Science builds our modern world because it can overcome intuitions, gut-feels, common sense, and prejudices that happen to be wrong. The scientific method allows us – flawed and limited human beings – to grasp objective reality from the infinitesimal to the cosmic.
But human nature fights science.
NPR recently did a piece on an important problem in science – ignoring negative results. Well-known research about bilingualism, for example, exposes “a flaw in how scientific research reaches the mainstream.”
The bilingual hypothesis proposes that speaking multiple languages has benefits to your brain that improve multitasking. Positive results come from a 2009 study. But a researcher involved in the study says there were actually four studies done – one showed a positive effect and the other three showed no effect. Only one was published. Guess which one?
“In fact, one of the authors of the 2009 paper tried to replicate the experiment that found a positive benefit for bilingualism. And that replication failed to work. So in other words, there were four experiments. Three did not show benefits, and they weren’t published. One showed a benefit and was published. But it couldn’t be reproduced. And then the reproduction was not published.”
Any reasonable reading of the results says the hypothesis that bilingualism benefits the brain in this way has been, if not falsified outright, at least not supported.
A well-designed and conducted study that yields negative results – that does not prove its hypothesis – is as valid, important, and useful as a positive study. But no one likes to publish negative studies – it’s not exciting, doesn’t get you tenure, doesn’t make the news, doesn’t fill the room at a conference. Even if the original paper had been withdrawn, not everyone would get the message. It would still be cited and the popular press usually ignores retractions. You and I – interested laymen – would still have the wrong information unless we go out of your way to check in places like RetractionWatch.com.
This problem holds science back and can have a direct impact on our lives. How can medicine advance if only part of the research is published? I’ve posted about this before. There’s a group in the UK calling “for all past and present clinical trials to be registered and their full methods and summary results reported.” I hope they succeed worldwide.
What we need are prestigious journals of negative results. Every major journal should start a subsidiary that does just that.
PS: Sometimes negative results do get into the news. As fivethirtyeight.com concludes, it’s very unlikely that cell phone use increases cancer rates. That’s good to know. I feel better.