Should neuroscience focus on big team science or the traditional model?

Funding for larger collaborations in neuroscience has increased over the past 10 to 15 years, with the advent of the BRAIN Initiative, U19 collaboration grants from the U.S. National Institutes of Health and similar programs. The goal has been to develop open-source resources and tackle projects not feasible for individual labs. But not everyone agrees that this shift is productive for the field. We polled our readers and contributors and engaged a market-research firm to conduct structured interviews with senior scientists from leading neuroscience programs around the world.

Most advocated for a balanced approach, though the ideal balance varied from person to person. Many noted that big teams can be valuable for building shared datasets, such as large-scale connectomics and sequencing projects, and can enable cross-disciplinary collaborations that might not happen otherwise. But some respondents were concerned that large team projects stifle innovation, arguing that the traditional model enables flexibility, creativity and independence. Others raised funding concerns; big team efforts can take support away from smaller initiatives and tend to favor elite institutions.

“It should mostly not focus on big team science. Large teams should be assembled and pursued with commercial and nonprofit funding for specific projects. The bulk of NIH funding should remain focused on the traditional model.”
—Selmaan Chettih, postdoctoral researcher, Columbia University

“I’ve been part of consortia, not small and not large. Maybe 20 people. Sometimes, I’ve been missing having a small lab of just two to three people working seriously. Large labs can generate many results, but not necessarily great results. Small groups can handle things. But I’m in a truly valuable consortia, on brain imaging, and it is so enriching to see perspectives across countries and labs on topics. Both [sizes] are necessary. I like the focus of a small group, but I also enjoy a large team.”
—Luana Colloca, professor of pain and translational symptom science, University of Maryland School of Nursing

“There will be a place for both and a need for both. I think there are certain ideas and projects that can only be done by very large teams. I would be worried if there was too much of a shift to large teams because of the risk of inefficiencies, the risk on the rigor of science. I think mainly neuroscience should still be driven by single labs or small collaborations.”
—Jorg Grandl, associate professor of neurobiology, Duke University

“Big team science will produce results that are otherwise difficult to achieve. On the other hand, the creative insight produced by individuals or small teams is priceless.”
—Gautam Kao, researcher, University of Gothenburg

“A balance must be struck between big science, which can hammer down already established theories, and innovative small working groups. While neuroscience could certainly benefit from more big science, I see a danger that these flashy ventures … may stifle small-scale innovation.”
—Friedrich Kling, doctoral student, Wellcome Optical Biology Programme, University College London

“I’ve seen both work tremendously well. We need both. Science needs bigger teams and investment, because it is worth it, but not at the expense of creative science. Chaos can be good. Labs challenge each other.”
—Drew Robson, research group leader, systems neuroscience and neuroengineering, Max Planck Institute for Biological Cybernetics

“Big science continues to grow. We need more funding mechanisms to support it without being too prescriptive—and we need to protect the small science that is, after all, where the best ideas originate.”
—Joshua Sanes, professor emeritus of molecular and cellular biology, Harvard University, contributing editor at The Transmitter

“I definitely think there’s room for team science, and there are opportunities to create new future directions by bringing people together that may not normally talk to each other. And I’m a big fan of those incentive structures. I do think that you want to be careful about not taking away autonomy. You still want PIs to be independent and running their own research programs with core lab funding, but I do think that these team grants can really add synergy and sort of catalyze things.”
—Jason Shepherd, professor of neurobiology, University of Utah

“There is room for both. I’ve been closely involved in the Allen Institute—big team, open science—since almost its founding. They’ve done something unique, which would not have been done otherwise, and there is a lot to be said for that. But I think that should be no more than 10 percent of the overall scientific effort. Most of it should be small labs run by independent investigators … We need some big science. Most advances are going to come from small groups trying new things.”
—Michael Stryker, professor of physiology, University of California, San Francisco

“I think identifying those questions [big science questions] and finding the teams that can do them is great. But it is never going to replace the individual ideas.”
—Anne West, professor of neurobiology, Duke University School of Medicine

“I’m not sure what traditional science means, but I’m not a huge fan of big teams. I’m against large consortium ideas where there are these big, amazing plans. They have good ideas and goals. There is a huge team with experts, who are amazing and can do anything. They get the grant; then who will really do the work? Nobody is in charge of making decisions; all are busy with other things. So, not a lot happens. I like it when someone is in charge. With nobody to take ownership of goals, to say, ‘We need to do this,’ to understand and know to make decisions, if it is all to be done by committee, then it’s very hard to make progress. Money gets invested, but not a lot gets done.”
—Martijn Cloos, associate professor of bioengineering, University of Queensland

“Outside of neuroscience, you think about things like particle accelerators or astrophysics where you have to build a giant telescope. You have to work as part of a large team because there’s this just massive, shared resource. But in neuroscience, there’s not anything like that. We don’t have the equivalent of particle accelerators or telescopes, so it tends to be extremely decentralized, and it’s kind of in the DNA of most neuroscientists that they’re all kind of doing their own thing at the lab level. And then it’s kind of a matter of opportunism to obtain funding for collaborative projects where the people are brought together nominally as a team, but my experience is that they rarely function as a true team.”
—Samuel Gershman, professor of psychology, Harvard University

“If the trend is towards funding more big team science, then only a few certain elites would do research and potentially rob us of smaller labs and more individual/creative thinking. A balance between the two seems wise.”
—Faranak Herrera, founder and CEO, 16p11.2 Genetic Foundation

“Big team science is fine, but there are two crucial considerations. (1) What fraction of total neuroscience funding goes to such projects? If there were infinite funding it would be great to support as many large-scale team projects as possible, but of course, this is not the case, and it’s important that a substantial fraction of funds still remain for individual/traditional approaches to scientific discovery. (2) Who is the labor force undertaking team science projects? The problem I have observed in these endeavors is that the people undertaking the majority of the labor—i.e., usually postdocs and grad students—have very little say in the scientific questions/direction. Thus, while they may in the end provide a significant contribution to the team goal, and possibly get their name on a widely read paper, to do so they sacrifice substantial opportunity to develop their own scientific thinking and experimental design skills, thus making them less prepared to lead cutting-edge research labs in the future.”
—Rich Pang, postdoctoral research associate, Princeton University

“In my view, only a small part of neuroscience should be dedicated to big team science. Large teams tend to dilute individual responsibility, which often leads to lower-quality data, weaker analyses, and fewer efforts to truly understand the phenomena under study.”
—Peter Rupprecht, junior group leader, University of Zurich