Competition in science

Driving science forward or a waste of resources?

Alexis Barr 10 May 2009

Neck-in-neck: parallel research is rampant

The scientific community is a fine example of survival of the fittest - only a select few have the attributes to be successful

Editor's note: We are pleased to present this essay by cancer researcher Alexis Barr, which was recently shortlisted for the BSCB's Writing Competition.

Have you ever had that heart-stopping moment during a routine search of recently published work when a paper appears that pretty much covers everything you have been working on for the past few years? When complaining about the fact that I had been beaten to publication to a non-academic friend of mine, my friend replied, “Why were you both working on the same thing in the first place?”

The answer in my case was that we did not know that another group was working on the same thing, but the question got me thinking. Why do we seemingly waste time, effort and money duplicating research? And is it a bad thing?

The aim of biological research is to increase our understanding of living systems. It thus seems bold that we would base our understanding of processes on the evidence provided by one experiment from one group. But if two groups independently reach the same conclusion then the evidence becomes harder to dispute. In a similar way, duplication of work can ensure that fraudulent publications are eliminated from the system. Care has to be exercised here: if one group’s data disagrees with that of another group does it necessarily mean that one set of data is false? Or is there a good reason why the two data sets should disagree? Indeed, this can lead to very interesting (and often entertaining) scientific debates about why the data may differ. However, when several groups disprove another group’s data, then you can start to question its quality.

This ultimately leads to the problem: if two groups are trying to solve the same problem, who will get the credit for it? One option is to publish back-to-back papers. However, not everyone is willing to ‘share the glory’. It can then become a race to publish first. In a bid to save time, corners may be cut in experiments and in the interpretation of data. Inevitably, in some cases this will lead to bad science and to the publication of false data. In other cases, the drive to publish first can help to concentrate the mind of the researcher and hence focus the project on the key experiments that need to be done. In this way, competition can be healthy, getting data out faster, disseminating knowledge and thus driving science forward.

The most galling situation is if you are unaware that you are in competition and are beaten to the answer. Attendance at conferences is the most useful way to find out what other people are working on. However, even then, people can be reluctant to share unpublished data. Much of the data presented at conferences is already in the final revision stages for publication, giving you, at most, a one month lead time over people who did not attend the conference. This is understandable to some extent – people do not want others to steal their ideas. On the other hand it can lead to extremely boring conferences where people are essentially either presenting data that is already out in the literature or presenting unpublished work in a completely useless fashion, where “gene X does this to gene Y” without revealing their actual identities. What conferences should be about is the presentation of early data and project ideas, so that researchers can get feedback and help and potentially set up useful collaborations to answer their research questions. In this way, conferences could promote collaboration instead of nurturing suspicion and competition.

So is it really a waste of resources to work on something you know someone else to be working on? The easy answer is yes. Why waste your precious, hard fought-for research grant on something someone else might take the glory for? It is simply too high a risk for most scientists – particularly those at the start of their career. If another group is much further on in their work and you are just beginning then of course it is futile to try and compete. However, if you only find out halfway through a project that you are competing with another lab, would you really want to give up all the hard work? This is a particularly difficult choice to make if you know your competitor has more money/people/resources at their disposal.

One choice is to try and collaborate, but this is not always an option. I would guess that most scientists would not give up – to leave a question unanswered when you have the tools to answer it is almost unspeakable. We all want that Eureka! moment for ourselves. And what if you did give it up and when your competitor did eventually publish it was inconsistent with your data? An alternative to trying to compete directly is to find a novel way of answering the question, perhaps by using, or developing, a new technology to give your data the edge over the competitor. Again, this will drive scientific progression forward by helping others to answer their research questions.

Inevitably, people will work on the same problems because we all want to answer the important questions. If people communicate they can potentially draw on a larger body of evidence and reach more solid conclusions. However, not everyone will play fair and someone always has to speak first.

The scientific community is a fine example of survival of the fittest. Only a select few have the attributes to be successful group leaders and competition in science is one way to determine who is the best. Inevitably, some people will lose out by not being able to publish their own work. However, those who are destined to be at the top of their field will have other ideas up their sleeves – not all of which will be being worked on by someone else.

Related information

The winning entry from the BSCB Writing Competition was also published on LabLit, along with more information about this year's contest.