Learning how to read people extends beyond one on one conversation into the realm of politics and media studies.
A recent study published in Science has sought to analyze the ways in which such false stories get promoted and circulated throughout social media. While many news commentators have pointed to the role of automated bots in spreading such “fake news”, Dr. Soroush Vosoughi and his team found that it’s humans themselves that are most responsible for this phenomena.
They looked at the idea of “rumor cascades” by tracking news stories from their origin along a series of “retweets” or shares on Twitter. This allowed them to first look at the origin of a social media story but also the path that story took among its online audience.
Stories were distinguished between fact-checked and verified stories and those debunked as hoaxes and “fake news”. These consisted of many topics, including political stories about elections and urban legends, with political stories being the most commonly shared and analyzed.
The political nature of many of these stories seems quite relevant, as they spiked during particularly salient times such as the 2012 and 2016 elections.
Moreover, false and true stories diffused in markedly different ways. False stories spread much faster in all of the topic categories, reaching more people more quickly than true stories. In fact, controlling for many factors, the average false story was 70 percent more likely to be retweeted than the average true story. This was especially true for political stories.
So far, this has matched the media narratives that warn of the vast diffusion of false news stories, especially during the 2016 election. However, when the media points to the role of bots in promoting these stories, they begin to drift from the truth themselves.
Instead, when excluding bots from the analysis, Dr. Vosoughi and his team found identical results, suggesting that it was not bots but humans who were spreading false stories. Perhaps we have nobody to blame but our own human judgment and inability to properly detect deception. The notion of confirmation bias may be important here, as our brains tend to emphasize what supports our convictions.
The authors declined to offer a firm reason for why this would be the case, but they ruled out the possibility that particularly influential users were promoting false stories. Rather, they emphasized the role of novelty in resulting in assessments that the story was more important. If everyone has already heard a story on CNN, what point is there in sharing it?
In attempting to evaluate this hypothesis, they found that false stories were more generally novel and also that many users expressed surprise in their response. Yet, this is not the whole story, as these same users frequently also showcased reactions of disgust.
While a great deal of questions remain, the proliferation of false news stories emphasizes the applicability of lie detection and people reading in a broader political environment. While we cannot use the same lie detection skills that this blog has discussed at length, Humintell’s Dr. David Matsumoto’s advice is always relevant.