Base Instinct
Latent Polarization
Klaus Desmet, Ignacio Ortuño-Ortín & Romain Wacziarg
NBER Working Paper, September 2025
Abstract:
We develop a new method to endogenously partition society into groups based on homophily in values, using fractional hedonic games as a theoretical foundation. The between-group differentiation that results from this partition provides a novel measure of latent polarization in society. We implement this method empirically using U.S. data from the World Values Survey. For the last forty years, the degree of latent polarization of the U.S. public has been high and relatively stable. In contrast, the degree of values polarization between voters of the two main political parties has steadily increased since the 1990s, and is now converging toward that of underlying values-based clusters. Thus, growing partisan polarization in the U.S. is a reflection of partisan views becoming increasingly aligned with the main values-based clusters in society.
Does Inflammatory Rhetoric Boost Support for Political Violence? Considering the Role of Geographic Context
Bryan Gervais & Irwin Morris
Political Research Quarterly, forthcoming
Abstract:
There has been significant speculation about the role that inflammatory elite rhetoric has played in sparking support for political violence in the United States. However, the extent to which uncivil and intolerant rhetoric contributes to support for political violence remains unclear. Similarly, the impact of racial/ethnic context on attitudes towards political violence is unclear. We report on the results from three experiments that included various measures of uncivil/intolerant rhetoric, geographic context, and political violence. Across our diverse set of experiments, we find that local racial heterogeneity is strongly associated with increased support for political violence. However, we find little evidence that inflammatory rhetoric, directly or indirectly, bolsters support for partisan violence. While our results cast doubt on the claim that inflammatory rhetoric from elites is responsible for bolstering broad support for partisan violence among Americans, they do suggest that increased heterogeneity may bolster support for political violence regardless of rhetoric elites adopt.
The effect of real-news party cues
Rasmus Skytte
American Journal of Political Science, forthcoming
Abstract:
News media routinely offer cues about the stances of party elites, but to what extent do these cues shape the policy opinions of the public? While numerous experiments find that partisans adopt the stances of their leaders, these findings may not generalize easily to the context of real news, which often contains richer policy information and more subtle, ambiguous party cues than the artificial stimuli commonly employed. To investigate the impact of party cues as they naturally appear in news coverage -- what I term real-news party cues -- I develop a new experimental paradigm. I first sample 70 articles covering political proposals from a large pool of real news and prepare two versions of each: one with and one without these cues. These articles then serve as stimuli in survey experiments among representative samples of Americans (n = 12,177). I find that real-news party cues have mostly modest effects on policy opinions.
Community notes reduce engagement with and diffusion of false information online
Isaac Slaughter et al.
Proceedings of the National Academy of Sciences, 23 September 2025
Abstract:
Social networks scaffold the diffusion of information on social media. Much attention has been given to the spread of true vs. false content on online social platforms, including the structural differences between their diffusion patterns. However, much less is known about how platform interventions on false content alter the engagement with and diffusion of such content. In this work, we estimate the causal effects of Community Notes, a novel fact-checking feature adopted by X (formerly Twitter) to solicit and vet crowd-sourced fact-checking notes for false content. We gather detailed time series data for 40,078 posts for which notes have been proposed and use synthetic control methods to estimate a range of counterfactual outcomes. We find that attaching fact-checking notes significantly reduces the engagement with and diffusion of false content. We estimate that, on average, the notes resulted in reductions of 46.1% in reposts, 44.1% in likes, 21.9% in replies, and 13.5% in views after being attached. Over the posts’ entire lifespans, these reductions amount to 11.6% fewer reposts, 13.3% fewer likes, 6.9% fewer replies, and 5.5% fewer views on average. In reducing reposts, we observe that diffusion cascades for fact-checked content are less deep and less “viral,” but not less broad, than synthetic control estimates for non-fact-checked content with similar reach. This structural difference contrasts notably with differences between false vs. true content diffusion itself, where false information diffuses farther, but with structural patterns that are otherwise indistinguishable from those of true information, conditional on reach.
Subdued but Unbroken: The Cohesion of Far-Right Extremist Followers after Deplatforming
Wei Zhong & Maggie Mengqing Zhang
Perspectives on Politics, forthcoming
Abstract:
While deplatforming has become an increasingly common strategy to combat online harm and far-right extremism, its effects on the followers of extremist groups — who are key supporters and play a crucial role in spreading and sustaining these ideologies — remain underexplored. On August 10, 2018, Twitter (now X) deplatformed one such far-right extremist group, the Proud Boys, along with their affiliated accounts. Leveraging this intervention, our research addresses a key knowledge gap by examining the impact of deplatforming on the cohesion of extremist group followers. Specifically, we investigate whether deplatforming leads to fragmentation or reinforces unity among the group’s followers. We assess cohesion through three theoretical lenses: task commitment, social commitment, and sense of belonging. By analyzing over 12 million tweets from approximately nine thousand Proud Boys supporters between August 1, 2017, and September 1, 2019, we find that deplatforming had a limited effect on reducing group cohesion. Instead, it may have prompted followers to seek broader networks and external interactions, leaving overall cohesion largely intact. This study offers important insights into the resilience of online extremist communities and the limitations of deplatforming as a strategy to disrupt them. Understanding these dynamics is essential for developing more effective approaches to counter online extremism and promote safer digital spaces.
Politically extreme individuals exhibit similar neural processing despite ideological differences
Daantje de Bruin & Oriel FeldmanHall
Journal of Personality and Social Psychology, forthcoming
Abstract:
The current state of political polarization in the United States encompasses a growing divide between partisans and a shift toward more extreme ideologies. Although rising ideological extremism poses societal challenges, the mechanisms supporting extreme views remain uncharacterized. Leveraging a combination of neurophysiological methods, we show that regardless of which side of the political aisle an individual is on, those with more extreme views show heightened neural activity to politically charged content in brain regions implicated in affective processing -- including the amygdala, periaqueductal gray, and posterior superior temporal sulcus. Moreover, we observe that those who share an extreme perspective -- even when they do not share an ideology -- exhibit increased neural synchronization in the broader posterior superior temporal sulcus region while consuming political content. For those on the most extreme ends of the ideological spectrum, this effect is further influenced by listening to extreme language. Finally, we find that shared arousal, measured through galvanic skin conductance responses, modulates the strength of coupling between shared extremity and neural synchrony. Together, our findings suggest a role for affect in shaping ideological extremity, which helps explain why those at the far ends of the political spectrum come to view the world through a shared, extreme lens.
How Mainstream Politicians Erode Norms
Vicente Valentim, Elias Dinas & Daniel Ziblatt
British Journal of Political Science, August 2025
Abstract:
How does anti-immigrant rhetoric by mainstream politicians affect norms of tolerance? How does this compare to similar statements made by radical-right politicians? Drawing on experimental evidence, we find that statements by mainstream politicians lead to more norm erosion than similar statements by radical-right politicians. Subsample analyses suggest that this is because statements by mainstream-right politicians erode norm perceptions of right-wing individuals, while those by radical-right politicians induce backlash among left-wing individuals, who hold closer to the norm in place. The latter effect (backlash by the left) disappears when similar statements are made by mainstream right politicians. We argue that this difference occurs because mainstream politicians represent the views of a larger part of the population or have a higher status. Our results highlight the pivotal role of mainstream politicians in enforcing or eroding democratic norms, and that similar political statements can have different effects depending on their sender.
A Curse or a Blessing? The Associations of Political Knowledge With Partisan Hostility and Condonation of Political Violence
Gabriel Miao Li
International Journal of Public Opinion Research, Autumn 2025
Abstract:
This study examines the complex associations of political knowledge with affective polarization and attitudes toward political violence in contemporary U.S. politics. Utilizing data from the 2024 American National Election Studies pre-election survey, the analysis highlights political knowledge's dual roles -- as both a potential catalyst for partisan animosity and a critical resource promoting democratic stability. Consistent with recent research, more politically knowledgeable individuals tend to express intensified negative feelings toward opposing political parties, and this tendency is particularly pronounced among individuals who attach less importance to their partisan identity. In the meantime, increased political knowledge is also significantly associated with reduced condonation of political violence, particularly among those who perceive their partisan identity as highly central to their self-concept. These findings highlight the dual-edged, context-dependent nature of political knowledge, underscoring the need for a nuanced understanding of its implications in contemporary democratic societies.
Biased LLMs Can Influence Political Decision-Making
Jillian Fisher et al.
Proceedings of the Association for Computational Linguistics, 2025, Pages 6559-6607
Abstract:
As modern large language models (LLMs) become integral to everyday tasks, concerns about their inherent biases and their potential impact on human decision-making have emerged. While bias in models are well-documented, less is known about how these biases influence human decisions. This paper presents two interactive experiments investigating the effects of partisan bias in LLMs on political opinions and decision-making. Participants interacted freely with either a biased liberal, biased conservative, or unbiased control model while completing these tasks. We found that participants exposed to partisan biased models were significantly more likely to adopt opinions and make decisions which matched the LLM’s bias. Even more surprising, this influence was seen when the model bias and personal political partisanship of the participant were opposite. However, we also discovered that prior knowledge of AI was weakly correlated with a reduction of the impact of the bias, highlighting the possible importance of AI education for robust mitigation of bias effects. Our findings not only highlight the critical effects of interacting with biased LLMs and its ability to impact public discourse and political conduct, but also highlights potential techniques for mitigating these risks in the future.
Ingroup solidarity drives social media engagement after political crises
Malia Marks et al.
Proceedings of the National Academy of Sciences, 2 September 2025
Abstract:
Social media are often said to exacerbate polarization by platforming hostility between groups. However, positive social emotions like ingroup solidarity may also drive social media engagement, particularly after major threats such as military invasions or terror attacks. In this preregistered study, we examine the socioemotional drivers of engagement following group threats in the context of the US 2024 presidential campaign trail, where both major political parties faced crises in July of 2024. We test how ingroup solidarity and outgroup hostility predicted social media users’ engagement with 62,118 posts by 484 US partisan accounts before and after the first Trump assassination attempt (July 13) and Biden’s re-election campaign suspension (July 21). We find that, while outgroup hostility is typically the dominant predictor of engagement, interactions with ingroup solidarity surged among Republicans after the Trump shooting and among Democrats after Biden’s withdrawal. We show that negativity toward other groups is not always key to going viral. Rather, positive ingroup emotions appear to play a leading role in times of crisis.
Friday Night Politics: Football Bans and Intraparty Political Gender Dynamics
Lindsey Kilpatrick, John Cluverius & Joshua Dyck
Social Science Quarterly, September 2025
Objective: To determine the effects of elite partisan cues on support or opposition to a hypothetical ban on tackle football for youth players.
Methods: Nationally representative survey experiment of 1800 American Adults, conducted by YouGov in 2019, utilizing four potential treatments: control, partisan Democratic cue, partisan Republican cue, and a bipartisan cue.
Results: Partisans respond to elite signaling about the issue. We find that Republican women, in particular, are most responsive to both partisan and bipartisan messages about football bans from political elites.