Sacha Altay is doing his PHD thesis at the Jean Nicod Institute on misinformation from a cognitive and evolutionary perspective. Emma de Araujo is a master student in Sociology doing an internship at the Jean Nicod Institute in the Evolution and Social Cognition Team. Hugo Mercier is a research scientist at the CNRS (Jean Nicod Institute) working on argumentation and how we evaluate communicated information.
Why do people share fake news? We believe others to be more easily swayed by fake news than us (Corbu et al., 2020; Jang & Kim, 2018), thus an intuitive explanation is that people share fake news because they are gullible. It makes sense that if people can’t tell truths from falsehoods, they will inadvertently share falsehoods often enough. In fact, laypeople are quite good at detecting fake news (Pennycook et al., 2019, 2020; Pennycook & Rand, 2019) and, more generally, they don’t get fooled easily (Mercier, 2020). However, despite this ability to spot fake news, people do share some news they suspect to be inaccurate (Pennycook et al., 2019, 2020). Why would they do that?
One explanation is that people share inaccurate information by mistake, because they are lazy or distracted (Pennycook et al., 2019; Pennycook & Rand, 2019). Indeed, a rational mind should only share accurate information, right? Not so fast. First, laypeople are not professional journalists, they share news for a variety of reasons, such as bonding with peers or having a laugh. Second, even when one’s goal is to inform others, accuracy alone is not enough.
How informative do you find the following (true) news “This morning a pigeon attacked my plants”? Now consider these fake news stories “COVID-19 is a bioweapon released by China” and “Drinking alcohol protects from COVID-19.” If true, the first one might start another world war, and the second one would end the pandemic in a massive and unprecedent international booze-up. Despite being implausible, as long as one is not entirely sure that such fake news is inaccurate, it has some relevance and sharing value.
In a recent article, we tested whether, as suggested above, the “interestingness-if-true” of a piece of information can in part make up for its questionable accuracy. In two pre-registered experiments, we had 600 online participants rating how willing they would be to share a series of true and fake news, and rate how accurate and interesting-if-true the pieces of news were. Participants were more willing to share news they found interesting-if-true, or accurate.
In addition, fake news was deemed much less accurate than true news but also more interesting-if-true.
Our results suggest that people may not share fake news because they are gullible, distracted or lazy, but instead because fake news has qualities that make up for its relative inaccuracy, such as being more interesting-if-true. Yet people likely share news they think might not be accurate for a nexus of reasons beside its interestingness-if-true (see, e.g., Kümpel et al., 2015; Petersen et al., 2018; Shin & Thorson, 2017). For instance, older adults share more fake news than younger adults despite being better than them at detecting fake news, probably because “older adults often prioritize interpersonal goals over accuracy” (Brashier & Schacter, 2020, p. 4).
In the end, we should keep in mind that (i) being accurate is not the same thing as being interesting: accuracy is only one part of relevance; (ii) sharing is not the same thing as believing, people share things they don’t necessarily hold to be true; (iii) sharing information is a social behavior motivated by a myriad of factors, and informing others is only one of them.