What does a choice look like?

The sealed and soundproofed room could be a recording studio. It is windowless except for a plate-glass panel looking into a control booth.But rather than recording albums in this basement lab on the Princeton University campus, a team of scientists is recording brains making decisions. Cords and cables snake in and out of a multi-million-dollar, liquid-helium-cooled MRI machine. A graduate student volunteer is fed into the mouth of the machine with the aim of gaining insight into a troubling modern problem: how, despite our best intentions, we end up watching too much TV.

For two hours, the subject remains strapped to the bed of the MRI. He is shown previews of Saturday Night Live sketches; he can choose to watch the sketches, but for each one he selects in the first hour, he will have a period of silence one and a half times as long during the second hour. Headphones pipe in the soundtrack and muffle the firing of the machine, which sounds like a jack hammer striking concrete at a carefully calibrated 17 times a second. In the control booth, monitors show a 3-D image of the subject's brain. This is a functional MRI (fMRI) study: in addition to showing the anatomy of the brain, the scanners collect data on activity in the brain as the subject makes decisions. When neurons fire, they need oxygenated blood. The blood rushing to those neurons shows up on the fMRI. Correlating the flow of blood with the moment the subject presses a button to play a sketch may help tease out how the prefrontal cortex, a reasoning part of the brain, negotiates with the nucleus accumbens, a region which is responsive to immediate gratification. Fundamentally, this experiment is aimed at increasing our understanding of choice, which, in the eyes of some, makes it an economics experiment, or more precisely, a neuroeconomics experiment.

A small group of economists are trying to peek inside the black box of the brain. For years economists have tried to understand decision making by using theory, constructing complex mathematical models, and mining data sets from real economic interactions. So how could mapping blood moving through the brain add to this work? "There are some economists who say brain data is just outside the domain of economics," says David Laibson, a professor of economics at Harvard University and one of the researchers working on the Saturday Night Live study. "Economists have a very spare framework for understanding human behavior. The framework is optimization. And that is a wonderful tool for understanding most economic behavior, but there is a lot of economic behavior that deviates from full optimization." Laibson doesn't want to abandon economics as it has been done but neither does he want to exclude potentially useful information.

"What has been and will be successful in economics are elegant, powerful mathematical models that explain the world and successfully predict what will happen next," Laibson says. "What has changed is the presumption that the good models will be perfectly rational and free of psychological mechanisms." He adds, "More and more economists are saying even though choice is the preeminent object of analysis we're interested in, to the extent they help us better understand choices, the underlying mechanisms that effect economic behavior belong in the economic models." And the activity in the brain in the seconds leading up to a decision may be the ultimate underlying mechanism.

Jonathan Cohen, a professor of psychology at Princeton University and co-director of the Princeton Neuroscience Institute, is working with Laibson. Cohen says psychology and neuroscience have traditionally been heavily experimental, but he shares an economist's presumption of optimization, with a caveat. Our neural architecture comes out of millions of years of evolution; what now appear to be irrational behaviors might have made perfect sense in other environments. For example, people have a difficult time saving enough for retirement. They seem to irrationally value the present over the future. "When we were lizards there were no bank accounts, there were no refrigerators, and there were no property rights, so there was no future. There was just what you had right now. We developed steep discount rates — use it or lose it," Cohen says. "Over the course of evolution we ended up developing, as humans, a part of our brain, the prefrontal cortex, that knows how to think about the future. I would contend that it is the very part of the brain that gave us things like bank accounts and refrigerators and property rights. It created a world in which there was a future."

In most circumstances, when there is a purely rational choice to be made, the appropriate region of the brain takes the lead making that choice. And when you touch a scalding object, a different, instinctive part of the brain takes over: there is no listing of pros and cons before you pull your hand away. But in some situations, what region of the brain should be taking charge is not as clear. Neuroeconomists are very curious about how our brain negotiates those situations that aren't black or white. But understanding the brain, whether at the level of interactions among specialized regions or the level of a single neurotransmitter, is in no way simple. Human brains have 100 billion neurons, and each makes contact with about a thousand others, creating more potential pathways for connections than there are stars in our galaxy.

Though neuroeconomics is perhaps a decade old, it is generating a tremendous amount of new information in forms that economists don't typically deal with. Some neuroeconomics looks at the molecular level, where the biochemistry of trust, which spills into the world as the social cooperation at the foundation of every financial transaction, appears to be controlled by oxytocin and the dopamine system seems to be a fastidious bean counter tracking rewards. Animal studies record the sound of a single neuron firing. Then there are eye-tracking studies, and an alphabet soup of fMRI, TMS, PET, EEG, and MEG data.

Camelia Kuhnen, a professor of finance at the Kellogg School of Management, has correlated risk profiles from neuroeconomic studies with genetic markers and found that two genes may predict a great deal about our risk tolerance or aversion. She quickly cautions that "genetic factors only account for about 30% of the variation across people in how much financial risk they take. The other 70% comes from non-genetic factors like culture, education, and experience in the marketplace." Parsing the roles of nature and nurture in financial behavior could help both individuals and policymakers. Neuroeconomics may help resolve the information inefficiencies and fairness issues that have made economically optimal decisions about public goods impossible. Whether the public good is national defense, clean air, or a town swimming pool, because everyone can share the benefits equally, each person has an incentive to understate its value to him or her, in order to minimize his or her contribution toward funding it. This leaves leaders without accurate information. Antonio Rangel, an economist at Caltech, led an experiment in which a group of volunteers assigned values to a mock public good while being scanned in an fMRI machine. Knowing that all of the participants' valuations would be corroborated with fMRI scans, each person gave very accurate information, eliminating the "free rider problem."

Though the technology is expensive and the applications largely experimental right now, Rangel looks into the future, imagining how this approach could be used to address a variety of problems. He speculates that a technology that provided ongoing neural feedback might let people making decisions with high uncertainty in high pressure environments, such as a financial trader or a commander on a battlefield, correct bad decisions before they are implemented.

The promise of neuroscience makes him willing to do exploratory work that may straddle disciplines awkwardly for a time. "I believe strongly that neuroscience is reaching a critical point with a set of tools and a set of theories that are allowing us to test for the first time the basis of human nature in a number of domains, especially how we make decisions. And I think over the next couple of decades that is going to fundamentally change how the world understands what a human being is," Rangel says. "At the same time, it is not obvious at all how this is going to change the way economics is done on a day to day basis."

The mix of excitement and uncertainty is common among those working in the field of neuroeconomics. Laibson says, "In the long run, it is going to be a very, very useful body of work. In the short run, it is impossible to tell whether anything we have done to date will have merit as insights unfold over the next 20 years."

Written by Ted O'Callahan