Martin Pettitt, via Flickr
Martin Pettitt, via Flickr

The Science of Big Science

Science is getting bigger. Just about every scientific discipline — astronomy, conservation, drug development, genetics, neuroscience, physics — is organizing massive collaborations of researchers in the name of reaching massive goals. These so-called Big Science efforts have big budgets, big lists of participating institutions, big press coverage, and big pronouncements. Big Science isn’t new (the term was around in 1961, if not before), but it does seem to be getting more popular.

Take a project that readers of this blog are probably familiar with: the BRAIN (Brain Research through Advancing Innovative Neurotechnologies) Initiative, which the world first heard about in February when President Obama mentioned it in his State of the Union address. The projected budget of BRAIN is $3 billion over 10 years, which will be divvied out by three federal agencies and several nonprofits. The project made headlines in the New York Times, the Washington Post, and every other major news outlet. Its goal, according to Obama, was to “unlock the mystery of the three pounds of matter that sits between our ears.” Could there be a larger project?

And BRAIN is just one of an increasingly long list of expensive, collaborative science projects. The Big Question, of course, is whether the Go-Big strategy is more effective than the typical model of funding individual labs.

Two commentaries came out last week about collaborations in biomedical sciences. One of them, about big-ticket neuroscience projects such as BRAIN, focused mostly on the expected payoffs of these efforts. The other, about medical research consortia, puts forth a more novel idea: that we need to treat the process of Big Science itself as a science. Creating a monster consortium might be trendy, but it’s not the right strategy for every scientific goal. Researchers need to figure out when, exactly, the approach is likely to be effective — and when it’s not.

It’s worth talking about why Big Science is popular and why it has the potential, at least, to do good. It’s hot partly because of the global economic crisis. Federal agencies (in the U.S. and many other countries) are giving out fewer and fewer grants to individual scientists. Big Science, though, is more resilient to cutbacks because its big teams can create a lobbying force, and make their pitch directly to legislators and the popular press. “Publicly high profile, sustained, big-project funding is an effective way of championing a discipline,” says Paul Matthews of Imperial College London in one of the new commentaries, published in Nature Reviews Neuroscience. “Pulling together a new level of public and political support for science demands an exciting vision.”

Matthews gives another (less cynical) reason Big Neuroscience is useful: It makes it easier to share data and ideas across disciplines and institutions. “There is so much in the present system that creates barriers between investigators and institutions and that slows (or even impedes) free flow of information,” he writes. Christof Koch, another contributor to that commentary, seems to agree. “Neuroscience is a splintered field, with circa 10,000 laboratories worldwide pursuing distinct questions with a dizzying variety of tools,” writes Koch, Chief Scientific Officer at the Allen Institute for Brain Science in Seattle. And each of those labs is in an intense competition to publish in top-tier journals. “To gain a competitive edge…hard-grained structural or functional information is hoarded and rarely made accessible online.” Researchers who belong to consortia, he argues, aren’t subject to the same financial and time constraints.

What experts don’t seem to agree on, though, is whether these big projects actually produce new, revolutionary ideas, or instead are more suited for the implementation of established ideas. Henry Markram, leader of the €1 billion, 86-institute Human Brain Project, obviously believes in group insights. “We are hampered by the general belief that we need an Einstein to explain how the brain works,” he writes in the Nature Reviews Neuroscience commentary. “What we actually need is to set aside our egos and create a new kind of collective neuroscience.”

Others point out, though, that the most famous successes in Big Science — the Human Genome project, the moon landing, the Manhattan Project — were essentially engineering projects, not basic discoveries. Take the Large Hadron Collider, an enormous particle collider that the European Organization for Nuclear Research built over 10 years in order to find the elusive Higgs particle. “It tested a hypothesis rather than developing it,” Matthews writes (in the paragraph directly following Markram’s Einstein comment, no less). “Recall that it was Peter Higgs — a single creative scientist — whose theory ‘discovered’ the Higgs boson.”

It’s not just this handful of scientists who quibble over the best way to set up consortia, or whether to set them up at all. The BRAIN announcement, for example, spurred all kinds of backlash from neuroscientists (on Twitter, in the blogosphere, and in traditional news outlets).

But rather than argue about whether these projects are useful or not, why not study the question rigorously and systematically? That’s the premise of the second new commentary, published in last week’s Science Translational Medicine and provocatively titled, “Curing Consortium Fatigue.”

The authors, Magdalini Papadaki and Gigi Hirsch, are from the Massachusetts Institute of Technology Center for Biomedical Innovation. You can tell they come from the business world; the commentary is littered with terms like “the innovation lifecycle”, “success-enablers”, and “synergistic initiatives”. Jargon aside, the ideas seem pretty smart to me. The first step in making a good Big Science project, they say, is to take a hard look at what’s worked in the past. “Understanding how and when various models of funding, intellectual property management, and leadership, for example, have proven useful can provide exemplars to inform how we design new collaborations,” they write.

But wait, you say. Surely people who put these giant, expensive things together have already done lots of logistical assessments? Not so much, according to Papadaki and Hirsch. “Despite the recent collaboration proliferation, little research has been conducted to assess the effectiveness of these alliances or to identify successful characteristics that can be applied to future ones.”

I’m all for Big Science and its lofty aims. But with funding in such short supply, perhaps it’s time to take some pointers from Big Business.