The study, published in the Proceedings of the National Academy of Sciences USA, suggests that scientists pay less attention to theories that are dense with mathematical detail.

Researchers in Bristol¡¯s School of Biological Sciences compared citation data with the number of equations per page in more than 600 evolutionary biology papers in 1998.

They found that most maths-heavy articles were referenced 50 per cent less often than those with little or no maths. Each additional equation per page reduced a paper¡¯s citation success by 28 per cent.

The size of the effect was striking, Tim Fawcett, research fellow and the paper¡¯s co-author, told Times Higher Education.

É«ºÐÖ±²¥

¡°I think this is potentially something that could be a problem for all areas of science where there is a tight link between the theoretical mathematical models and experiment,¡± he said.

The research stemmed from a suspicion that papers full of equations and technical detail could be putting off researchers who do not necessarily have much mathematical training, said Dr Fawcett.

É«ºÐÖ±²¥

¡°Even Steven Hawking worried that each equation he added to A Brief History of Time would reduce sales. So this idea has been out there for a while, but no one¡¯s really looked at it until we did this study,¡± he added.

Andrew Higginson, Dr Fawcett¡¯s co-author and a research associate in the School of Biological Sciences, said that scientists need to think more carefully about how they present the mathematical details of their work.

¡°The ideal solution is not to hide the maths away, but to add more explanatory text to take the reader carefully through the assumptions and implications of the theory,¡± he said.

But the authors say they fear that this approach will be resisted by some journals that favour concise papers and where space is in short supply.

É«ºÐÖ±²¥

An alternative solution is to put much of the mathematical details in an appendix, which tends to be published online.

¡°Our analysis seems to show that for equations put in an appendix there isn¡¯t such an effect,¡± said Dr Fawcett.

¡°But there¡¯s a big risk that in doing that you are potentially hiding the maths away, so it's important to state clearly the assumptions and implications in the main text for everyone to see.¡±

Although the issue is likely to extend beyond evolutionary biology, it may not be such a problem in other branches of science where students and researchers tend to be trained in maths to a greater degree, he added.

É«ºÐÖ±²¥

¡°People have brought attention to previous concerns that biologists might need to improve their maths education, to bring it up to the level of chemists and physicists. It would be a fantastic thing if there were a clear, concrete way this could be done,¡± he said.

Understanding of mathematical biology may also be improved through better presentation, such as by the use of figures and pictorial presentation of theories rather than equations, he added.

É«ºÐÖ±²¥

elizabeth.gibney@tsleducation.com