Primary tabs

There is a depressing quote concerning the ability of false information to outpace a more verifiable counterpart among human beings. Furthermore, it has many versions and variations, some of which are attributed to more than one author. However, it appears that including a wrongful citation would ensure that it would be repeated far earlier and far more often. This would be in keeping with the message of the quote – and would also be substantiated by scientific evidence. A new study has found that stories based on fallacious information makes its way to more Twitter users at a greater rate than a comparable piece of actual real news. These worrying findings, reported by researchers at MIT, indicates the risks posed by the now-firmly established phenomenon of ‘fake news’ and why it seems so influential in the world today.

The paper was published in Science and written by Soroush Vosoughi (a PhD student at the time of the study); Deb Roy, the director of MIT’s Laboratory for Social Machines (LSM) and Professor Sinan Aral of the institute’s School of Management. They decided to focus on how (supposedly) informative links spread through the social media platform Twitter, since it has become such a prominent source of news for so many people. They used the ‘flow’ of approximately 126,000 tweets presented as news (i.e. ones that asserted a claim of any kind) through the entirety of the site (i.e. the ‘twittersphere’). This ‘flow’ was defined by contiguous re-tweets of the same tweet.

 

 

The total number of ‘news tweets’ was associated with over 4 million cumulative tweets over 11 years by over 3 million users. They were tested for veracity using six prominent fact-checking online services: urbanlegends.about.com; truthorfiction.com; snopes.org; politifact.com; hoax-slayer.com and factcheck.org. The researchers found that the magnitude of ‘flow’ could be 10 to 20 times greater for tweets found to be false compared to that for verifiable ones. Furthermore, they found that ‘false’ posts were 70% more likely to be re-tweeted compared to ‘true’ ones. ‘Real’ news tweets also took six times as long to reach 1,500 users compared to ‘fake news’.

The researchers found that these dissemination rates were not affected by the topic or content (e.g. including images or links) of either tweet type. The researchers also reported that false tweet ‘dynamics’ were more likely to be influenced by virality compared to ‘real’ tweets. For example, false tweets of a political nature were much more viral. In addition, these false-news tweets were significantly superior in terms of the breadth and depth of their ‘diffusion’ into the Twittersphere.

This piece of information may lead to a conclusion that ‘false’ tweet dissemination may have been enhanced by ‘bots’. These artificial users are powered by semi-autonomous programmes that can carry out a large number of tasks, including the posting of tweets at specific or regular intervals. However, the MIT team were able to correct their analysis for bots and found that their results were not significantly affected by their exclusion.

Pictured (left to right): Seated, Soroush Vosoughi, a postdoc at the Media Lab's Laboratory for Social Machines; Sinan Aral, the David Austin Professor of Management at MIT Sloan; and Deb Roy, an associate professor of media arts and sciences at the MIT Media Lab, who also served as Twitter's Chief Media Scientist from 2013 to 2017. Photo: Melanie Gonick, MIT

Pictured (left to right): Seated, Soroush Vosoughi, a postdoc at the Media Lab's Laboratory for Social Machines; Sinan Aral, the David Austin Professor of Management at MIT Sloan; and Deb Roy, an associate professor of media arts and sciences at the MIT Media Lab, who also served as Twitter's Chief Media Scientist from 2013 to 2017. Photo: Melanie Gonick, MIT

Therefore, it may also be reasonable to expect that the type of ‘real’ Twitter user may also have influenced the spread of news-tweets. For example, the frequency of tweeting or number of followers may have been associated with the ability to disseminate individual posts. However, the researchers also corrected for these factors, as well as the number of users followed and age of Twitter account per user. In addition, the team also examined whether the possession of a blue tick affected the results. They found that those with significantly fewer followers and followees were more likely to spread ‘false’ news tweets. Those who took part in the dissemination of false tweets also spent significantly less time on Twitter and were significantly less likely to have blue ticks.

The re-tweeting of ‘false’ tweets was also not by the status (in terms of verification, follower count, follower count or age) of the user who originally posted the tweets in question. Therefore, the spread of false news tweets was not affected by user type, or by the type of Twitter network built up around individual users. Instead, the researchers explained their findings through the factor of novelty. In other words, there is an overwhelming drive to be the one to share unique or exceptionally interesting information online.

Therefore, this impulse was behind the tendency to tweet or re-tweet claims without pausing to verify them – or so the team reasoned. To test this, they extracted a random selection of 25,000 tweets shown to 5000 random users. They assessed the tweets with a novelty metric and found that ‘real’ tweets were significantly less likely to be novel than false ones. The team also found that false tweets were more likely to be met with replies indicating emotional responses such as surprise compared to real ones. False tweets were also more significantly more likely to elicit responses with an element of disgust, whereas real ones were significantly more likely to elicit responses including sadness and anticipation. However, they were also significantly more likely to evoke joy and trust based on the same metrics.

This new study from MIT may be deeply concerning for those involved in civic affairs and how information on Twitter affects them these days. It is also possible that its results may be exploited by those whose interests lie with the spread of what could be termed ‘fake news’. On the other hand, it could also inform efforts to improve the dissemination of accurate, fact-based information online. Finally, it indicates that humans, not bots, drive the spread of potential fallacies online, and that they do so for essentially human reasons.

Top image: Person using twitter. (Public Domain)

References

Vosoughi S, Roy D, Aral S. The spread of true and false news online. Science (New York, NY). 2018;359(6380):1146-51.

Remus S. On Twitter, false news travels faster than true stories. EurekAlert!. 2018. Available at: https://www.eurekalert.org/pub_releases/2018-03/miot-otf030718.php

Deirdre's picture

Deirdre O’Donnell

Deirdre O’Donnell received her MSc. from the National University of Ireland, Galway in 2007. She has been a professional writer for several years. Deirdre is also an experienced journalist and editor with particular expertise in writing on many areas of medical science. She is also interested in the latest technology, gadgets and innovations.Read More

No comment

Leave a Response