Experimental Economists Find Brain Regions That Govern Fear of the Economic Unknown

PASADENA, Calif.—Do you have second thoughts when ordering a strange-sounding dish at an exotic restaurant? Afraid you'll get fricasseed eye of newt, or something even worse? If you do, it's because certain neurons in the brain are saying that the potential reward for the risk is unknown. These regions of the brain have now been pinpointed by experimental economists at the California Institute of Technology and the University of Iowa College of Medicine.

In the December 9 issue of the journal Science, Caltech's Axline Professor of Business Economics Colin Camerer and his colleagues report on a series of experiments involving Caltech student volunteers and patients with specific types of brain damage at the University of Iowa. The object of the experiments was to see how the brain responded to degrees of economic uncertainty by having the test subjects make wagers while being scanned by a functional magnetic resonance imager (fMRI).

The results show that there is a definite difference in the brain when the wagers add a degree of ambiguity to the risk. In cases where the game involves a simple wager in which the chance of getting a payoff is very clearly known, the dorsal striatum tends to light up. But in a nearly identical game in which the chances of winning are unknown, the more emotional parts of the brain known as the amygdala and orbitofrontal cortex (OFC) are involved.

According to Camerer, this is a clear advancement in understanding the neural basis of economic decision making. Much is already known about how people deal with risk from the standpoint of social sciences and behavioral ecology, but greater understanding of how the brain structures are involved provides new insights on how certain behaviors are connected.

"The amygdala has been hypothesized as a generalized vigilance module in the brain," he explains. "We know, for example, that anyone with damage to the amygdala cannot pick up certain facial cues that normally allow humans to know whether they should trust someone else."

Problems with the amygdala are also known to be associated with autism, a brain disorder that causes sufferers to have trouble recognizing emotions in other people's faces. One of the authors of the paper, Ralph Adolphs, the Bren Professor of Psychology and Neuroscience at Caltech, has done extensive work in this area.

As for the OFC, the structure is associated with the integration of emotional and cognitive input. Therefore the OFC and amygdala presumably work together when a person is confronted with a wager for which the odds are unknown-the amygdala sends a "caution" message and the OFC processes the message.

The researchers set up the experiments so that the "risk" games and "ambiguity" games looked similar, to control for activity in the visual system so they could focus only on differences in decision making. In the "risk" games, each test subject was provided an opportunity to either choose a certain amount, like $3, or else choose a card that could be either red or blue. If the card was red, the test subject got $10, but if it came up blue, the test subject got nothing for that particular card.

In the risk games, each test subject was informed that the chance of drawing a red card was 50 percent, that there would be 10 of each color out of the total of 20 cards. Subjects made a series of 24 choices, with different sums of money at risk and different numbers of cards. In the ambiguity games, however, each test subject was told that the deck contained 20 cards, but was told nothing about how many were red and how many were blue.

As predicted from past experiments in which this type of risk was observed in test subjects, the researchers knew that the Caltech subjects with no brain damage would be more likely to draw cards in the risk game than in the ambiguity game, because people dislike betting when they do not know the odds. They were more likely to take sure amounts, which meant that their fear cost them money in expected value terms.

The patients at the University of Iowa Medical School, on the other hand, who had lesions to the OFC, played the game entirely differently. On average, these subjects with damage to the OFC were much more tolerant of risk and ambiguity.

Camerer says that the result with the brain-damaged test subjects fits well with the observation that many have suffered in their personal lives due to reckless financial decisions.

The research also addressed the intensity of the response in the brain as it correlates with degrees of risk. The results for the Caltech students showed more intense activity in the amygdala and OFC when the chance of winning is ambiguous, but there would be no such difference in patients with damage to those areas.

In sum, the results provide an important neurological understanding of how we humans handle risk in the real world, Camerer says.

"If you think about it, how often do you know the probability of success? Probably, the situation we modeled with the risk game is more the exception than the rule," he says. "In most situations, I think you are confronted with a risky choice in which you have little idea of the chances of different payoffs."

Does the study have any applications for society? Camerer says that our knowing what is happening at the most microscopic level in the neurons of the brain could lead to better understanding of bigger social effects. For example, a fear of the economic unknown will also create a strong preference for the familiar. In every country in the world, investors hold too many stocks they are familiar with, from their own countries, and do not diversify their stock holdings enough by buying ambiguous foreign stocks. The opposite of fear of the economic unknown may be driving entrepreneurs, who often thrive under uncertainty.

"It could be that aversion to ambiguity is like a primitive freezing response that we've had for millions of years," Camerer says. "In this case, it would be an economic freezing response."

The study is titled "Neural Systems Responding to Degrees of Uncertainty in Human Decision Making."

In addition to Camerer and Adolphs, the other authors are Ming Hsu and Meghana Bhatt, both graduate students in economics at Caltech; and Daniel Tranel of the University of Iowa College of Medicine.