Seeing Science From Trunk to Tail

The third part of an occasional series profiling big psychology grants travels to Chicago’s Center for Advanced Research in Behavioral Neurobiology, where researchers are working to integrate the vast knowledge of behavior into a bigger picture.

The 19th century lawyer and poet John Godfrey Saxe told a tale of six blind men who encounter an elephant. The first man felt the sturdy side of the beast and declared the elephant to be like a wall; the second felt its fearsome tusk and proclaimed that the creature is like a spear; the third felt the trunk and decided the animal is like a snake; the fourth felt a sturdy, rotund leg and concluded that an elephant is like a tree; the fifth confused an ear with a fan; and the sixth blind man, seizing the elephant’s tail, concluded the animal is like a rope.

Saxe, who based this work on a fable told in India, concluded:

So oft in theologic wars,
The disputants, I ween,
Rail on in utter ignorance
Of what each other mean,
And prate about an Elephant
Not one of them has seen!

Many psychologists argue that human behavior is so complex and so difficult to explore, that we haven’t really come close to understanding how human beings work as a whole. Like the blind men and the elephant, we know a lot about the individual components of behavior; the challenge is to combine those perspectives into a bigger picture.

APS Fellow and Charter Member Stephen W. Porges believes that recent grants to develop a new research facility, the Center for Advanced Research in Behavioral Neurobiology at the University of Illinois at Chicago, hold enormous promise for doing just that. The newly invigorated facility, using cutting-edge technologies such as an enormous network of computers linked together in a system dubbed the Grid, will allow vast analytical resources to be trained on a specific problem and will let researchers find answers that have thus far proved elusive. Porges is director of the new center.

“We really do have to approach research in a different way if we want to answer the big questions,” he said. “Science has just become too big to be conducted by one person, or even in one well-equipped lab. Today most researchers focus on one variable and apply the specific skills of a subdiscipline. Perhaps, they may discuss their research questions with another scientist who studies another variable within a similar research domain. Scientists keep missing the fact that even if they study different variables, the different variables are often mediated by a common mechanism. We believe that our integrated approach will develop new tools to change the way behavioral and biobehavioral research will be conducted in the future.”

For example, behavioral neuroscientists and psychophysiologists typically study behaviors including social behavior or regulation of emotion by measuring variables that are comparatively easy to measure, such as heart rate, cortisol levels, or brain function using EEG. What’s been missing is simultaneous tracking of the different systems. Matching up data from physiological and behavioral monitoring, particularly when different sampling rates are involved, is essentially impossible, so most studies only focus on a single response system. While this has allowed us to better understand specific processes, it also has made it much harder for researchers to develop broad theories, and often misses the complex interactions between systems. As a result, much of our analysis of human behavior is operating at a relatively low level; it’s the equivalent of being able to describe the simple particles that make up a hydrogen atom, while not being able to characterize the role of complex molecules such as water – a combination of hydrogen and oxygen – in keeping us alive.

Proximity and Parity
The new funding for the Center has the potential to change all that. The National Institute of Health’s National Center for Research Resources awarded $3.6 million to the Center and the University of Illinois at Chicago added another $3 million to the project. That money will be used to develop approximately 30,000 square feet of research space, including animal modeling facilities, human behavior labs, and computational neuroscience labs. “We’re going to have an opportunity to conduct biobehavioral science on a scale that has not been possible before,” said Porges.

The idea is to bring together researchers from very different fields to build upon each other’s work in a collaborative enterprise. The point, Porges said, is to bypass chokepoints in the existing research process, through what he describes as the use of proximity and parity. “If you want to integrate and translate research, you have to have the people working near each other, you have to have proximity,” he said. “And you have to offer the people involved parity; you can’t play games with things like resources based on people’s degrees, specialties or subdisciplines.”

Thanks to computer and networking technologies, proximity can be achieved without physically moving people. “High speed streaming of video and data through the Internet can enable active collaboration, even among people from different countries,” Porges said.

Convergent with the design and development of the new research facilities of the Center for Advanced Research is collaboration with scientists from other institutions in Chicago to develop new research technologies. In October, the National Science Foundation funded a two-year, $2 million grant for a project aimed at developing Cybertools, new tools for collecting and analyzing data on an unprecedented scale. This project joins scientists from the University of Chicago, Argonne National Laboratory, and the University of Illinois at Chicago. Headed by Bennett Bertenthal, a professor of psychology and computational neuroscience at the University of Chicago, the Cybertools grant will be used to build a distributed data warehouse called the Social Informatics Data Grid, which in turn will be connected to the TeraGrid, a $150 million computer project designed to let researchers attack problems with the next generation of computer tools.

Bertenthal, an APS Fellow, said the effort will be focused on developing tools to enable researchers to monitor human behavior and physiological measures both in individual and group settings. “It’s the culmination of a number of years of work and will provide a set of tools for psychological scientists that will help them begin to do research on a larger and faster scale than has been available before,” he said. “Science is going beyond the stage of individual labs and now requires large collaborations, often global collaborations, and software tools for automating research and enabling collaboration.”

The project has many objectives, but perhaps none as ambitious as Bertenthal’s hope that some day we will be able to understand human behavior by automatically extracting data from things like a videotape. “The need today is for large, integrated data sets that require multiple measures that exceed the data collection ability of any existing individual lab or even small groups of labs,” Bertenthal said. “We’re trying to provide the infrastructure that will allow a large group to share data.

But that’s not what is unique about this initiative, he explained. “What makes the project unprecedented is that, because behavior is changing all the time we need a way to store and model dynamic changes in behavior. So one goal is to store and annotate streaming data files.

“Think about each measure as a time series, whether a single physical measure or as rich and complex as a video of human behavior. We’re putting together a facility that will let us store, synchronize and analyze behavioral and physiological measures together. We’re about to convert all data into a common data model so we can look at and analyze the relationships between multiple measures. It’s something no one has ever been able to do before.”

C. Sue Carter, a behavioral neuroendocrinologist and co-director of the Center for Advanced Research in Behavioral Neurobiology at the University of Illinois at Chicago – said such new tools are critical if we are to ever really understand how human beings work. “Behavior is the most integrative science of all. It can’t be addressed on any single level of analysis.”

Carter sees the new Center as a way for science to escape that trap. “I work with prairie voles, which form long term pair bonds,” she said. She uses the animals to study social bonding. “What we do is not, for example, social psychology, but it is on the edge of lot of disciplines. The Center is a way to bring people together with others who have very different points of view. It encourages us to work on translational research or clinical research problems. For the first time we’re going to be able to see how several systems function in one organism. Right now, no one person or group has the expertise – intellectual or technical – to do that, but the Center will give us that ability.”

Porges agreed: “Scientific research is moving from the cottage industry model to a collaborative model requiring shared resources.” Collaboration doesn’t merely mean shipping data to a statistician. Rather, it means bringing scientists together from a wide variety of disciplines to investigate a common problem and to critique current methods and strategies. In some ways, he said, we can’t predict the products of the Center until we experience the interactions that are generated in this environment. We are moving into an uncharted area. “It’s very much a Field of Dreams: If you build it, they will come. But we’re committed to building it with a specific organizational structure and a set of constraints that will keep it from being co-opted by the same forces that partitioned science and dominate research today.”

He has high hopes. “The concept underlying the Center is to shift the organizing principles psychology uses to explain behavior. When we talk to social psychologists, physiological psychologists, or clinicians, each is going to assume that their orientation and skills will provide the most appropriate explanation. But when we put those people together with people from several other disciplines – medicine, computer science, genetics, neurophysiology, and neuroanatomy – the dynamics of the intellectual environment change and the dialog will shift from single variable models to more integrated explanations. Systems once believed to be unrelated to behavior, may be identified as playing a key role.”

“Our goal is not only to change our science, but also to change the way we train the next generation of psychologists. To do this we have to create the right arena. It’s not that the techniques are bad, but they are limited,” he said. “And if you’re going to build a collaborative research environment, you need the best tools. That’s what we are striving to build here.”