A fact-checker predicted which hoax would resurface — and beat it by an hour

September 12, 2019
Category: Fact-Checking,IFCN

Maarten Schenk has studied fake news and hoaxes so exactingly that he managed to predict a group of trolls’ next post.

The co-founder of the fact-checking site Lead Stories in Belgium keeps several Twitter columns open whenever tragedy strikes so he can study which claims are getting more attention than just a handful of likes or retweets.

He came to realize that every time a mass shooting occurred, trolls on the anonymous message board site 4chan would play a “joke” on the comedian Sam Hyde and frame him as the culprit.

On Aug. 15, after a gunman injured six officers and one pedestrian in Philadelphia, Schenk knew it would happen again. He posted a fact-check of the false Sam Hyde claim an hour before it started circulating again, he told the IFCN.

“As fact-checkers, when there’s something that happens like a hurricane or a shooting or a crisis, we immediately go on hoax-hunting alert,” he said.

In moments of tragedy, crisis or turmoil, misinformation is often heightened. Schenk is not the first fact-checker to note the repetitive nature of false news during such events. A growing body of research offers some explanations, and media junkies have long commented on the trend, but Schenk said he doesn’t think it’s too complicated: “People are lazy, and they don’t understand how Google works.”

When news breaks, people are desperate for information

Research from the University of Washington has confirmed that news with gripping emotional narratives — breaking events like natural disasters, political unrest or mass protests— are more likely to result in misinformation online as people rush to their communities on social media to make sense of things.

The study explains that “rumoring is a reaction to information scarcity and ambiguity, which are common characteristics of disasters; when information is scarce or ambiguous, people try to fill gaps in understanding by creating their own versions of truth.”

Mass shootings are just one example of cases in which misinformation is known to spike and repeat itself predictably. Schenk also noted that politicians or political groups tend to re-use the same old photos and videos to incriminate immigrants whenever xenophobic tensions flare up.

Recently in South Africa, researchers at the fact-checking site Africa Check debunked a video from 2017 that had resurfaced after recent attacks by mobs on foreign-owned businesses in the province of Gauteng.

The video was of the former police deputy minister Bongani Mkongi making false claims about the threat of foreign nationals in Johannesburg, and this was not the first time it’s resurfaced. Naphtali Khumalo, a researcher at Africa Check, told the IFCN that the video started making rounds last month when police clashed with market vendors in a separate incident in Johannesburg’s business district.

“The video is to some extent a political tool, particularly for people who are anti-immigration,” Khumalo said. “The re-sharing (of it) is part of long-standing grievances that most people have about not feeling safe within the inner city.”

Even though the ministry of police retracted and apologized for Mkongi’s comments in 2018, Khumalo expects the video will likely resurface and go viral again.

Not all news is photographed well 

After fires started raging in the Amazon forest of Brazil last month, several media outlets were quick to point out how many notable celebrities and politicians shared completely outdated photos in their posts urging followers to action.

“When there’s a crisis, you’re eager to throw something up, anything,” Schenk said.

He said he believes a big part of the problem is that people will simply search for something like “Amazon forest fire” on Google and post whatever they find without understanding that the results may not all be photos from the recent event.

“It is well known that social media posts with images or video tend to perform better than those without, so people tend to want to ‘spruce up’ their posts with some illustration,” he said. “And unfortunately, not everyone is equally diligent in making sure the images are current and presented in the right context.”

Schenk also pointed out that after the Notre Dame fire, which lasted 15 hours Aug. 15 at the medieval cathedral in Paris, there was noticeably little image-driven misinformation, probably because “it was in the heart of the most touristy place of Paris.” There were more than enough high-definition photos to go around, so internet users didn’t need to rely on fakes.

Pauline Moullot, a journalist from Check News France, confirmed to the IFCN that there had been few fake photos shared during the fire “because journalists and television stations came from all over the world, and were rapidly there.”

Even though real photos and videos were in circulation, these fueled abundant elaborate hoaxes, Moullot said. She noted that the team at Check News “hadn’t seen so many conspiracy theories from one event for a long time.”

So while a rapid availability of photos may lessen the severity of reused image-driven hoaxes, it’s not enough to entirely curb misinformation during a crisis.

When in doubt, repetition breeds familiarity

Recycled misinformation, however, poses a particularly dangerous threat: A 2018 study from Yale University confirmed that the more often someone sees a false headline, the more likely that person is to believe it.

The authors found that “when the truth is hard to come by, familiarity is an attractive stand-in.”

It doesn’t help that breaking news environments are chaotic, and often emotionally wrenching. Paul Levinson, a professor of communications and media studies at Fordham University, told the Las Vegas Sun that “people are hungry for any information” in the wake of a tragedy, so their critical sensors are down.

Schenk added that there are always pranksters, trolls or social media users desperate to raise their engagement who will deliberately post false or out-of-context images to try to get clicks.

“It’s a game of telephone,” he said. “Some guy tweets something, a screenshot of it goes viral, then the top and bottom parts get cut off (in re-posts).”

“People just keep sharing it, even if it’s not real. They just don’t care.”

Daniela Flamini is a freelancer for the IFCN. She can be reached at Factchecknet@poynter.org.