One of my favourite photos shows a van adrift in the middle of a Mongolian river. The unfortunate occupants, who came to no harm, were volunteers recording ultrasonic bat sounds for my research on bat populations.
This project, called iBats, began more than six years ago, when I scaled-up traditional bat-monitoring methods so that anyone could, with some hi-tech gadgetry, use them. At the time, most volunteers on science projects needed a good initial knowledge base, and so tended to be interested amateurs. Poor scientific design had given some citizen science projects a bad name among academics.
Today, all that has changed. Thanks to the increasing ubiquity of smartphones and tablets, the past 18 months have seen an extraordinary growth in applications that take images, sounds, and other recordings to feed into research projects. Social media and gaming are rapidly changing who can take part in science, and how they can participate. My Mongolian picture’s most recent outing was a fortnight ago at the Zoological Society of London, at a symposium entitled Smarter Science: The power of the crowd, which highlighted the boom in both the quantity and variety of citizen science.
Some projects use citizens as sensors. If your phone has the WideNoise app, you can use it to collect data for University College London’s map of noise pollution. If you’ve also got the London School of Economics and Politics’ Mappiness app, you can then record how that affects your wellbeing. The iBats app gathers ultrasonic sounds with the help of a special microphone.
Other projects use citizens as thinkers, to categorise data that is too difficult for a computer to process. At Zooniverse.org, there are opportunities to help make sense of everything from telescope images in astronomy, to whale calls, to ancient texts; over 700,000 people have joined in. Darwin Tunes allows participants to steer the cultural evolution of music, while fold.it lets them grapple with the complexities of protein folding.
More radical approaches blur the distinction between citizen and scientist. For example, in 2010 the respected international journal Biology Letters published a study on bumblebee vision coordinated by researchers at UCL and designed, carried out and written up by a class of primary school children in Devon.
The number of do-it-yourself scientists is also growing, as the skills to programme smartphones spread and amateurs begin designing their own hardware using customisable computers such as Raspberry Pi, which is the size of a credit card and costs about $30. I haven’t seen a cheap ultrasonic microphone for a Raspberry Pi yet, but I would bet that one is months not years away.
Projects can mobilise and organise the public with astonishing speed to provide data on topical issues. The development, by researchers at the University of East Anglia, of the Ashtag app to monitor ash dieback is a recent example. Governments are waking up to the importance of citizen science too, perhaps seeing it as a good deal in a climate of austerity—a guide (pdf) published by the UK Environmental Observation Framework last week highlights the approach’s quality and cost-effectiveness.
Although researchers’ perception of the value of citizen science has changed radically in the last six years, challenges remain. Such projects put scientists in unfamiliar territory: they have to generate, train and build capacity in users, while providing feedback and rewards, and designing a study that is feasible, scalable, and interesting enough to be published. With this in mind, I often collaborate with organisations experienced in engaging and retaining volunteers.
The growth of citizen science raises other issues. The proliferation of projects might mean that scientists have to compete for public attention, efforts may be duplicated, data standards may not be agreed, and projects might be developed without consideration of users or politics. The recent development of citizen science platforms, such as Zooiniverse or iNaturalist, which have common data standards and suites of projects or tools might ameliorate these issues.
Crowd-collected data also pose new analytical and visualisation challenges. Projects such as the ZSL’s Instant Wild, which shows images from automatic cameras, collect more data than they can analyse. Designing ways to recognise biodiversity images and sounds automatically is an exciting challenge.
The democratisation of research continues apace and I hope that eventually anyone will be able to ask and answer questions about their world. Just as it is becoming increasingly meaningless to distinguish between online society and the physical world, so in another six years, citizen science might be just science.