How to cool off hot results
In this week’s Nature, Vincent Detours raises the concern that the now-common practice of publishing papers online ahead of print (AOP in Nature-speak) has the potential to leave high profile results dangling out in public without any level-headed, calm and cool commentary to guide the ensuing media consumption of the hot result. He therefore suggests that “editorial comments” should accompany any controversial findings, to place the matter in context and to act as an informative back-up to the press release.
Many Nature papers are already accompanied by press releases, short “layman’s summaries” that are meticulously written by our hard-working and very talented press office. These concise blurbs strenuously implore the would-be reporter to take care in the interpretation of controversial results. Having been to internal meetings discussing these types of issues, I can assure you that the press office takes any potential for “spin” or misinformation very seriously. Hence, the organization of press briefings with the controversial manuscript’s authors upon the release of some papers. However, the way I see it, less scientifically-impactful papers that have a particular social appeal are more likely to be misrepresented in a subsequent news story.
Some of these studies may report, for example, a correlation between a human trait and brain activity (such as that measured using fMRI) and because of the lack of mechanism or clear physiological connection, the study ends up in a lower-tier journal, published as an interesting observation. This is until the headlines scream Love makes you light up — even in your brain, researchers say. Even though the article doesn’t necessarily state that scientists can read your mind, I am willing to bet that quite a few CNN.com readers walked away from that article wondering where they could get one of these newfangled fMRI machines to learn (once and for all) what their distant spouse truly felt about them. The danger of reading too much into such correlations using fMRI results is discussed here.
[As an aside, this same sort of gap between correlation and causality was a significant issue in the infamous NY Times Op-Ed piece on the mindset of the swing voter. I took this to task, as did a Nature editorial, Slate magazine, and an article in the Dana Foundation’s online journal Cerebrum, among others…]
Why would the risk be greater for these correlatory studies to be a victim of irresponsible reporting? Possibly because the society journals and other solid specialist journals that tend to publish them do not have the budget for a full-time press office, making it more likely that the results will be released without an accompanying editorial perspective.
But I am not convinced that having some editorial clarity, whether provided by the journal editors or any other expert in the field, necessarily deflects all sensationalist writing by media outlets. If a reporter finds an angle s/he likes and has a flexible editor, the spin within a story can easily take over. There are many different types of reporting gone awry, from the sensationalist to the lack of reporting potential conflicts of interest. Although I see the point of Dr. Detours in relation to the responsibility of the high-impact journal, I feel that it is only a very small part of a greater problem involving reporting responsibility. Fancy headlines and stories can sell more papers, but shouldn’t misrepresent the scientific facts or limitations for a lay public that I do not expect can “do their homework” and examine the raw data for themselves before formulating an opinion.