POWER to the people

Citizen science is experiencing a renaissance and with numerous projects available, non-scientists are proving that a PhD doesn’t have to be a prerequisite to the scientific endeavour. You can even map the most obscure parts of the universe without needing to get out of bed...

When Alice Sheppard looks back to her first few days of classifying galaxies, she remembers initial confusion and apprehension but also experiencing an incredible thrill. Back then, little did she know the impact citizen science would have on her life.

Alice was one of the first recruits to Galaxy Zoo – an online astronomy project that enlists members of the public to assist in the morphological classification of large numbers of galaxies. In 2007, the Sloan Digital Sky Survey telescope had imaged and released about two million objects – nearly half of which were galaxies. Oxford University PhD student Kevin Schawinski was looking for rare blue, elliptical galaxies in the images but was overwhelmed by the sheer amount of data he would have to look through. "I classified 50,000 galaxies myself in a week, it was mind-numbing." He told the BBC at the project’s launch. Kevin’s research was a classic example of a “data deluge” where modern research produces vast amounts of information but the teams involved don’t have enough time or resources to analyse it all.

[caption id="attachment_33335" align="alignleft" width="302" caption="Galaxy Zoo - The general public are shown images of galaxies from the Sloan Digital Sky Survey and asked to classify their morphology"][/caption]

With one million images to classify all by himself, Kevin was drowning in data and desperately in need of a pint. Galaxy Zoo was born after he and his team headed to The Royal Oak pub in Oxford for a few drinks with Chris Lintott (astrophysics researcher at Oxford and presenter of The Sky at Night). There, the academics realised that if they could attract members of the public to help them cope with the abundance of information, the project would be completed far more quickly and efficiently.  Web designers Phil Murray of Fingerprint Digital Media and Dan  Andreescu were enlisted bring the site to life and shortly afterwards, Galaxy Zoo was launched into cyberspace, instructing first-time visitors to take a short tutorial before sorting galaxy images into two main categories: spiral and elliptical. The site was inundated with visitors almost immediately; receiving 70,000 classifications per hour after the first day, leading to one blown circuit breaker and an absolutely astounded team of researchers who estimate there are now more than 250,000 active armchair astronomers from all ages and backgrounds. These individuals with no astrophysical background have created a more detailed map of the universe and their work has given rise to over 30 scientific papers, exciting discoveries and numerous online friendships. Galaxy Zoo has evolved to become the Zooniverse, a web portal which hosts more than a dozen citizen science projects covering a range of scientific disciplines for those who want to try their hand at real research.

Citizen science is simply science undertaken by members of the public that aren’t professional scientists; individuals who voluntarily contribute their time, efforts or resources to research without needing a science background. The presence of massive online scientific datasets and the availability of high-speed internet access to ordinary people are providing new opportunities for citizen scientists from all over the world to contribute to growing our knowledge. There is now a myriad of assignments that ‘citizens’ can get involved with. These could take many forms; some use smartphones and computers to classify images such as Cancer Research’s CellSlider where real images of tumour samples are analysed by users in the form of a simple game of snap. Video-game enthusiasts can choose to contribute to biochemistry with projects such as Fold-It where the objective of the game is to fold the structure of selected proteins using various gaming tools. Or, for a commitment-free contribution, volunteers can even lend their computers’ excess power to solving biological puzzles such as folding@home, or pioneering citizen science project SETI@Home, which harnesses the idle computing time of millions of participants in the search for extraterrestrial life. Some projects even require their participants to put down their modern gadgets and head into the big outdoors to undertake ecological research with do-it-yourself sampling kits.

While it has certainly experienced a renaissance in recent years, citizen science is not exactly a new concept. Amateur scientists have throughout history contributed to academic research1. Even Charles Darwin had correspondents that would write to him about the naturalist work they were undertaking. “Citizen science is in many senses an old idea – scientists have been asking volunteers to help with their work for centuries, whether it's amateur astronomers monitoring the skies for supernovae or birdwatchers providing data to ornithologists. What's changed recently is that in many fields the traditional pattern – amateur data collection and professional analysis – has been reversed in order to cope with the sheer volume, velocity and variety of modern datasets,” explains Chris Lintott, Citizen Science Project Lead.

However, more traditional forms of amateur science do exist with many projects still requiring citizens to go out and collect data. Dr David Jones from Imperial College London and the Natural History Museum runs the Soil and Earthworm Survey at the Open Air Laboratories Project (OPAL), a citizen science initiative led by Imperial College which, unlike Galaxy Zoo, requires people to leave their houses to rediscover the outdoors, motivating them to record local wildlife and habitat data for the first time. OPAL, made possible by a Big Lottery Fund grant, has led to scientists and the public working together to gather a wealth of new data about wildlife, their distribution across the UK and the condition of their habitats. Their networks of community scientists play vital roles in spreading the organisation’s philosophy to all sectors of society. The teams engage with schools, local groups and natural history societies as well as disadvantaged and hard to reach groups and organise local activities to allow people to contribute to ecological research as well as helping citizens to discover and value green areas on their doorsteps they may have not been aware of before. In January, The Community Environment report concluded that more than half a million people across the country have been inspired to discover their local environment through OPAL and the project has mapped more than 25,000 sites across England, including areas never sampled before by scientists.

[caption id="" align="alignleft" width="282" caption="OPAL's Soil and Earthworm survey"][/caption]

We get a whole host of advantages from citizen scientists, as they do a lot more fieldwork than scientists could do on their own,” says David. “These surveys are designed to ask researchers questions that we are interested in answering. And we get a lot of information from our standardised sampling approach.”

Helping professional scientists with data collection or analysis is obviously a substantial advantage, but what do the citizen scientists get out of it themselves? To find out, let’s go back to Alice, our intrepid galaxy classifier: “I think citizen science is important for two reasons,” she says. “Firstly, it massively increases the amount we can find out; secondly, science is something that involves everybody given the technical world we live in and therefore it’s only right that people get to take part.” Alice found out about Galaxy Zoo by accident. “I bought myself BANG! A Complete History of the Universe by Brian May, Patrick Moore and Chris Lintott. They had a website where I could write in with questions, which I did. Chris e-mailed me back with great answers, so of course I started reading his website. A few months later, Galaxy Zoo was announced. It didn't necessarily sound any more interesting than all the other amazing astronomy I'd started learning since reading BANG! And I didn't think I'd be any good at it, but it was definitely worth having a go.” Alice jumped into the Galaxy Zoo community and has never looked back. She now moderates the Galaxy Zoo forum, gives regular talks about citizen science, has a blog and runs skeptic events in London. She now juggles galaxy hunting and public engagement commitments with a Masters degree in Astrophysics. “If I hadn’t discovered Galaxy Zoo, I think I’d still be looking for a home. I know that sounds over the top, but finding it really was like finding a home: scientifically curious people and a real mission. It’s set me on my career path and found my greatest interests.”

Dr Arfon Smith, technical lead at Zooniverse believes citizen science allows the public to become aware of how the scientific process works. “Citizen science improves public scientific literacy and increases the transparency about how science really happens. With projects like the ones we coordinate at Zooniverse, the way the public see how science happens is inherently opened up.” In order to successfully design future citizen science projects, Arfon and his colleagues wanted to find out their participants’ motivation for taking part in Galaxy Zoo. The organisation added a survey to their website, encouraging participants to share why they wanted to be involved in the first place. “We discovered that they wanted to contribute to real research projects and this eventually led to the Zooniverse: projects where people would be interested in doing better than machines.” As Arfon points out, many citizen science projects, including Galaxy Zoo, exist because of the limitations of current technology – the fact that computer algorithms are currently not as good at recognising patterns as humans are. In several of these projects, researchers are using humans to teach machines, to bring them up to scratch. “The point at which we need people is a shifting land camp. We can come up with algorithms but only if we’ve used people to make the decisions cognitively first. There will come a time when projects may be stopped because computers are as good as human pattern-recognition, but there will always be areas that will lack algorithms or will need human intervention first,” says Arfon. "Computers will slowly get better at classifying galaxies, but looking at an image and asking, “What's that odd thing” remains uniquely human and we’ll always need that. Projects like these are a really good example of humans and machines working together.” Chris echoes this outlook and says that he’ll stop running projects when computers can complete the task adequately, because they’ve learnt from participants’ pattern-recognition. “In some ways the ideal is our supernova hunting project where new data provided by citizen scientists inspired the further development of machine learning to the point that the volunteers were no longer required. To me, that's a successful outcome. As for how long there will remain things computers can't do – I think that's a bet on the race between the speed with which data sets are growing and the increasing percentage that computers can handle. I think that we might see that as datasets grow we'll need humans to handle an ever smaller percentage but the total amount of work needed might stay the same.”

[caption id="attachment_33338" align="alignleft" width="300" caption="Cell Slider - Volunteers use images from Cancer Research UK to help to classify archived cancer samples"][/caption]

A symbiosis between humans and machines is certainly something to think about, but citizen science also raises questions about what it means to be a scientist these days. For all its advances, crowd-sourced science is a concept that has not been without criticism; David Weinberger from the Berkman Centre for the Internet and Society at Harvard has notably said: “These people are not doing the work of scientists…They are doing the work of scientific instruments.”

Arfon contests views similar to Weinberger’s: “I would disagree that being a scientist is solely about interpretation. A lot of the scientific process is about doing the hard work, which is what our volunteers are doing. It’s a really easy criticism to say these people aren’t really scientists, but I’m not entirely comfortable with any definition of a ‘scientist’ that I’ve come across yet.” Likewise, Alice is not hugely interested in getting bogged down with definitions and wants to focus on the long-term outcomes of citizen science: “What I really hope for the long run is that people will take more ownership of science. Science is something lots of people love, but a lot more remember as a miserable time at school – and they feel convinced they’re not clever enough to understand any of it."

Regardless of whether armchair astronomers and the like can be counted as ‘scientists’ amongst experts in the field, one thing is clear: these days you do not need to have a scientific background, have access to a laboratory or even leave your bedroom to contribute to real-world research. If executed effectively, citizen science can also be a great education and outreach tool, as well as providing space for a meaningful collaboration – something that conventional scientific research is built on. Chris says: “I think citizen science is a reversion to an old model in which it was possible for people from a wide range of backgrounds to feel like they were making a contribution to science – not just those of us with PhDs and university positions. I think the real magic of citizen science is that such authentic engagement is possible at every stage of learning about science, rather than just at the end.”

Reference: Silvertown, J. 2009. “A New Dawn for Citizen science,” Trends in Ecology and  Evolution, 24, 467