‘Negative’ findings — results that don’t confirm an expected outcome or original hypothesis — have a bad rep in modern science. They’re far less likely than so-called ‘positive’ results to pass peer-review since virtually all journals have a strong bias against such findings. In an effort to raise awareness of the perils such biases have on research, ECNP’s Preclinical Data Forum has announced the world’s first prize for publishing ‘negative’ scientific results.

negative science

Photo Credit: Nevit. CC-BY-SA.

Called the “ECNP Preclinical Network Data Prize”, the €10,000 prize will initially aim to encourage the publishing of negative research in neuroscience. However, if there’s rich interest, the award could be expanded to other fields of science by other organizations.

Unpublished negative scientific results can lead some research groups on unnecessary false tracks. Simply put, if you’d know there’s a good chance the research will come up blank before starting, you might choose not to engage and use limited resources in some other project.  Unpublished data is effectively a waste of valuable real and human capital, particularly in the face of the reproducibility challenge. According to a recent study, irreproducible biomedical research costs the US economy alone $28 billion each year. Over 50% of published biomedical data cannot be reproduced, as startling as that may sound.

“Science is historically self-correcting.  This process is most effective when both positive and negative results are published.  However, negative results are less likely to get published because they are often believed to generate less “value” for an individual scientist, organization or journal. Indeed, compared with the positive data, negative data may appear less exciting, are less likely to open new avenues of research and therefore new funding opportunities,” said Dr. Thomas Steckler, the ECNP Preclinical Data Forum co-Chair, Janssen Pharmaceutica NV, in a press release.

Publication bias ultimately provides a perverse incentive for researchers to either fit their data with a conclusion or abandon their research altogether because it doesn’t fit expectations. According to Nature, social and medical sciences are most affected. After shifting through 221 sociological studies conducted between 2002 and 2012, Stanford researchers found only 48% of the completed studies had been published. Of all the null studies, just 20% had appeared in a journal, and 65% had not even been written up. By contrast, roughly 60% of studies with strong results had been published.

Helping science out of the file drawer

Dr Anton Bespalov, ECNP Preclinical Data Forum Co-Chair, added:

“There are hundreds of drug trials which have failed in the last few years.  Analysis of the factors that led to these failures is very often compromised by the biased representation of the early, preclinical work. The prize aims to emphasize to scientists and academic publishers that there is real value in publishing all the results, not just the headline-grabbing positive results”.

The prize will be awarded by the ECNP’s Preclinical Data Forum. Scientists may apply for the prize starting from Wednesday 8th November 2017, with a closing date of 1st August 2018.

To become eligible, papers have to satisfy the following conditions:

  • Was written in English and was published or is accepted for publication in a peer-reviewed journal that is CURRENTLY listed by Web of Science and Scopus
  • Date of publication is not older than 1 January 2012
  • Reports results of a non-clinical study (or set of studies)
  • Is not a review
  • Is a full-length paper with detailed materials and methods section(s) (within the body of the paper or as supplementary information)

Previously, PLoS One launched its “Missing Pieces” collection of negative, null, and inconclusive studies. Among the featured works, one study found no evidence that suggests support groups help with postpartum disorder in Bangladeshi mothers. Another failed to replicate the findings of four previous experiments for the “depletion model” of self-control, which suggested self-control is a limited resource that runs out. Hopefully, more ‘negative science’ journals and awards will help tip the scales, even for just a little bit.