{"id":575854,"date":"2026-01-27T16:48:05","date_gmt":"2026-01-27T19:48:05","guid":{"rendered":"https:\/\/revistapesquisa.fapesp.br\/?p=575854"},"modified":"2026-01-27T16:48:05","modified_gmt":"2026-01-27T19:48:05","slug":"warnings-left-by-users-on-social-media-posts-containing-misinformation-have-little-effect","status":"publish","type":"post","link":"https:\/\/revistapesquisa.fapesp.br\/en\/warnings-left-by-users-on-social-media-posts-containing-misinformation-have-little-effect\/","title":{"rendered":"Warnings left by users on social media posts containing misinformation have little effect"},"content":{"rendered":"

When people come across a post containing misinformation on social media, is it effective to leave a comment warning others that it\u2019s fake news? A paper published in July in the Harvard Kennedy School Misinformation Review<\/em> found that fact-checking corrections provided by regular users have only a limited effect on people who already believe false claims about COVID-19. When those corrections include a link to verified information from a news outlet, the impact increases slightly, depending on the country. The authors emphasize the need for institutional strategies and functional platform improvements to make misinformation countermeasures more effective.<\/p>\n

The study surveyed 3,000 people in Brazil, India, and the United Kingdom, asking them to evaluate Facebook posts about COVID-19. One thousand participants in each country answered a questionnaire after being shown real-world misinformation posts that were making the rounds when the data were collected, in March 2021. Some of the posts promoted chloroquine\u2014even though large clinical trials had already ruled out its effectiveness\u2014or downplayed the severity of the health emergency.<\/p>\n

Participants in each country were divided into three subgroups of roughly 300 people each. The first, a control group, saw three true posts and six false ones\u2014with no comments attached to any of them. The second group saw the same set of posts, except that in some cases, researchers added a mock user comment flagging the post as false. In the third group, those warning comments also included a link to a news organization\u2019s verified information. Although the links weren\u2019t clickable, they showed a preview with the article\u2019s headline and the name of the outlet.<\/p>\n

Participants then rated how accurate each post seemed\u2014choosing among \u201cnot accurate,\u201d \u201csomewhat accurate,\u201d \u201creasonably accurate,\u201d or \u201cvery accurate.\u201d Each response was scored on a scale from 0 to 3. Using those scores, the researchers calculated an average \u201cmisinformation belief level\u201d for each country, starting with the control group. In the UK, the average score was 0.83\u2014indicating the lowest belief in misinformation. In Brazil, it was 1.03, and in India, 1.68, the highest of the three.<\/p>\n

The researchers then repeated the calculation for the groups that saw fact-checking comments. In Brazil, the effects were mild. Compared with the control group, belief in misinformation dropped by 6.9% among participants who saw fact-checking comments with links and by 5.7% among those who saw comments without links. In India, though, linked comments had a stronger effect: they reduced belief in misinformation by roughly 10% compared with the control group. Comments without links, by contrast, showed no meaningful impact. In the UK, differences among the three groups were minimal, since the control group was already highly skeptical of misinformation.<\/p>\n

Another phase of the experiment tested whether fact-checking comments could reduce participants\u2019 willingness to share false COVID-19 posts. The setup mirrored the first experiment: participants rated how likely they were to share the same posts. Responses were scored on a 0\u20133 scale, where 0 meant \u201cnot at all likely,\u201d 1 \u201cunlikely,\u201d 2 \u201csomewhat likely,\u201d and 3 \u201cvery likely.\u201d In the control group, the average likelihood to share misinformation posts was lowest in the UK (0.47), followed by Brazil (0.71) and India (1.61).<\/p>\n

Efforts to combat misinformation must be multilayered<\/p><\/blockquote>\n

In Brazil, exposure to fact-checking comments reduced people\u2019s willingness to share false posts by 7.4% to 11.2% compared with the control group. In India, unlinked comments had only a minor effect (a 5.7% drop), while those that included a link to a trusted information source reduced people\u2019s willingness to share false COVID-19 content by 11.6%.<\/p>\n

\u201cThese prompts work to some degree\u2014but their impact depends heavily on each country\u2019s context,\u201d says Camila Mont\u2019Alverne, a professor at the University of Strathclyde in Scotland and one of the study\u2019s authors. \u201cThat\u2019s why social and economic differences need to be factored in when designing strategies against misinformation.\u201d Mont\u2019Alverne suggests that social media platforms should give users more tools to help preserve information integrity. \u201cLabels are one example\u2014short text tags or visual cues that help users spot whether a post is credible or false,\u201d she says. \u201cPlatforms used them for a while during the pandemic, but later pulled back.\u201d<\/p>\n

Raquel Recuero, a researcher at the Federal University of Rio Grande do Sul (UFRGS) who was not involved in the study, argues that these measures are unlikely to solve the problem since misinformation is a systemic issue. \u201cPeople are bombarded with falsehoods across multiple channels,\u201d she explains. \u201cSomeone who already doubts vaccine efficacy might see a video of a foreign health official claiming vaccines aren\u2019t safe, then get a message from a neighbor repeating this false claim. All that reinforcement strengthens their belief,\u201d says Recuero, who authored the book A rede da desinforma\u00e7\u00e3o<\/em> \u2013 Sistemas, estruturas e din\u00e2micas nas plataformas de m\u00eddias sociais<\/em> (A network of misinformation: Social media platforms, structures, and dynamics; Editora Sulina, 2024). \u201cThat\u2019s why a warning comment or a fact-checking link, while useful, can only go so far amid the sheer volume of content moving through social media,\u201d she adds.<\/p>\n

\u201cThis study reminds us that efforts to combat misinformation must be multilayered,\u201d says Dayane Machado, a doctoral researcher at the Department of Science and Technology Policy at the University of Campinas (UNICAMP), who studies health misinformation and was not involved in the study. \u201cWe can\u2019t rely on just one approach.\u201d Machado coauthored a 2020 study in Frontiers in Communication <\/em>that examined Brazilian YouTube creators spreading false information about vaccines. The research found that YouTube was helping fuel the spread of misinformation by allowing those videos to be monetized\u2014bringing in ad revenue for both the creators and the platform itself. She argues that the burden of correcting misinformation shouldn\u2019t fall on individual users.<\/p>\n

Recuero agrees, saying the problem demands a coordinated set of actions. One approach, she says, is offline engagement\u2014working directly with trusted local actors like community health agents, who can share accurate information credibly\u2014as long as they are not themselves misinformed. \u201cIn a study we conducted in Maranh\u00e3o, we found that even health agents had doubts about vaccine effectiveness\u2014because they, too, were being bombarded with misinformation,\u201d she recalls. She also highlights the value of coalitions of universities, government agencies, and civil society groups to tackle misinformation in specific fields through coordinated action.<\/p>\n

In such a complex landscape, is it still worth commenting to fact-check falsehoods on social media? \u201cAbsolutely,\u201d Recuero says. \u201cAt the very least, you might discourage someone from sharing that lie.\u201d<\/p>\n

The story above was published with the title “A warning lost in the noise<\/strong>” in issue 356 of October\/2025.<\/p>\n","protected":false},"excerpt":{"rendered":"A study conducted in Brazil, India, and the United Kingdom evaluated data from 3,000 people","protected":false},"author":684,"featured_media":563467,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_exactmetrics_skip_tracking":false,"_exactmetrics_sitenote_active":false,"_exactmetrics_sitenote_note":"","_exactmetrics_sitenote_category":0,"footnotes":""},"categories":[166],"tags":[220],"coauthors":[2721],"class_list":["post-575854","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-policies-st-en","tag-communication"],"acf":[],"_links":{"self":[{"href":"https:\/\/revistapesquisa.fapesp.br\/en\/wp-json\/wp\/v2\/posts\/575854","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/revistapesquisa.fapesp.br\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/revistapesquisa.fapesp.br\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/revistapesquisa.fapesp.br\/en\/wp-json\/wp\/v2\/users\/684"}],"replies":[{"embeddable":true,"href":"https:\/\/revistapesquisa.fapesp.br\/en\/wp-json\/wp\/v2\/comments?post=575854"}],"version-history":[{"count":1,"href":"https:\/\/revistapesquisa.fapesp.br\/en\/wp-json\/wp\/v2\/posts\/575854\/revisions"}],"predecessor-version":[{"id":575857,"href":"https:\/\/revistapesquisa.fapesp.br\/en\/wp-json\/wp\/v2\/posts\/575854\/revisions\/575857"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/revistapesquisa.fapesp.br\/en\/wp-json\/wp\/v2\/media\/563467"}],"wp:attachment":[{"href":"https:\/\/revistapesquisa.fapesp.br\/en\/wp-json\/wp\/v2\/media?parent=575854"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/revistapesquisa.fapesp.br\/en\/wp-json\/wp\/v2\/categories?post=575854"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/revistapesquisa.fapesp.br\/en\/wp-json\/wp\/v2\/tags?post=575854"},{"taxonomy":"author","embeddable":true,"href":"https:\/\/revistapesquisa.fapesp.br\/en\/wp-json\/wp\/v2\/coauthors?post=575854"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}