Can Science Be Crowd-Sourced?
News recently broke that Stockholm University's Sven Hovmöller had discovered the atomic structure of complex crystals known as approximants -- a complicated chemistry riddle he spent eight years pondering. But the real story behind the story was that he credited his breakthrough to insights from his then 10-year-old son, Linus, who knew nothing about chemistry or crystals, but a great deal about Sudoku. In short, Linus perceived a pattern where his father did not, demonstrating that sometimes in science knowing too much about a problem can muddle the path to a solution, and a fresh, clear view from the outside makes all the difference.
There is no substitute for the rigorous training credentialed scientists undergo to tackle our most challenging problems, but this heartwarming story, and others like it, gives many observers the impression that anyone can "do" science. Indeed, much attention has been paid lately to the notion of "citizen science" -- members of the general public participating directly in the scientific research process. In fact, some scientists themselves have been championing the idea, seeing it as a way to increase public involvement and support for science.
But we need to think carefully about the appropriate role of citizens in science in order to harness the public's interest and energy while still preserving the integrity of the scientific process. As I see it, there are definitely opportunities for non-scientists to participate, but their roles must be carefully defined. Research in any domain of science today requires specialized training to build up knowledge and clinical competence. To make major breakthroughs, we need people with expertise who are engaged in sustained research over a long period of time -- in a word, scientists.
So, when and how should citizens be involved in science?
First, there will be occasions when citizens can participate in data analysis and provide direct input to professional scientists. There are now successful examples of this in astronomy and chemistry. One of them, Galaxy Zoo, invites the public to assist in classifying the shapes of over sixty million galaxy images. No knowledge of astronomy is required, and it turns out that the human brain is more suited to this activity than any advanced computer. More than 250,000 people have taken part in Galaxy Zoo so far, producing a wealth of valuable data and sending telescopes on Earth and in space chasing after their discoveries.
One such case centers on Hanny van Arkel, a Dutch biology schoolteacher who chanced upon a strange interstellar object that she could not match with any of the known galaxy types listed in the Galaxy Zoo classification tutorial. As it happens, this object, now known as Hanny's Voorwerp, is eminently unique: a light echo from the dying gasp of a black hole that was once active as a quasar. We knew that such objects ought to exist but this was the first one to be discovered. Van Arkel is now listed as a co-author with me on a scientific paper interpreting the discovery.
Another example is the computer game FoldIt, developed by the University of Washington, Seattle. Foldit drafts competitive video gamers and leverages their gaming experience and intuition to flesh out new structures for proteins, and it does so better than computers. At last count there were seven scientific publications that included FoldIt Players as co-authors.
Second, the public can contribute to the actual collection of scientific data -- but only once the scientific community has come to a consensus and defined the parameters of debate. We cannot decide on the efficacy of a medical treatment by counting the number of "Likes" an intervention receives on Facebook; no matter what, professionals will still need to conduct continued clinical trials and evaluate their outcomes carefully. But gathering the full range of side effects a drug may have upon use is a point where public input would be invaluable. Individuals reporting on their own experience with particular therapies would provide first-hand accounts that could be considered in the improvement of drug design.
Recent success stories make it clear that citizens may well play an increasingly important role in aiding science. But if the public gets involved at too early a stage in the scientific process, confusion can ensue. A real-world case is found in the climate crisis. Scientifically, the climate change problem is a complex one that has profound implications for each one of us. A deeper understanding of modeling future uncertainties is actively being developed and debated among scientists, but the terms of the debate have devolved from evidence and data to politics, due to the participation of citizens with specific agendas. As such, before we can even begin to explore a solution, we must convince large swathes of the public that there is a problem to begin with.
While some crowd-sourcing advocates will chafe at a limited role for public involvement in the scientific process, citing the buzz surrounding effective crowd-sourcing in its other applications, we ultimately still need experts. No one wants to walk across a crowd-sourced bridge. Citizen science is new territory for us all -- scientists and citizens -- and the possibilities for success are legion. But we must think very carefully about where the boundaries should be. What is clear is that there must be boundaries, and that is a truth we don't need to crowd-source.
This post appeared in the June 24, 2012 issue of Huffington.