Last year Twitter attempted to gazump its social media rival Facebook by banning political adverts on the site.  It was the latest shot in the battle to ensure that the content that’s shared via social media is reliable and robust.  A recent study from the University of Texas at Austin highlights the challenges involved, as Facebook users were generally very bad at identifying fake news.

Volunteers were kitted out with a wireless electroencephalography headset whilst they read various political news headlines in a Facebook-style interface.  Each volunteer was asked to determine the credibility of the stories.  Alarmingly, they were only able to accurately do so 44% of the time when the news was aligned with their own political beliefs.

“We all believe that we are better than the average person at detecting fake news, but that’s simply not possible,” the researchers say. “The environment of social media and our own biases make us all much worse than we think.”

Political blind spots

The volunteers were initially quizzed in order to ascertain their personal political beliefs, before then being fitted with the EEG headset and asked to read 50 different political news headlines.  Forty of these headlines were split between being true or false, with the remaining 10 headlines true.

To try and make the process as realistic as possible, the platform attempted to mimic the fake news flags that Facebook has incorporated into the site since 2016.  Whereas Facebook’s flags are determined by a fact-checking process, the volunteers were exposed to flags largely distributed at random.  Each volunteer was asked to rate the articles based upon its believability, credibility and truthfulness.

While the initial results weren’t positive, it was also evident that as the volunteers got into the task, they started spending more time and engaging their frontal cotices more in their analyses of each headline, especially when they believed it to be false.  It suggested a degree of cognitive dissonance at play as the headlines countered their beliefs.

Despite this dissonance however, relatively few participants changed their minds, with most believing that the headlines that conformed to their pre-existing beliefs were true, and those that did not were false.  What’s more, the flag (or not) associated with the headline seemed to make no difference at all.

Indeed, even political affiliation appeared to make no difference, which is in itself interesting as previous research has suggested that conservatives are slightly more susceptible to fake news that liberals.  In this instance however, it appeared to make no difference at all.

The Facebook interface itself didn’t seem to help matters, as it seemed to dilute the ability of participants to think critically as they were in a more passive, pleasure-seeking mindset.

“When we’re on social media, we’re passively pursuing pleasure and entertainment,” the researchers explain. “We’re avoiding something else.”

It’s well known that people often succumb to confirmation bias, but it appears that social media can help to worsen that vulnerability as we often seek a degree of escapism when we venture online.  This can result in a lower willingness to critically scrutinize content we see, and instead search out balm for the mind.

“The fact that social media perpetuates and feeds this bias complicates people’s ability to make evidence-based decisions,” the researchers conclude. “But if the facts that you do have are polluted by fake news that you truly believe, then the decisions you make are going to be much worse.”