National Science Foundation Update

With a budget of $6 billion, the National Science Foundation (NSF) is the only federal science agency dedicated solely to supporting basic research. That doesn’t mean NSF is the only agency that supports basic research – the National Institutes of Health (NIH) and the Department of Defense have significant basic research programs. But NSF is distinguished by its efficiency and its essential role in supporting scientific inquiry that often would not be funded by more mission-oriented agencies.
NSF’s Social, Behavioral, and Economic Sciences (SBE) Directorate funds basic research in human cognition, social behavior, language development, learning, and behavioral neuroscience, among other topics, to the (current) annual tune of $220 million. Typically overshadowed by NIH in Congress, this was supposed to be NSF’s year. In one of the most unpredictable budget climates in years, it looked like NSF might outpace NIH in budget growth for the Fiscal Year 2008. But recent developments have brought that scenario into question.
Congress passed the America COMPETES Act in the fall of 2007, which was aimed at boosting math and science education as well as basic research (see “The Federal Budget Season” in the November 2007 Observer). This was to translate into new money for all of NSF’s research directorates, including SBE, but the year-end Congressional struggle to pass the government’s spending bills resulted in a giant omnibus bill that fell short of funding expectations.
In support of the America COMPETES Act, President Bush requested an 8.7 percent increase, or a $506 million boost, for NSF in FY 2008, and the House and Senate both passed bills that would have added to that total of $6.07 billion. Across-the-board cuts mandated by the President, however, resulted in an additional $117 million, and the research directorates will receive 1.2 percent more than FY 2007.

While this is disappointing news and NSF will have to re-prioritize in the wake of the omnibus bill, there are several new initiatives that the behavioral and social science research community should keep its eye on.

The Transformative Train
Transdisciplinary research appears to have caught on at NSF. There is a new emphasis on “transformative” research in keeping with recommendations of the National Science Board (on which three APS Fellows sit; see December 2006 Observer). The Board defines transformative research as that which is “driven by ideas that have the potential to radically change our understanding of an important existing scientific or engineering concept or leading to the creation of a new paradigm or field of science or engineering.” The Board has made it clear that it wants NSF to weave this emphasis throughout its core values.
APS Fellow and NSB Member Alan Leshner said “the Board recognizes that many scientists believe all funding agencies, including but not limited to NSF, tend to be conservative, particularly when funding is tight. The Board wanted to make sure that although in general it is wise to bet on somewhat safer things, there also is a place for the more revolutionary or transformative kinds of research. So, the Board felt if one makes very clear that this kind of research is welcome, investigators will be more likely to submit their more innovative ideas.” Leshner, formerly the director of the National Institute on Drug Abuse, is executive officer of the American Association for the Advancement of Science. He previously served as a senior official at NSF.

The Foundation will encourage use of the Small Grants for Exploratory Research mechanism and a more prominent presence in the FY 2009 budget (meaning there may be new money set aside for projects specifically relating to transformative research). Even the intellectual merit review criterion has been modified to include this new direction (see Chapter III in the latest version of NSF’s Grant Proposal Guide).
NSF plans to test and assess existing mechanisms and experimental pilots on transformative research. A database and analysis of grants made with existing and new award mechanisms are being planned so the Foundation can compare applications and funding patterns. This initiative, by the way, isn’t being done as a whim – a large survey of applicants revealed the perception that NSF is welcoming of transformative research and that NSF is even preferable over other agencies as the best place to submit ideas for risky research (see below).
There is some debate as to whether a push for transformative research is really necessary. Some feel it’s best left to the academic institutions to foster revolutionary science, without being told to do so from the feds. Besides, the transformative nature of research often isn’t appreciated or even understood until 10 or 20 years after the fact.

“Other than the rare case in which one can anticipate field-altering insights (say in the case of a new hominid skeleton, or a new material, or some relatively obvious ‘marker’ for transformativeness), by and large one can only know post-hoc whether something yielded genuine transformation,” said David Poeppel, University of Maryland, and a member of the SBE Advisory Committee at NSF.

On the other hand, the National Science Board, which made the recommendation, is composed of working scientists who feel that the research community needs encouragement, both financial and ideological, to move in this direction. For now, the only change has been in the intellectual merit criterion, but there may be targeted money set aside in the future.

APS Past President Richard Thompson, also an NSB member, said “NSF has implemented new efforts to identify [transformative] research in all areas they fund, to the extent possible. This might sometimes involve funding a project that seems a bit chancey if the projected outcome might be transformative. Incidentally, NSF has always attempted to support tranformative research. The problem, as always, is how to identify it. It will be interesting to see how the new effort works out over the next few years.” Thompson is the Keck Professor of Psychology and Biological Sciences at the University of Southern California.

Grand Challenges
Together with other federal research agencies, NSF is planning to take on “Grand Challenges” in the social and behavioral sciences. The White House’s Office of Science Technology and Policy (OSTP) is currently reviewing a document that outlines these research objectives, and once it’s been approved NSF plans to go to town (literally) with it – that is, NSF representatives and OSTP will widely disseminate the document to Congress and policymakers throughout Washington. Increased exposure to the latest themes and breakthroughs in our science will hopefully translate to bigger slices of the appropriations pie in years to come (at the very least it will influence the FY 2009 budget and fend off future Congressional attacks).

The SBE sciences have as a common goal a deeper understanding of human beings at every level, and scientists in these fields have new technologies and expanded capabilities to tackle society’s most critical challenges. The Grand Challenges themes and related questions include:

1. Human Origins: Who are we and how did we get here?
2. Mind & Brain: How does behavior arise?
3. Complex Systems: How do people and natural and manmade systems interact?
4. Policy: How we can we take charge of our future? We need to enhance evidence-based policy and decision-making, and to foster better communication between SBE scientists and the public.

The research that these themes will engender will use cutting-edge tools, such as genomics, neuroimaging, and cyber infrastructure. The idea is to bring into the behavioral fold progressive technology that advances the field.

“It is important to note that the social and behavioral sciences have changed significantly over the last few decades because of new technology,”said David Lightfoot, assistant director of NSF and head of the SBE Directorate.

The SBE sciences have a unique role to play in understanding societal challenges because of the human element, so they should be integrated into policy and decision-making at every level and every sector, public and private.

Big Science
In light of the new 2008 funding levels, it’s not clear what the budgets will be for new programs the SBE Directorate has been working on.  Nevertheless, below are descriptions of several emerging initiatives.

The SBE Directorate is moving in the direction of Big Science. A first-ever standing inter-directorate program has been established, called Dynamics of Coupled Natural and Human Systems. Supported by three NSF directorates – SBE, Biology, and Geology – this program is looking for proposals that examine complex interactions among human and natural systems at diverse scales, such as land use, the role of institutions in decision-making, and social valuation of biodiversity. Other cross-cutting initiatives include the International Polar Year (yes, behavioral science has a role to play here, such as studying effects of long-term isolation in those icy outposts), and the Ecology of Infectious Diseases.

One program in which the SBE Directorate plans to play a prominent role is Cyber-enabled Discovery and Innovation, which will support research on computational thinking and complexity and interacting systems. It’s an ambitious initiative that NSF hopes will revolutionize the field and, depending on internal re-prioritization, it could get as much as $50 million. Programmatic themes include emergent phenomena and tipping points in human development, certainly topics behavioral science speaks to. The solicitation was released in late 2007 — look for it on NSF’s website.

Neuroscience is gaining new momentum at NSF. One idea revolves around refining the discipline’s tools, which could lead to a sort of “neurotech” initiative. If it materializes, it will be an agency-wide program, and coordinated within SBE by Douglas Whalen, the cognitive neuroscience program officer. The program would encourage applying complexity theory to imaging, an innovative approach that will help tackle the big question of how the brain creates thought and behavior.  Whalen personally believes that “without the social and behavioral sciences, you don’t have neuroscience, and your imaging results are only as good as your behavioral results.” This is an inclusive and forward-thinking perspective on neuroscience, and hopefully it is one that will stimulate great collaboration.

The Science of Science
The government often calls upon economists to help make decisions about policy – feeding cold, hard numbers into scales and indices that manufacture seemingly objective measures on which policymakers rely. In 2005, the President’s Science Advisor John Marburger called for a national science of science policy, asking for research on innovation and scientific discovery processes, as well as on how policymakers use science to shape policy.

As a result, NSF created the Science of Science and Innovation Policy (SciSIP) research program. By studying science as a social process, SciSIP’s goal is the development of an evidence-based platform for science policy. It hopes to accomplish this by generating usable knowledge and theories; improving science metrics; and cultivating a community of practice focusing on SciSIP across the academy, the public sector, and industry. So it’s a two-pronged process: scientifically understand the process of innovation, and then bring that knowledge to bear on policy.

One example of the kind of ideas blooming from this initiative is the measurement of well-being. Last fall, NSF convened a roundtable of the best psychologists (and other scientists, including an economist) in the field of well-being, including APS Fellow and editor of Perspectives in Psychological Science Ed Diener (who chaired the workshop), Martin Seligman, Norman Bradburn, and Nobel Prize-winner Daniel Kahneman. They tackled such questions as: How can science policy and science outcomes be evaluated by measuring societal well-being? Can scientific priorities be based on well-being? Does well-being as an outcome lead to different science priorities than considering other outcomes? What about national competitiveness and productivity in relation to science and well-being? Addressing these questions has implications for health and the economy, both of which are linked to well-being.

In its first year, the SciSIP program received 60 proposals, 19 of which it funded for a total of $6.8 million. The grants are interdisciplinary and international, and they aim to show how behavioral models can explain how policymakers choose what to fund. The evaluation of SciSIP is key and data will be available to future researchers. We don’t know how long it will be before we see results at the national level, but the hope is that the usefulness of the project will be evident through stakeholder meetings and a “community of practice.”

The second solicitation is out, and it focuses on three emphasis areas: analytical tools, model building, and data development and augmentation. Go to NSF’s website and look up “NSF 08-520” – the deadline is March 18, 2008.

NSF Studies Itself
While NIH is revamping peer review (see the December 2007 Observer), NSF recently took a closer look at its proposal and award management mechanisms (see the IPAMM Report on NSF’s website). The broad picture won’t surprise anyone: success rates went down between 2002 and 2005. Though the budget increased by 44 percent during that time, the average size of an award went up 41 percent and applications were up 50 percent. There’s also a ripple effect of NIH here – during its doubling between 1998 and 2003, NIH doled out a lot of infrastructure money and this naturally led to a surge in grant applications all around.

NSF also surveyed applicants, grantees, and reviewers. 56 percent believe that NSF welcomes transformative research proposals and that NSF is more receptive to risky research than are other federal research agencies. Reviewers concur that the peer review system is stressed – applications are up 50 percent but the reviewer pool has increased only by 15 percent. Interestingly, respondents underestimated NSF’s funding rates – 49 percent thought it was 10 percent or lower, but in reality it’s 20 percent. A variety of remedies are under consideration, such as limiting proposal submissions, or increasing the number of awards. Stay tuned for developments.