Search

Peer reviewed: True

Most content consumed online is curated by proprietary algorithms deployed by social media platforms and search engines. In this article, we explore the interplay between these algorithms and human agency. Specifically, we consider the extent of entanglement or coupling between humans and algorithms along a continuum from implicit to explicit demand. We emphasize that the interactions people have with algorithms not only shape users’ experiences in that moment but because of the mutually shaping nature of such systems can also have longer-term effects through modifications of the underlying social-network structure. Understanding these mutually shaping systems is challenging given that researchers presently lack access to relevant platform data. We argue that increased transparency, more data sharing, and greater protections for external researchers examining the algorithms are required to help researchers better understand the entanglement between humans and algorithms. This better understanding is essential to support the development of algorithms with greater benefits and fewer risks to the public.

Numerous indicators suggest that democracy is under threat, including in Europe. In 2020, The Economist’s democracy index determined that one EU member state, Hungary, was no longer a democracy. Throughout Europe, populist movements –mainly, but not exclusively, on the political right– have pitted “the people” against a presumed “elite” that is variously constructed as including mainstream media, politicians, experts, scientists, and academics. The COVID-19 pandemic has put further pressure on societies by requiring restrictions on social behaviours to control the pandemic that are unprecedented in democracies and that may facilitate autocratization. Although symptoms and causes of these trends are intertwined and difficultto tease apart, there is little doubt that wilful disregard of evidence and expertise, accompanied by a flood of misinformation –on social media, in hyperpartisan news sites, and in political discourse– are at the heart of the challenge to democracies. Misinformation matters: Exposure has been shown to make a causal contribution to populist voting in Italy, to triggering ethnic hate crimes in Germany and Russia, and it has been shown to set political agendas. Misinformation is particularly problematic because it has longer-term consequences: false information lingers in memory even if people acknowledge, believe, and try to adhere to a correction. Lingering misinformation, in turn, can be politically consequential, for example when corrections of politicians’ falsehoods do not affect people’s feeling about the politician or their voting intention. Misinformation, however, does not exist in a vacuum: misinformation is disseminated(sometimes intentionally, in which case it is best referred to as disinformation) and it is consumed and shared by the public. To understand the effects of misinformation on democracy thus requires an understanding of the processes of dissemination and consumption. In this chapter, I focus on twoimportant drivers of misinformation spread and how they interact with human cognition: First, I examine the role of demagogues; that is, political leaders who rely on false claims and promises, and emotive exploitation of people’s prejudices, in order to gain power. How do they exercise power, and why do peopleaccept demagoguery? Second, I examine the role of social media, focusing in particular on the role of algorithms. How does social media capture human attention? Why do people participate in sharing of misinformation?

An increasing number of real-world interventions aim to preemptively protect or “inoculate” people against misinformation. Inoculation research has demonstrated positive effects on misinformation resilience when measured immediately after treatment via messages, games, or videos. However, very little is currently known about their long-term effectiveness and the mechanisms by which such treatment effects decay over time. We report five pre-registered longitudinal experiments (Ntotal = 11,759) that investigate the effectiveness of psychological inoculation interventions over time. We find that text-based and video-based inoculation interventions can remain effective for one month (whereas game-based interventions appear to decay more rapidly), and that memory-enhancing “booster interventions” can boost the diminishing effects of counter-misinformation interventions to the original effect size. We conclude that misinformation researchers would benefit from integrating knowledge from the cognitive science of memory to design a new generation of psychological interventions that can counter misinformation durably over time and at-scale.

This study is designed to test the efficacy of nine different interventions (and a control condition).

In recent years, the UK has become divided along two key dimensions: party affiliation and Brexit position. We explored how division along these two dimensions interacts with the correction of political misinformation. Participants saw accurate and inaccurate statements (either balanced or mostly inaccurate) from two politicians from opposing parties but the same Brexit position (Experiment 1), or the same party but opposing Brexit positions (Experiment 2). Replicating previous work, fact-checking statements led participants to update their beliefs, increasing belief after fact affirmations and decreasing belief for corrected misinformation, even for politically aligned material. After receiving fact-checks participants had reduced voting intentions and more negative feelings towards party-aligned politicians (likely due to low baseline support for opposing party politicians). For Brexit alignment, the opposite was found: participants reduced their voting intentions and feelings for opposing (but not aligned) politicians following the fact-checks. These changes occurred regardless of the proportion of inaccurate statements, potentially indicating participants expect politicians to be accurate more than half the time. Finally, although we found division based on both party and Brexit alignment, effects were much stronger for party alignment, highlighting that even though new divisions have emerged in UK politics, the old divides remain dominant.

Supplemental material, sj-docx-1-scx-10.1177_10755470231162634 for Science Communication as a Collective Intelligence Endeavor: A Manifesto and Examples for Implementation by Dawn Holford, Angelo Fasce, Katy Tapper, Miso Demko, Stephan Lewandowsky, Ulrike Hahn, Christoph M. Abels, Ahmed Al-Rawi, Sameer Alladin, T. Sonia Boender, Hendrik Bruns, Helen Fischer, Christian Gilde, Paul H. P. Hanel, Stefan M. Herzog, Astrid Kause, Sune Lehmann, Matthew S. Nurse, Caroline Orr, Niccolò Pescetelli, Maria Petrescu, Sunita Sah, Philipp Schmid, Miroslav Sirota and Marlene Wulf in Science Communication

This pre-registration is part of a larger international study of HCPs that includes other variables and questions about vaccines. Methods for the data collection in each of the five countries involved (UK, Germany, France, Portugal, Finland) are pre-registered in separate documentation prior to data collection. This pre-registration deals specifically with planned analyses for one DV in the data (HCP's perception of the difficulty to rebut anti-vaccination arguments.)

Binational study (Portugal and the UK) to assess the relationship between family cohesion, vaccine-related behaviours, and the Pro-VC-Be.

This pre-registration specifies hypotheses and method for the study prior to data collection.

Supplemental material, sj-docx-1-scx-10.1177_10755470231162634 for Science Communication as a Collective Intelligence Endeavor: A Manifesto and Examples for Implementation by Dawn Holford, Angelo Fasce, Katy Tapper, Miso Demko, Stephan Lewandowsky, Ulrike Hahn, Christoph M. Abels, Ahmed Al-Rawi, Sameer Alladin, T. Sonia Boender, Hendrik Bruns, Helen Fischer, Christian Gilde, Paul H. P. Hanel, Stefan M. Herzog, Astrid Kause, Sune Lehmann, Matthew S. Nurse, Caroline Orr, Niccolò Pescetelli, Maria Petrescu, Sunita Sah, Philipp Schmid, Miroslav Sirota and Marlene Wulf in Science Communication

This is a pre-registration of the hypotheses and methods prior to data collection.

Preregistration of analyses related to HCPs’ attitudes to Covid-19 mandates (JITSUVAX WP1)

Experimental materials, data and analysis scripts underlying the paper "TITLE".

This is a pre-registration of the hypotheses and method prior to data collection.

This study builds on previous work that finds that the sharing of partisan political content and misinformation on social media is often driven by the perceived 'political usefulness' of the content, and affective polarization. The current study draws on inoculation theory to develop an 'inoculation' video that aims to demotivate the sharing of highly partisan political content online by informing viewers of the consequences of growing polarization, how they contribute to growing polarization with the sharing of this content, the techniques used to manipulate them (i.e. scapegoating, ad-hominem attacks and emotional language), and reminds them that shared humanity transcends political lines. The study will test the effectiveness of this inoculation in reducing the sharing of synthetic highly partisan derogatory political content centered around the highly polarized issue of Brexit.

Pre-registration for JITSUVAX WP2.1 Angelo Fasce, Dawn Holford, Stephan Lewandowsky, Philipp Schmid (stephan.lewandowsky@bristol.ac.uk) Introduction Rationale Describe the rationale for the review in the context of existing knowledge. Many studies have catalogued anti-vaccine arguments and misinformation (e.g., Hoffman et al., 2019; Hughes et al. 2021), grouping them into thematic categories. However, few studies seem to differentiate between the "tropes"—broad categories that are sometimes also referred to as "themes", or "narratives"—of a message and its specific method of persuasion. Moreover, there are few discussions of attitudinal roots—underlying fears or ideologies that sustain and motivate specific attitudes (Hornsey & Fielding, 2017)—that may underlie the catalogued anti-vaccine arguments and misinformation. Given that some studies suggest that understanding the attitudinal roots of arguments, and differentiating the narrative content and the method of persuasion can help mitigate the potential damage of anti-vaccination arguments (Hornsey & Fielding, 2017, Schmid & Betsch, 2019), there is a need for a more sophisticated taxonomy of anti-vaccination arguments. Thus, the primary aim of this review is to systematically identify existing tropes and rhetorical strategies present in studies of anti-vaccination content in the literature to date. Based on this, we aim to develop a unifying taxonomy of anti-vaccination content that links tropes and rhetorical strategies to the psychological roots of anti-vaccination attitudes. Objectives Provide an explicit statement of the objective(s) or question(s) the review addresses. The primary aim of this review is to systematically identify existing tropes and rhetorical strategies present in existing studies of anti-vaccination content in the literature to date. Based on this, we aim to develop a taxonomy of anti-vaccination content that links tropes and rhetorical strategies to the psychological roots of anti-vaccination attitudes. Methods Eligibility criteria Specify the inclusion and exclusion criteria for the review and how studies were grouped for the syntheses. Inclusion criteria • Published articles in peer-reviewed journals or on scientific preprint servers (e.g., medrXiv) • In English, although future searches may be conducted in German, Portuguese, Finnish, Swedish, Spanish, and French. • Reports analyses of anti-vaccine statements in public circulation (e.g., on a website, social media, blog etc.). We define "analyses" broadly, including descriptive analyses, conceptual and philosophical analysis, text modelling, etc. Exclusion criteria • Studies focusing on vaccination of non-human subjects Information sources Specify all databases, registers, websites, organisations, reference lists and other sources searched or consulted to identify studies. Specify the date when each source was last searched or consulted. We will be searching the following major scientific search engines: Scopus, Web of Science, PubMed, and PsycInfo. These sources were selected for their coverage of the relevant literature (e.g., psychology, health studies and argumentation) and their ability to satisfy relevance, transparency, and reproducibility criteria in systematic search (Gusenbauer & Haddaway, 2020). Because much of the research on COVID-19 will still be emerging, we will also search three preprint servers relevant to the field: • PsyArXiv • ArXiv • MedrXiv We will only include articles published in these sources prior to 1 July 2021. We will be recording the date on which the search took place for each database. Search strategy Present the full search strategies for all databases, registers and websites, including any filters and limits used. We will use the following search logic to encompass the different variations in terminology used in the literature: ("vacc* refus*" OR "anti-vacc*" OR "anti-vax*" OR "vacc* oppon*" OR "vacc* oppos*" OR "vacc* skeptic*" OR "vacc* sceptic*") AND (trope* OR meme* OR narrative* OR argumen* OR theme* OR rhetoric) Search keys will be adapted to the capabilities of each search engine as needed. Specifically, the Boolean logic represented by our search string may exceed the capability of search engines for, e.g., preprint servers, in which case we will seek to approximate the logic via partial repeated searches. Selection process Specify the methods used to decide whether a study met the inclusion criteria of the review, including how many reviewers screened each record and each report retrieved, whether they worked independently, and if applicable, details of automation tools used in the process. Articles will be collated in an Excel spreadsheet and titles inspected to remove duplicates. Two reviewers will independently review the title and abstracts of each article and categorise them as "relevant", "not relevant", or "uncertain". Where there are discrepancies between reviewers or one reviewers has indicated "uncertain", the reviewers will meet to agree on the categorisation. Two reviewers will then review the full text of articles that were determined to be relevant on the basis of title and abstract and categorise them as "accept", "reject", or "uncertain". If there are discrepancies between reviewers, they will discuss and reach a consensus. If the two reviewers still disagree after discussion, a third reviewer will review the article to determine if it should be accepted. The criteria for acceptance will be based on the eligibility criteria and the reasons for rejection will be recorded. Data collection process Specify the methods used to collect data from reports, including how many reviewers collected data from each report, whether they worked independently, any processes for obtaining or confirming data from study investigators, and if applicable, details of automation tools used in the process. Where possible, we will use database tools that allow automated extraction of article metadata (title, authors, year, doi/URL, journal/preprint server). However, this may not be possible for articles found on preprint servers that do not offer such tools (e.g., PsyArXiv). In this case, reviewers will manually complete this extraction. For each article, a single reviewer will independently complete extraction for all the content items mentioned in "data items". Data items List and define all outcomes for which data were sought. Specify whether all results that were compatible with each outcome domain in each study were sought (e.g. for all measures, time points, analyses), and if not, the methods used to decide which results to collect. List and define all other variables for which data were sought (e.g. participant and intervention characteristics, funding sources). Describe any assumptions made about any missing or unclear information. For each article accepted for inclusion in the review, we will record the following metadata: • Title • Authors • Year • doi/URL • Journal/preprint server We will also extract the following content from each article: • Content platform o This category refers to the platform, or channel, that was studied. Multiple platforms may be reported in the same article. Common platforms include: websites, Twitter, Facebook, YouTube. • Content type (as described by article authors) o This category refers to the individual articles' authors' terminology of the tropes or strategies reported. Common terms used are "tropes", "themes", "narratives", "tactics", "strategies". • List of tropes and/or strategies identified in the article o All tropes and strategies mentioned in the article will be extracted. Study risk of bias assessment Specify the methods used to assess risk of bias in the included studies, including details of the tool(s) used, how many reviewers assessed each study and whether they worked independently, and if applicable, details of automation tools used in the process. We do not assess the risk of bias or certainty because the objective is to collect all existing tropes and rhetorical strategies identified in the literature, thus there are no interventions or effect strength analyses. To counter publication bias, our search includes three well-known preprint servers, which have grown in popularity since the onset of the pandemic and provide a means for publication ahead of peer review. Based on a pre-screening of articles on anti-vaccine sentiment, we also expect to reach saturation point soon because most articles in the domain already contain substantial overlap in the tropes and rhetorical strategies they report. Competing interests This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 964728 (JITSUVAX). References Gusenbauer, M., & Haddaway, N. R. (2020). Which academic search systems are suitable for systematic reviews or meta-analyses? Evaluating retrieval qualities of Google Scholar, PubMed, and 26 other resources. Research Synthesis Methods, 11(2), 181–217. https://doi.org/10.1002/jrsm.1378 Hoffman, B. L., Felter, E. M., Chu, K.-H., Shensa, A., Hermann, C., Wolynn, T., Williams, D., & Primack, B. A. (2019). It’s not all about autism: The emerging landscape of anti-vaccination sentiment on Facebook. Vaccine, 37(16), 2216–2223. https://doi.org/https://doi.org/10.1016/j.vaccine.2019.03.003 Hornsey, M. J., & Fielding, K. S. (2017). Attitude roots and Jiu Jitsu persuasion: Understanding and overcoming the motivated rejection of science. In American Psychologist (Vol. 72, Issue 5, pp. 459–473). American Psychological Association. https://doi.org/10.1037/a0040437 Hughes, B., Miller-Idriss, C., Piltch-Loeb, R., White, K., Creizis, M., Cain, C., & Savoia, E. (2021). Development of a Codebook of Online Anti-Vaccination Rhetoric to Manage COVID-19 Vaccine Misinformation. MedRxiv, 2021.03.23.21253727. https://doi.org/10.1101/2021.03.23.21253727 Schmid, P., & Betsch, C. (2019). Effective strategies for rebutting science denialism in public discussions. Nature Human Behaviour, 3(9), 931–939. https://doi.org/10.1038/s41562-019-0632-4

This is the preregistration of a secondary data analysis exploring citizens' metacognitive accuracy in detecting political misinformation. It is the first step of a two-step preregistration procedure where the second step will be the preregistration of the actual analysis code.

Preregistration of methodology for document preparation Stephan Lewandowsky, Final version, 25 June 2020 stephan.lewandowsky@bristol.ac.uk Debunking2020 aims to be a consensually-developed document authored by leading researchers in the field of the behavior and cognition of misinformation and its debunking. It seeks to build on the Debunking Handbook, authored by John Cook and Stephan Lewandowsky in 2011 (http://sks.to/debunk), which is no longer current. This document is intended to preregister the process for creating Debunking2020.

This is an updated registration of the project that can be found at https://osf.io/te8hq/ - Our earlier registration contained the wrong p value for the cluster-significance threshold (page 11) Our aim is to gain a better understanding of the cognitive mechanisms underlying continued reliance on misinformation. To this end we will present subjects with brief fictional news reports that may or may not require their memory to be updated. Reliance on misinformation will be assessed through a decision task involving the clarification of a picture as a 'match' or 'mismatch' for the preceding report.

An experiment investigating the effects of different framings of COVID-19 on participants' climate change concern and support for climate mitigation.