This post is also available in: Spanish
I spoke about cognitive biases in my last post, highlighting how important is to double-check anything before publishing. Well, those among you who enjoy the search for reliable information surely share the feeling that it is like walking blindfolded along a path full of obstacles. I admit that even having a scientific career, I’ve found myself falling like a novice into some information tricks that I thought I was over with. And believe me, I try hard to avoid them because I’m rather convinced that disrupted information is to blame for the immaturity and oversimplification that threaten the well-being of our society. Many of the facts we take as certain are an accumulation of errors that build up with each step they follow from their origin to us as recipients. So this is a three-posts dive into the dark side of the information triangle (origin-messager-recipient), starting with the origin: Science.
But I don’t want to start lambasting science without making clear that it is our most reliable source of information, and the best available system to improve human knowledge and dust off biases and beliefs from observational facts: It boasts a solid methodology, it’s got ways to prevent biases and test itself, and it’s continuously updated and scrutinized. Thanks to science we’ve managed to dominate and understand our environment as well as anticipate disasters to take measures against them. Human history can’t be understood without science.
Furthermore, it is important to be aware that science doesn’t provide undeniable truths. It is knowledge that evolves continuously, it isn’t perfect nor fixed. Actually, a whopping 9 out of 10 studies are estimated to have relevant mistakes, and some of them are unavoidable and part of the normal development of science. Besides, some science products, such as models, are necessarily wrong because they are simplifications of complex processes, but that doesn’t mean they’re not useful.
But there’s room for improvement in many areas. Sadly, some are rooted in the system itself, that presses harshly to publish, not minding as much the quality as the number of publications in terms of funding or boosting a scientist’ professional career. Negligence or misconduct among scientists, journals or reviewers find there a fertile soil to grow.
As if that were not enough, the independent reproduction of experiments is rarely funded, as the results for that kind of test aren’t likely to be published unless they disprove the original paper. Negative results (like a vaccine that doesn’t work) are also neglected by journals, although they would complete the information given by positive studies and would prevent other scientists from repeating the same experiments. Fortunately, there have been some improvements in the latter case.
From a wider perspective, there is an imbalance in the targets of research and in the publications. There are many studies about some specific aspects that we lack from others at the same level, and that might affect the way we perceive the information. Cheap/easier studies, those that provide direct benefits, or that are culturally relevant or in agreement with our views (as an example, clinical trials have been done traditionally on white males and the results were extrapolated to the rest of the population) are prioritized and more likely to be published. The hotter a study is, the more funding and quality is expected, and wherever there’s an economical prospect there are private studies from partial entities.
Will all that in mind, it’s natural and healthy to feel skeptic about scientific studies, because not everything in science is reliable. However, filtering science involves searching specific information full of jargon in journals that you often have to pay for, and that isn’t everybody’s cup of tea. If despite that you’re determined to dig down to the roots, here’s a chart that might be helpful:
At the end of the day, each of us will find the shortcuts that are more in-line with what we need from the triangle science-messager-recipient, depending on our interests. There are tools that can be handy for relevant fields where we want to go deeper, such as reviews (that condense the state-of-the-art of a topic in a paper), scientific reports, or specific outreach from scientists about their area of expertise (like http://www.realclimate.org/ ). A more attractive approach is given by popular science writers, that provide a general and easy-to-read approach to research studies and play an essential role as middle-men between science and the general public….but that’s a story for another post.