‘True’ fact checks bring ‘cognitive pleasure’
Fact checks that label a piece of content as “true” have a higher chance of going viral than those labeled “false.”
This was just one of the conclusions of an independent study on fact-checking in Argentina during that country’s last presidential election.
In 2019, researchers from the University of Maryland partnered with a team from Universidad Nacional de Quilmes to analyze fact checks published by Chequeado — Argentina’s most recognized fact-checking organization — on social media.
Researchers Ernesto Calvo, Natalia Aruguete and Tiago Ventura wanted to know how voters perceived, accepted, consumed and shared fact checks. Conducted with 2,040 participants, the study included five experiments, all based on social media data. The results were impressive and have the potential to impact not just the fact-checking community but also projects aiming to fight mis/disinformation, like Facebook’s Third Party Fact-Checking program.
“The results of this study show that people would rather share an article considered ‘true’ than a ‘false’ one. This research also proves that each time we (Chequeado) point out to someone that something is not what they thought it was, their opinion about our brand or our organization is diminished,” wrote Laura Zommer, Chequeado’s CEO, in an opinion piece published in the newspaper La Nación (available in English here).
In a Zoom meeting held Tuesday, Calvo and Aruguete discussed their methodology and summarized their findings.
“When the fact check validates as being true something that people already believed was true, the chances of sharing that fact check are higher than when there is a double false, when something you thought was wrong is considered wrong,” Calvo said.
Among fact-checkers, it could be surprising that data shows the so-called “I was right effect” is more frequent with true content, but researchers offered a simple explanation.
“Something that is positive twice — ‘I thought it as true and it actually is’ — generates cognitive pleasure. Double falses, on the other hand, cause cognitive harm and the conversation tends to stop.”
In “Chequeado in Argentina: Fact-Checking and the Spread of Disinformation on Social Media,” researchers highlighted that “between July and December (2019), Chequeado published twice as many ‘false’ adjudications. This implies that the fact-checking organization is consistently causing ‘cognitive harm’ to its readers,” the researchers alerted.
The study also found, both numerically and visually, that Chequeado isn’t politically biased. A social media analysis (that can surely be replicated in other parts of the world) showed that Chequeado’s fact checks are evenly shared on Twitter by those in favor and against the party controlling the government.
Finally, the research proved that fact-checking can play a central role in the fight against electoral mis/disinformation. Researchers concluded that people might not change their opinions after reading Chequeado’s fact checks, but they saw evidence of a change in behavior. Many of the study participants became less prone to share unreliable information after receiving a fact check.
- Mafindo: “These images don’t show kids addicted to online games” (In Indonesian)
- What might have started as a joke has become a viral hoax in Indonesia. Mafindo’s team alerts that an Instagram filter called “Crazy Eyes” is being used to disseminate the false idea that playing online games can give you weird-looking eyes. A single post on Facebook has been shared more than 75,000 times, spreading fear among parents.
- Maldita: “Here is a list of 100 phishing hoaxes you might have seen during the pandemic” (In Spanish)
- Attention! Please! There is no “anonymous mode” on WhatsApp. Airlines aren’t offering “free flights.” McDonald’s isn’t giving away “100 euros coupons.” Maldita’s fact-checkers have caught 100 hoaxes spreading URLs that actually aim to steal your personal data without you noticing. Be aware.
From the news:
- “Naked Truth” RSF campaign to defend reliable reporting in Brazil, from Reporters Without Borders. RSF launched Monday a multilingual social media campaign called “Naked Truth.” The initiative aims to draw attention to Brazilian President Jair Bolsonaro’s use of “lies and attacks on the media to mask his inability to address the COVID-19 pandemic’s devastating impact.” The campaign features a photoshopped portrait of a naked Bolsonaro covered by a placard with the latest COVID-19 case and death figures.
- “Australian Code of Practice on Disinformation and Misinformation,” from DIGI. Twitter, Google, Facebook, Microsoft, Redbubble and TikTok announced Monday that they now have a code to fight mis/disinformation. All companies agreed to be transparent about their actions, act against paid content that disinforms and to partner with universities, allowing them to analyze content on their platforms. “The code was developed in response to the Australian Government policy announced in December 2019, where the digital industry was asked to develop a voluntary code of practice on disinformation.”
- “How memes became a major vehicle for misinformation,” from Axios. Memes have become a more pernicious and difficult to detect way to spread misinformation. Axios reporters Sara Fischer and Alison Snyder explained how artificial Intelligence used by tech platforms to detect falsehoods has a difficult time recognizing text layered on top of an image.
- “Biden lies less than Trump, fact-checkers say. But he’s not perfect,” from CNN. Daniel Dale and PolitiFact editor-in-chief Angie Holan talked to Brian Stelter about how fact-checking in the U.S. has changed during the Biden administration. Both said they are able to do more fact-checking now that they’re not faced with a torrent of claims from former President Donald Trump.
From/for the community:
- Turkish Minister of Communication Fahrettin Altun announced that Teyip Erdogan’s government will launch a fact-checking initiative called “Dogru mu?” The logo revealed by Altun on Twitter resembles the one used by Doğruluk Payı, one of the Turkish IFCN’s verified signatories.
- The Duke Reporters’ Lab announced Monday that its fact check tagging system ClaimReview has surpassed 100,000 fact checks. Reporters’ Lab director Bill Adair along with ClaimReview research and outreach coordinator Joel Luther discussed the potential future applications of the schema including automated live fact-checking and as an added feature for smart speakers.
If you are a fact-checker and you’d like your work/projects/achievements highlighted in the next edition, send us an email at email@example.com by next Tuesday.
Any corrections? Tips? We’d love to hear from you: firstname.lastname@example.org.
Thanks for reading Factually.