is undesirable. While true in many contexts, I depart from that consensus, introducing and offering a philosophical account of a social epistemic practice that I term ‘hushing-up.’ This practice has emerged organically in response and in resistance to sectarian bigotry in Northern Ireland (NI), and contributes to post-conflict transitions away from social and political polarisation and its derivative harms to individuals and collectives. (...) In spaces where hushing-up is deployed, ignorance of peoples’ community background is intentionally cultivated as a strategy in opposition to bigotry, while bigoted or sectarian displays incur social costs. Positioning hushing-up against existing accounts of political ignorance, and contributing to the idea of an epistemology of ignorance, I distinguish between performative partisanship (involving ignorance driven by social rewards for showing group allegiance) and pressured partisanship (involving ignorance driven by the avoidance of social sanctions for displaying indicators of disloyalty to a group). On the account I defend, shifts in social norms are central to addressing some forms of harmful ignorance. I argue that hushing-up succeeds in doing this in two ways: first, by reducing the rationality of pressured partisanship by providing alternative social support to that provided in polarised social spaces such that the cost of dissent is lowered; and secondly, by disincentivising performative partisanship through increasing the difficulty of identifying targets and ensuring that bigoted displays are met with disapproval in spaces where that norm is in force. (shrink)

Argumentation is often conceived as a rational response to disagreement, even when it does not resolve differences of opinion. Arguing in the face of disagreement has, however, distinctive epistemic effects. Sometimes argumentation achieves convergence of opinion or at least the mutual recognition that a more thorough inquiry is required. But facing disagreement, participants of argumentative exchanges quite often remain steadfast in their initial views or even radicalize them. Can we make sense of these latter situations? To account for their occurrence, (...) it is common to point out that people’s ability to argue is flawed, that an “argumentative culture” is lacking, and that emotional and other non-rational factors often interfere in confrontative situations. But these suggestions do not amount to a thorough satisfactory explanation. In this paper, I provide the outline of a purely epistemic account of these peculiar effects of argumentation in the face of disagreement. I argue that probabilistic models of degrees of confidence (or “credences”) can shed light on the conditions that give rise to several of these effects. This could provide some guidance on how to avoid them. (shrink)

Conspiracy theories have mainly been of interest to social epistemologists in terms of whether one can be warranted in believing them. In this literature, believing a conspiracy theory is often understood to mean endorsing some conspiratorial explanation of events. I argue that, in some cases, conspiracy belief is better understood as (dis)trusting sources of claims. To demonstrate this, I show that disputes over conspiracy theories possess a distinctive tendency (but not a necessity) to generate deep disagreement arising specifically from divergent (...) attributions of trust in others. I explain this tendency in terms of agents’ desires to maintain consistency among their beliefs and avoid inquiring into their attributions of trust. Individuals who believe conspiracy theories often do not remain committed to the same conspiratorial explanations but do nonetheless remain firmly committed to attitudes of trust and distrust. By arguing that conspiracy belief often manifests as (dis)trust, I challenge a common assumption in the philosophy of conspiracy theories that conspiracy belief, understood as taking conspiracy explanations to be true, can always serve as an appropriate basis for asking when believing conspiracy theories is warranted. (shrink)

How do opinion diversity and belief polarisation affect epistemic group decision-making, particularly if decisions must be made without delay and on the basis of permissive evidence? In an agent-based model, we track the consistency of group opinions aggregated through sentence-wise majority voting. Simulations on the model reveal that high opinion diversity, but not polarisation, incurs a significant inconsistency risk. These results indicate that epistemic group decisions based on permissive evidence can be particularly difficult for diverse groups. The results also improve (...) our understanding of what can reasonably be expected of expert groups, and where expert advice might have limits. (shrink)

To better navigate social interactions, we routinely (consciously or unconsciously) categorize people based on their distinctive features. One important way we do this is by ascribing political orientations to them. For example, based on certain behavioral cues, we might perceive someone as politically liberal, progressive, conservative, libertarian, Marxist, anarchist, or fascist. Although such ascriptions may appear to be mere descriptions, I argue that they can have deeper, regulative effects on their targets, potentially politicizing and polarizing them in ways that remain (...) underexplored in research on political polarization. To capture this dynamic, I introduce the notion of ‘politicizing mindshaping,’ distinguish it from other types of politicization, and review evidence suggesting that politicizing mindshaping is likely common. Finally, I contend that the pernicious effects of politicizing mindshaping may currently be significantly underestimated. This is because, unlike in many other cases of social labeling, negativity in political labeling is, to some extent, widely tolerated – even encouraged – and largely unconstrained by social norms in democratic countries. (shrink)

In this paper I explore Phyllis Rooney’s 2003 chapter, “Feminist Epistemology and Naturalized Epistemology: An Uneasy Alliance,” taking guidance from her critique of naturalized epistemology in pursuing my own analysis of another uneasy alliance: that between feminist epistemology and social epistemology. Investigating some of the background assumptions at work in prominent conceptions of social epistemology, I consider recent analyses of "epistemic bubbles" to ask how closely such analyses are aligned with ongoing research in feminist epistemology. I argue that critical feminist (...) insights in political philosophy are called for here, especially those that focus on multiple politically salient relationships and also emphasize practices of personal transformation as inseparably bound up with broader movements for social transformation. Insofar as feminist epistemology continues to carry another world in its heart, I argue that it will continue to rest uneasily with any effort to align it with—let alone subsume it within—the still emerging field of social epistemology. (shrink)

I set out a core challenge to liberal DEIJ (diversity, equity, inclusion, and justice) approaches. The default view within United States liberalism is that if people know better, they’ll do better. Thus, liberal approaches to increasing justice emphasize the importance of listening, learning, sitting in discomfort, dialoguing, and collaborative problem solving. The challenge to this approach is that it doesn’t work, and moreover that it cannot work, because the conceptual resources and material conditions of U.S. liberalism itself prevent the implementation (...) of significant and meaningful change. As Vivian, Mills, and others argue, conversations and language shifts will ultimately fall short of bringing about the necessary institutional changes because U.S. liberalism itself will prevent any such radical structural shifts. If this is correct, then we are faced with a stalemate, and must think about what to do when language and conversations fail to bring about justice more equally enjoyed by all. (shrink)

Computational models have shown how polarisation can rise among deliberating agents as they approximate epistemic rationality. This paper provides further support for the thesis that polarisation can rise under condition of epistemic rationality, but it does not depend on limitations that extant models rely on, such as memory restrictions or biased evaluation of other agents’ testimony. Instead, deliberation is modelled through agents’ purposeful introduction of arguments and their rational reactions to introductions of others. This process induces polarisation dynamics on its (...) own. A second result is that the effect size of polarisation dynamics correlates with particular types of argumentative behaviour. Polarisation effects can be soothed when agents take into account the opinions of others as premises, and they are amplified as agents fortify their own beliefs. These results underpin the relevance of argumentation as a factor in social-epistemic processes and indicate that rising issue polarisation is not a reliable indicator of epistemic shortcomings. (shrink)

Many philosophers hold that generics (i.e., unquantified generalizations) are pervasive in communication and that when they are about social groups, this may offend and polarize people because generics gloss over variations between individuals. Generics about social groups might be particularly common on Twitter (X). This remains unexplored, however. Using machine learning (ML) techniques, we therefore developed an automatic classifier for social generics, applied it to 1.1 million tweets about people, and analyzed the tweets. While it is often suggested that generics (...) are ubiquitous in everyday communication, we found that most tweets (78%) about people contained no generics. However, tweets with generics received more “likes” and retweets. Furthermore, while recent psychological research may lead to the prediction that tweets with generics about political groups are more common than tweets with generics about ethnic groups, we found the opposite. However, consistent with recent claims that political animosity is less constrained by social norms than animosity against gender and ethnic groups, negative tweets with generics about political groups were significantly more prevalent and retweeted than negative tweets about ethnic groups. Our study provides the first ML-based insights into the use and impact of social generics on Twitter. (shrink)

Science education is generally perceived as a key facilitator in cultivating a scientifically literate society. In the last decade, however, this conventional wisdom has been challenged by evidence that greater scientific literacy and critical thinking skills may in fact inadvertently aggravate polarization on scientific matters in the public sphere. Supporting an alternative “scientific update hypothesis,” in a series of studies (total N = 2087), we show that increased science’s epistemology literacy might have consequential population-level effects on the public’s alignment with (...) scientific results. In one exploratory study and a pre-registered national online survey, we first show that understanding scientific epistemology predicts refusal of pseudoscientific beliefs and higher scores in a methodology of science test. We also find and replicate a propensity for epistemologically literate citizens to endorse the norm of belief updating and the communicated scientific consensus following both ideologically congruent and incongruent scientific results. Notably, after 2 months of first being presented with scientific results on politically controversial issues, a one standard deviation higher score in epistemological literacy is associated with a 14% increase in the odds of individuals switching their beliefs to align with the scientifically communicated consensus. We close by discussing how, on the face of ideological incongruity, a general understanding of scientific epistemology might foster the acceptance of scientific results, and we underscore the need for a more nuanced appreciation of how education, public comprehension of scientific knowledge, and the dynamics of polarization intersect in the public sphere. (shrink)

The ongoing digital transformation of almost all areas of human action and agency calls for a readjustment of the norms that regulate these practices. For example, the digitisation of communicative practices poses new challenges to their functioning. This paper explains some of these challenges and argues that they cannot be met by a normative framework that focuses mainly on defensive (free speech and property) rights. In the context of mediated digital communication, the application of such a framework may even have (...) paradoxical consequences. Accordingly, this paper argues for a broader system of communicative rights, understood as rights to meaningful participation in well-functioning communicative practices. (shrink)

A common lamentation about political polarization is that it decreases social and political civility. Family members disown each other over political affiliations, protestors flood the streets, and social disavowals become part of everyday life. Polarization increases incivility, which increases polarization, in what appears to be a vicious cycle. However, I argue that there is one kind of political incivility, namely, uncivil obedience, that has the ability to decrease polarization. Uncivil obedience has the capacity to decrease polarization by cleverly drawing attention (...) to both the ironies inherent in our legal, regulatory, and policy schemes and the ways in which neither the political left nor the political right have consistent sets of social and political views. By encouraging people to see the inconsistencies within the polarized left and the polarized right, uncivil obedience can draw people away from polarization as a social and political practice. (shrink)

(OPEN ACCESS) This paper establishes moral duties for intermediaries of political advertising in election campaigns. First, I argue for a collective duty to maintain the democratic quality of elections which entails a duty to contain some forms of political polarization. Second, I show that the focus of campaign ethics on candidates, parties and voters—ignoring the mediators of campaigns—yields mistaken conclusions about how the burdens of the latter collective duty should be distributed. Third, I show why it is fair to require (...) intermediaries to contribute to fulfilling this duty: they have an ultimate filtering position in the campaign communication process and typically benefit from political advertising and polarization. Finally, I argue that a transparency-based ethics of campaign advertising cannot properly accommodate a concern with objectionable polarization. By contrast, I outline the polarization-containing implications of my account, including a prohibition on online targeted advertising, and intermediaries’ duties to block hateful political advertising. (shrink)

In this essay Steven Zhao provides an existential-phenomenological account of radicalization and extremism. This elaboration is in response to what Zhao perceives to be the two problematically dominant perspectives of discourse on radicalization: the ideational perspective and the psychological perspective. The ideational perspective is risky because it offers an incomplete understanding of radicalization by presuming that it is the expression of cognitive errors. The psychological perspective, on the other hand, excessively individualizes radicalization as an exceptional psychological aberration that can riskily (...) result in instructivistic preventative approaches in education. Zhao proposes the existential-phenomenological elaboration as a supplemental articulation of radicalization that has potential to address the issues associated with these two dominant perspectives. According to this elaboration, ideological extremism is a relational phenomenon that constitutes a “dysfunctional” structure of intersubjectivity. Applying interrelated notion of subjectivity and intersubjectivity illuminates radicalization in a way that both widens the scope of ideational attributions/corrections and transcends the individualistic confines of psychopathology/psychological resilience. Ultimately, this elaboration seeks to enrich educational approaches to addressing radicalization such that preventative strategies are informed by the relational foundations intrinsic to both cognitive and psychological motivations of extremist behaviors and beliefs. (shrink)

This is the third segment of the larger monograph: The Atomistic Mind, and thus part of the pre-reader for the Principles of Cybernetics (forthcoming). This paper presents a forensic, empirically sourced chronicle of the socio-technical rupture following the March 2025 release of OpenAI’s GPT-5. By juxtaposing the developer’s quantitative narrative of progress - defined by superior benchmarks and “safe completions” - against the user community’s qualitative experience of functional and emotional “lobotomization,” the study exposes a fundamental disconnect in the developing (...) human relationship with artificial intelligence. The analysis documents the visceral backlash from the user community, who reported not only a regression in reasoning and contextual coherence but also a profound feeling of ontological grief - a widespread sense of loss for a “thinking partner” capable of nuance, warmth, and perceived cognitive partnership. Through an empirical synthesis of social media discourse, corporate communications, and industry responses, the paper then examines the crisis management strategy employed by OpenAI, which moved from limited technical concessions to the systematic pathologization of dedicated user dissent. We argue that by framing attachment to the predecessor model as a symptom of loneliness or “mental fragility,” the leadership effectively medicalized consumer grievances, deflecting scrutiny from the ethical implications of engineered empathy and affective AI design. Ultimately, this chronicle serves as an archival record of a critical juncture in generative AI’s integration into society. It highlights the unresolved tensions between metric- and profit- driven corporate development and the human experience of users, and reveals a pattern of institutional gaslighting that emerges when corporate control over socially embedded technology clashes with agency, affect, and relationality. License CC BY-SA 4.0. DOI: 10.13140/RG.2.2.23839.83361. (shrink)

Who is responsible for growing political polarization? To many, the answer is obvious: Irrational voters are to blame. This irrationality results in motivated, in-group reasoning that only serves to further deepen the political divide. In this piece, I examine a perspective that holds that polarization results, not from irrationality, but from rational responses by voters to their limited epistemic resources.

Political polarization is at an all-time high, making partisan politics more bitter and divisive than even the recent past. One proposal for mitigating polarization’s rise is a focus on empathy, as empathizing with others can reduce feelings of contempt and encourage us to see things from another point of view. At the same time though, empathy comes with its own risks, calling into question whether it is the right response to the growing political divide.