Have you ever been surprised at yourself after reviewing what you brought home from a shopping spree? Perhaps you bought a certain brand of chocolate only because a television ad for it showed warm tropical beaches and palm trees. Or you purchased an overpriced pastry at a bakery because it somehow reminded you of a treat from your childhood. Maybe you sprang for an electronic gadget you knew you didn't need just because your colleagues already have it.

Emotions, memories, herd instincts and other intangibles all influence our buying decisions. And none of these factors involves the classic cost-benefit analysis we generally pride ourselves on: that we won't buy something unless what we get seems worth what we must pay. What is going on inside our heads when we make such decisions? Marketers would certainly like to know. With modern neurotechnology, they are beginning to find out.

A Matter of Taste
Neuroscientists around the world are assessing why we respond to certain product advertising. Early investigations have examined traditional beliefs such as the one that macho imagery sells cars to men. Henrik Walter, a neurologist and psychiatrist at University Hospital in Ulm, Germany, showed 66 black-and-white photographs of sports cars, sedans and small cars to 12 young males while mapping their brain activity with functional magnetic resonance imaging (fMRI). The scans showed that a brain structure called the nucleus accumbens was significantly more active when the men viewed the sports cars. This tiny region is a center for self-reward. It is activated by the signaling molecule dopamine and releases endogenous opiates--substances linked to lust and pleasure.

Usually the nucleus accumbens is active when it receives signals that are important to survival--for instance, sexual stimulants or the prospect of food. Even a huge car fan would not claim that a Ferrari is critical to survival, however, so why does seeing such a vehicle trigger the release of dopamine? Because the object is one of perceived desire. Advertising has taught men that powerful sports cars are highly desirable.

Walter's study, part of work he is doing for DaimlerChrysler, is just one of a growing number designed to probe how we react to advertising and therefore how marketers might tailor their ads to target buyers. In the future, carmakers could ostensibly test which models or design variations within a model prompt the strongest emotional response. Indeed, General Motors and Ford, not to mention Camelot, the U.K.'s national lottery operator, have begun to examine how neurological surveys could augment traditional market research.

So far most so-called neuromarketing studies have used fMRI, which shows how metabolism changes in various brain regions. Perhaps the most notable test was a 2004 exploration by neuroscientist P. Read Montague of the Baylor College of Medicine and his colleagues. Montague turned the fMRI machine on one of America's most classic brand-name duels: Coke versus Pepsi.

For years, marketing experts have wondered why Coca-Cola remains the best-selling cola even though its archrival, Pepsi-Cola, triumphs regularly in blind taste tests. Montague invited 67 participants to help him find an answer. Subjects were given unlabeled samples of each beverage while inside an fMRI scanner. Both sodas did about equally well. Both drinks elicited strong but similar reactions in the ventral putamen, a brain structure believed to be the registry of satisfaction.

Montague then repeated the test but identified the samples as Coke or Pepsi. This time three quarters of the participants stated that Coke tasted better. The fMRI images showed that while the ventral putamen had still been activated equally by the two brands, when participants were told they were drinking Coke they also showed heightened activity in the medial prefrontal cortex. This region is linked to complex thoughts, evaluations and self-image. Obviously, impressions associated with Coke overpowered the primary taste sensation. Montague concluded that Coke's long-term advertising had succeeded in getting cultural messages to affect areas of the brain that influence personal preferences.

Herds and Memory
That men favor sports cars and cola drinkers favor Coke is not news. So what does neuromarketing bring to the party? Objectivity--proof of a mechanism that links stimulus and effect. For decades, companies have employed traditional market research tools in their attempts to determine why customers prefer one product over another: standardized questionnaires, individual interviews with open questions, and focus groups of potential buyers. The problem with these methods is that they rely on participants to describe their motives in words and to do so without bias. But not every cola drinker knows what drives her to her favorite drink, nor can she objectively describe the impulse, given the bias of the marketing influence she has already been exposed to. Neuroscience's tools, in contrast, skip all emotion, introspection and ego.

These tools probe beyond the herd behavior of consumers, too. Gregory S. Berns, a neuroscientist at Emory University, asked 30 participants to compare 54 pairs of abstract three-dimensional figures and to decide if the paired items were similar or different. He also provided them with answers that he said other participants had given--some were correct, but some he changed to incorrect. The fMRI scans recorded during these tests showed that subjects who trusted the majority's opinion more than their own had greater activity in the prefrontal cortex, where decision making occurs. They chose wrong answers more frequently, because they believed the rigged group responses. Individuals who stuck to their own analyses--and got more right answers--showed less activity in the same region.

Another team, led by Richard B. Silberstein of Swinburne University of Technology near Melbourne, Australia, examined why some advertisements remain in people's memory better than others do. Silberstein showed women a televised documentary that was interrupted several times by commercials and recorded their brain waves during the presentation. One week later he gave the women a memory test about the commercials; the ads they remembered best were the ones that had triggered an unusually fast surge in the electrical activity of the left frontal lobe. Marketers could potentially use this finding to test prototype commercials--a strong surge could imply which commercial would stay in long-term memory.

The Neural Shopping Center
Neuroscientists are admittedly fascinated with such early results. Most of them caution, however, that conclusions are preliminary at best. Nevertheless, advertisers are proving impatient and are looking into how brain tests can enhance traditional market research. Some self-proclaimed neuromarketing experts also seem to be more interested in making money than in serious science. A few consulting groups already promise they can help potential clients build loyal, long-term relationships with their customers.

Walter says too many people are overestimating the power of brain scans. For one thing, he notes, the impressive, colorful images need careful interpretation. They do not directly reflect cognition; they are statistical analyses, and their predictive power depends highly on which statistical variables a researcher chooses to show.

Gerald Zaltman, professor emeritus at Harvard Business School and author of How Consumers Think (Harvard Business School Press, 2003), sees another danger: a widespread misunderstanding that it is possible to identify specific brain areas that react exclusively to certain marketing stimuli. The idea that there is a specific buying center in the brain may be alluring from a marketing point of view, he says, but it is as wrong as the long-discredited phrenology thesis that a person's mental disposition can be understood by mapping the bumps on his or her skull.

To fully explain what is going on in a person's head as he or she watches a commercial or eats a chocolate bar requires an understanding of many different brain regions and how they interact. Neuroscientists can say little about this topic even for well-studied brain diseases, much less ephemeral phenomena such as reactions to ads. Neuromarketing faces practical limitations, too, notably the high cost of using a sophisticated fMRI machine for even a handful of research subjects. Distortion of results must be considered as well; individuals are not likely to react to a soda, hamburger or car when lying inside a narrow fMRI tube or when they have electrodes pasted to their heads in exactly the same way as they would when lounging on their sofa.

Still, given how hard corporations fight to find an edge for their product, researchers such as Walter imagine that imaging technologies will probably become standard tools for assessing marketing strategies. Silberstein adds optimistically that the trend could even benefit consumers: if companies can indeed find out what entices people, then perhaps they will offer products that will better satisfy customers. Neuromarketing has also drawn criticism from some consumer advocates who worry that the technique could lead to manipulative strategies by corporations--a modern-day equivalent of the subliminal messages discovered in movie reels of the 1950s. But Montague takes an optimistic view: studies like his on Coke and Pepsi could actually empower consumers by making them aware of their susceptibility to advertising messages and images.