blog

A guest post from Yajing Xu, a Neuroscience PhD student at UCL, funded by the Wellcome Trust Programme. Her research interests focus on the role of microglia in the postnatal development of the dorsal horn pain circuitries. She is keen to learn more about open science and how to improve the way we are doing science.

The process of how science is produced, from grant applications to publishing can seem very complex. But as a budding scientist, I wanted to know how to navigate these waters, and particularly how to navigate peer review. So when a Peer review: The nuts and bolts workshop by Sense about Science came along, I gladly participated.
The main focus of the workshop was about the need for, the problems of, and the alternatives to peer review. We discussed how this self-correcting process works; it acts as a quality stamp for science – faulty results and bad research designs are filtered out by scientists in the same field. We also discussed the various challenges associated with the process. I wanted to share my thoughts with other early career researchers who might be new to peer review.

Pre- vs post-publication reviews

Traditional peer review takes place before articles are published. There can be several rounds of revisions as other experts scrutinise the findings and sometimes even request additional experiments, meaning this process can be very long – in some cases over a year!. This is not only frustrating for scientists under huge pressure to publish, but also slows down progress and can delay important data from reaching the medical world that could impact treatment of patients.

Some publication platforms, such as F1000Research have now adopted post-publication review to speed up this process. To the scientific community I believe this is a welcome change as most scientists know their field very well, and can make judgements based on the quality of research design and data.

Single-blinded, double-blinded or open review?

The traditional peer-review format has been single-blinded, where authors are known but reviewers are anonymous. Anonymity protects reviewers, but could also allow them to avoid accountability or potentially even stall articles that do not align with their own research. Equally problematic is that the prestige of well-known authors might influence reviewers into more easily agreeing with their findings.

To overcome these problems, journals are experimenting with double-blinded and open peer review. Even with double-blinding, it is often still possible to guess the authors while at the same time failing to address problems of reviewers’ anonymity. Open peer-review might solve these problems but comes with caveats of its own – reviewers are unprotected, and junior scientists might be particularly wary to criticize established scientists.

To me, this highlights the need to change the way people think and treat each other in science, and I hope we arrive some day at the stage where people are respected for their opinion, because they are sound and logical, and not because of their publishing record or status in the science hierarchy.

Moving towards open science

Scientists are under pressure to publish in high impact journals, seen as the main currency of scientific success. This brings with it many problems including the over-emphasis of results and limited resources for replication studies, essential to ensure robust science. Also associated is the problem of access in science, as many high impact journals require a subscription fee. Even the wealthiest universities struggle to keep their subscriptions. I feel science that is financed mostly through public money should ultimately belong to the public. At least, every student should have access to that knowledge. This would not only reduce research inequality between countries, but would also help research to reach a higher impact by reaching more people. I think there is a general trend towards open access, with Wellcome now financing their own open access publications, including a new platform powered by F1000, Wellcome Open Research and I think this process could be accelerated if we reduce the emphasis of ‘impact factor’ of journals as the measurement of scientific importance and success.

In conclusion, peer review is a key part of science, but there is still much to improve with the current system. To me, it seems that many of its problems are linked, whether it is peer review, publication practices, or grant systems. I believe we are moving in the right direction towards more openness within the system and it’s encouraging to see the experiments and innovations I have highlighted being explored. But if there is one thing to pick out that needs urgent change, it is how science is rewarded – we need to reward good lab practices, encourage replication studies and encourage honesty and openness within science.

If you’re an early career researcher and want to know more about peer review and debate challenges to the system, you could attend to the next Peer Review: the nuts & bolts workshop at Glasgow Caledonian University on Friday 4th November. Apply now!