Jonah Berger, a professor of marketing at the University of Pennsylvania, made news a few months ago when he published the results of a study demonstrating that where people cast ballots affects how they vote. Although voters think they are making rational decisions based solely on the issues and facts, they are actually subtly influenced by a long list of other variables, most of which operate at an unconscious level. Mind Matters editor Jonah Lehrer chats with Berger about what this new research can teach us about elections, expensive clothing and the human brain.

LEHRER: Your most recent paper found that voters are more likely to approve a sales tax increase when voting in a school. Why? What do you think is driving this effect?

BERGER: We build on behavioral priming research, which finds that cues or stimuli in the environment, such as the things we see, can activate related concepts in our mind that carry over to influence behavior, even outside our awareness. In a classic study, for example, participants exposed to elderly related words ended up walking more slowly leaving the experiment. The idea is that the words activated the elderly stereotype, which includes walking slowly, and such thoughts influenced behavior. Similarly, in the case of polling locations, seeing lockers, desks and other things associated with schools might activate norms (such as the urge to take care of children) or identities (that is, being a parent) that then shift people to vote to support school funding.

LEHRER: Given the significance of this effect—it's enough to sway even a moderately close election—what steps do you think should be taken? Should we be more careful when choosing polling places?

BERGER: Policy makers should definitely pay more attention to where people vote and, if possible, be more careful in the types of places selected. Choosing polling places is already a tough task, though—they need to be centrally located, handicap accessible, et cetera, so we are not arguing to eliminate churches and schools altogether. Rather, if such places are used, there are ways to minimize their potential influence. For example, people should vote in a general multipurpose room rather than a classroom filled with children’s drawings or a church room filled with religious imagery. These steps should help reduce the influence.

LEHRER: In another recent paper, you begin by noting that after NASA landed the Pathfinder spacecraft on the surface of Mars, there was an unusual increase in sales of the Mars candy bar, even though the candy bar has nothing to do with the planet. What's going on here?

BERGER: The main idea here is similar to the voting paper. The planet and the candy bar share the same name, so repeated exposure to information about the planet, such as newspaper articles, should make the candy bar more active or accessible in consumer minds. Then when people are choosing what food or candy to buy, they may be more likely to select a Mars bar because it is at the top of their mind. This effect can also be used prosocially to improve health. In another study, we show that linking a message to eat more fruits and vegetables to a prevalent cue in the environment can act as a reminder, and lead people to make healthier choices.

LEHRER: You've pointed out that expensive brands come with a subtle paradox: consumers spend lots of money on these brands (such as Prada or Armani) but these high-end products are actually less likely to contain clear brand markers. Although that $20 Tommy Hilfiger polo shirt will almost certainly contain a Hilfiger logo, that $100 Prada t-shirt won't advertise its provenance. Why, then, do consumers spend so much money on expensive brands?

BERGER: Communication. Consumers want other people to think certain things about them, and so they wear clothes that communicate particular identities. Conspicuous consumption, or spending lots of money on visible goods, is a good way to try and communicate wealth, but this signal breaks down when any wannabe can buy a certain car or handbag. What were once status symbols become just aspirational markers rather than the real thing. Consequently, insiders may engage in more inconspicuous consumption to signal only to others in the know. Such subtle signals may be almost invisible to the mainstream, and this helps maintain their cache.

LEHRER: In a 2007 paper, you note that "Manhattanites stopped wearing mesh trucker hats when the bridge-and-tunnel crowd adopted them." In other words, the urbane crowd abandoned a fad once too many of the "wrong" people adopted it. You argue that this kind of switch is because consumers choose products that communicate a "desired identity." Is this true of all product domains? Or are certain consumer items more intertwined with our identity than others? If so, why? Why are hats more reflective of our identity than, say, backpacks?

BERGER: Certain domains are definitely more symbolic of identity and this fact has a lot to do with utility and function. People buy detergent based on what cleans the best, and consequently that choice doesn’t say much about who they are. Shirt choice, however, is based much less on function. Two different colored shirts do an equally good job of being shirts, so observers are more likely to attribute someone’s choice to something about their identity. Similarly, things that violate functionality are more likely to be seen as identity-relevant. Wearing sunglasses indoors makes it harder to see. Wearing a scarf in the summer is unnecessary. Consequently, these behaviors are more likely to be seen as signals of identity because they have little functional value.