Warning: strip_tags() expects parameter 1 to be string, array given in /afs/ir.stanford.edu/group/knowledgebase/cgi-bin/wp-includes/formatting.php on line 631
by Kathleen O’Toole
Magnetic resonance imaging machines are widely used to diagnose physical illnesses. Now they also hunt down abnormalities in theories of the mind. One of the granddaddies of those theories concerns the rational self-interested investor who knows his or her own needs best and will do a better job of investing for retirement than, say, the professionals who do it for California state employees.
Or at least that was what many economists thought before behavioral scientists teamed up with neuroscientists to watch blood flow through the brains of people making investment decisions inside MRI machines.
In separate experiments using functional magnetic resonance imaging—MRI for short, Camelia Kuhnen, a doctoral candidate in economics at the Business School, and Baba Shiv, an associate professor of marketing, found evidence that emotions can interfere with so-called rational investment decision-making.
Kuhnen teamed up with Brian Knutson, an assistant professor of psychology at Stanford, to study students asked to decide between two stocks and a bond during separate games. The volunteers were supposed to figure out which stock was good and which was bad by watching the market while in the MRI machine. The researchers found that the nucleus accumbens, a peanut-sized part of the brain that exhibits excitement when someone expects an immediate reward—such as water when thirsty—lit up with blood flow two seconds before volunteers invested in a stock even though it had a bad history. This suggested they took pleasure from taking risk. In contrast, the anterior insula, a part of the brain linked to anxiety when people are subjected to repulsive stimuli such as pictures of mutilated bodies, lit up just before volunteers decided to invest in the safe but suboptimal-performing bond.
Shiv and colleagues at Carnegie Mellon University and the University of Iowa, studied wagering decisions of people with normal IQs, some of whom had lesions in areas of the brain associated with emotion. Each participant was given $20 to risk $1 at a time on 20 coin tosses. They could decline to participate in all or any rounds and keep their money. Those who participated earned $2.50 if they won but had to give up $1 if they lost.
“From a logical standpoint, the right thing to do was to invest in every round,” Shiv says. “With a 50-50 chance of winning, the expected value of playing each round was $1.25, while the expected value of not playing was just $1.”
Those with brain lesions invested in 84 percent of the rounds, earning an average of $25.70. Normal participants invested in just 58 percent of the rounds, earning an average of $22.80.
Fear seemed to play a large role in risk-avoidance behavior of the normal participants, Shiv says. Over time, the normal participants grew more cautious, declining to play almost as often as agreeing to risk $1 on the coin toss. “And what we found out, through additional analysis, is that normal individuals were reacting emotionally to the outcome of the previous round,” he said. “If they lost money, they got scared and had the tendency to fall back and decline to play further.”
Studies such as these may shed light on the reasons for what financial experts call the “equity premium puzzle,” the large number of individuals who prefer to invest in bonds rather than stocks even though stocks have historically provided a much higher rate of return. According to Shiv, there is widespread evidence that when the stock market starts to decline, people shift their retirement savings—that is, their long-term, not short-term, investments—from stocks to bonds. “Whereas all research suggests that, even after taking into account fluctuations in the market, overall people are better off investing in stocks in the long term,” Shiv says.
“People are not as rational as we would like them to be,” Kuhnen says. “I’m happy we got these results because I believe this is evidence that economists should take [emotion] into consideration when we write [investing] models for individuals.”
The research by Kuhnen and Knutson was published last September in the journal Neuron. The research by Shiv and colleagues appeared in June in Psychological Science.
Also on Stanford Knowledgebase:
