Exoplanets, Life, and the Danger of a Single Study

There’s value in covering new research advances, even when the underlying science is unsettled. But there are also risks.

Republish
  • An artist’s impression of K2-18b, a super-Earth exoplanet whose habitability has been debated by scientists.


VIEWPOINTS: Partner content, op-eds, and Undark editorials.


The recent announcement that scientists discovered water on the planet K2-18b, 110 light years away, prompted a media swoon. News stories, including a piece written by me, billed it as the first detection of water on a “potentially habitable” planet outside our solar system.

The blowback from the astronomy community was swift. A chorus of critics stated on Twitter that, although K2-18b orbits its host star within a distance range astronomers call the habitable zone, the planet is most likely too hot and under too much pressure to support life.

The sentiments expressed in a Scientific American essay by Harvard University astronomer Laura Kreidberg were typical of many in the community. Kreidberg’s piece suggested that news outlets were “crying wolf” and that scientists, press officers, and the press had all contributed to the misreporting of the story.

But in describing K2-18b as a potentially habitable planet, journalists were accurately reporting the views of the scientists who led one of the research studies. Those scientists repeatedly stated to reporters that the planet was “potentially habitable” — and they continued to say so when the specific criticisms of their peers’ were put to them.

The episode highlights a longstanding issue: How should we science journalists cover incremental research advances, especially when the underlying science is unsettled?

Recent history gives us numerous examples of how these so-called single-study stories can go wrong. Among the most notable was the coverage of a 1998 Lancet paper in which Andrew Wakefield and coauthors proposed that a combined measles, mumps, and rubella vaccine was linked to autism and bowel disease. Another high profile case involved papers published in Science in 2004 and 2005 in which a group led by Hwang Woo-Suk claimed to have cloned human embryonic stem cells for the first time.

Both research efforts were later discredited. In the case of the Wakefield paper, the selection of the sample group was biased, there were undisclosed vested interests, and the authors made several claims that did not stand up to scrutiny. The paper was retracted in 2010. Woo-Suk’s papers were simply fraudulent.

I’d suggest that both of these cases are outliers; the media was duped — as were the journals themselves — by researchers who had an agenda or were simply dishonest about their results. Even when scientists act in good faith, however, things can go wrong. It is legitimate to ask those at the highest level of their profession to give their view on their own work, even if that view is speculative and at odds with what their rivals and colleagues have to say. But there are a few provisos.

First, a journalist should always reflect dissenting views. Sadly, much if not most coverage of the K2-18b story failed to do this, as is all too often the case with other single-study stories. (My own K2-18b coverage came under criticism for initially not including enough dissenting voices.) Many journalists believe that if research has been published in a peer-reviewed journal, it must be credible, and they make the mistake of reporting the research uncritically.

As I suggested in an article for BBC News, it is our job as science journalists to challenge what we are told. This is especially the case now that more of us report on controversial topics such as genetically modified crops, cloning, and climate change, which have complex political, as well as scientific, dimensions.

But even in the case of basic science discoveries — in the realm, say, of anthropology or dark energy — there is often plenty of debate. The standard narrative, “people used to think x and now, because of this discovery they think y,” is not the way science works, and it quite frankly makes for boring copy.

A second proviso is that all voices are not equal. The views of people who are not qualified in the particular area of research in question carry less weight than those of people who are. Like many serious journalism platforms, BBC News, where I work, has a strict policy of balance and impartiality. In the 1990s and 2000s, that policy led many of our programs to “balance” scientific voices warning of climate change or reassuring people about the safety of vaccines with the voices of people arguing the opposite.

The contrarian views would often come from pressure groups such as climate change deniers and anti-vaxxers or from scientists commenting on a specialty different than their own. The BBC’s science and health correspondents argued strongly against the policy, which had been at the heart of the organization’s journalism. In 2010, this led to a change in editorial guidelines, the new stance being that “due weight” should be given to the scientific consensus on any given subject before reflecting contrary views. 

A third and final proviso is to beware of science journalism’s oldest enemy: hype. Journalists should be wary both of researchers’ natural enthusiasm and of their sometimes-deliberate efforts to drum up publicity in order to secure research funding. In academia, cures for cancer, a new understanding of physics, and bottomless sources of clean energy are seemingly always between five and 10 years away.

The most recent example of a new technology being touted as an answer to all our problems is gene editing. In August 2017, Shoukhrat Mitalipov and colleagues reported in the journal Nature that they had successfully repaired a gene associated with a rare heart condition in a human embryo. The research has since been challenged, but in reporting the development, Britain’s bestselling daily newspaper, The Sun, loaded on the hyperbole, writing that “the revolutionary work … could help end 10,000 hereditary illnesses including cancer” and that “scientists say it could signal the end to inherited diseases.”

Seventeen years earlier, British tabloids were saying the same thing about gene therapy. A story in The Daily Express’s Sunday Review in July 2000 asked: “Could we be on the verge of the greatest medical advance ever seen, even greater than the defeat of smallpox or cholera — the defeat of time itself?” Similarly rosy predictions were made in 2007 about how microbes would provide an “endless” supply of biofuel.

Such credulous reporting may be becoming more prevalent, ironically, because of scientific institutions’ efforts to be helpful to science journalists. Organizations such as the U.K.’s Science Media Center have sprouted up to coordinate the dissemination of press releases and other resources to journalists, often with the stated aim of countering misinformation. We science journalists get ideas, our editors get happy, uplifting stories, and the public gets a warm glow in its heart.

But an information pipeline that runs uninterrupted from scientists to press officers to the news media puts us at risk of another kind of misinformation. A great science story counts for nothing if it gives readers a misleading impression or paints a cartoonish, one-dimensional picture of how science works. Such stories are their own brand of fake news. In writing them, we do neither the scientists, their press officers, nor our readers any favors.

When I started as a science journalist in the 1980s, single-study stories were the norm. Our job was to translate complex scientific information and artfully explain its significance to a non-scientific audience.

But many of us saw a responsibility to do more: to challenge, weigh, and assess the tablets of stone we were handed from omniscient researchers and to put them in a societal context. In other words, we became journalists, using our own skills and experience to add value and provide an important civic service.

My sense, though, is that because of staff and budget cuts, the extra time and effort needed to fulfil that role are seen by editors as luxuries, and so single-study stories are on the increase.

Perhaps in 20 years’ time scientists will have confirmed that K2-18b really is habitable. Until then, let’s hope that science journalists will have the time and the self-confidence to listen to a range of views, and to give their own perspectives.


Pallab Ghosh is an award-winning science correspondent with BBC News. He works across television, radio, online, and digital platforms.

See What Others Are Saying

6 comments / Join the Discussion

    I really miss the emails from Charles Knight, the human aggregator. 1. Here are the stories. 2. Here is how they were covered. 3. This is who got it right.

    Reply

    American mass media is full of really awful science journalism, often it is contorted into marketing. Most editors have no science background at all, they are marketers first and foremost.

    Reply

    I think people know the score with artists impressions. I think its those of science journalists that we are discussing. What do you think? Please join in the conversation.

    Reply

    I’m not so sure that people do know the score with ‘graphic artists’, and that’s because the graphic images often come directly from the science groups or agencies themselves. Take the recent story: “Supernova Explosion Captured In Video Form For The First Time.” The most impressive part was the short NASA video of the blast wave – in colour – surging away from the core. This turned out (but was not widely reported) to have been reconstructed from single-frame B&W images captured by the Kepler telescope. Which it turns out takes frames at thirty-minute intervals. According to the lead researcher, “the shock breakout itself lasts only about 20 minutes”. Yet these data allowed NASA to produce a high-definition colour video of “the initial moments of a supernova.”

    Reply

    Mr Leifert in not correct when he says “we have not yet seen a single exoplanet”. Direct imaging is an extremly difficult method of observing exoplanets but it is doable and NASA now has 47 planets listed in its archive as being discovered by direct imaging. The list can be viewed at this URL by putting Ïmaging” in the filter for Discovery Method

    https://exoplanetarchive.ipac.caltech.edu/cgi-bin/TblView/nph-tblView?app=ExoTbls&config=planets

    To see a typical image Leifert should Google HR 8799e.

    I don’t begrudge graphic artists their fun in imagining what planet is like.

    Reply

    My gripe is with the illustrations that often accompany reports of newly discovered exoplanets. I am sure that typical average readers think of them as photographs (not having digested the caption or credit line) and do not know that we have not yet seen a single exoplanet, much less know what kind of topography and atmosphere it may have. I don’t know why such “artist’s impressions” are commissioned.

    Reply
Join the discussion

Your email address will not be published. Required fields are marked *

Top

Whistleblowers
& Tipsters

Corruption in science?
Academic discrimination?
Research censorship?
Government cover-ups?

Undark wants to hear about it.

Email us at tips@undark.org, or visit our contact page for more secure options.