Functional MRI snoops out how, why we decide to buy

Beneath the layers of rational thought that insulate humanity from other animals lie the engines of our discontent, the clusters of neurons unevolved from pre-Neanderthal days whose firings bring fear and anger, wants and desires. Beside them are the centers that distinguish us by our empathy, charity, and good will.

But the line between the present and our evolutionary past is often blurred by an internal battle raging among these different regions of the brain. Spoken words and actions, visions of the real world as well as those artificially created, trigger these brain regions with neither our knowledge nor our control.

Functional MRI is peeling back the layers, exposing why we do what we do, finding the realities beneath the concept of free will, helping us understand ourselves, making us more resistant-or maybe more vulnerable-to the manipulations of others.

With a 3T scanner at the University Hospital of Ludwig-Maximilians University in Munich, radiologist Dr. Christine Born is trying to understand how the mind perceives and processes brand names. The focus of her research is on the automotive industry.

Born has found that well-known brands of cars activate positive emotional responses in our brains, that these responses occur in specific parts of the brain, and that they do not depend on the type of car, just the brand.

In research reported at the 2006 RSNA meeting and expanded upon at the European Congress of Radiology in 2007, Born studied the neural responses of 20 adult men and women who, while in the scanner, viewed a series of three-second visual stimuli containing the logos of well- and lesser known brands of car manufacturers and insurance companies. A brief question was included with each stimulus to evaluate perception of the brand. The volunteers pressed a button to respond using a four-point scale ranging from "disagree" to "agree strongly." The fMRI recorded areas of the brain responding to the different stimuli.

Strong brands activated a network of brain areas involved in positive emotional processing and associated with self-identification and rewards, namely the inferior frontal gyrus, anterior insula, and anterior cingulate. A small cluster of neurons also fired predominantly in the left hemisphere of the precuneus.

Weak brands, however, generated a different response, causing bilateral activation in the precuneus. They produced higher levels of activation in areas of working memory and a negative emotional response.

"The brain is for marketing experts a kind of black box," Born said.

She and her colleagues are using fMRI to open it.

"Brain branding," as Born calls it, is on the periphery of a new kind of medical imaging that uses advanced imaging technologies to understand how we make decisions. This could have enormous implications for how we view ourselves, how we act, and how others-including corporations and governments-interact with us.

Functional MRI goes beyond the traditional means for understanding decision making, such as questionnaires, surveys, and focus groups, to gather information about how the brain responds at a very basic level. It color-codes different levels of blood oxygenation, which correlate with changes in neural activity, stripping away the filtered responses of subjects who may consciously-or subconsciously-tell investigators something other than what they really feel.

This was made clear in 2003, when Dr. P. Read Montague Jr., a professor of neuroscience at Baylor College of Medicine, used fMRI to gauge consumer reactions to Pepsi and Coca- Cola. The research grew out of television commercials some 30 years earlier in which the two soft drinks went head to head in taste tests. Pepsi always won. But if Pepsi really tasted better, Montague reasoned, then several decades later, Pepsi should have been dominating the marketplace-and it wasn't. So Montague did his own taste test, tracking subjects' brain responses with fMRI.

Taste tests were split about evenly in subject preference between Pepsi and Coke, until the subjects were informed which was which. When Coke drinkers were informed, three-quarters of them stated a preference for its taste, at which time the fMRI showed a marked change in brain activation to the medial prefrontal cortex. This part of the brain, believed to control rational thought, sprang to life, possibly through the recall of advertisements that had tied the brand to positive associations.

The publication of these results in October 2004 in the peer-reviewed journal Neuron is widely hailed as the moment when science began to replace conjecture as the basis for marketing. This newfound ability has also raised questions about how such research should be done.

"With new imaging technologies, researchers are able to conduct experiments that have the potential to predict behavior, consciousness, and pathology," said Judy Illes, Ph.D., director of the neuroethics program at the Stanford Center for Biomedical Ethics. "These types of experiments raise ethical concerns about how to handle the sensitive data and how much individuals want to know-or want other people to know-about their mental state."

Marketing magnates have no such reservations, as they spend millions of dollars in efforts to understand the mind of the consumer. Many would be thrilled if the brain was found to house a "buy button" they could push, a collection of neurons that, when activated, launches the consumer to make a designated purchase.

While no such button may ever be found and, in fact, is not likely to exist, research indicates that some people's purchasing decisions may be very simple. Impulsive people who feel compelled to buy or acquire things immediately show brain activity in a particular reward center of the brain, according to research published in the Dec. 20, 2006, issue of The Journal of Neuroscience. Ahmad Hariri, Ph.D., director of the developmental imaging genomics program at the University of Pittsburgh, and colleagues showed that the ventral striatum, a core component of the brain's reward circuitry, lights up when subjects express such impulsiveness.

Hariri gave 45 subjects a choice between receiving an amount from 10¢ to $105 that day and receiving $100 at one of seven points up to five years in the future. Individuals with the strongest preference for immediate rather than delayed rewards showed the most activity in the ventral striatum. This part of the brain, Hariri concluded, appears to play a key role in striking a balance between gratification and delay, impulsive action and prudent choice.

"This could have far-reaching implications for our current and future well-being," Hariri said.

Most purchasing decisions, however, are much more complex, as seen in the activation patterns of subjects participating in fMRI studies conducted by Brian Knutson, Ph.D., at Stanford University. Knutson, an assistant professor of psychology and neuroscience, gave $20 to each of 26 adults who were given the choice of keeping the money or spending it on various products. The researchers monitored several regions of the brain believed to be involved in making this decision.

"None of these regions is like a 'buy button' or an all-or-none predictor," Knutson said. "They are all probabilistic; they are all working together to help you make the decision about whether you want to buy that thing or not."

Images of the products appearing on a computer screen visible in an MR scanner activated the nucleus accumbens, a subcortical region of the brain associated with the anticipation of pleasure when products appealed to the subjects. Excessive prices, however, activated a different region: the insula, a part of the brain associated with balancing gains and losses. Simultaneously, excessive prices deactivated a third region of the brain, the medial prefrontal cortex.

As much as these various regions seem to work together, they also conflict. When the nucleus accumbens and insula lit up, the subject chose to buy the product. But when the insula lit up and the medial prefrontal cortex shut down, the subject chose to keep the money.

The study, published in Neuron, casts purchase decisions as a balancing act between the pleasure of buying something and the pain of paying for it. The results might explain the tendency of consumers to overspend when purchasing items with credit cards instead of cash, because they reap the immediate reward but don't immediately suffer the penalty.

The processes behind economic decisions can get even more complex when they involve conflicting wants and needs, as occurs when people must weigh their emotions as well as the uncertain facts.

Montague, whose research into Pepsi and Coke gave birth to the field of neuroeconomics, has since looked into the effect of "what might have been" on decisions involving investments. In results that appeared in the May 29 Proceedings of the National Academy of Sciences, Montague and colleagues reported the discovery of a "fictive" or "what if" signal in the brain's standard reward signal that reflects actual experience.

Researchers found that fictive learning, the result of thinking "what might have happened," played an important role in subjects' decisions. This effect appeared as a distinct selective activation signal in the ventral caudate nucleus.

This fictive learning signal discovered by Montagu does not necessarily produce a conscious "feeling," he said, but contributes to the brain's computation and planning operations in a way that can be experimentally probed. It provides a window into the emotion of regret for a path taken or not taken, and it can exert a strong influence on future decision-making.

The brain's response to mistakes was synaptically palpable at the University of Michigan, Ann Arbor, when Dr. Stephan Taylor, an associate professor of psychiatry, and colleagues used fMRI to capture increased activity in the rostral anterior cingulate cortex (rACC), a part of the brain believed to be involved with emotional responses. In the test, correct answers carried various monetary rewards, incorrect answers incurred penalties, and some tests had neither rewards nor penalties. Subjects had nanoseconds to press a button when the visual cue was displayed. They received immediate feedback regarding their choice. Results were published in the April 12, 2006, Journal of Neuroscience.

The rACC lit up when the subjects realized they had made a mistake and that they had lost money because of it. This part of the brain did not light up, however, if the mistake carried no penalty.

The researchers compared the results to data from a similar study of people diagnosed with obsessive-compulsive disorder. These patients showed activation of the rACC when they made mistakes-even when the mistakes carried no consequences. Patients suffering from OCD may have a hyperactive response to making errors, Taylor said, with increased worry and concern about having done something wrong.

DECISION-MAKING PROCESS

Much of the research into neuroeconomics is undertaken in the hope of better understanding the human mind. The decision-making process that is revealed in economically based choices may be similar to, if not the same as, that when other types of choices are involved. Montague believes that the fictive learning signal he found embedded in the reward signal might provide insights into why people choose to abuse drugs even though they can imagine the bad consequences.

Not all research has such lofty goals. The fMRI studies conducted by Dr. Joshua Freedman, an assistant clinical professor of psychiatry at the University of California, Los Angeles, and colleagues last January gauged subjects' responses to Super Bowl commercials. It was the second year Super Bowl commercials had been evaluated using fMRI.

The research was conducted at the behest of FKF Applied Research, a marketing firm, cofounded by Freedman, that counts Fortune 500 companies among its clients. The group claims its methods "uncover previously hidden or suppressed thoughts and emotions." Stimuli presented to subjects can be TV ads, print ads, Internet images, logos, slogans, sounds, even smells.

The Coca-Cola "video game" ad, an animated fantasia of Rube Goldberg-like mechanisms triggered by the deposit of money in a Coke machine, topped the list of the most influential ads during the 2007 Super Bowl, lighting up neurons associated with positive emotions, according to FKF. Commercials that didn't do as well stoked regions of the brain associated with anxiety, notably the amygdala. This happened in the FedEx ad, when a caveman was crushed by a dinosaur.

Functional MRI provides very precise measurements, but is not the only way to tap into the unconscious. David Lewis, Ph.D., and colleagues at Neuroco, a U.K.-based market research firm, outfitted subjects with a helmet composed of quantitative EEG sensors and an "eye cam" that correlates what is seen with the accompanying thought processes. Neuroco has used QEEG to examine consumer brain responses, for example, while shopping in retail stores, reporting their findings to commercial sponsors.

These studies lack the precision of fMRI, because the skull diffuses brain signals, said Knutson, who relates the EEG experience to being outside a stadium during a baseball game.

"You hear the crowd roar. You know something happened. But you don't know who hit what to where," he said.

OLD AGE MEETS NEW AGE

Proponents of this new age of research point to the positives that can come from understanding how we make decisions, even when that understanding is focused purely on consumerism. Born's research in Germany is aimed at better understanding the needs of people, she said, and creating markets that are more oriented toward the satisfaction of those needs.

She and her colleagues are now looking at what happens in the brains of people age 50 and older. One question is whether the advertising done in Germany resonates with this age group, which has amassed a significant portion of the country's wealth, Born noted.

"Research aimed at finding ways to address individual needs may contribute to a higher quality of life," she said.

Both the National Institutes of Health and National Science Foundation are getting interested in how aging affects financial decision making, Knutson said.

"Very little is known about this. And it's important to understand, because we have an aging population and there may be issues in terms of financial reserves," he said.

Greg Freiherr is business editor of Diagnostic Imaging magazine.