The View From Within

APS Member Abrams is Director of the Office of Behavioral and Social Sciences Research (OBSSR) in the Office of the Director of the National Institutes of Health (NIH). Prior to joining OBSSR in April 2005, Abrams was professor of psychiatry and human behavior and professor of community health at Brown University Medical School, and co-director of Transdisciplinary Research at Brown-affiliated Butler Hospital. Abrams holds a bachelor’s degree in computer science and psychology from the University of Witwatersrand, Johannesburg, South Africa and masters and doctoral degrees in clinical psychology from Rutgers University. He joined Brown University in 1978 and he was the founding director of the Centers for Behavioral and Preventive Medicine for the past 16 years. Abrams is a licensed clinical psychologist, specializing in health psychology/behavioral and preventive medicine. His research is on addictive behaviors and lifestyle risk factors for chronic disease. His research has been funded by NIH and he has served both as an external scientific advisor and as a consulting program staff at the National Cancer Institute.

APS Executive Director Alan Kraut recently sat down with Abrams to talk about the status of behavioral science at NIH and his initial experience on the job.

First Impressions
KRAUT: It’s not really fair to ask you for a first impression here at NIH, because you have so much background as a grantee and former staffer.

ABRAMS: There were some surprises. In this position you see how everything comes together within and across the system. It’s a very different picture from what you see on the outside, or from within a single institute. My expectations of getting things done quickly have to be tempered by the reality of the culture and the large size of the system. But there are some incredibly talented and dedicated people here who truly care about advancing the knowledge base and helping improve the public’s health. You just don’t realize how much they do behind the scenes to try to make things work well for research, and how committed and caring they are.

Among the institutes and centers (ICs), there are some solid supporters and traditional homes for behavioral and social science research. But, times and priorities do change. There’s still a lot of work to do in bringing everybody up to speed in terms of the discoveries, knowledge base, measures and methodologies and the statistical innovations largely derived from behavioral and social sciences that can be helpful to ICs, for example, to addressing challenges in biology, like systems biology or genetics, or that are being used to look at simulation modeling of the factors of infectious disease spreading through the world, like Avian Flu.

KRAUT: How do you go about doing that, bringing people up to speed about behavioral and social sciences research methods?

ABRAMS: One way is to have a new strategic plan at OBSSR. It’s been ten years since the office was first staffed and the times have changed dramatically. Part of that strategic planning process involves identifying what are the current and future extraordinary opportunities and the gaps in our knowledge. Another part is to raise the visibility, credibility, and value of behavioral and social science among institute directors and others in leadership positions and the public.

KRAUT: What would be some examples?

ABRAMS: One example is the connection between genes and the social environment. We’ve commissioned an Institute of Medicine (IOM) report to look at genes in the social environment.

Another example would be informatics and communications technologies, and how behavioral and social science can contribute, such as interactive, tailored communications for health promotion — improving adherence, reducing medical errors — and for improved management of chronic disease such as diabetes, asthma or heart failure.

A third example would be imaging and early detection, specifically how imaging and behavior and social science can work together.

And finally, there are enormous opportunities at the basic science level and at the applied level to understand brain, mind, periphery and behavior — both in physical and mental illnesses and in interactions between behavior, social stressors, and biology.

KRAUT: You are the third director of this office. Do you see your role as building on what was done before or as starting from scratch?

ABRAMS: It’s a bit of both. On the one hand behavioral and Social sciences are now a presence at NIH; building on and complementing what’s already there makes the job a little easier. On the other hand, there are still formidable challenges, new directions and the need to prevent backsliding during tight fiscal times. There’s also new discoveries and unanswered questions that deserve every bit of any new investment we can make in behavioral and social science. As our science matures we also need to convince the NIH leadership and others that there’s enormous value to continuing to invest as much as possible in the behavioral and social sciences, not only to improve health but also to inform basic science. In fact, I believe strong investment in basic behavioral and social science is going to be critical to biomedical research moving forward.

KRAUT: I know you’re just starting your rounds of visits with institute and center directors. How have you been received by them?

ABRAMS: My sense is that they have been very welcoming and most are very interested in behavioral and social science research and more transdisciplinary collaboration between biology, behavioral, social and population sciences.

KRAUT: There are several institutes that already have a good base of behavioral science, but how about institutes where behavioral science has yet to have a foothold. Any particular differences in how behavior is viewed at those institutes?

ABRAMS: A core difference may be that behavior is taken for granted or seen as common sense or intuitive, and behavioral and social science is not yet seen as a hard science with significant contributions to make in collaboration with biomedical science.

Balancing Basic and Applied Research
KRAUT: When you talk about behavioral and social science already being appreciated, you mention things like methodologies and measures. Similarly, I hear from various institutes that social and behavioral factors are obviously important in issues like compliance, or behavioral change at the community level. What I hear less of is the importance of behavior at the basic level. On the biomedical side, there’s recognition that if you’re going to develop a vaccine, you need research on molecular genetics, on virology, and on up the line until you get the vaccine. My sense is that NIH people at the top will say, okay, now let’s bring in the behavioral scientists to tell us how to convince mothers to bring their kids in for vaccinations, not appreciating that there’s a parallel process — a whole series of basic behavioral and social science questions that have to be addressed before you get to the intervention that would convince mothers to bring their kids in. Before you get to the applied level, you actually have to know how it is that health information is processed, and what perceptions are about, and the development of thinking skills, and so on.

ABRAMS: I’ve thought a lot about the balance between basic and applied science and the fact that the pendulum seems to swing one way or the other. We’re in a position now where the pendulum and the pressures are so much toward applications, results, and impact on society that we have to be particularly sensitive that we don’t shortchange the basic foundation on which the next generation of innovations would be built. The innovations we see at the applied level come from decades of basic behavioral science work.

KRAUT: What kinds of things are you thinking about doing?

ABRAMS: It requires quite a lot of educating about what basic behavioral and social science is. The mechanisms of basic behavioral science are every bit as challenging to tease apart, as they are in molecular biology. It is interesting that the methodologies of behavioral science, particularly, the statistical methodologies, are now being used in genetics where you have thousands of variables, for example in gene chip technology and in looking for patterns among multiple low-penetrance polygenetic contributions to common disease. We’re used to having very complicated interactions between thousands of variables in the behavioral and social sciences as well as having nested, cluster-level interactions.

Perhaps everyone is beginning to realize that future progress will involve more and more transdisciplinary “team science.” The more we learn about the genetics of common diseases the more we recognize a need for forging partnerships among the biomedical, behavioral and social sciences. It’s not about genes versus environment but about genes AND the behavioral and social environment interacting over lifespan developmental transitions, during sensitive periods (e.g. in utero, puberty) and across generations.

Genes and environment reciprocally interact and each domain influences and is influenced by the other; the causal pathways are neither unidirectional nor linear. It’s not about some sciences being “hard science” and others being “soft science” but rather it’s about respecting all disciplines and collaborating to execute the best and most rigorous science possible given the complex subject matter under study and the different approaches needed to understand mechanisms of action.

One of the most compelling examples I’ve encountered is the work of [Nobel Laureate] Eric Kandel in revealing the biological bases of memory. Kandel couldn’t have done his work and won the Nobel Prize without probably a half a century of learning theory work on memory. That in turn came from a great deal of work in animal research, which is one of the mainstays of basic behavioral and social science. I actually discussed this with Eric recently, and he said that in the 1950s and 60s, he became attracted to the behavioral science ways of looking at cognitive and mental processes, and he went from there to the biological underpinnings. But his first experiences in methodology and in measurement were in the behavioral sciences. So there’s no way that he could have looked at memory without the tools and the basic research that was done in classical and operant conditioning, for example.

As another example, let’s take a complex disorder like schizophrenia. We know that very different mechanisms are involved in the various subtypes of the general schizophrenia syndrome. Some have to do with processing of emotions, some have to do with attentional mechanisms, some have to do with memory or speed of information processing.

To understand even the basic genes involved, you’re going to need behavioral measures of each subtype. Because the chances are, if you’re talking about multiple genes, you’re going to have clusters of those genes associated with the subtypes. And, if you don’t tease that apart, you’re going to have difficulty examining how genes and environment interact to produce the variations in subtypes of the schizophrenia syndrome. As someone once said, behavior is the leading edge of the expression of the gene-environment interaction.

KRAUT: NIMH has a series of projects aimed at further understanding the cognitive aspects of schizophrenia because of the data that says that it’s the cognitive issues around schizophrenia that better predict relapse or severity of the illness, versus things like hallucinations, which are so much more dramatic apparent in the clinical aspects and so, would first seem to be a major factor in determining issues.

ABRAMS: Right. And, you might also argue that the cognitive processes are more malleable and modifiable, giving hope for improved treatment.

Robert Plomin said at the APS Annual Convention [in May 2005] that maybe we’ve got this the wrong way around, and we should be looking from behavior to genes, as opposed to cells up to behavior. That may be accurate.

A Home for Basic Behavioral Science
KRAUT: I want to raise the issue that has been an important one for APS over the last several years, and that is the funding of basic behavioral science at the National Institute of General Medical Science. Last December, a work group staffed by OBSSR reported to the NIH Director’s advisory council that basic behavior and social science needed a more solid home, pointing among other things to changes taking place at NIMH as an example. The workgroup’s recommendation basically was to take a look at an institute like the National Institute of General Medical Sciences (NIGMS), as a place that certainly could support more basic research in behavioral and social science. That follows years of support by Congress on this same issue. There is legislation within NIGMS that would allow this. And, just last month, the Institute of Medicine issued its periodic report on how NIH spends its research training dollars, and they, too, noted that NIGMS could go a better job of supporting training in basic behavioral and social science.

You’ve only been here a couple of months, but I know you’ve thought a lot about this. What’s being done to provide a stable home for basic behavioral science research at NIH?

ABRAMS: We’ve started working more actively to find homes for basic behavioral science research, including some movement towards adding some training protocols and some discussion of other initiatives at NIGMS. It’s really too early to say how it might play out, but I’m encouraged that there is a much deeper awareness in the NIH leadership of the responsibility to take care of this issue and be sure that basic behavioral and social science research is as strongly supported as basic biological research is, wherever it’s housed.

Cross Training
KRAUT: Earlier, you mentioned the biological mechanisms underlying memory. I can remember NIH presentations at the beginnings of imaging, where you would see these wonderful color pictures of the brain at work, but I knew that what was being characterized were outdated learning theories from the 1950s and 60s. It was only when cognitive science was brought in that progress on where memories actually were in the brain, and how they’re made, and how they last, and what takes place when learning takes place, what structures change in the brain, that it was there that progress was made.

What you’re suggesting is that there are many more of these kinds of biological mechanisms that need to be informed by what we already know in behavior.

ABRAMS: Right. But the other thing I want to say is that behavioral and social sciences also have an obligation to reach out to the biological sciences and to educate themselves in a transdisciplinary manner. Not superficially, but to truly take the time to learn the tools and ways of thinking of biology.

You brought up a great example. New technologies in imaging allow us to look at dynamic pathways and transmission in the brain, as opposed to static photographs like the “this is your brain on drugs” kind of picture. We can go to another generation of research that involves dynamic feedback loops, but it means that the behavioral and social sciences need to understand and be able to use the tools and technology in ways that we were not able to do even five years ago.

To take this a bit further, I feel really strongly that you have to have the teams of the future all together in one place; physical proximity is important. What I learned from my work at Brown is that it’s the conversation in the corridor or over coffee that stimulates transdisciplinary integration. That doesn’t happen if people are not physically in the same location.

KRAUT: But don’t you think that our current university structure works against that?

ABRAMS: Yes. The traditional silos of medical schools, like the focus on organ systems and disease groups at NIH, have their place but they are also loosening their boundaries. An excellent example of 21st century health research is in [NIH Director] Elias Zerhouni’s “Roadmap” initiative.* The leading, early-adopter institutions are transforming to facilitate transdisciplinary, team science.

It seems to me we’re in a stage of unraveling the old systems and increasingly being dissatisfied with the old synthesis and conceptual models, but we’re not quite ready to put together the new revolutionary next step. Although everywhere we turn, we’re seeing hints of it. Those hints bode very well for behavioral and social science, especially if we can be very articulate about what is good and what’s solid about our science and how we can make contributions to 21st century visions of improvements in health and quality of life. Perhaps we are witnessing the early stages of a true scientific revolution in the Kuhnian sense — of giving up old theories for a new vision and conceptual synthesis across disciplines and levels of analysis from molecules to macro-environments.

KRAUT: You said NIH needs to have a foundation of basic behavioral science to support the next generation of research. That brings me to what the next generation of researchers in behavior is going to look like, the training for this generation in behavioral sciences. What role does OBSSR have in that?

ABRAMS: OBSSR should play a very strong role in training. Not just training in the new generation of scientists, but cross training scientists at every level of career trajectories in other disciplines that they need to understand at a much deeper level, so they can move from multi-disciplinary science to true trans-disciplinary science.

KRAUT: Would you say it’s virtually given that training in psychology requires post-doctoral training?

ABRAMS: Yes. I would say it’s lifelong, but at least post-doctoral. And, I would say, one of the strengths of the profession has been that type of training. Why is it that you find psychologists in almost everything, from genetics to culture to violence to homeland security and behavioral economics? Something about our training allows us to have that foundation, but also to branch out and to use our skills in a variety of different disciplines and settings.

Defending Behavioral Science
KRAUT: Earlier this year, the House of Representatives voted on the annual budget for NIH. In doing so, they approved an amendment from a member of Congress from Texas to de-fund two behavioral science research grants that were supported by the National Institute of Mental Health (NIMH). These were grants that you could easily make the case for how important they are to public health at a lot of different levels.

As an aside, I should note that there was tremendous support from NIH Director Zerhouni who made strong supportive statements, and the director of NIMH was very active in trying to get this turned around. And ultimately, the assumption on Capitol Hill is that this amendment will be dropped in the final version of the budget when the House and Senate meet to reconcile their different versions. Still, it’s very discouraging for many of us who are advocates for behavioral and social sciences.

This is probably your first experience within NIH at having grants targeted in this way. What were your thoughts about this?

ABRAMS: At the core of this is the peer review system. The NIH is geared towards ensuring that we have the best procedure in the world for allocating grant money. That gives all of us confidence that we are funding the very best science and that the integrity of the process is preserved. In turn, that allows everybody to stand firm and defend questioned grants, whether it’s biological science that’s threatened, or behavior or social science.

KRAUT: And not only were these two grants defended on the basis of peer review. But, they were defended on the basis that they’re important research issues for finding out about public health and improving the public health.

ABRAMS: Right. You’ve got to keep your eye on the prize. And, the prize is that we’re doing all of this to advance knowledge to improve health, well-being, longevity, and quality of life.