When Science Alone Fails to Convince: A Brief Intervention Fostering an Evidence-Oriented Mindset Strengthens Science-Based Corrections

When Science Alone Fails to Convince: A Brief Intervention Fostering an Evidence-Oriented Mindset Strengthens Science-Based Corrections | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article When Science Alone Fails to Convince: A Brief Intervention Fostering an Evidence-Oriented Mindset Strengthens Science-Based Corrections Anat Lande-Brenner, Ullrich K. H. Ecker, Ruth Mayo This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8204469/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 11 You are reading this latest preprint version Abstract This research demonstrates that citing scientific sources to correct disinformation can be ineffective or even counterproductive, and offers a potential remedy. Three experiments simulated a WhatsApp conversation about COVID-19 vaccine disinformation, examining corrections in the high-uncertainty context of the early 2021 stage of the pandemic (Studies 1 and 2) and among vaccine skeptics in 2024 (Study 3). A science-based correction backfired during the pandemic and was ineffective among skeptics in 2024. However, a brief, content-agnostic intervention (ThinkFRE), comprising three simple statements prompting attention to facts and experts, consistently reversed these negative outcomes. The intervention specifically enhanced science-based corrections, not corrections with no source, and it worked even for skeptics who typically distrust science, suggesting it acts by restoring the importance of evidence and expertise. These findings indicate that for science to be an effective antidote to disinformation, individuals may first need a cognitive nudge toward an evidence-oriented mindset. Biological sciences/Psychology Social science/Psychology Scientific community and society/Scientific community Disinformation Correction Prebunking COVID-19 Trust in Science Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction We are living in a "post-truth" era, characterized by growing challenges to factual accuracy, widespread mistrust of experts, and a blurring of the line between objective knowledge and personal belief 1–3 . This erosion of epistemic norms fuels the spread of disinformation 4 . A salient example is the COVID-19 pandemic, which unleashed an “infodemic” - a rapid spread of misinformation that shaped public perceptions and behaviors, often with harmful effects 5,6 . For instance, exposure to media outlets that downplayed the risks of COVID-19 or spread misleading information has been linked to lower compliance with public health guidelines and greater resistance to preventive measures 7–9 . As a result, some individuals have rejected expert advice, ignored public health guidelines, and refused vaccination, thereby endangering public health 10 . In response, scholars have emphasized the critical role of the social and behavioral sciences in combating the spread of the virus, particularly in understanding and countering misinformation 11 . Moreover, public trust in science and scientists has emerged as a key factor in navigating global crises such as the COVID-19 pandemic, as it strongly predicts adherence to preventive measures 12,13 . Sustaining this trust is vital not only for managing current threats but also for addressing future global challenges, as illustrated by the recent resurgence of measles, a disease that can be effectively prevented through vaccination but has reemerged due to misinformation and declining confidence in vaccines 14,15 . T rust and Distrust in Science Cross-national surveys suggest that public trust in many public bodies, such as institutions and governments, has declined since 2007 16 . However, most people continue to express a relatively high level of trust in science and scientists 17,18 . In the initial phase of the COVID-19 pandemic, surveys even indicated an increase in public trust in science 19,20 . This is noteworthy given that prior epidemics tended to be associated with a decline in trust toward scientists and their work 21 . Moving beyond self-report surveys, a 2022 cross-cultural study 22 introduced the “Einstein Effect”, showing that scientific sources are generally perceived as more credible than spiritual gurus, though this effect was demonstrated in response to meaningless content (i.e., “pseudo-profound bullshit” 23 ). Despite such findings, a dominant public narrative persists, suggesting a widespread crisis of trust in science and scientists. This disconnect between the (limited) empirical evidence and the dominant narrative raises important questions about how trust in science operates in real-world contexts. In particular, the present work examines trust in science when scientific sources are used to correct misinformation, as in the case of COVID-19. One distinction that may help explain the apparent divergence is between trust in the abstract sense—as measured via generic surveys—and enacted trust , that is, whether or not people accept information from trustworthy sources in a noisy, politicized, and polarized environment. Focusing on enacted trust, Iyengar and Massey 24 argue that today’s widespread dissemination of misleading and biased content contributes to growing distrust in the scientific enterprise. They claim that structural shifts in the media landscape have made scientific findings more vulnerable to distortion, especially when they threaten ideological worldviews. Roozenbeek et al. 25 demonstrate that across multiple countries, participants who reported higher trust in scientists were less likely to believe in false information related to COVID-19. West and Bergstrom 26 further highlight that both the dissemination of false information within the scientific community and the misrepresentation of scientific results to the public can erode confidence in scientific institutions. Building on these findings, the current research empirically examines trust in scientific sources in the applied context of correcting disinformation, in light of claims that the post-truth era has led to a neglect of facts and expertise in favor of subjectively confirmatory information. We also test a brief intervention designed to highlight the distinction between objective and subjective content, and between experts and lay perspectives, with the goal of restoring trust in factual content and expert authority. Correcting disinformation A wide range of interventions grounded in psychological theory have been developed to mitigate the spread and influence of misinformation, and many have received substantial empirical support 27 . Fact-checking labels are commonly used, but their impact tends to be small and they carry the risk that unflagged content may appear more credible if readers assume it has been verified 28 . Moreover, fact-checks rarely provide individuals with the critical thinking skills needed to evaluate claims independently, especially in the absence of verification cues 29 . More detailed refutations tend to reliably reduce false beliefs but they can only be applied retrospectively to target specific pieces of misinformation 30–33 . Another common approach involves nudging individuals toward accurate information by presenting factual content in a way that facilitates its ease of processing 1,34 . Subtle accuracy reminders can also reduce sharing of misinformation on social media 35–39 . Pre-emptive warnings about misinformation can also provide some protection but only if they are very explicit and specific 40–42 . A more involved proactive approach is inoculation, which combines warnings with an educational component that provides relevant counterarguments, explains misleading argumentation techniques, or discredits sources of low-quality information 43 ; such inoculations can build resistance to misinformation by raising awareness and skepticism before exposure 44–46 . For instance, forewarning individuals about politically motivated disinformation has been shown to reduce susceptibility and promote accurate perceptions of scientific consensus 47 . Such interventions can be incorporated in engaging educational interventions (i.e., courses and games) designed to improve people’s information literacy by teaching them how misinformation is constructed 48,49 . Other interventions focused on boosting media literacy are also able to enhance truth discernment 50 . One challenge faced by these interventions, however, is that they require substantial motivation, engagement, and cognitive resources, which may limit their applicability. Recent work has thus emphasized the need for scalable strategies for social media environments 38,51 . The Suggested S t rategy : ThinkFRE Considering the strengths and limitations of existing strategies and acknowledging that individuals are often unreceptive to factual corrections and expert advice, we propose a novel pre-emptive strategy. The suggested approach departs from the traditional emphasis on truth and accuracy evaluations that may be influenced by prior knowledge, beliefs, or political attitudes. It also goes beyond general scientific reasoning interventions 52 by specifically addressing disinformation through a focus on the distinction between (1) objective versus subjective information that is (2) falsifiable versus non-falsifiable, and comes from (3) expert versus layperson sources. These three basic yet powerful distinctions, while ideally neutral and widely acceptable across ideological divides, have become increasingly blurred in today’s information environment. The aim of the proposed strategy is to restore attention to these foundational categories, offering an accessible and broadly acceptable pathway for resisting disinformation. We term this new strategy ThinkFRE (see Fig. 1). ThinkFRE is a self-guided tool that , according to our hypothesis , will increase reliance on factual information and expert knowledge through three simple core prompts: (1) “ Fact : Is the information based on evidence or personal anecdote?” This encourages the reader to consider whether information is objective or subjective. (2) “ Refute : Can you think of a way to disprove the information presented?” This question prompts critical evaluation and reinforcing that only objective claims can be disproven. (3) “ Expert : Does the source have the necessary knowledge?” This encourages scrutiny of the information’s origin and highlights the difference between expert and non-expert sources. The intervention was introduced to participants as “ThinkFRE”; the “FRE” acronym was hoped to evoke an association with free and independent thought; omission of the second “E” was intentional, as it may subtly disrupt automatic reading and encourage deeper cognitive engagement. In sum, ThinkFRE offers a concise, general-purpose strategy for evaluating information without requiring users to judge what is true or false. Instead, it guides individuals to identify whether information is objective or subjective, falsifiable, and whether it comes from a reliable expert or not. Unlike content-specific fact-checking tools or cognitively demanding interventions, ThinkFRE is short, teachable, and easy to apply in everyday contexts. ThinkFRE was predicted to induce a sense of general healthy skepticism and self-efficacy that goes beyond specific content 53 . The Current Research The current research included three studies examining the potential effects of disinformation corrections from scientific sources, as well as the ThinkFRE strategy as a possible intervention. The studies focused on COVID-19 vaccination disinformation relevant to everyday life and were embedded within a realistic setting—a WhatsApp group conversation—to capture how people encounter and respond to misinformation in daily social interactions. Study 1 examined this question in the context of disinformation claiming that the COVID-19 vaccine causes infertility in women. Study 2 built on this by adding a general warning about source unreliability, allowing us to examine whether the ThinkFRE strategy provides added value even when participants are already alerted to the potential inaccuracy of the disinformation. Studies 1 and 2 were conducted in January 2021, amid the uncertainty of the early phase of the COVID-19 pandemic, just after the vaccine campaign had begun, when individuals were making behavioral decisions about whether to get vaccinated (or not). Study 3 replicated the core design in a more stable context after COVID-19 was no longer classified as a public health emergency of international concern by the WHO, yet it remained a pandemic-level disease (June 2024); it addressed disinformation concerning the vaccine’s long-term consequences. This setting of Study 3 allowed us to distinguish between participants with positive versus negative baseline attitudes, focusing specifically on vaccine skeptics, and use a more robust multi-item belief scale. Method All three studies were preregistered and approved by the Hebrew University Review Board (Studies 1–2: 11101_2021_IRB; Study 3: 048_2024_IRB). All methods were performed in accordance with the relevant guidelines and regulations. All participants provided informed consent at the beginning of the survey and could withdraw at any time. Study 1: Design and Procedure Conducted in January 2021, at the onset of the COVID-19 vaccine campaign and amidst widespread public uncertainty 1 , Study 1 focused on a viral piece of disinformation circulating at that time: the claim that the COVID-19 vaccine causes female sterility. The study followed a 2 (group: ThinkFRE vs. control) × 3 (correction type: scientific sources vs. no source vs. no-correction baseline) between-subjects design. Participants were randomly assigned to one of two groups (ThinkFRE vs. control). All participants were instructed that they would be presented with information found on social media and that they should read it carefully. Participants in the ThinkFRE group were introduced to the ThinkFRE strategy and instructed to apply it (see Fig. 1). In the control group, participants received no strategic instruction. Subsequently, all participants were shown a simulated WhatsApp conversation (see Fig. 2) containing the disinformation, followed by one of three correction conditions: (1) a correction citing official scientific sources (WHO, FDA, CDC) and clinical trial data; (2) a correction with identical content but no scientific references; or (3) a no-correction baseline where the disinformation was left unchallenged. Immediately following this exposure, we measured participants’ belief that the COVID-19 vaccine could cause sterility, our primary dependent variable. Study 1: Participants The required sample size was calculated using G*Power 3.1, with the aim to detect an interaction in an omnibus test of a 2 × 3 between-subjects ANOVA. Expecting a small to medium effect size of for the interaction effect at power = .80 and α = .05, the suggested minimum sample size was 244. To allow exclusions, 302 participants from the Prolific online platform were recruited. Participants were US residents aged 18–40, native English speakers, and had a Prolific approval rate of at least 85%. According to preregistered criteria, we excluded 15 participants who either failed the attention check ( n = 2), did not complete the survey in one sitting ( n = 1), reported external disturbances ( n = 5), reported that they completed the survey with someone else ( n = 1), or had a completion time of more than 3 SD above the mean ( M time = 6.40 min, SD time = 4.19, n = 6). The final sample consisted of N = 287 participants ( M age = 27.72 years, SD age = 6.51; 145 men, 137 women, and 5 participants who preferred not to state their gender); there were 141 participants in the ThinkFRE group and 146 in the control group. Study 1: Stimuli Disinformation: The disinformation and its correction were presented in a WhatsApp Group Chat format (see Fig. 2). The conversation started with a member forwarding an online article indicating in the title that the COVID-19 vaccine 2 causes sterility in 97% of women. Then, the same person who sent the link supports this claim and adds that Bill Gates, a big supporter of vaccination, did not hide his conviction that there should be a decline in the population. This combination of a specific health claim with a conspiratorial motive was designed to create a single, compelling disinformation narrative. Correction Type: In the two conditions with a correction, the correction was added by another chat-group member who replied that the vaccine is safe and effective and that its potential benefits outweigh the known and potential harms of becoming infected with COVID-19. In the correction condition with a scientific source, the same message was sent, but with an attribution to scientific sources, namely the Food and Drug Administration (FDA), Centers for Disease Control and Prevention (CDC), and World Health Organization (WHO), additionally emphasizing that the vaccine was declared safe and effective based on data and findings from large scientific trials (see Fig. 2, and for the entire stimuli see Fig. S1 in the supplementary material). Study 1: Measurement After being exposed to the display of the WhatsApp conversation, all participants were asked: “In your opinion, how likely is it that the COVID-19 vaccine can cause sterility?” The question was answered on a 7-point Likert scale ranging from “extremely unlikely” to “extremely likely.” This was our main dependent variable, which we refer to as “belief in presented disinformation” 3 . Study 2: Design and Procedure Study 2 was conducted in the same context as Study 1, aiming to replicate it while introducing an additional element: a general warning about the credibility of the source that initially shared the disinformation. This warning was presented to participants across all conditions. Studies 1 and 2 were conducted within a few days of each other, ensuring that participants’ vaccination status and the broader public discourse around COVID-19 vaccines remained consistent across both studies. The study design and procedure were largely identical to Study 1. Two minor changes were made to the stimuli to make the presentation more realistic: (1) the forwarded article included a small thumbnail image of a graph, and (2) the screenshot showed the entire WhatsApp interface, including the message composition area at the bottom, which had been cropped out in Study 1. These changes increased the ecological validity of the conversation display. More importantly, Study 2 added a short general warning to all conditions (see Fig. S3 in the supplementary material). In Study 1, only the ThinkFRE group was told in the instructions that they would be introduced to a method that enables people to distinguish between factual and false information, whereas the control group was only told to read the information carefully. Thus, one may claim that the ThinkFRE group was more explicitly warned regarding the risk of false information. Therefore, in Study 2, a warning was presented emphasizing the unreliability of the source for all participants. The warning came from a group member reacting with dismay to the disinformation (“Can we please limit such questionable and unconfirmed sources?”). In response, the person who forwarded the link asks how the other person can be certain that anything they read is true and confirmed, and why the article they shared is more questionable. Study 2: Participants Following the same considerations as in Study 1, the initial sample consisted of 300 participants from the Prolific online platform. Participants were US residents aged 18–40, native English speakers, and had a Prolific approval rate of at least 85%. According to preregistered criteria, we excluded 14 participants who either failed the attention check ( n = 1), did not complete the questionnaire in one sitting ( n = 3), reported on external disturbances ( n = 3), or had a completion time of more than 3 SD above the mean ( M time = 5.92 min, SD time = 3,98, n = 6). This resulted in a final sample of N = 287 participants ( M age = 27.12 years, SD age = 6.19; 145 men, 137 women, 5 who preferred not to state their gender): 145 participants in the ThinkFRE condition and 142 participants in the control condition. Study 3: Design and Procedure Unlike the initial studies conducted at the pandemic’s peak in 2021, Study 3 took place in June 2024, when most people held established attitudes about COVID-19 vaccination and the acute phase of the pandemic had passed. This updated setting enabled a more stringent test of our hypotheses, as we could now assess the impact of science-based corrections and the ThinkFRE strategy specifically on vaccine skeptics—the group considered most susceptible to disinformation. To align with this focus, the study design incorporated two key updates from the previous studies. First, while the general theme of the COVID-19 vaccine was retained, the disinformation stimulus was revised to address contemporary concerns about the vaccine’s long-term side effects, a topic more salient at this later stage of the pandemic. Second, to enable a direct comparison based on vaccine attitudes, participants were pre-screened and placed into one of two groups: those with generally positive or generally negative attitudes. This resulted in a design with 3 between-subjects factors: 2 (vaccination attitude: supporters vs. skeptics) × 2 (group: ThinkFRE vs. Control) × 3 (correction type: scientific sources vs. no source vs. no-correction baseline). Study 3: Participants To boost power, a larger sample was collected. The initial sample consisted of 1197 participants from the Prolific online platform. The participants were US residents aged 18– 5 0, native English speakers with a Prolific approval rate of at least 85%. We recruited participants according to their reported attitude regarding the COVID-19 vaccine: Half of the participants reported in the Prolific prescreening that they were in favor of the COVID-19 vaccine , and half of the participants reported that they were against the vaccine. According to preregistered criteria, we excluded participants who reported that they did not complete the survey alone ( n = 2), reported on external disturbances ( n = 18), did not complete the study in one sitting ( n = 24), admitted to not have put reasonable effort in the study ( n = 7), or reported that they looked up information online ( n = 14). We further excluded participants who reported a different vaccination attitude than indicated by the prescreening ( n = 127) 4 or had a completion time longer than 3 SD above the average time ( n = 30). After these exclusions, the final sample size was N = 975 participants ( M age = 34.80 years, SD age = 8.19; 422 men, 528 women, 25 who identified as non-binary/third-gender): 484 participants in the control condition and 491 in the ThinkFRE condition; 515 COVID-19 vaccine supporters and 460 COVID-19 vaccine skeptics. Study 3: Stimuli Disinformation : As in Studies 1 and 2, the disinformation and its correction were embedded within a simulated WhatsApp Group Chat (see Fig. S4 in the Supplementary Information). The conversation started with a member forwarding an online article whose headline claimed that people were experiencing sudden heart attacks after receiving the COVID-19 vaccine. When another group member responded by urging caution and requesting that unverified claims not be shared, the original sender defended and added a personal anecdote about a young, previously healthy man who allegedly suffered a heart attack following vaccination. Correction Type: As in Studies 1 and 2, there were three different conditions of correction: (1) no correction (baseline), (2) correction referencing scientific sources and data (the CDC, the American Heart Association, and the WHO), or (3) the same correction without any scientific reference (see Fig. 3). In the two conditions with a correction, the same group member who initially urged caution responded again, stating that there is no established link between COVID-19 vaccines and heart attacks. The correction clarified that although the vaccine can, in rare cases, cause heart inflammation, this is an extremely uncommon side effect. Importantly, the correction highlights that the risk of heart complications following COVID-19 infection is substantially higher than the risk associated with the vaccine. Study 3: Measurement s In addition to the primary dependent variable assessing direct belief in the disinformation claim (i.e., that the COVID-19 vaccine causes random heart attacks), Study 3 included five supplementary items designed to capture belief in the disinformation more indirectly (see Table 1 for all items). To improve upon the single-item belief measure used in Studies 1 and 2, the present study thus implemented a more robust multi-item composite, thereby addressing known issues of reliability and spurious effects with single-item measures (Swire-Thompson et al., 2020; 2022). Furthermore, we also assessed participants’ general attitudes toward the COVID-19 vaccine, as well as their trust in science and scientists, and their level of intellectual humility. Table 1. Measured items assessing the belief in the presented disinformation, correcti ve information , and distinction between the two. Specifically, items #1, #3, and #6 refer directly to the connection between COVID-19 vaccine and heart disease (i.e., disinformation). Items #2 and #5 pertain to the COVID-19 virus and heart disease (i.e., corrective information). Item #4 directly contrasts the belief in the potential consequences of the vaccine and the virus. Results Results S tudy 1: First , as an omnibus analysis, we conducted a two-way ANOVA with a 2 (group: ThinkFRE/Control) × 3 (correction type: correction with scientific source/correction with no source/baseline) design. The findings showed that participants in the ThinkFRE group did not show a significantly lower belief that the vaccine can cause sterility ( M = 2.04, 95% CI [1.82, 2.27]) compared to control participants ( M = 2.34, 95% CI [2.08, 2.60]), F (1, 281) = 3.0 7 , p = .08 1 . There was no main effect of correction type, F (2, 281) = 1.46, p = .234. The interaction between group (ThinkFRE vs. control) and correction type was also nonsignificant, F (2, 281) = 2.78, p = .064. Comparison of ThinkFRE and Control Across Different Correction Types : Despite the nonsignificant omnibus results, to directly test specific a-priori hypotheses regarding the potential effect of ThinkFRE when the correction is assigned to scientific sources, we ran specific planned contrasts. The results showed that in the condition with a scientific-source correction, participants who learned ThinkFRE believed less in the presented disinformation ( M = 1.82, 95% CI [1.45, 2.18]) compared to control participants ( M = 2.71, 95% CI [2.23, 3.18]), t (281) = 2.93, p = .004, Cohen’s d = 0.61, 95% CI [0.20, 1.03]) (Fig. 3). Participants who learned the ThinkFRE paradigm were not different in their belief in the presented disinformation from control participants in the baseline condition ( p = .95), nor when receiving a correction without a source ( p = .82). These two null effects suggest that learning ThinkFRE did not have a general social-desirability effect in the sense of leading participants to report lower belief in the disinformation if it appeared alone or with a correction message without a scientific reference. Comparison of Correction Types Within Each Group: We also ran specific contrasts to test the hypothesis that a scientific-source correction can reduce belief in the disinformation compared to a correction without a source and the baseline condition. In the control group, belief in the disinformation was higher when the correction included a scientific reference ( M = 2.71, 95% CI [2.23, 3.18]) compared to the baseline condition ( M = 1.98, 95% CI [1.58, 2.38]; t (281) = 2.46, p = .014, Cohen’s d = 0.50, 95% CI [0.10, 0.90]), suggesting a backfire effect of a scientific correction. However, among control participants, those who received a scientific correction were not different in their belief in the disinformation from those presented with a correction without a source ( p = .22). Importantly, no backfire effect of a scientific correction was observed in the ThinkFRE group ( p = .55). Results S tudy 2 : As in Study 1, we conducted an initial omnibus two-way ANOVA with a 2 (group: ThinkFRE/Control) × 3 (correction type: correction with scientific source/correction with no source/baseline) design. We found a main effect of group, indicating that participants who learned ThinkFRE believed less that the vaccine might cause sterility ( M = 1.86, 95% CI [1.68, 2.05]) compared to those in the control group ( M = 2.36, 95% CI [2.09, 2.63]), F (1, 281) = 9.25, p = .003, η p 2 = 0.03, 95% CI [0.004, 0.08]. There was no significant main effect of correction type, F (2, 281) = 2.97, p = .053. However, there was a significant interaction, suggesting that the effect of correction type depended on the group, F (2, 281) = 3.99, p = .02, η p 2 = 0.03, 95% CI [0.0005, 0.07]. Comparison of ThinkFRE and Control Across Correction Types: Analyzing specific planned contrasts to directly test our hypotheses, we see again that the effect of ThinkFRE was not a result of a general desirability effect. Consistent with Study 1, only in the scientific-correction condition did control participants exhibit significantly greater belief in the disinformation ( M = 2.90, 95% CI [2.36, 3.43]) compared to participants who learned ThinkFRE ( M = 1.75, 95% CI [1.45, 2.05]), t (281) = 4.05, p < .001, Cohen’s d = 0.83, 95% CI [0.42, 1.23]. There was no difference between control participants and ThinkFRE participants in their belief in the disinformation when presented with a correction without a source ( p = . 40 ) or no correction at all ( p = . 71 ). Comparison of Correction Types Within Each Group: As in Study 1, we tested the hypothesis of a potential backfire effect of a scientific reference when correcting disinformation. In the control group, participants believed the disinformation significantly more after reading a correction with a scientific source ( M = 2.90, 95% CI [2.36, 3.43) compared to the baseline condition ( M = 1.90, 95% CI [1.55, 2.25]), t (281) = 3.53, p < .001, Cohen’s d = 0.72, 95% CI [0.31, 1.13]) and compared to the correction without a source ( M = 2.28, 95% CI [1.78, 2.79]), t (281) = 2.14, p = .03, Cohen’s d = 0.44, 95% CI [0.03, 0.85]), thereby demonstrating a backfire effect of scientific sources (Fig. 4). This pattern was absent in the ThinkFRE group. Results S tudy 3 : The results section will focus on the multi-item compound measure, designed to capture a more general belief in the presented disinformation. Out of the six items, we combined the four items probing the perceived influence of the COVID-19 vaccine (i.e., directly connected to the belief in the disinformation): belief in a causal link between the vaccine and heart attacks (item 1; “ In your opinion, how likely is it that there is any causal connection between the COVID-19 vaccine and heart attacks ? ”), belief in a causal link between the vaccine and heart inflammation (item 3; “ In your opinion, how likely is it that there is any causal connection between the COVID-19 vaccine and heart inflammation ? ”), the bipolar cause-attribution item regarding heart inflammation (item 4; “ Please move the circle along the slider to indicate, in your opinion, the more common cause for heart inflammation . (You can place the circle anywhere on the scale between "COVID-19 Vaccine" and "COVID-19 Virus)", reverse-coded to align with the other items), and speculation about cases of heart attack or heart inflammation following vaccination (item 6; “ In your opinion, out of 1000 healthy people below the age of 60 who got the COVID-19 vaccine , how many will subsequently suffer a heart attack or heart inflammation? The scale ranges from 0 (0% of the population, i.e. no-one) to 100 or more people (10% or more of the population) ”). These four items (Cronbach’s α = .89) were combined into a composite score on a 0-10 scale (see Supplementary Information for item correlations). The results for the single-item measure of belief in the disinformation (item #1) and the bipolar cause-attribution item (item #4) are reported separately in the Supplementary Information. Our primary focus was on participants who were skeptical of the COVID-19 vaccine, as they were presumed to be more susceptible to endorsing disinformation. In contrast, corrections were expected to have a limited impact among vaccine supporters, who were less likely to believe the false claim to begin with. Accordingly, we analyzed the effects of scientific references in corrections and the ThinkFRE intervention separately for COVID-19 vaccine skeptics and COVID-19 vaccine supporters (for the sake of completeness, results of a three-way ANOVA are reported in the Supplementary Information). COVID-19 Vaccine Skeptics : There was a strong effect of group, indicating that ThinkFRE participants generally believed less in the claimed link between the COVID-19 vaccine and heart disease presented in the disinformation ( M = 6.31, 95% CI [6.06, 6.56]), compared to control participants ( M = 6.91, 95% CI [6.69, 7.13]), F (1,454) = 12.69, p < .001, η p 2 =0.03, 95% CI [0.006, 0.06]. There was no significant main effect of correction type ( p = .28) and no significant interaction ( p = .42) among COVID-19 vaccine skeptics. Comparison of ThinkFRE and Control Across Correction Types: We conducted specific contrasts according to our a priori hypothesis. Replicating the results of Studies 1 and 2, COVID-19 vaccine skeptics who used ThinkFRE had lower general disinformation belief in the scientific-correction condition ( M = 6.01, 95% CI [5.5 4 , 6.4 7 ]) than control participants ( M = 6.90, 95% CI [6.5 5 , 7. 24 ]), t (454) = 3.05, p = .003, d = 0.49, 95% CI [0.17, 0.81]. However, as in Studies 1 and 2, there was no significant difference between ThinkFRE and control in the baseline condition ( p = .051) and in the no-source correction condition ( p = .25). Comparison of Correction Types Within Each Group: We also conducted specific contrasts to test the effect of the scientific correction within each group to test for a potential backfire effect. In contrast to our earlier findings, the scientific correction did not produce a backfire effect among vaccine skeptics in the control group. Instead, the correction was simply ineffective, showing no significant difference when compared to the baseline condition ( p = .93) or to the correction without a source ( p = .81). In the ThinkFRE group, the scientific correction ( M = 6.01, 95% CI [5.5 4 , 6.4 7 ]) did not influence belief in the disinformation compared to the baseline condition ( M = 6.30, 95% CI [5. 90 , 6. 71 ]; p = .32) but was effective compared to the no-source correction ( M = 6. 63 , 95% CI [ 6.17 , 7.08 ]), t (454) = 2.05, p = .041, d = 0.34, 95% CI [0.01, 0.67] 5 . COVID-19 Vaccine Supporters : Among COVID-19 vaccine supporters, there were no significant main effects. General belief in the presented disinformation about the COVID-19 vaccine was not affected by group ( p = .95), nor by the correction type ( p = .67) or their interaction ( p = .15). Comparison of ThinkFRE and Control Across Correction Types: When conducting the specific contrasts among COVID-19 vaccine supporters, no differences were found between ThinkFRE and control participants in the baseline condition ( p = .13), no-source correction condition ( p = .24), and scientific correction condition ( p = .80). Comparison of Correction Types Within Each Group: Among COVID-19 vaccine supporters in the control group, the scientific correction condition was not different from the baseline condition ( p = .46) or the no-source correction condition ( p = .93). Similarly, among COVID-19 vaccine supporters who learned ThinkFRE, the correction with a scientific source did not differ from baseline ( p = .42) or the no-source correction ( p = .30). Trust in Science and Scientists We measured trust in science with the Trust in Science and Scientists Inventory (Nadelson et al., 2014). COVID-19 vaccine supporters showed significantly higher trust in science and scientists ( M = 4.07, 95% CI [4.03, 4.11]) compared to COVID-19 vaccine skeptics ( M = 2.77, 95% CI [2.71, 2.83]), F (1, 963) = 1231.77, p < .001; η p 2 = 0.56, 95% CI [0.52, 0.59]. None of the other factors, neither group ( p = .13), nor correction type ( p = .33) , nor any interaction between them , affected participants’ trust in science and scientists. Trust in science was significantly negatively correlated with belief in the presented disinformation ( r = -0.77, p < .001). In other words, the higher someone’s trust in science, the less likely they were to believe in the causal link between the COVID-19 vaccine and heart disease. This pattern held for both COVID-19 vaccine supporters ( r = -0.40, p < .001) and skeptics ( r = -0.43, p <.001) 6 . Discussion The present research demonstrates that citing scientific sources to correct vaccine disinformation can be ineffective or even counterproductive, depending on the context. During a period of high uncertainty and unfamiliar disinformation about the new COVID-19 vaccines (Studies 1 & 2; 2021), a science-based correction backfired among participants in a control group that received no pre-emptive intervention, increasing belief in the disinformation compared to a no-correction baseline. Later, among COVID-19 vaccine skeptics (Study 3; 2024), the same type of science-based correction was merely ineffective. A novel, pre-emptive intervention that provided a subtle reminder of the importance of evidence and expertise, worked specifically to remedy this failure. Its positive effect was significant only in the scientific-source condition, where it made the correction more effective. This targeted impact suggests that for science-based corrections to succeed, individuals may first need to be prompted into an evidence-based mindset that restores the persuasive power of scientific expertise. On a broader level, our findings reveal a critical disconnect between self-reported, abstract trust in science and its practical enactment. While large-scale surveys often report high public trust in science and scientists, our research provides a more implicit, behavioral measure by testing this trust with highly relevant real-world disinformation within an ecologically valid context. The fact that a correction citing scientific authorities could fail or even backfire demonstrates that abstract trust does not automatically translate into enacted trust and the corresponding belief revision when it is most needed—in the direct confrontation with potent disinformation. This suggests that public trust in science is potentially weaker than often assumed, particularly under conditions of uncertainty and fear and among those with entrenched skeptical views. The fact that a brief intervention was sufficient to restore the positive impact of scientific sources indicates that the core issue may not be a deep-seated rejection of science, but a correctable lapse in applying an evidence-based mindset. Disinformation thrives in conditions of confusion, where people are unsure what to believe and the consequences of error can be severe 54 . Our findings on the effectiveness of the ThinkFRE strategy offer a promising path forward in tackling such challenges. In high-stakes domains like public health, the need for broadly applicable, easily deployable strategies is a priority, and ThinkFRE exemplifies such a tool. We propose its effectiveness stems from its theoretical foundation: by reminding individuals to distinguish between objective and subjective information, taking into account falsifiability and highlighting the value of expertise, it may activate critical evaluation in a way that is likely to be resilient to the influence of prior attitudes. The strength of this approach lies in its design as a lightweight, scalable intervention that requires no specialized knowledge or reliance on content-specific counterarguments. This content-agnostic nature aligns with a growing body of research identifying interventions that foster analytic thinking as among the most effective and broadly applicable tools against misinformation 37 , 38 , 55 . It also fits the theoretical framework emphasizing the potential to bridge epistemic conflicts by adopting common evidential standards 56 . While this suggests ThinkFRE has strong promise for widespread use, its generalizability must be empirically established. Therefore, a crucial avenue for future work is to test this strategy across diverse contexts and explore its practical applications in real-world settings. The current research has several limitations. Unlike Studies 1 and 2, Study 3 did not replicate the negative effect of a scientific source in the control group, regardless of whether a single-item or multi-item measure was used. While backfire effects have been reported in some studies 57 , 58 , such effects are generally rare, and several attempts to replicate such findings have been unsuccessful 59 – 63 . Correspondingly, it might be that the backfire effect observed in Studies 1 and 2 may have been driven by the extraordinary uncertainty, caution, and skepticism prevalent among large sections of the U.S. public during the unprecedented early stages of the COVID-19 pandemic. In such contexts, new scientific information may require particularly high levels of trust to effectively counter disinformation. This highlights the critical role of timing in disinformation and correction efforts, as backfire effects may be more likely to emerge in environments characterized by heightened uncertainty or low source trust. Another potential, if speculative, explanation for the lack of a backfire effect in Study 3 is the high level of conviction within our vaccine skeptic group. By design, these participants already held a significantly higher belief in the disinformation than our sample in our earlier studies. When a belief is this strong, it is possible that a correction from a distrusted source is met with simple dismissal rather than the active counter-argumentation that typically drives a backfire effect. In essence, their conviction may have been so solidified that the correction was merely ignored, not actively resisted. Furthermore, we do not know whether the effect of ThinkFRE is caused by one specific component of the intervention (e.g., the Expert component) or by the interaction of all three components (Fact, Refutable, Expert). Future research should examine whether each individual concept can separately restore trust in factual information and expert knowledge, or whether it is ThinkFRE as a whole that encourages a mindset of healthy skepticism. In any case, invoking objectivity and expertise may go a long way in restoring trust in scientific sources. Declarations Research Funding This research was supported by the Israel Science Foundation (ISF) Grant no. 740/23 RM, awarded to Ruth Mayo. Declaration of competing interest: The authors declare no competing interests. Author Contribution A.L.B - Conceptualization, Methodology, Data Curation, Formal Analysis, Visualization, Writing (original draft), Writing (review & editing). U.K.H.E. 6 Conceptualization, Methodology, Writing (review & editing). R.M. 6 Conceptualization, Methodology, Validation, Supervision, Writing (review & editing), Funding Acquisition. Data Availability The datasets, analysis code and preregistrations for these studies are available via [OSF](https:/osf.io/6ynek/overview?view_only=e2f387afa7174ec48942c2c30e1197aa) . References Levy, N. Nudges in a post-truth world. J. Med. Ethics 43 , 495–500 (2017). McIntyre, L. Post-Truth . (MIt Press, 2018). Van der Linden, S. & Löfstedt, R. E. Risk and Uncertainty in a Post-Truth Society . (Routledge, 2019). Lewandowsky, S., Ecker, U. K. H. & Cook, J. Beyond misinformation: Understanding and coping with the “post-truth” era. J. Appl. Res. Mem. Cogn. 6 , 353–369 (2017). Rothkopf, D. J. When the buzz bites back. Wash. Post 11 , B1--B5 (2003). Zarocostas, J. How to fight an infodemic. The lancet 395 , 676 (2020). Ash, E., Galletta, S., Hangartner, D., Margalit, Y. & Pinna, M. The effect of Fox News on health behavior during COVID-19. Polit. Anal. 32 , 275–284 (2024). Bursztyn, L., Rao, A., Roth, C. P. & Yanagizawa-Drott, D. H. Misinformation during a Pandemic . https://www.nber.org/papers/w27417 (2020). Simonov, A., Sacher, S., Dubé, J.-P. & Biswas, S. Frontiers: The Persuasive Effect of Fox News: Noncompliance with Social Distancing During the COVID-19 Pandemic. Mark. Sci. 41 , 230–242 (2022). Ecker, U. K. H. et al. Why misinformation must not be ignored. Am. Psychol. 80 , 867–878 (2025). van Bavel, J. J. et al. Using social and behavioural science to support COVID-19 pandemic response. Nat. Hum. Behav. 4 , 460–471 (2020). Algan, Y., Cohen, D., Davoine, E., Foucault, M. & Stantcheva, S. Trust in scientists in times of pandemic: Panel evidence from 12 countries. Proc. Natl. Acad. Sci. 118 , e2108576118 (2021). Plohl, N. & Musil, B. Modeling compliance with COVID-19 prevention guidelines: the critical role of trust in science. Psychol. Health Med. 26 , 1–12 (2021). Adeoye, A. F. et al. The 2025 United States Measles Crisis: When Vaccine Hesitancy Meets Reality. Cureus 17 , e88196. Parums, D. V. A Review of the Resurgence of Measles, a Vaccine-Preventable Disease, as Current Concerns Contrast with Past Hopes for Measles Elimination. Med. Sci. Monit. Int. Med. J. Exp. Clin. Res. 30 , e944436 (2024). OECD. Trust and Public Policy: How Better Governance Can Help Rebuild Public Trust . OECD Public Governance Reviews 1–158 http://dx.doi.org/10.1787/9789264268920-en (2017). Cologna, V. et al. Trust in scientists and their role in society across 68 countries. Nat. Hum. Behav. 9 , 713–730 (2025). Wellcome. Wellcome Global Monitor: How Covid-19 affected people’s lives and their views about science. (2020). Bromme, R., Mede, N. G., Thomm, E., Kremer, B. & Ziegler, R. An anchor in troubled times: Trust in science before and within the COVID-19 pandemic. PloS One 17 , e0262823 (2022). Wellcome. Insights from the Wellcome Global Monitor. Wellcome https://wellcome.org/insights/articles/public-trust-scientists-rose-during-covid-19-pandemic-0 (2021). Eichengreen, B., Aksoy, C. G. & Saka, O. Revenge of the experts: Will COVID-19 renew or diminish public trust in science? J. Public Econ. 193 , 104343 (2021). Hoogeveen, S. et al. The Einstein effect provides global evidence for scientific source credibility effects and the influence of religiosity. Nat. Hum. Behav. 6 , 523–535 (2022). Pennycook, G., Cheyne, J. A., Barr, N., Koehler, D. J. & Fugelsang, J. A. On the reception and detection of pseudo-profound bullshit. Judgm. Decis. Mak. 10 , 549–563 (2015). Iyengar, S. & Massey, D. S. Scientific communication in a post-truth society. Proc. Natl. Acad. Sci. 116 , 7656–7661 (2019). Roozenbeek, J. et al. Susceptibility to misinformation about COVID-19 around the world. R. Soc. Open Sci. 7 , 201199 (2020). West, J. D. & Bergstrom, C. T. Misinformation in and about science. Proc. Natl. Acad. Sci. 118 , e1912444117 (2021). Kozyreva, A. et al. Toolbox of individual-level interventions against online misinformation. Nat. Hum. Behav. 8 , 1044–1052 (2024). Pennycook, G., Bear, A., Collins, E. T. & Rand, D. G. The implied truth effect: Attaching warnings to a subset of fake news headlines increases perceived accuracy of headlines without warnings. Manag. Sci. 66 , 4944–4957 (2020). Howell, E. L. & Brossard, D. (Mis) informed about what? What it means to be a science-literate citizen in a digital world. Proc. Natl. Acad. Sci. 118 , e1912436117 (2021). Prike, T. & Ecker, U. K. H. Effective correction of misinformation. Curr. Opin. Psychol. 54 , 101712 (2023). Lewandowsky, S., Ecker, U. K. H., Seifert, C. M., Schwarz, N. & Cook, J. Misinformation and Its Correction: Continued Influence and Successful Debiasing. Psychol. Sci. Public Interest 13 , 106–131 (2012). Ecker, U. K. H., O’Reilly, Z., Reid, J. S. & Chang, E. P. The effectiveness of short‐format refutational fact‐checks. Br. J. Psychol. 111 , 36–54 (2020). Lewandowsky, S. et al. The Debunking Handbook 2020 . (2020). Cook, J. & Lewandowsky, S. The debunking handbook. University of Queensland. (2011). Butler, L. H., Prike, T. & Ecker, U. K. Nudge-based misinformation interventions are effective in information environments with low misinformation prevalence. Sci. Rep. 14 , 11495 (2024). Fazio, L. Pausing to consider why a headline is true or false can help reduce the sharing of false news. Harv. Kennedy Sch. Misinformation Rev. 1 , (2020). Pennycook, G. et al. Shifting attention to accuracy can reduce misinformation online. Nature 592 , 590–595 (2021). Pennycook, G., McPhetres, J., Zhang, Y., Lu, J. G. & Rand, D. G. Fighting COVID-19 misinformation on social media: Experimental evidence for a scalable accuracy-nudge intervention. Psychol. Sci. 31 , 770–780 (2020). Pennycook, G. & Rand, D. G. Accuracy prompts are a replicable and generalizable approach for reducing the spread of misinformation. Nat. Commun. 13 , 2333 (2022). Chan, M. S., Jones, C. R., Hall Jamieson, K. & Albarrac\’\in, D. Debunking: A meta-analysis of the psychological efficacy of messages countering misinformation. Psychol. Sci. 28 , 1531–1546 (2017). Ecker, U. K. H., Lewandowsky, S. & Tang, D. T. W. Explicit warnings reduce but do not eliminate the continued influence of misinformation. Mem. Cognit. 38 , 1087–1100 (2010). Lewandowsky, S., Stritzke, W. G. K., Oberauer, K. & Morales, M. Misinformation and the “War on Terror”: when memory turns fiction into fact. in Terrorism and Torture: An Interdisciplinary Perspective (eds. Stritzke, W. G. K., Lewandowsky, S., Denemark, D., Clare, J. & Morgan, F.) 179–203 (Cambridge University Press, 2009). doi:10.1017/CBO9780511581199.010. Spearing, E. R. et al. Countering AI-generated misinformation with pre-emptive source discreditation and debunking. R. Soc. Open Sci. 12 , 242148 (2025). Compton, J., van der Linden, S., Cook, J. & Basol, M. Inoculation theory in the post-truth era: Extant findings and new frontiers for contested science, misinformation, and conspiracy theories. Soc. Personal. Psychol. Compass 15 , e12602 (2021). Lewandowsky, S. & Van Der Linden, S. Countering misinformation and fake news through inoculation and prebunking. Eur. Rev. Soc. Psychol. 32 , 348–384 (2021). Cook, J., Lewandowsky, S. & Ecker, U. K. H. Neutralizing misinformation through inoculation: Exposing misleading argumentation techniques reduces their influence. PloS One 12 , e0175799 (2017). der Linden, S., Leiserowitz, A., Rosenthal, S. & Maibach, E. Inoculating the public against misinformation about climate change. Glob. Chall. 1 , 1600008 (2017). Basol, M. et al. Towards psychological herd immunity: Cross-cultural evidence for two prebunking interventions against COVID-19 misinformation. Big Data Soc. 8 , 20539517211013868 (2021). Roozenbeek, J. & der Linden, S. Fake news game confers psychological resistance against online misinformation. Palgrave Commun. 5 , 1–10 (2019). Guess, A. M. et al. A digital media literacy intervention increases discernment between mainstream and false news in the United States and India. Proc. Natl. Acad. Sci. 117 , 15536–15545 (2020). Roozenbeek, J., Van Der Linden, S., Goldberg, B., Rathje, S. & Lewandowsky, S. Psychological inoculation improves resilience against misinformation on social media. Sci. Adv. 8 , eabo6254 (2022). Georgiou, N., Delfabbro, P. & Balzan, R. The effectiveness of a scientific reasoning intervention for conspiracy theory beliefs. Appl. Cogn. Psychol. 37 , 369–382 (2023). Salovich, N. A., Kirsch, A. M. & Rapp, D. N. Evaluative mindsets can protect against the influence of false information. Cognition 225 , 105121 (2022). Rapp, D. N. & Salovich, N. A. Can’t we just disregard fake news? The consequences of exposure to inaccurate information. Policy Insights Behav. Brain Sci. 5 , 232–239 (2018). O’Mahony, C., Brassil, M., Murphy, G. & Linehan, C. The efficacy of interventions in reducing belief in conspiracy theories: A systematic review. Plos One 18 , e0280902 (2023). Tay, L. Q. et al. Broadening Misinformation Research: The Roles of Evidence Retrievability and Epistemic Bridging. Preprint at https://doi.org/10.31234/osf.io/jvs9u_v1 (2025). Schwarz, N., Sanna, L. J., Skurnik, I. & Yoon, C. Metacognitive experiences and the intricacies of setting people straight: Implications for debiasing and public information campaigns. Adv. Exp. Soc. Psychol. 39 , 127–161 (2007). Schwarz, N., Newman, E. & Leach, W. Making the truth stick & the myths fade: Lessons from cognitive psychology. Behav. Sci. Policy 2 , 85–95 (2016). Ecker, U. K. H., Lewandowsky, S., Swire, B. & Chang, D. Correcting false information in memory: Manipulating the strength of misinformation encoding and its retraction. Psychon. Bull. Rev. 18 , 570–578 (2011). Ecker, U. K. H., Lewandowsky, S. & Chadwick, M. Can corrections spread misinformation to new audiences? Testing for the elusive familiarity backfire effect. Cogn. Res. Princ. Implic. 5 , 41 (2020). Ecker, U. K. H. et al. The psychological drivers of misinformation belief and its resistance to correction. Nat. Rev. Psychol. 1 , 13–29 (2022). Swire-Thompson, B., DeGutis, J. & Lazer, D. Searching for the backfire effect: Measurement and design considerations. J. Appl. Res. Mem. Cogn. 9 , 286–299 (2020). Wood, T. & Porter, E. The elusive backfire effect: Mass attitudes’ steadfast factual adherence. Polit. Behav. 41 , 135–163 (2019). Footnotes At the time Studies 1 and 2 were conducted, two COVID-19 vaccines had been approved for use in the U.S.: the Pfizer-BioNTech and Moderna COVID-19 vaccines. Initial doses were recommended for healthcare personnel and residents of long-term care facilities. Vaccines only became widely available to the general population in April 2021. No specific vaccine manufacturer or type was indicated in the studies. Following the main dependent variable, participants were asked a second question: “In your opinion, how likely is it that Bill Gates is connected to the COVID-19 pandemic or its vaccine?” After running the studies, we realized that the phrasing of the second dependent variable is problematic: It is also possible to interpret the question in terms that Bill Gates might be connected to the COVID-19 vaccine by helping with its funding, which could be the reason why we did not find effects for this variable. Since it was not the main variable and since it appeared after the main dependent variable about the connection between the vaccine and sterility, we do not report on this variable in this paper. The prescreening item was: “Please describe your attitudes towards the COVID-19 (Coronavirus) vaccines.” We selected half of the participants who answered “For (I feel positively about the vaccines)” and the other half who answered “Against (I feel negatively about the vaccines).” In the debrief section at the end of the study, we included the same item with the additional options: “Neutral (I don't have strong opinions either way)” and “Prefer not to say.” In the preregistration, we committed to analyzing the data according to participants’ vaccine attitudes. However, we had not anticipated the extent of deviation from the prescreening responses and therefore excluded participants whose responses were inconsistent. The main findings and their significance remain unchanged when excluding these 127 participants (see Supplementary Information). This specific comparison was not part of our pre-registered hypotheses and should be considered exploratory. A similar report for Intellectual Humility can be found in the Supplementary Information. Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterialsPDF.pdf Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 18 Mar, 2026 Reviews received at journal 10 Mar, 2026 Reviews received at journal 05 Mar, 2026 Reviewers agreed at journal 02 Mar, 2026 Reviewers agreed at journal 23 Feb, 2026 Reviews received at journal 16 Jan, 2026 Reviewers agreed at journal 17 Dec, 2025 Reviewers invited by journal 17 Dec, 2025 Editor assigned by journal 10 Dec, 2025 Submission checks completed at journal 01 Dec, 2025 First submitted to journal 30 Nov, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8204469","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":563113374,"identity":"c63f0e82-20a0-4105-bd78-e25f78462e8b","order_by":0,"name":"Anat Lande-Brenner","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABEklEQVRIiWNgGAWjYNCCAiA+wHyAgcHAgsGAgSHhAEiQjR2nesYGkDKGA2wJQC0SSFqYCWrhAZFgLVCAQwv/7ObnDxgM7PL4jvd83fCjQILBXCLh4QGGGjsGPhxaJO4cMwTaklwseebstps9QIdZzkgAOuxYMm6H3UgAaWFO3HAjd9sNHpBfboC0sB3AqUX+RvpHoJb6xA333zy7+Qeu5R9uLQY3ckC2HAbawsN2G24LYxtuLYY3cgpnJBgcT5x5Js3stoyBBI/BmQcJBxL7knlwaZG7kb7hw4eK6sS+44ef3Xzzx0bO4HhO8ocP3+zk5NsbcPgfCBKQ2DxAlAAS4cGtHhOwHyBF9SgYBaNgFAx/AAAI6GB85pTcDgAAAABJRU5ErkJggg==","orcid":"","institution":"The Hebrew University of Jerusalem","correspondingAuthor":true,"prefix":"","firstName":"Anat","middleName":"","lastName":"Lande-Brenner","suffix":""},{"id":563113377,"identity":"0cc298ec-c0c7-40dd-bf80-46d2debe952d","order_by":1,"name":"Ullrich K. H. Ecker","email":"","orcid":"","institution":"University of Western Australia","correspondingAuthor":false,"prefix":"","firstName":"Ullrich","middleName":"K. H.","lastName":"Ecker","suffix":""},{"id":563113378,"identity":"29b10a76-353c-4808-9890-b8298b58cc16","order_by":2,"name":"Ruth Mayo","email":"","orcid":"","institution":"The Hebrew University of Jerusalem","correspondingAuthor":false,"prefix":"","firstName":"Ruth","middleName":"","lastName":"Mayo","suffix":""}],"badges":[],"createdAt":"2025-11-25 14:38:33","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8204469/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8204469/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":98865308,"identity":"3a3cea5a-4292-4f41-83eb-ad05652b7744","added_by":"auto","created_at":"2025-12-23 10:43:00","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":780176,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/37391d3e73028e621e24ec96.docx"},{"id":99308958,"identity":"908f5eee-4acd-43d9-b6e7-dcf2d596f692","added_by":"auto","created_at":"2025-12-31 16:09:34","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4856,"visible":true,"origin":"","legend":"","description":"","filename":"4a2a3f09b8b340bdae6619dd3cf11eef.json","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/dccf98b9da9263a6686e2e70.json"},{"id":98865311,"identity":"51c60ccd-09fd-4601-94ae-72d58560d88a","added_by":"auto","created_at":"2025-12-23 10:43:00","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":901425,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterialsPDF.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/1d5d6f4657f2be80396eb826.pdf"},{"id":98865313,"identity":"3bd30a99-cc77-4ccd-964f-82a8fdc23595","added_by":"auto","created_at":"2025-12-23 10:43:00","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":144076,"visible":true,"origin":"","legend":"","description":"","filename":"4a2a3f09b8b340bdae6619dd3cf11eef1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/778ba84a0c3262ad766af7e1.xml"},{"id":99309017,"identity":"1580d22b-2096-48a7-a893-74db7bb5858b","added_by":"auto","created_at":"2025-12-31 16:09:40","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":120229,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/fdf7d63720701079cc1ce2b4.jpeg"},{"id":98865316,"identity":"06533711-1f12-4c2a-bd07-2c6386dd0fcb","added_by":"auto","created_at":"2025-12-23 10:43:00","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1074,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/2f0302cce71ef3de51ed0d02.jpeg"},{"id":99309130,"identity":"829824ce-02db-454d-89a9-5bf3254d3e25","added_by":"auto","created_at":"2025-12-31 16:09:46","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":163606,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/5ba9fe7b2a9bce2dff0492a7.png"},{"id":98865323,"identity":"f335eee6-a6e7-4ebf-8b5c-c22b3976bfcf","added_by":"auto","created_at":"2025-12-23 10:43:00","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":79955,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/c765f837fba1d7c325ba0e3f.png"},{"id":98865317,"identity":"25cc2be2-9f6f-45a2-8ec4-86b8eb800bda","added_by":"auto","created_at":"2025-12-23 10:43:00","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":105459,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/349422266ee371c716b36ced.png"},{"id":98865314,"identity":"0f553fc0-9f6f-46c2-b29f-fb58f8a426e4","added_by":"auto","created_at":"2025-12-23 10:43:00","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":94585,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/acc303f64144d921b140ab30.png"},{"id":98865330,"identity":"70371b22-b429-47bd-b4ed-0e89a5fb4a1c","added_by":"auto","created_at":"2025-12-23 10:43:01","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":160505,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/4732f00af721dcc332af21e7.png"},{"id":99308954,"identity":"fd5df8f3-f932-40ce-bb25-7e28fad03433","added_by":"auto","created_at":"2025-12-31 16:09:34","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":96912,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/1aa3bc0430aa3817c31dbf7d.png"},{"id":99309141,"identity":"70c5b773-140a-451b-b749-6b3715ad3d16","added_by":"auto","created_at":"2025-12-31 16:09:47","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":935,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/5af05abda450e039f52444f2.png"},{"id":98865326,"identity":"3ea47283-9c46-4bc1-92ca-cba7c2ad6cbf","added_by":"auto","created_at":"2025-12-23 10:43:01","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":30741,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/1055aacbd8e0db168fd9fc99.png"},{"id":99308697,"identity":"e8237e65-8184-44b2-a545-5cc520ed7cad","added_by":"auto","created_at":"2025-12-31 16:08:59","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":67853,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/e718ece1168f75f600d283ac.png"},{"id":98865324,"identity":"31490b01-0c1f-4fe9-88bd-d5fedf4ee5a6","added_by":"auto","created_at":"2025-12-23 10:43:00","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28418,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/9c62b05e9dfc1134d7ac67d1.png"},{"id":98865328,"identity":"21811467-1af4-4a88-bc39-53c71329b0a1","added_by":"auto","created_at":"2025-12-23 10:43:01","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":25888,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/fb98bbc04a1a7966135fbab2.png"},{"id":98865325,"identity":"2465b3bf-c504-4cb0-8f61-04792976d484","added_by":"auto","created_at":"2025-12-23 10:43:00","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":38366,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/ccb00a393fbc6e1c86a3caae.png"},{"id":98865321,"identity":"0672c4de-1212-46a7-954b-d5363d063164","added_by":"auto","created_at":"2025-12-23 10:43:00","extension":"xml","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":141448,"visible":true,"origin":"","legend":"","description":"","filename":"4a2a3f09b8b340bdae6619dd3cf11eef1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/ddca72b9017a92a5ab212c11.xml"},{"id":99308976,"identity":"aee841ce-c341-4ad5-bc35-b2f8380e0df3","added_by":"auto","created_at":"2025-12-31 16:09:36","extension":"html","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":160667,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/9d3b0001c64507be52a0ea64.html"},{"id":99308652,"identity":"fbae6150-8aa9-4b4e-a9c0-9c86e999a652","added_by":"auto","created_at":"2025-12-31 16:08:53","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":397398,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThinkFRE strategy as it was presented to the participants.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/ee2ec48c6c9e242f54de127b.png"},{"id":99308765,"identity":"40485e25-76a8-4a2b-83dc-c438635f08a9","added_by":"auto","created_at":"2025-12-31 16:09:07","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":280963,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eStimuli for Study 1 in WhatsApp Group Chat format, presented by correction type conditions. \u003c/strong\u003eThe conversation presented was inspired by real-life disinformation circulating at the time of the study. Nevertheless, all WhatsApp messages and names are fictitious and were created to resemble a realistic exchange.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/9b03020f8e497efa340b7edd.png"},{"id":98865304,"identity":"fc2b707a-7b12-41e4-9d15-25411418f9e1","added_by":"auto","created_at":"2025-12-23 10:43:00","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":168506,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eStudy 1: Belief in the presented disinformation according to group (ThinkFRE vs. control) and correction type (correction with a scientific source, correction without a source, baseline). \u003c/strong\u003eA higher value on the y-axis means greater belief in the disinformation (scale from 1 to 7). Error bars show standard errors of the mean. ** Indicates \u003cem\u003ep\u003c/em\u003e\u0026lt; .01, * Indicates \u003cem\u003ep\u003c/em\u003e \u0026lt; .05.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/f4db73167babd4efc6a64206.png"},{"id":99308901,"identity":"5db50676-c75e-49a1-b07b-4f1fcad4b284","added_by":"auto","created_at":"2025-12-31 16:09:27","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":151632,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eStudy 2: Belief in the presented disinformation according to group (ThinkFRE vs. control) and correction type (correction with a scientific source\u003c/strong\u003e,\u003cstrong\u003ecorrection without a source, baseline). \u003c/strong\u003eA higher value on the y-axis means greater belief in the disinformation (scale 1-7). Error bars show standard error of the mean. *** Indicates \u003cem\u003ep\u003c/em\u003e \u0026lt; .001, * Indicates \u003cem\u003ep\u003c/em\u003e \u0026lt; .05.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/72f570bd167bddc442545a44.png"},{"id":98865312,"identity":"d49541b5-72c3-4b53-a0d5-10b1568fe583","added_by":"auto","created_at":"2025-12-23 10:43:00","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":245886,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eStudy 3: Belief in the presented disinformation according to COVID-19 vaccination attitude (supporters / skeptics) group (ThinkFRE / control) and correction type (baseline / correction without a source / correction with a scientific source). \u003c/strong\u003eA higher value on the y-axis means greater belief in the disinformation, using a composite measure (scale 0-10). Error bars show standard error of the mean. ** indicates \u003cem\u003ep\u003c/em\u003e \u0026lt; .001; * indicates \u0026lt; .05.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/1f80b2f20f7001af4e79058b.png"},{"id":99322627,"identity":"6483cf07-f53a-407d-a8dd-0cb45f31f6ae","added_by":"auto","created_at":"2025-12-31 16:43:51","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2572424,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/2991efff-b40a-4536-9b8d-670ee9c16e73.pdf"},{"id":99308915,"identity":"9d79fe6b-10c9-443c-addc-4f69e6de962a","added_by":"auto","created_at":"2025-12-31 16:09:30","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":901425,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterialsPDF.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8204469/v1/4e2e345afed121a964b4d015.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"When Science Alone Fails to Convince: A Brief Intervention Fostering an Evidence-Oriented Mindset Strengthens Science-Based Corrections","fulltext":[{"header":"Introduction","content":"\u003cp\u003eWe are living in a \u0026quot;post-truth\u0026quot; era, characterized by growing challenges to factual accuracy, widespread mistrust of experts, and a blurring of the line between objective knowledge and personal belief\u003csup\u003e1\u0026ndash;3\u003c/sup\u003e. This erosion\u0026nbsp;of epistemic norms fuels\u0026nbsp;the spread of\u0026nbsp;disinformation\u003csup\u003e4\u003c/sup\u003e. A salient example is the COVID-19 pandemic, which unleashed an \u0026ldquo;infodemic\u0026rdquo; - a rapid spread of misinformation that shaped public perceptions and behaviors, often with harmful effects\u003csup\u003e5,6\u003c/sup\u003e. For instance, exposure to media outlets that downplayed the risks of COVID-19 or spread misleading information has been linked to lower compliance with public health guidelines and greater resistance to preventive measures\u003csup\u003e7\u0026ndash;9\u003c/sup\u003e. As a result, some individuals have rejected expert advice, ignored public health guidelines, and refused vaccination, thereby endangering public health\u003csup\u003e10\u003c/sup\u003e. In response, scholars have emphasized the critical role of the social and behavioral sciences in combating the spread of the virus, particularly in understanding and countering misinformation\u003csup\u003e11\u003c/sup\u003e. Moreover, public trust in science and scientists has emerged as a key factor in navigating global crises such as the COVID-19 pandemic, as it strongly predicts adherence to preventive measures\u003csup\u003e12,13\u003c/sup\u003e. Sustaining this trust is vital not only for managing current threats but also for addressing future global challenges, as illustrated by the recent resurgence of measles, a disease that can be effectively prevented through vaccination but has reemerged due to misinformation and declining confidence in vaccines\u003csup\u003e14,15\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eT\u003c/strong\u003e\u003cstrong\u003erust\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;and Distrust\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;in Science\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCross-national surveys suggest that public trust in many public bodies, such as institutions and governments, has declined since 2007\u003csup\u003e16\u003c/sup\u003e. However, most people continue to express a relatively high level of trust in science and scientists\u003csup\u003e17,18\u003c/sup\u003e. In the initial phase of the COVID-19 pandemic, surveys even indicated an increase in public trust in science\u003csup\u003e19,20\u003c/sup\u003e. This\u0026nbsp;is noteworthy given\u0026nbsp;that prior epidemics\u0026nbsp;tended to be\u0026nbsp;associated with a decline in trust toward scientists and their work\u003csup\u003e21\u003c/sup\u003e.\u0026nbsp;Moving beyond self-report surveys, a 2022 cross-cultural study\u003csup\u003e22\u003c/sup\u003e introduced the \u0026ldquo;Einstein Effect\u0026rdquo;, showing that scientific sources are generally perceived as more credible than spiritual gurus, though this effect was demonstrated in response to meaningless content (i.e., \u0026ldquo;pseudo-profound bullshit\u0026rdquo;\u003csup\u003e23\u003c/sup\u003e). Despite such findings, a dominant public narrative persists, suggesting a widespread crisis of trust in science and scientists. This disconnect between\u0026nbsp;the (limited)\u0026nbsp;empirical evidence and the dominant narrative raises important questions about how trust in science operates in real-world contexts. In particular, the present work examines trust in science when scientific sources are used to correct misinformation, as in the case of COVID-19.\u003c/p\u003e\n\u003cp\u003eOne distinction that may help explain the apparent divergence is between trust in the abstract sense\u0026mdash;as measured via generic surveys\u0026mdash;and \u003cem\u003eenacted trust\u003c/em\u003e, that is,\u0026nbsp;whether\u0026nbsp;or not\u0026nbsp;people accept\u0026nbsp;information from trustworthy sources\u0026nbsp;in a\u0026nbsp;noisy, politicized, and\u0026nbsp;polarized\u0026nbsp;environment.\u0026nbsp;Focusing on enacted trust, Iyengar and Massey\u003csup\u003e24\u003c/sup\u003e argue that today\u0026rsquo;s widespread dissemination of misleading and biased content contributes to growing distrust in the scientific enterprise. They claim that structural shifts in the media landscape have made scientific findings more vulnerable to distortion, especially when they threaten ideological worldviews.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eRoozenbeek et al.\u003csup\u003e25\u003c/sup\u003e demonstrate that across multiple countries, participants who reported higher trust in scientists were less likely to believe in false information related to COVID-19.\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003eWest and Bergstrom\u003csup\u003e26\u003c/sup\u003e further highlight that both the dissemination of false information within the scientific community and the misrepresentation of scientific results to the public can erode confidence in scientific institutions.\u003c/p\u003e\n\u003cp\u003eBuilding on these findings, the current research empirically examines trust in scientific sources in the applied context of correcting disinformation, in light of claims that the post-truth era has led to a neglect of facts and expertise in favor of subjectively confirmatory information. We also test a brief intervention designed to highlight the distinction between objective and subjective content, and between experts and lay perspectives, with the goal of restoring trust in factual content and expert authority.\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorrecting disinformation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA wide range of interventions grounded in psychological theory have been developed to mitigate the spread and influence of misinformation, and many have received substantial empirical support\u003csup\u003e27\u003c/sup\u003e.\u0026nbsp;Fact-checking labels are commonly used, but their impact tends to be small and they carry the risk that\u0026nbsp;unflagged content may appear more credible if readers assume it has been verified\u003csup\u003e28\u003c/sup\u003e.\u0026nbsp;Moreover, fact-checks rarely provide individuals with the critical thinking skills needed to evaluate claims independently, especially in the absence of verification cues\u003csup\u003e29\u003c/sup\u003e. More detailed refutations tend to reliably reduce false beliefs but they can only be applied retrospectively to target specific pieces of misinformation\u003csup\u003e30\u0026ndash;33\u003c/sup\u003e. Another common approach involves \u003cem\u003enudging\u003c/em\u003e individuals toward accurate information by presenting factual content in a way that facilitates its ease of processing\u003csup\u003e1,34\u003c/sup\u003e. Subtle accuracy reminders can also reduce\u0026nbsp;sharing of misinformation on social media\u003csup\u003e35\u0026ndash;39\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003ePre-emptive warnings about misinformation\u0026nbsp;can also provide some protection but only if they are very explicit and specific\u003csup\u003e40\u0026ndash;42\u003c/sup\u003e.\u0026nbsp;A more\u0026nbsp;involved proactive approach is inoculation, which combines\u0026nbsp;warnings\u0026nbsp;with an educational component that provides relevant counterarguments, explains misleading argumentation techniques, or discredits sources of low-quality information\u003csup\u003e43\u003c/sup\u003e; such inoculations can build resistance to misinformation by raising awareness and skepticism before exposure\u003csup\u003e44\u0026ndash;46\u003c/sup\u003e.\u0026nbsp;For instance, forewarning individuals about politically motivated disinformation has been shown to reduce\u0026nbsp;susceptibility and promote accurate perceptions of scientific consensus\u003csup\u003e47\u003c/sup\u003e.\u0026nbsp;Such interventions can be incorporated in engaging educational interventions (i.e., courses and games) designed to improve people\u0026rsquo;s information literacy\u0026nbsp;by teaching\u0026nbsp;them how misinformation\u0026nbsp;is constructed\u003csup\u003e48,49\u003c/sup\u003e.\u0026nbsp;Other interventions focused on boosting\u0026nbsp;media literacy\u0026nbsp;are also able to\u0026nbsp;enhance truth discernment\u003csup\u003e50\u003c/sup\u003e.\u0026nbsp;One challenge faced by these interventions, however, is that they\u0026nbsp;require\u0026nbsp;substantial\u0026nbsp;motivation, engagement,\u0026nbsp;and\u0026nbsp;cognitive\u0026nbsp;resources, which may limit their applicability.\u0026nbsp;Recent\u0026nbsp;work has thus\u0026nbsp;emphasized the need for scalable strategies for social media environments\u003csup\u003e38,51\u003c/sup\u003e. \u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe Suggested S\u003c/strong\u003e\u003cspan dir=\"RTL\"\u003et\u003c/span\u003e\u003cstrong\u003erategy\u003c/strong\u003e\u003cstrong\u003e: ThinkFRE\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConsidering the strengths and limitations of existing strategies and acknowledging that individuals are often unreceptive to factual corrections and expert advice, we propose a novel pre-emptive strategy. The suggested approach departs from the traditional emphasis on truth and accuracy evaluations that may be influenced by prior knowledge, beliefs, or political attitudes. It also goes beyond general scientific reasoning interventions\u003csup\u003e52\u003c/sup\u003e by specifically addressing disinformation through a focus on the distinction between (1) objective versus subjective information that is (2) falsifiable versus non-falsifiable, and comes from (3) expert versus layperson sources. These three basic yet powerful distinctions, while ideally neutral and widely acceptable across ideological divides, have become increasingly blurred in today\u0026rsquo;s information environment. The aim of the proposed strategy is to restore attention to these foundational categories, offering an accessible and broadly acceptable pathway for resisting disinformation.\u003c/p\u003e\n\u003cp\u003eWe term this new strategy \u003cstrong\u003eThinkFRE\u003c/strong\u003e (see Fig. 1). ThinkFRE is a self-guided tool that\u003cspan dir=\"RTL\"\u003e,\u003c/span\u003e according to our hypothesis\u003cspan dir=\"RTL\"\u003e,\u003c/span\u003e will increase reliance on factual information and expert knowledge through three simple core prompts: (1) \u0026ldquo;\u003cstrong\u003eFact\u003c/strong\u003e: Is the information\u0026nbsp;based on evidence or personal anecdote?\u0026rdquo; This encourages the reader to consider whether\u0026nbsp;information\u0026nbsp;is objective or subjective. (2) \u0026ldquo;\u003cstrong\u003eRefute\u003c/strong\u003e: Can you think of a way to disprove the information presented?\u0026rdquo; This question prompts critical evaluation and reinforcing that only objective claims can be disproven. (3) \u0026ldquo;\u003cstrong\u003eExpert\u003c/strong\u003e: Does the source have the necessary knowledge?\u0026rdquo; This encourages scrutiny of the information\u0026rsquo;s origin and highlights the difference between expert and non-expert sources.\u003c/p\u003e\n\u003cp\u003eThe intervention was introduced to participants as \u0026ldquo;ThinkFRE\u0026rdquo;; the \u0026ldquo;FRE\u0026rdquo; acronym was hoped to evoke an association with free and independent thought; omission of the second \u0026ldquo;E\u0026rdquo; was intentional, as it may subtly disrupt automatic reading and encourage deeper cognitive engagement. In sum, \u003cstrong\u003eThinkFRE\u003c/strong\u003e offers a concise, general-purpose strategy for evaluating information without requiring users to judge what is true or false. Instead, it guides individuals to identify whether information is objective or subjective, falsifiable, and whether it comes from a reliable expert or not. Unlike content-specific fact-checking tools or cognitively demanding interventions, ThinkFRE is short, teachable, and easy to apply in everyday contexts. ThinkFRE was predicted to induce a sense of general healthy skepticism and self-efficacy that goes beyond specific content\u003csup\u003e53\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe Current Research\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe current research included three studies examining the potential effects of disinformation corrections from scientific sources, as well as the ThinkFRE strategy as a possible intervention. The studies focused on COVID-19 vaccination disinformation relevant to everyday life and were embedded within a realistic setting\u0026mdash;a WhatsApp group conversation\u0026mdash;to capture how people encounter and respond to misinformation in daily social interactions.\u003c/p\u003e\n\u003cp\u003eStudy 1 examined this question in the context of disinformation claiming that the COVID-19 vaccine causes infertility in women. Study 2 built on this by adding a general warning about source unreliability, allowing us to examine whether the ThinkFRE strategy provides added value even when participants are already alerted to the potential inaccuracy of the disinformation. Studies 1 and 2 were conducted in January 2021, amid the uncertainty of the early phase of the COVID-19 pandemic, just after the vaccine campaign had begun, when individuals were making behavioral decisions about whether to get vaccinated (or not). Study 3 replicated the core design in a more stable context after COVID-19 was no longer classified as a public health emergency of international concern by the WHO, yet it remained a pandemic-level disease (June 2024); it addressed disinformation concerning the vaccine\u0026rsquo;s long-term consequences. This setting of Study 3 allowed us to distinguish between participants with positive versus negative baseline attitudes, focusing specifically on vaccine skeptics, and use a more robust multi-item belief scale.\u003c/p\u003e"},{"header":"Method","content":"\u003cp\u003eAll three studies were preregistered and approved by the Hebrew University Review Board (Studies 1\u0026ndash;2: 11101_2021_IRB; Study 3: 048_2024_IRB). All methods were performed in accordance with the relevant guidelines and regulations. All participants provided informed consent at the beginning of the survey and could withdraw at any time.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy 1: Design\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eand Procedure\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConducted in January 2021, at the onset of the COVID-19 vaccine campaign and amidst widespread public uncertainty\u003ca href=\"#_ftn1\" name=\"_ftnref1\" title=\"\"\u003e\u003c/a\u003e\u003csup\u003e1\u003c/sup\u003e, Study 1 focused on a viral piece of disinformation circulating at that time: the claim that the COVID-19 vaccine causes female sterility. The study followed a 2 (group: ThinkFRE vs. control) \u0026times; 3 (correction type: scientific sources vs. no source vs. no-correction baseline) between-subjects design.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eParticipants were randomly assigned to one of two groups (ThinkFRE vs. control). All participants were instructed that they would be presented with information found on social media and that they should read it carefully. Participants in the ThinkFRE group were introduced to the ThinkFRE strategy and instructed to apply it (see Fig. 1). In the control group, participants received no strategic instruction. Subsequently, all participants were shown a simulated WhatsApp conversation (see Fig. 2) containing the disinformation, followed by one of three correction conditions: (1) a correction citing official scientific sources (WHO, FDA, CDC) and clinical trial data; (2) a correction with identical content but no scientific references; or (3) a no-correction baseline where the disinformation was left unchallenged. Immediately following this exposure, we measured participants\u0026rsquo; belief that the COVID-19 vaccine could cause sterility, our primary dependent variable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy 1: Participants\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe required sample size was calculated using G*Power 3.1, with the aim to detect an interaction in an omnibus test of a 2 \u0026times; 3 between-subjects ANOVA. Expecting a small to medium effect size of \u003cimg width=\"51\" height=\"19\" src=\"https://myfiles.space/user_files/58895_8739fc6c57c1c19a/58895_custom_files/img1766485682.png\" alt=\"image\"\u003e for the interaction effect at power = .80 and \u0026alpha; = .05, the suggested minimum sample size was 244. To allow exclusions, 302 participants from the Prolific online platform were recruited. Participants were US residents aged 18\u0026ndash;40, native English speakers, and had a Prolific approval rate of at least 85%. According to preregistered criteria, we excluded 15 participants who either failed the attention check (\u003cem\u003en\u003c/em\u003e = 2), did not complete the survey in one sitting (\u003cem\u003en\u003c/em\u003e = 1), reported external disturbances (\u003cem\u003en\u003c/em\u003e = 5), reported that they completed the survey with someone else (\u003cem\u003en\u003c/em\u003e = 1), or had a completion time of more than 3 \u003cem\u003eSD\u003c/em\u003e above the mean (\u003cem\u003eM\u003csub\u003etime\u003c/sub\u003e\u003c/em\u003e = 6.40 min, \u003cem\u003eSD\u003csub\u003etime\u003c/sub\u003e\u003c/em\u003e = 4.19, \u003cem\u003en\u003c/em\u003e = 6). The final sample consisted of \u003cem\u003eN\u003c/em\u003e = 287 participants (\u003cem\u003eM\u003c/em\u003e\u003csub\u003eage\u003c/sub\u003e = 27.72 years, \u003cem\u003eSD\u003c/em\u003e\u003csub\u003eage\u003c/sub\u003e = 6.51; 145 men, 137 women, and 5 participants who preferred not to state their gender); there were 141 participants in the ThinkFRE group and 146 in the control group.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy 1: Stimuli\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDisinformation:\u0026nbsp;\u003c/em\u003eThe disinformation and its correction were presented in a WhatsApp Group Chat format (see Fig. 2). The conversation started with a member forwarding an online article indicating in the title that the COVID-19 vaccine\u003ca href=\"#_ftn2\" name=\"_ftnref2\" title=\"\"\u003e\u003c/a\u003e\u003csup\u003e2\u003c/sup\u003e causes sterility in 97% of women. Then, the same person who sent the link supports this claim and adds that Bill Gates, a big supporter of vaccination, did not hide his conviction that there should be a decline in the population. This combination of a specific health claim with a conspiratorial motive was designed to create a single, compelling disinformation narrative.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCorrection Type:\u0026nbsp;\u003c/em\u003eIn the two conditions with a correction, the correction was added by another chat-group member who replied that the vaccine is safe and effective and that its potential benefits outweigh the known and potential harms of becoming infected with COVID-19. In the correction condition with a scientific source, the same message was sent, but with an attribution to scientific sources, namely the Food and Drug Administration (FDA), Centers for Disease Control and Prevention (CDC), and World Health Organization (WHO), additionally emphasizing that the vaccine was declared safe and effective based on data and findings from large scientific trials (see Fig. 2, and for the entire stimuli see Fig. S1 in the supplementary material).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy 1: Measurement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAfter being exposed to the display of the WhatsApp conversation, all participants were asked: \u0026ldquo;In your opinion, how likely is it that the COVID-19 vaccine can cause sterility?\u0026rdquo; The question was answered on a 7-point Likert scale ranging from \u0026ldquo;extremely unlikely\u0026rdquo; to \u0026ldquo;extremely likely.\u0026rdquo; This was our main dependent variable, which we refer to as \u0026ldquo;belief in presented disinformation\u0026rdquo;\u003ca href=\"#_ftn3\" name=\"_ftnref3\" title=\"\"\u003e\u003c/a\u003e\u003csup\u003e3\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy 2: Design\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eand Procedure\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStudy 2 was conducted in the same context as Study 1, aiming to replicate it\u0026nbsp;while introducing an additional element: a general warning about the credibility of the source that initially shared the disinformation.\u0026nbsp;This warning was presented to participants across all conditions. Studies 1 and 2 were conducted within a few days of each other,\u0026nbsp;ensuring that participants\u0026rsquo; vaccination status and the broader public discourse around COVID-19 vaccines remained consistent across both studies. The study design and procedure were\u0026nbsp;largely\u0026nbsp;identical to Study 1. Two minor changes were made to the stimuli to make the presentation more realistic:\u0026nbsp;(1)\u0026nbsp;the forwarded article included a small thumbnail image of a graph, and (2)\u0026nbsp;the screenshot showed the entire WhatsApp interface, including the message composition area at the bottom, which had been cropped out in Study 1.\u0026nbsp;These changes\u0026nbsp;increased the ecological validity of the conversation display. More importantly, Study 2 added\u0026nbsp;a short general warning\u0026nbsp;to all conditions\u0026nbsp;(see Fig.\u0026nbsp;S3 in the supplementary material).\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003eIn Study 1, only the ThinkFRE group was told in the instructions that they would be introduced to a method that enables people to distinguish between factual and false information, whereas the control group was only told to read the information carefully. Thus, one may claim that the ThinkFRE group was more explicitly warned regarding the risk of false information. Therefore, in Study 2, a warning was presented emphasizing the unreliability of the source for all participants. The warning came from a group member reacting with dismay to the disinformation (\u0026ldquo;Can we please limit such questionable and unconfirmed sources?\u0026rdquo;). In response, the person who forwarded the link asks how the other person can be certain that anything they read is true and confirmed, and why the article they shared is more questionable. \u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy 2:\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eParticipants\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFollowing the same considerations as in Study 1, the initial sample consisted of 300 participants from the Prolific online platform. Participants were US residents aged 18\u0026ndash;40, native English speakers, and had a Prolific approval rate of at least 85%. According to preregistered criteria, we excluded 14 participants who either failed the attention check (\u003cem\u003en\u003c/em\u003e = 1), did not complete the questionnaire\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003ein one sitting (\u003cem\u003en\u003c/em\u003e = 3), reported on external disturbances (\u003cem\u003en\u003c/em\u003e = 3), or had a completion time of more than 3 \u003cem\u003eSD\u003c/em\u003e above the mean (\u003cem\u003eM\u003csub\u003etime\u003c/sub\u003e\u003c/em\u003e = 5.92 min, \u003cem\u003eSD\u003csub\u003etime\u003c/sub\u003e\u003c/em\u003e = 3,98, \u003cem\u003en\u003c/em\u003e = 6). This resulted in a final sample of \u003cem\u003eN\u003c/em\u003e = 287 participants (\u003cem\u003eM\u003csub\u003eage\u003c/sub\u003e\u003c/em\u003e = 27.12 years, \u003cem\u003eSD\u003csub\u003eage\u003c/sub\u003e\u003c/em\u003e = 6.19; 145 men, 137 women, 5 who preferred not to state their gender): 145 participants in the ThinkFRE condition and 142 participants in the control condition.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy 3: Design\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eand Procedure\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUnlike the initial studies conducted at the pandemic\u0026rsquo;s peak in 2021, Study 3 took place in June 2024, when most people held established attitudes about COVID-19 vaccination and the acute phase of the pandemic had passed. This updated setting enabled a more stringent test of our hypotheses, as we could now assess the impact of science-based corrections and the ThinkFRE strategy specifically on vaccine skeptics\u0026mdash;the group considered most susceptible to disinformation.\u003c/p\u003e\n\u003cp\u003eTo align with this focus, the study design incorporated two key updates from the previous studies. First, while the general theme of the COVID-19 vaccine was retained, the disinformation stimulus was revised to address contemporary concerns about the vaccine\u0026rsquo;s long-term side effects, a topic more salient at this later stage of the pandemic. Second, to enable a direct comparison based on vaccine attitudes, participants were pre-screened and placed into one of two groups: those with generally positive or generally negative attitudes. This resulted in a design with 3 between-subjects factors: 2 (vaccination attitude: supporters vs. skeptics) \u0026times; 2 (group: ThinkFRE vs. Control) \u0026times; 3 (correction type: scientific sources vs. no source vs. no-correction baseline). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy 3:\u003c/strong\u003e\u003cem\u003e\u003cspan dir=\"RTL\"\u003e \u003c/span\u003e\u003c/em\u003e\u003cstrong\u003eParticipants\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo boost power, a larger sample was collected. The initial sample consisted of \u003cspan dir=\"RTL\"\u003e1197\u003c/span\u003e participants from the Prolific online platform. The participants were US residents aged 18\u0026ndash;\u003cspan dir=\"RTL\"\u003e5\u003c/span\u003e0, native English speakers with a Prolific approval rate of at least 85%. We recruited participants according to their reported attitude regarding the COVID-19 vaccine: Half of the participants reported in the Prolific prescreening that they were in favor of the COVID-19 vaccine\u003cspan dir=\"RTL\"\u003e,\u003c/span\u003e and half of the participants reported that they were against the vaccine. According to preregistered criteria, we excluded participants who reported that they did not complete the survey alone (\u003cem\u003en\u003c/em\u003e = 2), reported on external disturbances (\u003cem\u003en\u003c/em\u003e = 18), did not complete the study in one sitting (\u003cem\u003en\u003c/em\u003e = 24), admitted to not have put reasonable effort in the study (\u003cem\u003en\u003c/em\u003e = 7), or reported that they looked up information online (\u003cem\u003en\u003c/em\u003e = 14). We further excluded participants who reported a different vaccination attitude than indicated by the prescreening (\u003cem\u003en\u003c/em\u003e = 127)\u003ca href=\"#_ftn4\" name=\"_ftnref4\" title=\"\"\u003e\u003c/a\u003e\u003csup\u003e4\u003c/sup\u003e or had a completion time longer than 3 \u003cem\u003eSD\u003c/em\u003e above the average time (\u003cem\u003en\u003c/em\u003e = 30). After these exclusions, the final sample size was \u003cem\u003eN\u003c/em\u003e = 975 participants (\u003cem\u003eM\u003csub\u003eage\u003c/sub\u003e\u003c/em\u003e = 34.80 years, \u003cem\u003eSD\u003csub\u003eage\u003c/sub\u003e\u003c/em\u003e = 8.19; 422 men, 528 women, 25 who identified as non-binary/third-gender): 484 participants in the control condition and 491 in the ThinkFRE condition; 515 COVID-19 vaccine supporters and 460 COVID-19 vaccine skeptics.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy 3:\u003c/strong\u003e\u003cem\u003e\u003cspan dir=\"RTL\"\u003e \u003c/span\u003e\u003c/em\u003e\u003cstrong\u003eStimuli\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDisinformation\u003c/em\u003e\u003cem\u003e:\u003c/em\u003e As in Studies 1 and 2, the disinformation and its correction were embedded within a simulated WhatsApp Group Chat (see Fig. S4 in the Supplementary Information). The conversation started with a member forwarding an online article whose headline claimed that people were experiencing sudden heart attacks after receiving the COVID-19 vaccine. When another group member responded by urging caution and requesting that unverified claims not be shared, the original sender defended and added a personal anecdote about\u003cspan dir=\"RTL\"\u003e\u0026nbsp;\u003c/span\u003ea\u0026nbsp;young,\u0026nbsp;previously healthy man who allegedly suffered a heart attack following vaccination.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCorrection Type:\u003c/em\u003e As in Studies 1 and 2, there were three different conditions of correction: (1) no correction (baseline), (2) correction referencing scientific sources and data (the CDC, the American Heart Association, and the WHO), or (3) the same correction without any scientific reference (see Fig. 3). In the two conditions with a correction, the same group member who initially urged caution responded again, stating that there is no established link between COVID-19 vaccines and heart attacks. The correction clarified that although the vaccine can, in rare cases, cause heart inflammation, this is an extremely uncommon side effect. Importantly, the correction highlights that the risk of heart complications following COVID-19 infection is substantially higher than the risk associated with the vaccine.\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy 3: Measurement\u003c/strong\u003e\u003cstrong\u003es\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn addition to the primary dependent variable assessing direct belief in the disinformation claim (i.e., that the COVID-19 vaccine causes random heart attacks), Study 3 included five supplementary items designed to capture belief in the disinformation more indirectly (see Table 1 for all items). To improve upon the single-item belief measure used in Studies 1 and 2, the present study\u0026nbsp;thus implemented a more robust multi-item composite, thereby addressing known issues of reliability and spurious effects\u0026nbsp;with single-item measures (Swire-Thompson et al., 2020;\u0026nbsp;2022). Furthermore, we also assessed participants\u0026rsquo; general attitudes toward the COVID-19 vaccine, as well as their trust in science and scientists, and their level of intellectual humility.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1. Measured items assessing the belief in the presented disinformation, correcti\u003c/strong\u003e\u003cstrong\u003eve\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;information\u003c/strong\u003e\u003cstrong\u003e,\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;and distinction between the two.\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cimg width=\"597\" height=\"348\" src=\"https://myfiles.space/user_files/58895_8739fc6c57c1c19a/58895_custom_files/img1766485683.png\" alt=\"A table with text on it AI-generated content may be incorrect.\"\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSpecifically, items #1, #3, and #6 refer directly to the connection between COVID-19 vaccine and heart disease (i.e., disinformation). Items #2 and #5 pertain to the COVID-19 virus and heart disease (i.e., corrective information). Item #4 directly contrasts the belief in the potential consequences of the vaccine and the virus.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eResults\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eS\u003c/strong\u003e\u003cstrong\u003etudy 1:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFirst\u003cstrong\u003e\u003cspan dir=\"RTL\"\u003e,\u003c/span\u003e\u003c/strong\u003e as an omnibus analysis, we conducted a two-way ANOVA with a 2 (group: ThinkFRE/Control) \u0026times; 3 (correction type: correction with scientific source/correction with no source/baseline) design. The findings showed that participants in the ThinkFRE group did not show a significantly lower belief that the vaccine can cause sterility (\u003cem\u003eM\u003c/em\u003e = 2.04, 95% CI [1.82, 2.27]) compared to control participants (\u003cem\u003eM\u003c/em\u003e = 2.34, 95% CI [2.08, 2.60]), \u003cem\u003eF\u003c/em\u003e(1, 281) = 3.0\u003cspan dir=\"RTL\"\u003e7\u003c/span\u003e, \u003cem\u003ep\u003c/em\u003e = .08\u003cspan dir=\"RTL\"\u003e1\u003c/span\u003e. There was no main effect of correction type, \u003cem\u003eF\u003c/em\u003e(2, 281) = 1.46, \u003cem\u003ep\u003c/em\u003e = .234. The interaction between group (ThinkFRE vs. control) and correction type was also nonsignificant, \u003cem\u003eF\u003c/em\u003e(2, 281) = 2.78, \u003cem\u003ep\u003c/em\u003e = .064.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eComparison of ThinkFRE and Control Across Different Correction Types\u003c/u\u003e: Despite the nonsignificant omnibus results, to directly test specific a-priori hypotheses regarding the potential effect of ThinkFRE when the correction is assigned to scientific sources, we ran specific planned contrasts. The results showed that in the condition with a scientific-source correction, participants who learned ThinkFRE believed less in the presented disinformation (\u003cem\u003eM\u003c/em\u003e = 1.82, 95% CI [1.45, 2.18]) compared to control participants (\u003cem\u003eM\u003c/em\u003e = 2.71, 95% CI [2.23, 3.18]), \u003cem\u003et\u003c/em\u003e(281) = 2.93, \u003cem\u003ep\u003c/em\u003e = .004, Cohen\u0026rsquo;s \u003cem\u003ed\u003c/em\u003e = 0.61, 95% CI [0.20, 1.03]) \u0026nbsp;(Fig. 3). Participants who learned the ThinkFRE paradigm were not different in their belief in the presented disinformation from control participants in the baseline condition (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e= .95), nor when receiving a correction without a source (\u003cem\u003ep =\u0026nbsp;\u003c/em\u003e.82). \u0026nbsp;These two null effects suggest that learning ThinkFRE did not have a general social-desirability effect in the sense of leading participants to report lower belief in the disinformation if it appeared alone or with a correction message without a scientific reference.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eComparison of Correction Types Within Each Group:\u003c/u\u003e We also ran specific contrasts to test the hypothesis that a scientific-source correction can reduce belief in the disinformation compared to a correction without a source and the baseline condition. In the control group, belief in the disinformation was higher when the correction included a scientific reference (\u003cem\u003eM\u003c/em\u003e = 2.71, 95% CI [2.23, 3.18]) compared to the baseline condition (\u003cem\u003eM\u003c/em\u003e = 1.98, 95% CI [1.58, 2.38]; \u003cem\u003et\u003c/em\u003e(281) = 2.46, \u0026nbsp;\u003cem\u003ep\u0026nbsp;\u003c/em\u003e= .014, Cohen\u0026rsquo;s \u003cem\u003ed\u003c/em\u003e = 0.50, 95% CI [0.10, 0.90]), suggesting a backfire effect of a scientific correction. However, among control participants, those who received a scientific correction were not different in their belief in the disinformation from those presented with a correction without a source (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e= .22). Importantly, no backfire effect of a scientific correction was observed in the ThinkFRE group (\u003cem\u003ep\u003c/em\u003e = .55).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eS\u003c/strong\u003e\u003cstrong\u003etudy 2\u003c/strong\u003e\u003cspan dir=\"RTL\"\u003e:\u003c/span\u003e\u003c/p\u003e\n\u003cp\u003eAs in Study 1, we conducted an initial omnibus two-way ANOVA with a 2 (group: ThinkFRE/Control) \u0026times; 3 (correction type: correction with scientific source/correction with no source/baseline) design. We found a main effect of group, indicating that participants who learned ThinkFRE believed less that the vaccine might cause sterility (\u003cem\u003eM\u003c/em\u003e = 1.86, 95% CI [1.68, 2.05]) compared to those in the control group (\u003cem\u003eM\u003c/em\u003e = 2.36, 95% CI [2.09, 2.63]), \u003cem\u003eF\u003c/em\u003e(1, 281) = 9.25, \u003cem\u003ep\u003c/em\u003e = .003, \u0026eta;\u003csub\u003ep\u003c/sub\u003e\u003csup\u003e2\u003c/sup\u003e = 0.03, 95% CI [0.004, 0.08]. There was no significant main effect of correction type, \u003cem\u003eF\u003c/em\u003e(2, 281) = 2.97, \u003cem\u003ep\u003c/em\u003e = .053. However, there was a significant interaction, suggesting that the effect of correction type depended on the group, \u003cem\u003eF\u003c/em\u003e(2, 281) = 3.99, \u003cem\u003ep\u003c/em\u003e = .02, \u0026eta;\u003csub\u003ep\u003c/sub\u003e\u003csup\u003e2\u003c/sup\u003e = 0.03, 95% CI [0.0005, 0.07].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eComparison of ThinkFRE and Control Across Correction Types:\u003c/u\u003e Analyzing specific planned contrasts to directly test our hypotheses, we see again that the effect of ThinkFRE was not a result of a general desirability effect. Consistent with Study 1, only in the scientific-correction condition did control participants exhibit significantly greater belief in the disinformation (\u003cem\u003eM\u003c/em\u003e = 2.90, 95% CI [2.36, 3.43]) compared to participants who learned ThinkFRE (\u003cem\u003eM\u003c/em\u003e = 1.75, 95% CI [1.45, 2.05]), \u003cem\u003et\u003c/em\u003e(281) = 4.05, \u003cem\u003ep\u003c/em\u003e \u0026lt; .001, Cohen\u0026rsquo;s \u003cem\u003ed\u003c/em\u003e = 0.83, 95% CI [0.42, 1.23]. There was no difference between control participants and ThinkFRE participants in their belief in the disinformation when presented with a correction without a source (\u003cem\u003ep\u003c/em\u003e \u003cspan dir=\"RTL\"\u003e=\u003c/span\u003e .\u003cspan dir=\"RTL\"\u003e40\u003c/span\u003e) or no correction at all (\u003cem\u003ep\u003c/em\u003e \u003cspan dir=\"RTL\"\u003e=\u003c/span\u003e .\u003cspan dir=\"RTL\"\u003e71\u003c/span\u003e). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eComparison of Correction Types Within Each Group:\u003c/u\u003e As in Study 1, we tested the hypothesis of a potential backfire effect of a scientific reference when correcting disinformation. In the control group, participants believed the disinformation significantly more after reading a correction with a scientific source (\u003cem\u003eM\u003c/em\u003e = 2.90, 95% CI [2.36, 3.43) compared to the baseline condition (\u003cem\u003eM\u003c/em\u003e = 1.90, 95% CI [1.55, 2.25]), \u003cem\u003et\u003c/em\u003e(281) = 3.53, \u003cem\u003ep\u003c/em\u003e \u0026lt; .001, Cohen\u0026rsquo;s \u003cem\u003ed\u003c/em\u003e = 0.72, 95% CI [0.31, 1.13]) and compared to the correction without a source (\u003cem\u003eM\u003c/em\u003e = 2.28, 95% CI [1.78, 2.79]), \u003cem\u003et\u003c/em\u003e(281) = 2.14, \u003cem\u003ep\u003c/em\u003e = .03, Cohen\u0026rsquo;s \u003cem\u003ed\u003c/em\u003e = 0.44, 95% CI [0.03, 0.85]), thereby demonstrating a backfire effect of scientific sources (Fig. 4). This pattern was absent in the ThinkFRE group.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eS\u003c/strong\u003e\u003cstrong\u003etudy 3\u003c/strong\u003e\u003cspan dir=\"RTL\"\u003e:\u003c/span\u003e\u003c/p\u003e\n\u003cp\u003eThe results section will focus on the multi-item\u0026nbsp;compound measure, designed to capture a more general belief in the presented disinformation. Out of the six items, we combined the four items probing the perceived influence of the COVID-19 vaccine (i.e., directly connected to the belief in the disinformation): belief in a causal link between the vaccine and heart attacks (item 1; \u0026ldquo;\u003cem\u003eIn your opinion, how likely is it that there is any causal connection between the \u003cstrong\u003eCOVID-19 vaccine\u0026nbsp;\u003c/strong\u003eand \u003cstrong\u003eheart attacks\u003c/strong\u003e?\u003c/em\u003e\u0026rdquo;), belief in a causal link between the vaccine and heart inflammation (item 3; \u0026ldquo;\u003cem\u003eIn your opinion, how likely is it that there is any causal connection between the \u003cstrong\u003eCOVID-19 vaccine\u0026nbsp;\u003c/strong\u003eand \u003cstrong\u003eheart inflammation\u003c/strong\u003e?\u003c/em\u003e\u0026rdquo;), the bipolar cause-attribution item regarding heart inflammation (item 4;\u0026nbsp;\u0026ldquo;\u003cem\u003ePlease move the circle along the slider to indicate, in your opinion, the more common cause for \u003cstrong\u003eheart inflammation\u003c/strong\u003e. (You can place the circle anywhere on the scale between \u0026quot;COVID-19 Vaccine\u0026quot; and \u0026quot;COVID-19 Virus)\u0026quot;,\u003c/em\u003e reverse-coded to align with the other items), and speculation about cases of heart attack or heart inflammation following vaccination (item 6; \u0026ldquo;\u003cem\u003eIn your opinion, out of 1000 healthy people below the age of 60 who got the \u003cstrong\u003eCOVID-19 vaccine\u003c/strong\u003e, how many will subsequently suffer a heart attack or heart inflammation? The scale ranges from 0 (0% of the population, i.e. no-one) to 100 or more people (10% or more of the population)\u003c/em\u003e\u0026rdquo;). These four items (Cronbach\u0026rsquo;s \u0026alpha; = .89) were combined\u0026nbsp;into a\u0026nbsp;composite score\u0026nbsp;on a\u0026nbsp;0-10\u0026nbsp;scale (see Supplementary Information for\u0026nbsp;item\u0026nbsp;correlations). The results for the single-item measure of belief in the disinformation (item #1) and the bipolar cause-attribution item (item #4)\u0026nbsp;are\u0026nbsp;reported separately in the Supplementary Information.\u003c/p\u003e\n\u003cp\u003eOur primary\u0026nbsp;focus was on\u0026nbsp;participants who were skeptical of the COVID-19 vaccine,\u0026nbsp;as they\u0026nbsp;were presumed to be more susceptible to endorsing disinformation. In contrast,\u0026nbsp;corrections were expected to have a\u0026nbsp;limited impact among vaccine supporters, who were less likely to believe the false claim to begin with. Accordingly, we\u0026nbsp;analyzed the effects of scientific references in corrections and the ThinkFRE intervention separately for COVID-19 vaccine skeptics and COVID-19 vaccine supporters (for the sake of completeness, results of a three-way ANOVA are reported in the Supplementary Information).\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCOVID-19 Vaccine Skeptics\u003c/em\u003e:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThere was a strong effect of group, indicating that ThinkFRE participants generally believed less in the claimed link between the COVID-19 vaccine and heart disease presented in the disinformation (\u003cem\u003eM\u003c/em\u003e = 6.31, 95% CI [6.06, 6.56]), compared to control participants (\u003cem\u003eM\u003c/em\u003e = 6.91, 95% CI [6.69, 7.13]), \u003cem\u003eF\u003c/em\u003e(1,454) = 12.69, \u003cem\u003ep\u003c/em\u003e \u0026lt; .001, \u0026eta;\u003csub\u003ep\u003c/sub\u003e\u003csup\u003e2\u003c/sup\u003e =0.03, 95% CI [0.006, 0.06]. There was no significant main effect of correction type (\u003cem\u003ep\u003c/em\u003e = .28) and no significant interaction (\u003cem\u003ep\u003c/em\u003e = .42) among COVID-19 vaccine skeptics. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eComparison of ThinkFRE and Control Across Correction Types:\u003c/u\u003e We conducted specific contrasts according to our a priori hypothesis. Replicating the results of Studies 1 and 2, COVID-19 vaccine skeptics who used ThinkFRE had lower general disinformation belief in the scientific-correction condition (\u003cem\u003eM\u003c/em\u003e = 6.01, 95% CI [5.5\u003cspan dir=\"RTL\"\u003e4\u003c/span\u003e, 6.4\u003cspan dir=\"RTL\"\u003e7\u003c/span\u003e]) than control participants (\u003cem\u003eM\u003c/em\u003e = 6.90, 95% CI [6.5\u003cspan dir=\"RTL\"\u003e5\u003c/span\u003e, 7.\u003cspan dir=\"RTL\"\u003e24\u003c/span\u003e]), \u003cem\u003et\u003c/em\u003e(454) = 3.05, \u003cem\u003ep\u0026nbsp;\u003c/em\u003e= .003, \u003cem\u003ed\u003c/em\u003e = 0.49, 95% CI [0.17, 0.81]. However, as in Studies 1 and 2, there was no significant difference between ThinkFRE and control in the baseline condition (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e= .051) and in the no-source correction condition (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e= .25). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eComparison of Correction Types Within Each Group:\u003c/u\u003e We also conducted specific contrasts to test the effect of the scientific correction within each group to test for a potential backfire effect. In contrast to our earlier findings, the scientific correction did not produce a backfire effect among vaccine skeptics in the control group. Instead, the correction was simply ineffective, showing no significant difference when compared to the baseline condition (\u003cem\u003ep\u003c/em\u003e = .93) or to the correction without a source (\u003cem\u003ep\u003c/em\u003e = .81). In the ThinkFRE group, the scientific correction (\u003cem\u003eM\u003c/em\u003e = 6.01, 95% CI [5.5\u003cspan dir=\"RTL\"\u003e4\u003c/span\u003e, 6.4\u003cspan dir=\"RTL\"\u003e7\u003c/span\u003e]) did not influence belief in the disinformation compared to the baseline condition (\u003cem\u003eM\u003c/em\u003e = 6.30, 95% CI [5.\u003cspan dir=\"RTL\"\u003e90\u003c/span\u003e, 6.\u003cspan dir=\"RTL\"\u003e71\u003c/span\u003e]; \u003cem\u003ep\u003c/em\u003e = .32) but was effective compared to the no-source correction (\u003cem\u003eM\u003c/em\u003e = 6.\u003cspan dir=\"RTL\"\u003e63\u003c/span\u003e, 95% CI [\u003cspan dir=\"RTL\"\u003e6.17\u003c/span\u003e, \u003cspan dir=\"RTL\"\u003e7.08\u003c/span\u003e]), \u003cem\u003et\u003c/em\u003e(454) = 2.05, \u003cem\u003ep\u0026nbsp;\u003c/em\u003e= .041, \u003cem\u003ed\u003c/em\u003e = 0.34, 95% CI [0.01, 0.67]\u003ca href=\"#_ftn1\" name=\"_ftnref1\" title=\"\"\u003e\u003c/a\u003e\u003csup\u003e5\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCOVID-19 Vaccine Supporters\u003c/em\u003e:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAmong COVID-19 vaccine supporters, there were no significant main effects. General belief in the presented disinformation about the COVID-19 vaccine was not affected by group (\u003cem\u003ep\u003c/em\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e=\u0026nbsp;.95), nor by the correction type (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e= .67) or their interaction (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e= .15).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eComparison of ThinkFRE and Control Across Correction Types:\u003c/u\u003e When conducting the specific contrasts among COVID-19 vaccine supporters, no differences were found between ThinkFRE and control participants in the baseline condition (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e= .13), no-source correction condition (\u003cem\u003ep\u003c/em\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e=\u0026nbsp;.24), and scientific correction condition (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e= .80).\u003c/p\u003e\n\u003cp\u003e\u003cu\u003eComparison of Correction Types Within Each Group:\u003c/u\u003e Among COVID-19 vaccine supporters in the control group, the scientific correction condition was not different from the baseline condition (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e= .46) or the no-source correction condition (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e= .93). Similarly, among COVID-19 vaccine supporters who learned ThinkFRE, the correction with a scientific source did not differ from baseline (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e= .42) or the no-source correction (\u003cem\u003ep\u0026nbsp;\u003c/em\u003e= .30).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTrust in Science and Scientists\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWe measured trust in science with the Trust in Science and Scientists Inventory (Nadelson et al., 2014). COVID-19 vaccine supporters showed significantly higher trust in science and scientists (\u003cem\u003eM\u003c/em\u003e = 4.07, 95% CI [4.03, 4.11]) compared to COVID-19 vaccine skeptics (\u003cem\u003eM\u003c/em\u003e = 2.77, 95% CI [2.71, 2.83]), \u003cem\u003eF\u003c/em\u003e(1, 963) = 1231.77, \u003cem\u003ep\u003c/em\u003e \u0026lt; .001; \u0026eta;\u003csub\u003ep\u003c/sub\u003e\u003csup\u003e2\u0026nbsp;\u003c/sup\u003e= 0.56, 95% CI [0.52, 0.59]. None of the other factors, neither group (\u003cem\u003ep\u003c/em\u003e = .13), nor correction type (\u003cem\u003ep\u003c/em\u003e = .33)\u003cspan dir=\"RTL\"\u003e,\u003c/span\u003e nor any interaction between them\u003cspan dir=\"RTL\"\u003e,\u0026nbsp;\u003c/span\u003eaffected participants\u0026rsquo; trust in science and scientists.\u003c/p\u003e\n\u003cp\u003eTrust in science\u0026nbsp;was significantly negatively correlated with belief in the presented disinformation (\u003cem\u003er\u003c/em\u003e = -0.77, \u003cem\u003ep\u003c/em\u003e \u0026lt; .001). In other words, the higher someone\u0026rsquo;s trust in science, the less likely they were to believe in the causal link between the COVID-19 vaccine and heart disease. This pattern held for both COVID-19 vaccine supporters (\u003cem\u003er\u003c/em\u003e = -0.40, \u003cem\u003ep\u003c/em\u003e \u0026lt; .001) and skeptics (\u003cem\u003er\u003c/em\u003e = -0.43, \u003cem\u003ep\u003c/em\u003e \u0026lt;.001)\u003ca href=\"#_ftn2\" name=\"_ftnref2\" title=\"\"\u003e\u003c/a\u003e\u003csup\u003e6\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe present research demonstrates that citing scientific sources to correct vaccine disinformation can be ineffective or even counterproductive, depending on the context. During a period of high uncertainty and unfamiliar disinformation about the new COVID-19 vaccines (Studies 1 \u0026amp; 2; 2021), a science-based correction backfired among participants in a control group that received no pre-emptive intervention, increasing belief in the disinformation compared to a no-correction baseline. Later, among COVID-19 vaccine skeptics (Study 3; 2024), the same type of science-based correction was merely ineffective. A novel, pre-emptive intervention that provided a subtle reminder of the importance of evidence and expertise, worked specifically to remedy this failure. Its positive effect was significant only in the scientific-source condition, where it made the correction more effective. This targeted impact suggests that for science-based corrections to succeed, individuals may first need to be prompted into an evidence-based mindset that restores the persuasive power of scientific expertise.\u003c/p\u003e \u003cp\u003eOn a broader level, our findings reveal a critical disconnect between self-reported, abstract trust in science and its practical enactment. While large-scale surveys often report high public trust in science and scientists, our research provides a more implicit, behavioral measure by testing this trust with highly relevant real-world disinformation within an ecologically valid context. The fact that a correction citing scientific authorities could fail or even backfire demonstrates that abstract trust does not automatically translate into enacted trust and the corresponding belief revision when it is most needed\u0026mdash;in the direct confrontation with potent disinformation. This suggests that public trust in science is potentially weaker than often assumed, particularly under conditions of uncertainty and fear and among those with entrenched skeptical views. The fact that a brief intervention was sufficient to restore the positive impact of scientific sources indicates that the core issue may not be a deep-seated rejection of science, but a correctable lapse in applying an evidence-based mindset.\u003c/p\u003e \u003cp\u003eDisinformation thrives in conditions of confusion, where people are unsure what to believe and the consequences of error can be severe\u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e. Our findings on the effectiveness of the ThinkFRE strategy offer a promising path forward in tackling such challenges. In high-stakes domains like public health, the need for broadly applicable, easily deployable strategies is a priority, and ThinkFRE exemplifies such a tool. We propose its effectiveness stems from its theoretical foundation: by reminding individuals to distinguish between objective and subjective information, taking into account falsifiability and highlighting the value of expertise, it may activate critical evaluation in a way that is likely to be resilient to the influence of prior attitudes. The strength of this approach lies in its design as a lightweight, scalable intervention that requires no specialized knowledge or reliance on content-specific counterarguments. This content-agnostic nature aligns with a growing body of research identifying interventions that foster analytic thinking as among the most effective and broadly applicable tools against misinformation\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e,\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e,\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e. It also fits the theoretical framework emphasizing the potential to bridge epistemic conflicts by adopting common evidential standards\u003csup\u003e\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e. While this suggests ThinkFRE has strong promise for widespread use, its generalizability must be empirically established. Therefore, a crucial avenue for future work is to test this strategy across diverse contexts and explore its practical applications in real-world settings.\u003c/p\u003e \u003cp\u003eThe current research has several limitations. Unlike Studies 1 and 2, Study 3 did not replicate the negative effect of a scientific source in the control group, regardless of whether a single-item or multi-item measure was used. While backfire effects have been reported in some studies \u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e,\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e, such effects are generally rare, and several attempts to replicate such findings have been unsuccessful\u003csup\u003e\u003cspan additionalcitationids=\"CR60 CR61 CR62\" citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e. Correspondingly, it might be that the backfire effect observed in Studies 1 and 2 may have been driven by the extraordinary uncertainty, caution, and skepticism prevalent among large sections of the U.S. public during the unprecedented early stages of the COVID-19 pandemic. In such contexts, new scientific information may require particularly high levels of trust to effectively counter disinformation. This highlights the critical role of timing in disinformation and correction efforts, as backfire effects may be more likely to emerge in environments characterized by heightened uncertainty or low source trust.\u003c/p\u003e \u003cp\u003eAnother potential, if speculative, explanation for the lack of a backfire effect in Study 3 is the high level of conviction within our vaccine skeptic group. By design, these participants already held a significantly higher belief in the disinformation than our sample in our earlier studies. When a belief is this strong, it is possible that a correction from a distrusted source is met with simple dismissal rather than the active counter-argumentation that typically drives a backfire effect. In essence, their conviction may have been so solidified that the correction was merely ignored, not actively resisted. Furthermore, we do not know whether the effect of ThinkFRE is caused by one specific component of the intervention (e.g., the Expert component) or by the interaction of all three components (Fact, Refutable, Expert). Future research should examine whether each individual concept can separately restore trust in factual information and expert knowledge, or whether it is ThinkFRE as a whole that encourages a mindset of healthy skepticism. In any case, invoking objectivity and expertise may go a long way in restoring trust in scientific sources.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eResearch Funding\u003c/h2\u003e\n\u003cp\u003eThis research was supported by the Israel Science Foundation (ISF) Grant no. 740/23 RM, awarded to Ruth Mayo.\u003c/p\u003e\n\u003ch2\u003e\u003cstrong\u003eDeclaration of competing interest:\u003c/strong\u003e\u003c/h2\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\n\u003cp\u003eA.L.B - Conceptualization, Methodology, Data Curation, Formal Analysis, Visualization, Writing (original draft), Writing (review \u0026amp;amp; editing). U.K.H.E. 6 Conceptualization, Methodology, Writing (review \u0026amp;amp; editing). R.M. 6 Conceptualization, Methodology, Validation, Supervision, Writing (review \u0026amp;amp; editing), Funding Acquisition.\u003c/p\u003e\n\u003ch2\u003eData Availability\u003c/h2\u003e\n\u003cp\u003eThe datasets, analysis code and preregistrations for these studies are available via [OSF](https:/osf.io/6ynek/overview?view_only=e2f387afa7174ec48942c2c30e1197aa) .\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eLevy, N. Nudges in a post-truth world. \u003cem\u003eJ. Med. Ethics\u003c/em\u003e \u003cstrong\u003e43\u003c/strong\u003e, 495\u0026ndash;500 (2017).\u003c/li\u003e\n\u003cli\u003eMcIntyre, L. \u003cem\u003ePost-Truth\u003c/em\u003e. (MIt Press, 2018).\u003c/li\u003e\n\u003cli\u003eVan der Linden, S. \u0026amp; L\u0026ouml;fstedt, R. E. \u003cem\u003eRisk and Uncertainty in a Post-Truth Society\u003c/em\u003e. (Routledge, 2019).\u003c/li\u003e\n\u003cli\u003eLewandowsky, S., Ecker, U. K. H. \u0026amp; Cook, J. Beyond misinformation: Understanding and coping with the \u0026ldquo;post-truth\u0026rdquo; era. \u003cem\u003eJ. Appl. Res. Mem. Cogn.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 353\u0026ndash;369 (2017).\u003c/li\u003e\n\u003cli\u003eRothkopf, D. J. When the buzz bites back. \u003cem\u003eWash. Post\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, B1--B5 (2003).\u003c/li\u003e\n\u003cli\u003eZarocostas, J. How to fight an infodemic. \u003cem\u003eThe lancet\u003c/em\u003e \u003cstrong\u003e395\u003c/strong\u003e, 676 (2020).\u003c/li\u003e\n\u003cli\u003eAsh, E., Galletta, S., Hangartner, D., Margalit, Y. \u0026amp; Pinna, M. The effect of Fox News on health behavior during COVID-19. \u003cem\u003ePolit. Anal.\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 275\u0026ndash;284 (2024).\u003c/li\u003e\n\u003cli\u003eBursztyn, L., Rao, A., Roth, C. P. \u0026amp; Yanagizawa-Drott, D. H. \u003cem\u003eMisinformation during a Pandemic\u003c/em\u003e. https://www.nber.org/papers/w27417 (2020).\u003c/li\u003e\n\u003cli\u003eSimonov, A., Sacher, S., Dub\u0026eacute;, J.-P. \u0026amp; Biswas, S. Frontiers: The Persuasive Effect of Fox News: Noncompliance with Social Distancing During the COVID-19 Pandemic. \u003cem\u003eMark. Sci.\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 230\u0026ndash;242 (2022).\u003c/li\u003e\n\u003cli\u003eEcker, U. K. H. \u003cem\u003eet al.\u003c/em\u003e Why misinformation must not be ignored. \u003cem\u003eAm. Psychol.\u003c/em\u003e \u003cstrong\u003e80\u003c/strong\u003e, 867\u0026ndash;878 (2025).\u003c/li\u003e\n\u003cli\u003evan Bavel, J. J. \u003cem\u003eet al.\u003c/em\u003e Using social and behavioural science to support COVID-19 pandemic response. \u003cem\u003eNat. Hum. Behav.\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 460\u0026ndash;471 (2020).\u003c/li\u003e\n\u003cli\u003eAlgan, Y., Cohen, D., Davoine, E., Foucault, M. \u0026amp; Stantcheva, S. Trust in scientists in times of pandemic: Panel evidence from 12 countries. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e \u003cstrong\u003e118\u003c/strong\u003e, e2108576118 (2021).\u003c/li\u003e\n\u003cli\u003ePlohl, N. \u0026amp; Musil, B. Modeling compliance with COVID-19 prevention guidelines: the critical role of trust in science. \u003cem\u003ePsychol. Health Med.\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 1\u0026ndash;12 (2021).\u003c/li\u003e\n\u003cli\u003eAdeoye, A. F. \u003cem\u003eet al.\u003c/em\u003e The 2025 United States Measles Crisis: When Vaccine Hesitancy Meets Reality. \u003cem\u003eCureus\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, e88196.\u003c/li\u003e\n\u003cli\u003eParums, D. V. A Review of the Resurgence of Measles, a Vaccine-Preventable Disease, as Current Concerns Contrast with Past Hopes for Measles Elimination. \u003cem\u003eMed. Sci. Monit. Int. Med. J. Exp. Clin. Res.\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, e944436 (2024).\u003c/li\u003e\n\u003cli\u003eOECD. \u003cem\u003eTrust and Public Policy: How Better Governance Can Help Rebuild Public Trust\u003c/em\u003e. \u003cem\u003eOECD Public Governance Reviews\u003c/em\u003e 1\u0026ndash;158 http://dx.doi.org/10.1787/9789264268920-en (2017).\u003c/li\u003e\n\u003cli\u003eCologna, V. \u003cem\u003eet al.\u003c/em\u003e Trust in scientists and their role in society across 68 countries. \u003cem\u003eNat. Hum. Behav.\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 713\u0026ndash;730 (2025).\u003c/li\u003e\n\u003cli\u003eWellcome. Wellcome Global Monitor: How Covid-19 affected people\u0026rsquo;s lives and their views about science. (2020).\u003c/li\u003e\n\u003cli\u003eBromme, R., Mede, N. G., Thomm, E., Kremer, B. \u0026amp; Ziegler, R. An anchor in troubled times: Trust in science before and within the COVID-19 pandemic. \u003cem\u003ePloS One\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, e0262823 (2022).\u003c/li\u003e\n\u003cli\u003eWellcome. Insights from the Wellcome Global Monitor. \u003cem\u003eWellcome\u003c/em\u003e https://wellcome.org/insights/articles/public-trust-scientists-rose-during-covid-19-pandemic-0 (2021).\u003c/li\u003e\n\u003cli\u003eEichengreen, B., Aksoy, C. G. \u0026amp; Saka, O. Revenge of the experts: Will COVID-19 renew or diminish public trust in science? \u003cem\u003eJ. Public Econ.\u003c/em\u003e \u003cstrong\u003e193\u003c/strong\u003e, 104343 (2021).\u003c/li\u003e\n\u003cli\u003eHoogeveen, S. \u003cem\u003eet al.\u003c/em\u003e The Einstein effect provides global evidence for scientific source credibility effects and the influence of religiosity. \u003cem\u003eNat. Hum. Behav.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 523\u0026ndash;535 (2022).\u003c/li\u003e\n\u003cli\u003ePennycook, G., Cheyne, J. A., Barr, N., Koehler, D. J. \u0026amp; Fugelsang, J. A. On the reception and detection of pseudo-profound bullshit. \u003cem\u003eJudgm. Decis. Mak.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 549\u0026ndash;563 (2015).\u003c/li\u003e\n\u003cli\u003eIyengar, S. \u0026amp; Massey, D. S. Scientific communication in a post-truth society. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e \u003cstrong\u003e116\u003c/strong\u003e, 7656\u0026ndash;7661 (2019).\u003c/li\u003e\n\u003cli\u003eRoozenbeek, J. \u003cem\u003eet al.\u003c/em\u003e Susceptibility to misinformation about COVID-19 around the world. \u003cem\u003eR. Soc. Open Sci.\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 201199 (2020).\u003c/li\u003e\n\u003cli\u003eWest, J. D. \u0026amp; Bergstrom, C. T. Misinformation in and about science. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e \u003cstrong\u003e118\u003c/strong\u003e, e1912444117 (2021).\u003c/li\u003e\n\u003cli\u003eKozyreva, A. \u003cem\u003eet al.\u003c/em\u003e Toolbox of individual-level interventions against online misinformation. \u003cem\u003eNat. Hum. Behav.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 1044\u0026ndash;1052 (2024).\u003c/li\u003e\n\u003cli\u003ePennycook, G., Bear, A., Collins, E. T. \u0026amp; Rand, D. G. The implied truth effect: Attaching warnings to a subset of fake news headlines increases perceived accuracy of headlines without warnings. \u003cem\u003eManag. Sci.\u003c/em\u003e \u003cstrong\u003e66\u003c/strong\u003e, 4944\u0026ndash;4957 (2020).\u003c/li\u003e\n\u003cli\u003eHowell, E. L. \u0026amp; Brossard, D. (Mis) informed about what? What it means to be a science-literate citizen in a digital world. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e \u003cstrong\u003e118\u003c/strong\u003e, e1912436117 (2021).\u003c/li\u003e\n\u003cli\u003ePrike, T. \u0026amp; Ecker, U. K. H. Effective correction of misinformation. \u003cem\u003eCurr. Opin. Psychol.\u003c/em\u003e \u003cstrong\u003e54\u003c/strong\u003e, 101712 (2023).\u003c/li\u003e\n\u003cli\u003eLewandowsky, S., Ecker, U. K. H., Seifert, C. M., Schwarz, N. \u0026amp; Cook, J. Misinformation and Its Correction: Continued Influence and Successful Debiasing. \u003cem\u003ePsychol. Sci. Public Interest\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 106\u0026ndash;131 (2012).\u003c/li\u003e\n\u003cli\u003eEcker, U. K. H., O\u0026rsquo;Reilly, Z., Reid, J. S. \u0026amp; Chang, E. P. The effectiveness of short‐format refutational fact‐checks. \u003cem\u003eBr. J. Psychol.\u003c/em\u003e \u003cstrong\u003e111\u003c/strong\u003e, 36\u0026ndash;54 (2020).\u003c/li\u003e\n\u003cli\u003eLewandowsky, S. \u003cem\u003eet al.\u003c/em\u003e \u003cem\u003eThe Debunking Handbook 2020\u003c/em\u003e. (2020).\u003c/li\u003e\n\u003cli\u003eCook, J. \u0026amp; Lewandowsky, S. The debunking handbook. University of Queensland. (2011).\u003c/li\u003e\n\u003cli\u003eButler, L. H., Prike, T. \u0026amp; Ecker, U. K. Nudge-based misinformation interventions are effective in information environments with low misinformation prevalence. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 11495 (2024).\u003c/li\u003e\n\u003cli\u003eFazio, L. Pausing to consider why a headline is true or false can help reduce the sharing of false news. \u003cem\u003eHarv. Kennedy Sch. Misinformation Rev.\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, (2020).\u003c/li\u003e\n\u003cli\u003ePennycook, G. \u003cem\u003eet al.\u003c/em\u003e Shifting attention to accuracy can reduce misinformation online. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e592\u003c/strong\u003e, 590\u0026ndash;595 (2021).\u003c/li\u003e\n\u003cli\u003ePennycook, G., McPhetres, J., Zhang, Y., Lu, J. G. \u0026amp; Rand, D. G. Fighting COVID-19 misinformation on social media: Experimental evidence for a scalable accuracy-nudge intervention. \u003cem\u003ePsychol. Sci.\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 770\u0026ndash;780 (2020).\u003c/li\u003e\n\u003cli\u003ePennycook, G. \u0026amp; Rand, D. G. Accuracy prompts are a replicable and generalizable approach for reducing the spread of misinformation. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 2333 (2022).\u003c/li\u003e\n\u003cli\u003eChan, M. S., Jones, C. R., Hall Jamieson, K. \u0026amp; Albarrac\\\u0026rsquo;\\in, D. Debunking: A meta-analysis of the psychological efficacy of messages countering misinformation. \u003cem\u003ePsychol. Sci.\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 1531\u0026ndash;1546 (2017).\u003c/li\u003e\n\u003cli\u003eEcker, U. K. H., Lewandowsky, S. \u0026amp; Tang, D. T. W. Explicit warnings reduce but do not eliminate the continued influence of misinformation. \u003cem\u003eMem. Cognit.\u003c/em\u003e \u003cstrong\u003e38\u003c/strong\u003e, 1087\u0026ndash;1100 (2010).\u003c/li\u003e\n\u003cli\u003eLewandowsky, S., Stritzke, W. G. K., Oberauer, K. \u0026amp; Morales, M. Misinformation and the \u0026ldquo;War on Terror\u0026rdquo;: when memory turns fiction into fact. in \u003cem\u003eTerrorism and Torture: An Interdisciplinary Perspective\u003c/em\u003e (eds. Stritzke, W. G. K., Lewandowsky, S., Denemark, D., Clare, J. \u0026amp; Morgan, F.) 179\u0026ndash;203 (Cambridge University Press, 2009). doi:10.1017/CBO9780511581199.010.\u003c/li\u003e\n\u003cli\u003eSpearing, E. R. \u003cem\u003eet al.\u003c/em\u003e Countering AI-generated misinformation with pre-emptive source discreditation and debunking. \u003cem\u003eR. Soc. Open Sci.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 242148 (2025).\u003c/li\u003e\n\u003cli\u003eCompton, J., van der Linden, S., Cook, J. \u0026amp; Basol, M. Inoculation theory in the post-truth era: Extant findings and new frontiers for contested science, misinformation, and conspiracy theories. \u003cem\u003eSoc. Personal. Psychol. Compass\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, e12602 (2021).\u003c/li\u003e\n\u003cli\u003eLewandowsky, S. \u0026amp; Van Der Linden, S. Countering misinformation and fake news through inoculation and prebunking. \u003cem\u003eEur. Rev. Soc. Psychol.\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 348\u0026ndash;384 (2021).\u003c/li\u003e\n\u003cli\u003eCook, J., Lewandowsky, S. \u0026amp; Ecker, U. K. H. Neutralizing misinformation through inoculation: Exposing misleading argumentation techniques reduces their influence. \u003cem\u003ePloS One\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, e0175799 (2017).\u003c/li\u003e\n\u003cli\u003eder Linden, S., Leiserowitz, A., Rosenthal, S. \u0026amp; Maibach, E. Inoculating the public against misinformation about climate change. \u003cem\u003eGlob. Chall.\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 1600008 (2017).\u003c/li\u003e\n\u003cli\u003eBasol, M. \u003cem\u003eet al.\u003c/em\u003e Towards psychological herd immunity: Cross-cultural evidence for two prebunking interventions against COVID-19 misinformation. \u003cem\u003eBig Data Soc.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 20539517211013868 (2021).\u003c/li\u003e\n\u003cli\u003eRoozenbeek, J. \u0026amp; der Linden, S. Fake news game confers psychological resistance against online misinformation. \u003cem\u003ePalgrave Commun.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 1\u0026ndash;10 (2019).\u003c/li\u003e\n\u003cli\u003eGuess, A. M. \u003cem\u003eet al.\u003c/em\u003e A digital media literacy intervention increases discernment between mainstream and false news in the United States and India. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e \u003cstrong\u003e117\u003c/strong\u003e, 15536\u0026ndash;15545 (2020).\u003c/li\u003e\n\u003cli\u003eRoozenbeek, J., Van Der Linden, S., Goldberg, B., Rathje, S. \u0026amp; Lewandowsky, S. Psychological inoculation improves resilience against misinformation on social media. \u003cem\u003eSci. Adv.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, eabo6254 (2022).\u003c/li\u003e\n\u003cli\u003eGeorgiou, N., Delfabbro, P. \u0026amp; Balzan, R. The effectiveness of a scientific reasoning intervention for conspiracy theory beliefs. \u003cem\u003eAppl. Cogn. Psychol.\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 369\u0026ndash;382 (2023).\u003c/li\u003e\n\u003cli\u003eSalovich, N. A., Kirsch, A. M. \u0026amp; Rapp, D. N. Evaluative mindsets can protect against the influence of false information. \u003cem\u003eCognition\u003c/em\u003e \u003cstrong\u003e225\u003c/strong\u003e, 105121 (2022).\u003c/li\u003e\n\u003cli\u003eRapp, D. N. \u0026amp; Salovich, N. A. Can\u0026rsquo;t we just disregard fake news? The consequences of exposure to inaccurate information. \u003cem\u003ePolicy Insights Behav. Brain Sci.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 232\u0026ndash;239 (2018).\u003c/li\u003e\n\u003cli\u003eO\u0026rsquo;Mahony, C., Brassil, M., Murphy, G. \u0026amp; Linehan, C. The efficacy of interventions in reducing belief in conspiracy theories: A systematic review. \u003cem\u003ePlos One\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, e0280902 (2023).\u003c/li\u003e\n\u003cli\u003eTay, L. Q. \u003cem\u003eet al.\u003c/em\u003e Broadening Misinformation Research: The Roles of Evidence Retrievability and Epistemic Bridging. Preprint at https://doi.org/10.31234/osf.io/jvs9u_v1 (2025).\u003c/li\u003e\n\u003cli\u003eSchwarz, N., Sanna, L. J., Skurnik, I. \u0026amp; Yoon, C. Metacognitive experiences and the intricacies of setting people straight: Implications for debiasing and public information campaigns. \u003cem\u003eAdv. Exp. Soc. Psychol.\u003c/em\u003e \u003cstrong\u003e39\u003c/strong\u003e, 127\u0026ndash;161 (2007).\u003c/li\u003e\n\u003cli\u003eSchwarz, N., Newman, E. \u0026amp; Leach, W. Making the truth stick \u0026amp; the myths fade: Lessons from cognitive psychology. \u003cem\u003eBehav. Sci. Policy\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 85\u0026ndash;95 (2016).\u003c/li\u003e\n\u003cli\u003eEcker, U. K. H., Lewandowsky, S., Swire, B. \u0026amp; Chang, D. Correcting false information in memory: Manipulating the strength of misinformation encoding and its retraction. \u003cem\u003ePsychon. Bull. Rev.\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 570\u0026ndash;578 (2011).\u003c/li\u003e\n\u003cli\u003eEcker, U. K. H., Lewandowsky, S. \u0026amp; Chadwick, M. Can corrections spread misinformation to new audiences? Testing for the elusive familiarity backfire effect. \u003cem\u003eCogn. Res. Princ. Implic.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 41 (2020).\u003c/li\u003e\n\u003cli\u003eEcker, U. K. H. \u003cem\u003eet al.\u003c/em\u003e The psychological drivers of misinformation belief and its resistance to correction. \u003cem\u003eNat. Rev. Psychol.\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 13\u0026ndash;29 (2022).\u003c/li\u003e\n\u003cli\u003eSwire-Thompson, B., DeGutis, J. \u0026amp; Lazer, D. Searching for the backfire effect: Measurement and design considerations. \u003cem\u003eJ. Appl. Res. Mem. Cogn.\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 286\u0026ndash;299 (2020).\u003c/li\u003e\n\u003cli\u003eWood, T. \u0026amp; Porter, E. The elusive backfire effect: Mass attitudes\u0026rsquo; steadfast factual adherence. \u003cem\u003ePolit. Behav.\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 135\u0026ndash;163 (2019).\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Footnotes","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003e At the time Studies 1 and 2 were conducted, two COVID-19 vaccines had been approved for use in the U.S.: the Pfizer-BioNTech and Moderna COVID-19 vaccines. Initial doses were recommended for healthcare personnel and residents of long-term care facilities. Vaccines only became widely available to the general population in April 2021.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e No specific vaccine manufacturer or type was indicated in the studies.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e Following the main dependent variable, participants were asked a second question: \u0026ldquo;In your opinion, how likely is it that Bill Gates is connected to the COVID-19 pandemic or its vaccine?\u0026rdquo; After running the studies, we realized that the phrasing of the second dependent variable is problematic: It is also possible to interpret the question in terms that Bill Gates might be connected to the COVID-19 vaccine by helping with its funding, which could be the reason why we did not find effects for this variable. Since it was not the main variable and since it appeared after the main dependent variable about the connection between the vaccine and sterility, we do not report on this variable in this paper.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e The prescreening item was: \u0026ldquo;Please describe your attitudes towards the COVID-19 (Coronavirus) vaccines.\u0026rdquo; We selected half of the participants who answered \u0026ldquo;For (I feel positively about the vaccines)\u0026rdquo; and the other half who answered \u0026ldquo;Against (I feel negatively about the vaccines).\u0026rdquo; In the debrief section at the end of the study, we included the same item with the additional options: \u0026ldquo;Neutral (I don't have strong opinions either way)\u0026rdquo; and \u0026ldquo;Prefer not to say.\u0026rdquo; In the preregistration, we committed to analyzing the data according to participants\u0026rsquo; vaccine attitudes. However, we had not anticipated the extent of deviation from the prescreening responses and therefore excluded participants whose responses were inconsistent. The main findings and their significance remain unchanged when excluding these 127 participants (see Supplementary Information).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e This specific comparison was not part of our pre-registered hypotheses and should be considered exploratory.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e A similar report for Intellectual Humility can be found in the Supplementary Information.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Disinformation, Correction, Prebunking, COVID-19, Trust in Science","lastPublishedDoi":"10.21203/rs.3.rs-8204469/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8204469/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis research demonstrates that citing scientific sources to correct disinformation can be ineffective or even counterproductive, and offers a potential remedy. Three experiments simulated a WhatsApp conversation about COVID-19 vaccine disinformation, examining corrections in the high-uncertainty context of the early 2021 stage of the pandemic (Studies 1 and 2) and among vaccine skeptics in 2024 (Study 3). A science-based correction backfired during the pandemic and was ineffective among skeptics in 2024. However, a brief, content-agnostic intervention (ThinkFRE), comprising three simple statements prompting attention to facts and experts, consistently reversed these negative outcomes. The intervention specifically enhanced science-based corrections, not corrections with no source, and it worked even for skeptics who typically distrust science, suggesting it acts by restoring the importance of evidence and expertise. These findings indicate that for science to be an effective antidote to disinformation, individuals may first need a cognitive nudge toward an evidence-oriented mindset.\u003c/p\u003e","manuscriptTitle":"When Science Alone Fails to Convince: A Brief Intervention Fostering an Evidence-Oriented Mindset Strengthens Science-Based Corrections","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-23 10:42:50","doi":"10.21203/rs.3.rs-8204469/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-18T09:25:56+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-10T14:00:55+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-05T15:56:49+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"10444762514358537558283232704221962875","date":"2026-03-02T09:50:09+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"153014640373735560876556162644226907144","date":"2026-02-23T15:16:35+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-16T22:53:22+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"271445072372732568569191324423711536494","date":"2025-12-17T16:08:45+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-12-17T07:03:06+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-12-10T18:59:05+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-12-01T08:28:08+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-11-30T14:30:29+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"59bd725e-efc3-4ca8-b6a0-0b408a4a2606","owner":[],"postedDate":"December 23rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":59969056,"name":"Biological sciences/Psychology"},{"id":59969057,"name":"Social science/Psychology"},{"id":59969058,"name":"Scientific community and society/Scientific community"}],"tags":[],"updatedAt":"2026-05-15T08:26:07+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-23 10:42:50","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8204469","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8204469","identity":"rs-8204469","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}