Crossing Boundaries: The Growing Enterprise of Interdisciplinary Research

Anyone who followed this past election season — and, considering the voter turnout records, that’s pretty much everyone — no doubt grew familiar with, and likely a bit tired of, each candidate’s avowed mission of “reaching across the aisle.” Almost immediately upon winning the presidency, Barack Obama set out to do just that, inviting a handful of Republicans to a Super Bowl party. Still he was able to rally only meager cross-party support for his historic stimulus bill, failing, in some eyes, to validate his call for a bipartisan era — which in turn prompted The New Yorker to point out that eight days in office was, after all, “a tight schedule for era-delivering.”

In the sciences, the era of interdisciplinary study has been delivering for some time. The past 50 years have seen researchers engaged in their own version of aisle reaching, extending a hand or a methodology or a graduate student across campus and, in some cases, across the globe, to advance some form of basic understanding. A recent National Academy of Sciences committee, charged with summarizing the state of scientific study across disciplines, reeled off an impressive list of achievements, from genome sequencing to neuroimaging to the Manhattan Project.

Psychologists have not been strangers to this trend. Rather, they have been in the vanguard, according to a paper published in Science (Wuchty, 2007). In the second half of the 20th century, the average size of a psychology research team increased 75 percent — the top rate of increase among social sciences.

As research teams have expanded, their composition has diversified. Economists and political scientists, in particular, have teamed with psychologists at a progressive rate, the Science authors found. More importantly, the citation impact of these larger teams seems to have increased with their added size and breadth. This heightened influence holds true even when adjusting for the increase in self-citation that comes with a greater number of researchers per study.

New fields have already begun to emerge from these meetings of minds—neuroscience, political psychology, cognitive science, and evolutionary psychology, to name a handful. Such instances distinguish true interdisciplinary work from multi-disciplinary efforts, which, as APS Past President John Cacioppo pointed out in a previous Observer column, require “only that one share an established procedure with an investigator in another field.” Ideally, interdisciplinary collaborations lead to more than a parlor game of pass the procedure. They don’t just shift eyes onto the question at hand; they ask completely new questions. The goal here, it would seem, is not to reach across the aisle, but rather to eliminate it.

Still, despite their head start over the Aisle Reacher-in-Chief, collaborative scientists also face many challenges when it comes to working outside their comfort zone. An additional workload, communication breakdowns, and tenure-track requirements are some the interdisciplinary scientist’s heaviest burdens. But most consider the evolution of psychology well worth the growing pains. “When psychology departments were forming, it was experimental, social, clinical, developmental — as if any one of these things can be studied independent of the other,” says APS Fellow and Past Board Member Elizabeth Phelps, who is part of the interdisciplinary Center for Neuroeconomics at New York University, of the way psychology operated up through the first half of the 20th century. “I think we had divided up how we understand human behavior. … I see a lot of those barriers starting to be broken.”

A Better Model

A recent paper published by Phelps offers instant evidence of these broken barriers and the efforts taken to overstep the rubble: Two of the four coauthors worked in psychology departments and two worked in economics departments (Delgado, 2008). The study’s immediate aim was to investigate the age-old question of why people at auctions sometimes overbid. The research had a larger objective, however. It hoped to shed light on the emerging debate of “whether data from neuroscience can inform economic theory.” In short, it would be a barometer of interdisciplinary success.

Economists have two traditional explanations for overbidding. The first is general aversion to risk. The second says that victorious bids give people a rush of emotion — a “joy of winning,” in the literature. Using functional imaging techniques, Phelps and her neuroscience colleague, Mauricio Delgado of Rutgers, studied the brains of subjects engaged in an auction. They noticed a type of neural activity in a part of the brain that has been associated with loss in previous research. Instead of confirming one of the prior economic theories, the results seemed to suggest a third possible reason for overbidding: a fear of losing.

To further test this new explanation, the researchers designed another experiment. They separated participants into three auction groups. A control group would approach the auction in a routine manner. Another group was told that if they won their auction they would receive a bonus. A third group was given a bonus and told they could keep it if they won their auction, but had to return it if they lost. The results found that the third group, the one established to validate the new explanatory wrinkle, bid significantly higher than the others.

Once again, the researchers had supported the notion that a fear of losing was driving the decision to overbid. The traditional “joy of winning” idea wasn’t supported by the data, and the notion of risk aversion needed some qualification. By applying the techniques of neuroscience to an economic model, the interdisciplinary research team had created an entirely fresh, nuanced explanation of why people at auctions overbid.

“Our results provide evidence of how an understanding of the neural systems of economic behavior might inform economic theory,” the authors conclude in the September 26, 2008, issue of Science. In this instance, at least, neuroeconomics was more than a clever hybrid of fields best left to themselves — it was a valid new line of thinking.

“We were actually trying to support an economic model,” said Phelps recently, “and we ended up finding perhaps a different economic model that explains it better.”

Forced Socialization

What isn’t clear from reading the work on overbidding is the preparation that went into making the collaboration a success. Exhaustive meetings were needed to iron out methodological differences. Certain disciplinary constraints — economists can’t use deception, for instance — forced the researchers to design the research in a way that would enable both fields, economics and psychology, to learn from the conclusions. “If they want to talk to their colleagues about this data we have to satisfy…what I’d call their peculiarities,” Phelps joked about her economist coauthors.

As Christopher Federico knows, such peculiarities, or whatever one calls them, aren’t satisfied overnight. He deals with them everyday as director of the Center for the Study of Political Psychology at the University of Minnesota. The institution was created in 1995 specifically to facilitate the connections between psychologists and political scientists.

It’s one thing for researchers with a common interest to work together, says Federico. It’s quite another to foster a harmonious interdisciplinary setting. That’s why the center sponsors weekly meetings where students of psychology and political science come together for discussion. Speakers who appeal to both fields are brought in. And, when the intellectual urge to cooperate arises, the center has special funding for interdisciplinary projects. Over time, such efforts help both sides recognize the scientific problems facing one another and converse in terms all can understand.

Of course, guidance alone isn’t enough without individual investment. Years of additional training are needed to understand just the rudiments of several areas of science. An interdisciplinary scientist has daily responsibilities those in a single field might not have, from reading a wider range of journals to attending meetings in multiple departments. Federico, for instance, gave an interview for this Observer story from his office in the political science building.

Still, he considers the result worth the extra effort. “It’s given me ways of looking at problems that I’m interested in that I never would have had as much access to as a straight up-and-down social psychologist,” he says.

The most recent of these problems found Federico collaborating with political scientist Paul Goren to investigate why certain people identify with different ends of the political spectrum. A long history of psychological work has connected certain types of personalities with certain political ideologies. For example, the desire for closure has been linked with conservative leanings. Political scientists, however, believe that political preference is derived more from social factors than from traits of the individual. Instead of building up a political mindset from one’s own characteristics, this mindset trickles down from an elite establishment of politicians, academics, and pundits.

Federico, a trained social psychologist heavily versed in political science, and Goren, the reverse, attacked the problem with an interdisciplinary fervor. They analyzed recent political surveys and came up with what they consider a more complex picture of human behavior than either of the fields discovered alone. Individuals with a high need for closure tended toward conservative politics, but the degree of this correlation depended on one’s familiarity with the elite political establishment, the authors write in a chapter of Social and Psychological Bases of Ideology and System Justification (2009). Put simply, people lacking access to political experts choose their ideologies more haphazardly, regardless of personality traits.

It’s precisely this type of elevated analysis the interdisciplinary center for political psychology seeks to nurture. “[T]hat has to start somewhere,” Federico said of his good working-relationship with Goren. “And unless people are having contact with each other on a frequent basis, that’s not going to happen.”

‘Tweeners

For APS Fellow and former Board Member Douglas Medin and Scott Atran, their interdisciplinary contact comes during long walks, often taken in the fields of whatever indigenous population they’re studying at the time. Many years ago, Medin, a psychologist at Northwestern University, and Atran, a cognitive anthropologist at the University of Michigan, decided the best way to combine their interests in human behavior and the environment was to study them in tandem. Today they are co-directors of Northwestern’s Cognitive Studies of the Environment.

The success of the interdisciplinary program is a testament to the ability of a smorgasbord of scientists to converge on a single problem. In one of their many ongoing collaborative efforts, Medin, Atran, and their colleagues have tackled what’s known as commons behavior. This line of inquiry asks why certain populations don’t destroy a shared environment to fulfill selfish needs.

Previously, it was thought that only centralized governments or private property laws could prevent a large group from ravaging the environment into an unsustainable state. By working with ecologists, biologists, and linguists, in addition to psychologists and anthropologists, Medin and Atran have studied Mayan populations and their land and found that even without these perceived cultural necessities the natives collectively use the commons without draining its resources.

Part of the reason for this behavior, the researchers have found over the years, is that some Mayans wish to protect forest spirits. “It was an inherent, self-policing form of management, a stunning finding, that forest spirits were real in this sense,” said Atran recently. “This was only possible because of the collaboration.”

In addition to long walks, the interdisciplinary success achieved by Medin and Atran owes much to the graduate students often brought to the field. After a day’s work, says Medin, there’s plenty of time each evening for students and study leaders alike to work out whatever methodological or terminological kinks linger from their different backgrounds.

This practice highlights perhaps the greatest challenge of working across disciplines—a challenge emphasized by Medin and Atran, separately, and echoed by every other scientist contacted for this article. That is, the potentially difficult position interdisciplinary work puts students in once they graduate and look for a job. In a theoretical but common scenario, students might spend years learning to conduct complex, interdisciplinary research in the field, then appear to have spread their expertise too thin for any particular department to hire. In this sense, the very interdisciplinary work that stands to help a developing science stands to harm the developing scientist.

Should a student of multiple disciplines—a “tweener,” as Elizabeth Phelps fashioned herself—catch on as an assistant professor, then the problem becomes publishing enough articles in a specialty journal to earn tenure. This intense, narrow focus can stifle progress in the areas of research an up-and-coming academic might consider most exciting.

How to correct this institutional dilemma is, for the moment, unclear. “Maybe that just means we need to evolve different standards for tenure,” says Medin. “If our field will be more interdisciplinary, then maybe you should get some points for demonstrating that.” Atran proposes a model espoused by Google, whose managers, he says, are required to devote a significant chunk of time to pursuits outside their main area of expertise.

Federico sees these early hurdles as down-payments on a richer future: “There are challenges in the early part of your career,” he says, “then a freedom that other people might not have.”

Problems and Promises

Measuring interdisciplinary success can be a bit tricky. Take the case of cognitive science. On one hand, it seems like the exemplar of collaborative achievement—a well-stirred concoction of psychologists, computer scientists, linguists, philosophers, and neuroscientists. Its emergence, largely within the past 50 years, blossomed a few decades ago with the unveiling of several journals, meetings, and societies devoted solely to the new field.

But this superficial glance can be a bit deceiving, says Christian Schunn, University of Pittsburgh. Beneath this surface, cognitive science, as measured by its journals, is dominated by psychologists and, to a slightly lesser extent, computer scientists. These two disciplines far outrank the others with regard to author affiliations, citations, and approaches, Schunn and his colleagues report in a chapter of Problems and Promises of Interdisciplinary Collaboration (2005). Sometimes cognitive science takes the appearance of true interdisciplinary effort, the authors conclude, and sometimes it does not.

The difficulties inherent in close collaboration across fields will continue, perhaps indefinitely, for the same reasons it will always be a bit harder to, say, convince conservatives to support government spending. Although interdisciplinary work might benefit our understanding of behavior in the long-term, in the short-term it slows down the already slow scientific process. “Even if you can overcome barriers, you add a lot more time to figure out where your miscommunication comes from,” says Schunn. “It’s just much slower.”

As study teams continue to expand, so do the problems of collaboration. A study published in a 2004 issue of Social Studies of Science found that research across multiple universities encountered significantly more organizational problems than research involving a single institution.

Michele Gelfand, University of Maryland at College Park, has engineered solutions to the problem of reaching a widespread team. As principal investigator of Project InterAction — a multidisciplinary research initiative comprised of academics at several universities, not to mention countries, that aims to better understand the cultural processes of negotiation in the Middle East — Gelfand uses technology to bring her far-flung collaborators together more often. She holds “virtual brown bag” meetings, e-mails regular newsletters, and has set-up a website where news and research updates will be posted. And because electronic collaboration can only bring people so close, Gelfand and her colleagues are holding a workshop, in the summer of 2009, so that the project’s expansive catalog of computer scientists, cognitive and social psychologists, political scientists, and economists can swap scientific ins and outs — and improve each other’s performance in the process.

“Negotiation is incredibly complex. It’s helpful to have different perspectives on this topic,” says Gelfand, “Psychologists aren’t just the ones with ideas to offer.”

For one thing seems clear: Given the proper time and attention, the benefits of interdisciplinary collaboration are powerful, both for those doing the research, and to the public benefiting from it. Schunn and his colleagues have found that a greater percentage of scientists working on interdisciplinary projects considered the partnership a “stimulating relationship” and felt their ideas were being constructively challenged than did those working exclusively within their primary field.

Such a finding is particularly encouraging when one considers that the interdisciplinary trend won’t be going away anytime soon. “It’s an evolution of the field,” says Phelps. “When you’re trying to understand any behavior, and you’re not doing it solely for the benefit for other psychologists, you want to think about the bigger picture. You have to break things down.” And, so goes the hope, one day build them up again into something somewhat more than the sum of their parts.