(A Fowl Language Comic by Brian Gordon)

Did you know green tea can cure cancer?

Scientists have found evidence that chimps believe in god!

A group of teenagers have invented a condom that changes colour when it detects STDs!

These three statements have something in common: they’re all a load of crap and yet are the kinds of headlines that are widely reported and circulated by a number of news outlets. Yes, even that one about scientists finding evidence that chimps believe in god.

Websites such as the HuffPo, Buzzfeed, and Cracked are often cited as examples of publications that have built media empires around a strategy of clickbait, with social media platforms such as Twitter and Facebook – which encourage people to widely share headlines and images that are often misleading or decontextualized – fanning the flames of sensationalism and misinformation.

From civic engagement to popular music, many topics have been described by concerned commentators as undergoing a “meme-ification,” which refers to an apparent ‘dumbing down’ of whatever material is at stake. Online, shareable media content, the argument goes, tends to overly simplify information. The term ‘meme-ification’ references Internet memes, which are often defined as any digital artifact that gains a high level of popularity.[i] When commentators draw attention to ‘meme-ification,’ they typically identify blog posts, list and picture-based web articles, and other online content as contributing to misrepresentation, decontextualization, or oversimplification.

For issues relating to science and health this kind of misinformation can have particularly grave consequences. What are the implications of increasingly constructing scientific knowledge and discourse according to a ‘spreadable’ format?

(IFLScience logo)

A case study that can shed some light on this: the IFLScience Facebook page and website.

IFLScience was created by Elise Andrew In 2012, a Facebook page that was initially used to share “bizarre facts and cool pictures” with a small number of the creator’s friends.[ii] Within a day the page had accumulated over 1,000 “likes,” and after seven months it had received over one million. Today it has almost 20 million, and the website IFLScience.com – created by Andrew in November 2013 – receives 45 million unique visitors per month.

Initially the IFLScience Facebook page mostly distributed jokes, cartoons and memes with a science theme, but quickly began “including newsier finds” such as “the discovery of an enzyme that could produce hangover-free beer…” or a study showing that 14 adults had been ‘functionally cured of HIV’”. This content is widely shared on Facebook, ‘liked,’ commented on, and spread across other social media platforms such as Twitter; the IFLScience twitter account has over 144,000 followers. In light of the ever-increasing popularity of the cross-platform IFLScience brand, including a series of programs being produced for the Science Channel and a partnership with Discovery Communications to create online videos through YouTube, it becomes apparent why Andrew has been described as “journalism’s first self-made brand.”[iii]

This self-made brand, however, has been facing mounting criticism for its often misleading stories and headlines.

Thanks To Reduced Solar Activity, We Could Be Heading For A ‘Mini Ice Age’ In 2030.”

No, We Aren’t Heading Into A ‘Mini Ice Age.‘”

There Probably Won’t Be A ‘Mini Ice Age’ In 15 Years.”

The ‘Mini Ice Age’ Hoopla Is A Giant Failure Of Science Communication.”

Those are all headlines from IFLScience. The research presented in the first story – which claimed there was an impending mini ice age – did not actually support this sensationalist headline.

The IFLScience headline, “Marijuana Contains “Alien DNA” From Outside Of Our Solar System, NASA Confirms” was actually an article discussing the fact that many people shared articles from the site after just reading the headline, and not the actual content. Yet despite apparently being aware of this problem, IFLScience has continued building its brand on clickbait and sensationalism.

Clickbait science journalism is often not a matter of shoddy science; typically, the culprit is a sensationalised headline or a writer misinterpreting the results of the original study – a kind of broken telephone problem.

Case in point: an article in The Guardian (just to stop picking on IFLScience for a moment, and to show how pervasive this problem is) was headlined “Indigenous Australians most ancient civilisation on Earth, DNA study confirms.”

The original title of the study: “A genomic history of Aboriginal Australia.”

The original abstract:

Here we generate high-coverage genomes for 83 Aboriginal Australians and 25 Papuans. We find that Papuan and Aboriginal Australian ancestors diversified 25–40 thousand years ago. However, all of the studied Aboriginal Australians descend from a single founding population that differentiated ~10–32 kya. We infer a population expansion in northeast Australia during the past 10,000 years associated with limited gene flow from this region to the rest of Australia. We estimate that Aboriginal Australians and Papuans diverged from Eurasians 51–72 kya, following a single out-of-Africa dispersal, and subsequently admixed with archaic populations. Finally, we report evidence of selection in Aboriginal Australians potentially associated with living in the desert.

Of course, sometimes the science itself is to blame. One researcher, demonstrating how this kind of practice unfolds, recounts how he “fooled millions into thinking chocolate helps weight loss.” The author conducted a real study, designed as a sort of exposé on the junk-science diet industry, in which real human subjects were put on different diets. People put on a diet including a daily dose of chocolate lost weight 10 percent faster, a statistically significant difference; this was widely reported by a number of outlets.

The problem? As the author describes, measuring a large number of variables from a small number of people pretty much guarantees that you’ll get a statistically significant result. The study included 18 measurements, from weight and cholesterol to sleep quality and blood protein levels, in 15 people. The researchers didn’t know whether the chocolate test group would show improved cholesterol, lowered blood pressure, or improved sleep, but they knew at least one “statistically significant” result would pan out.

Misinformation spread through the news media is nothing new of course, but websites and Facebook pages such as IFLScience, and social media platforms such as Twitter, exacerbate this problem.

As a final point, here’s a link to a study that shows how less than 2.5% of people actually click on links to original scientific publications in news stories.

(Image by Brendan Boughen)

[i] Marwick, A. (2013). Memes. Contexts, 12(4), 12-13.

[ii] Hudson, R. (2012, September 24). Interview with Elise Andrew. Retrieved February 18, 2015, from http://www.thechemicalblog.co.uk/interview-with-elise-andrew/

[iii] Fitts, A.S. (2014, September 2). Do You Know Elise Andrew? Columbia Journalism Review. Retrieved from http://www.cjr.org/cover_story/elise_andrew.php?page=all