Economic theory is built on the assumption that human behavior is rational and predictable. At its most basic, economic decision making involves three steps:
- Obtain information regarding possible actions.
- Evaluate those actions.
- Choose between them.
Yet we know that in practice, our allocation of scarce resources (in this case, money) is frequently hardly rational at all. Individuals spend large sums of money on lottery tickets, fully aware that their chance of winning back their investment is virtually nil. Meanwhile, these same individuals are apt to decline the opportunity to contribute to their 401(k), even when their employer will contribute a matching amount. And how does one explain such behavior as starting the year by opening a Christmas savings account that pays no interest and charges a penalty for early withdrawal, and then deciding in June to take the money and splurge on a high-definition television?
According to surveys, roughly half of us reach retirement age with no financial assets other than our Social Security payments. It’s not as though we lack the time to think about our long-term needs. Most of us are fully aware of the need to watch our diets and save for retirement, but all too frequently, momentary temptation overwhelms well-considered goals.
To be fair, economists know that their assumption is imperfect — that we are not always rational and predictable. In 1871, the economist William Jevons observed, “I hesitate to say that men will ever have the means of measuring directly the feelings of the human heart. It is from the quantitative effects of the feelings that we must estimate the comparative amounts.” While recognizing the potential predictive value of feelings, economists knew that feelings themselves were incalculable and had to be left out of the equation entirely. And the results have been relatively successful.
But with the advent of brain imaging, neuroscience is now acquiring the ability to make direct measurements of thoughts and feelings. Using magnetic-resonance imaging (MRI) and other brain-imaging technology, neuroscientists had established the specific roles of the different parts of the brain by the mid-1990s. By the late 1990s this research had come to the attention of a growing number of economists and neurologists, who began studying brain activity in subjects as they were presented with a variety of economic problems and games.
A central tenet of neuroeconomics is that the brain has several systems that usually work cooperatively in decision making. Sometimes they compete with one another. For example, suppose you and Joe are approached by a third party, who offers $10 and an ultimatum: the two of you can split the money any way the Joe chooses, as long as you go along with the decision. If you don’t accept the decision, nobody gets any money. Joe decides he’ll keep $8 and let you have $2. What do you do?
According to game theorists, you should accept any offer in this situation because otherwise you get nothing. Yet people playing this game consistently reject offers they consider unfair. MRIs of subjects receiving unfair offers revealed activity in the dorsolateral prefrontal cortex (DLPFC), a part of the brain associated with planning; the insula, which is associated with such negative emotions as pain or disgust; and the anterior cingulate (ACC), an “executive function” area that receives input from many areas and resolves conflicts. The three regions appear to argue: the DLPFC wants the money, the insula appears to be “disgusted,” and the ACC tries to resolve the conflict. Whether players accept or reject an offer can be predicted by the extent of their insula activity.
While findings such as this clearly demonstrate the value of neuroeconomics in advancing our understanding of economic decision making, not everyone is convinced. Many neuroscientists are distrustful of the results, observing that MRIs operate too slowly and with insufficient resolution to measure minute, lightning-fast neural activity. Others are doubtful that two different systems — rational and emotional — exist in the brain.
Another potential problem lies in the data-gathering process itself. Subjects are required to lie in a claustrophobia-inducing MRI machine, sometimes for an hour or more. They may have to make hundreds of economic choices. Can real-world behavior be deduced from data collected in such a setting? Will neuroeconomics simply replace one unrealistic economic model with another?