False news travels far faster and extends further than the truth on social media, researchers have found in the largest scientific study of the way stories spread online.
Scientists at Massachusetts Institute of Technology found that falsehoods were 70 per cent more likely to be retweeted on Twitter than the truth — and true stories took six times longer on average to reach 1,500 people than falsehoods.
The study analysed 126,000 stories between 2006 and 2017, verified by six independent fact-checking organisations as either true or false news, which were tweeted 4.5m times by 3m people. The results are published in the journal Science.
Sinan Aral of MIT, the senior author, told the Financial Times: “I was really surprised by the sheer magnitude of the way false news travels faster than the truth.”
But the study also found that automated bots [web robots] do not increase the spread of false news over true stories.
“We found that, contrary to conventional wisdom, bots accelerate the spread of true and false stories at the same rate,” said Professor Aral. “False news spreads more than the truth because humans, not robots, usually tend to unfold it.”
Twitter, which is investing $10m in MIT analysis over 5 years, funded the challenge.
“The corporate gave us entry to the info however that they had no enter into the research design and evaluation,” mentioned Prof Aral. “They learn our [Science] paper when it was completed however weren’t concerned within the analysis.”
Though Twitter was not ready to touch upon the MIT researchers’ particular findings on the quick unfold of false information, the corporate mentioned the research was an instance of the best way it wished partnerships with extra with outdoors specialists “to assist us establish how we measure the well being of Twitter”.
It added: “Twitter’s open and real-time nature is a robust antidote to the spreading of all sorts of false info.”
In line with MIT, the challenge was much more intensive than earlier research of the unfold of false information, which had been both case research of specific tales or “analyses of small advert hoc samples”. The researchers intentionally averted utilizing the contentious time period “faux information” in favour of “false information”, which is less complicated to confirm objectively.
Twitter customers had been extra prone to unfold false information in the event that they had been usually much less lively on the positioning and had fewer followers than common, the MIT research discovered. Falsehood subtle sooner and additional than fact “regardless of these variations moderately than due to them”.
The researchers concluded from intensive evaluation of tweets that the dissemination of false information is principally pushed by a love of novelty. Individuals are extra prone to repeat information that surprises them — and false tales are usually stunning and novel in contrast with the reality.
The MIT authors say: “Understanding how false information spreads is step one towards containing it.” One conclusion from the research is that the present US debate on fake news is mistaken to focus a lot on bots, when human behaviour is definitely much more necessary. Options may embody discovering a dependable approach to label information as true or false which individuals would belief.
One other paper in the identical subject of Science by 15 US social media specialists requires a co-ordinated and large-scale investigation into the social, psychological and technological forces behind faux information. “What we wish to convey most is that faux information is an actual drawback, it’s a tricky drawback, and it’s an issue that requires severe analysis to unravel,” mentioned Filippo Menczer, founding father of the Indiana College Observatory on Social Media.