Rumor clustering, consensus, and polarization: Dynamic social impact and self-organization of hearsay

Knowledge through social networks: Accuracy, error, and polarisation

This paper examines the fundamental problem of testimony. Much of what we believe to know we know in good part, or even entirely, through the testimony of others. The problem with testimony is that we often have very little on which to base estimates of the accuracy of our sources. Simulations with otherwise optimal agents examine the impact of this for the accuracy of our beliefs about the world. It is demonstrated both where social networks of information dissemination help and where they hinder. Most importantly, it is shown that both social networks and a common strategy for gauging the accuracy of our sources give rise to polarisation even for entirely accuracy motivated agents. Crucially these two factors interact, amplifying one another's negative consequences, and this side effect of communication in a social network increases with network size. This suggests a new causal mechanism by which social media may have fostered the increase in polarisation currently observed in many parts of the world.

From pandemic to Plandemic: Examining the amplification and attenuation of COVID-19 misinformation on social media

This study examines the proliferation of COVID-19 misinformation through Plandemic-a pseudo-documentary of COVID-19 conspiracy theories-on social media and examines how factors such as (a) themes of misinformation, (b) types of misinformation, (c) sources of misinformation, (d) emotions of misinformation, and (e) fact-checking labels amplify or attenuate online misinformation during the early days of the pandemic. Using CrowdTangle, a Facebook API, we collected a total of 5732 publicly available Facebook pages posts containing Plandemic-related keywords from January 1 to December 19, 2020. A random sample of 600 posts was subsequently coded, and the data were analyzed using negative binomial regression to examine factors associated with amplification and attenuation. Overall, the extended an extended Social Amplification of Risk Framework (SARF) provided a theoretical lens to understand why certain misinformation was amplified, while others were attenuated. As for posts with misinformation, results showed that themes related to private firms, treatment and prevention of virus transmission, diagnosis and health impacts, virus origins, and social impact were more likely to be amplified. While the different types of misinformation (manipulated, fabricated, or satire) and emotions were not associated with amplification, the type of fact-check labels did influence the virality of misinformation. Specifically, posts that were flagged as false by Facebook were more likely to be amplified, while the virality of posts flagged as containing partially false information was attenuated. Theoretical and practical implications were discussed.

Adaptive cost-sensitive stance classification model for rumor detection in social networks

As online social networks are experiencing extreme popularity growth, determining the veracity of online statements denoted by rumors automatically as earliest as possible is essential to prevent the harmful effects of propagating misinformation. Early detection of rumors is facilitated by considering the wisdom of the crowd through analyzing different attitudes expressed toward a rumor (i.e., users’ stances). Stance detection is an imbalanced problem as the querying and denying stances against a given rumor are significantly less than supportive and commenting stances. However, the success of stance-based rumor detection significantly depends on the efficient detection of “query” and “deny” classes. The imbalance problem has led the previous stance classifier models to bias toward the majority classes and ignore the minority ones. Consequently, the stance and subsequently rumor classifiers have been faced with the problem of low performance. This paper proposes a novel adaptive cost-sensitive loss function for learning imbalanced stance data using deep neural networks, which improves the performance of stance classifiers in rare classes. The proposed loss function is a cost-sensitive form of cross-entropy loss. In contrast to most of the existing cost-sensitive deep neural network models, the utilized cost matrix is not manually set but adaptively tuned during the learning process. Hence, the contributions of the proposed method are both in the formulation of the loss function and the algorithm for calculating adaptive costs. The experimental results of applying the proposed algorithm to stance classification of real Twitter and Reddit data demonstrate its capability in detecting rare classes while improving the overall performance. The proposed method improves the mean F-score of rare classes by about 13% in RumorEval 2017 dataset and about 20% in RumorEval 2019 dataset.

Validation of the COVID-19 Transmission Misinformation Scale and Conditional Indirect Negative Effects on Wearing a Mask in Public

The SARS-CoV-2 (COVID-19) pandemic devastated the world economy. Global infections and deaths altered the behaviors of generations. The Internet acted as an incredible vehicle for communication but was also a source of unfounded rumors. Unfortunately, this freedom of information sharing and fear of COVID-19 fostered unfounded claims about transmission (e.g., 5G networks spread the disease). With negligible enforcement to stop the spread of rumors and government officials spouting unfounded claims, falsities became ubiquitous. Organizations, public health officials, researchers, and businesses spent limited resources addressing rumors instead of implementing policies to overcome challenges (e.g., speaking to defiant mask wearers versus safe reopening actions). The researchers defined COVID-19 transmission misinformation as false beliefs about the spread and prevention of contracting the disease. Design and validation of the 12-item COVID-19 Transmission Misinformation Scale (CTMS) provides a measure to identify transmission misinformation believers. Indirect COVID-19 transmission misinformation beliefs with a fear of COVID-19 decreased wearing a mask in public intentions. Callousness exacerbated COVID-19 transmission misinformation beliefs as a moderator.

The Psychology of Cultural Dynamics: What Is It, What Do We Know, and What Is Yet to Be Known?

The psychology of cultural dynamics is the psychological investigation of the formation, maintenance, and transformation of culture over time. This article maps out the terrain, reviews the existing literature, and points out potential future directions of this research. It is divided into three parts. The first part focuses on micro-cultural dynamics, which refers to the social and psychological processes that contribute to the dissemination and retention of cultural information. The second part, on micro-macro dynamics, investigates how micro-level processes give rise to macro-cultural dynamics. The third part focuses on macro-cultural dynamics, referring to the distribution and long-term trends involving cultural information in a population, which in turn enable and constrain the micro-level processes. We conclude the review with a consideration of future directions, suggesting behavior change research as translational research on cultural dynamics.

Science audiences, misinformation, and fake news

Concerns about public misinformation in the United States-ranging from politics to science-are growing. Here, we provide an overview of how and why citizens become (and sometimes remain) misinformed about science. Our discussion focuses specifically on misinformation among individual citizens. However, it is impossible to understand individual information processing and acceptance without taking into account social networks, information ecologies, and other macro-level variables that provide important social context. Specifically, we show how being misinformed is a function of a person's ability and motivation to spot falsehoods, but also of other group-level and societal factors that increase the chances of citizens to be exposed to correct(ive) information. We conclude by discussing a number of research areas-some of which echo themes of the 2017 National Academies of Sciences, Engineering, and Medicine's Communicating Science Effectively report-that will be particularly important for our future understanding of misinformation, specifically a systems approach to the problem of misinformation, the need for more systematic analyses of science communication in new media environments, and a (re)focusing on traditionally underserved audiences.

How shared reality is created in interpersonal communication

Communication is a key arena and means for shared-reality creation. Most studies explicitly devoted to shared reality have focused on the opening part of a conversation, that is, a speaker's initial message to an audience. The aspect of communication examined by this research is the evaluative adaptation (tuning) of the messages to the audience's attitude or judgment. The speaker's shared-reality creation is typically assessed by the extent to which the speaker's evaluative representation of the topic matches the audience-tuned view expressed in the message. We first review research on such audience-tuning effects, with a focus on shared-reality goals and conditions facilitating the generalization of shared reality. We then review studies using other paradigms that illustrate factors of shared-reality creation in communication, including mere message production, grounding, validation responses, and communication about commonly known information (including stereotypes) in intragroup communication. The different lines of research reveal the potency, but also boundary conditions, of communication effects on shared reality.

Porous Chambers, Echoes of Valence and Stereotypes

Selective exposure to one-sided news coverage, especially of controversial geopolitical events, may contribute to growing social polarization. Existing research on “echo chambers”—fragmented information environments that amplify homogeneous perspectives—focuses on the degree to which individuals and social media platforms shape informational segregation. Here, we explore whether news organizations directly contribute to echo chambers through the hyperlinks they embed in online articles. Using network and text analysis, we examined coverage of the 2014 shooting of Michael Brown in Ferguson, MO, and found that online news media exhibited weak community structure and high connectivity across news outlets. However, analyses also indicated that media sources were more likely to link to coverage that was similar to their own in terms of emotional valence and stereotype-relevant aspects of the events. While hyperlinking to diverse news sources may ameliorate fragmented information environments, selectively linking to similar coverage may contribute to growing polarization.