Not as rational as we think we are

March 17, 2008 BY danariely

USA Today By Russ Juskalian, Special for USA TODAY

Can thinking about an arbitrary number influence how much you’re willing to pay for a computer keyboard, a bottle of wine or a box of chocolates? Apparently so – and the degree of influence may shock you.In Predictably Irrational, Dan Ariely, a professor at Massachusetts Institute of Technology’s Media Laboratory and the Sloan School of Management, put the question to the test in an experiment involving a group of MBA students.

The experiment began with students being asked to write down the last two digits of their Social Security number. When the experiment ended, it revealed a pattern – that students with Social Security numbers ending in the highest-ending digits (80-99) were willing to pay more for items (the wine, the chocolates, etc.) than students with the lowest-ending Social Security numbers (01-20) were willing to pay.

Experiments such as this make up the foundation of Ariely’s book. Ariely argues that while economists continue to base theories on the idea that humans are rational – that we make optimal economic choices based the information we have – the notion is fundamentally flawed. Not only are we irrational, says Ariely, but when and in what form irrationality surfaces is predictable.

Economics can be a tough subject to tackle, but Predictably Irrational is surprisingly entertaining. While the book belongs in the same family as Freakonomics, don’t expect the same kind of theoretical hand-waving. Ariely is less interested in regression analysis and more interested in simple behavioral experiments such as trying to determine if the first person to order a beer at the table is happiest with his choice (yes).
The book may be easy to read, but not everything Ariely writes about should be dismissed with a chuckle about how foolish we are. In a particularly racy chapter, he details an experiment in 2001 involving 25 male students at the University of California, Berkeley. They were asked to imagine being sexually aroused, and to answer questions related to sexual preferences and “the likelihood of engaging in immoral behaviors such as date rape.” In this case, a participant was said to be in a “cold” state.

Next, the students were asked to look at erotic pictures in a private setting, then answer the same questions as before. In both cases, the students’ answers were rated on a scale between “no” (zero) and “yes” (100). The findings shocked Ariely. “The conclusions were consistent and clear – overwhelmingly clear, frighteningly clear.” When in the “cold” state, the average score was “5” for the question: Would you slip a woman a drug to increase the chance that she would have sex with you? For those in the aroused state, the average score had increased by 420% to “26.”

In question after question, the participants in an aroused state said they were more likely to engage in immoral behavior. In effect, the participants could not predict how they would act once aroused. Our irrationality (the propensity to calculate risks and benefits differently in various emotional, social and analytical states) can have wide-ranging consequences, Ariely says. And we’re better off understanding how humans really function than how we should function as ideal economists or moralists.

Ariely has written a book so user-friendly that he only scrapes the surface of what the implications of irrationality really mean. Minor disparities in how you or I calculate the worth of a keyboard probably doesn’t matter in the long run. But when these problems are elevated to a macro level – to nations dealing with carbon emissions and global warming or to the world trying to alleviate poverty and disease – our irrationality becomes glaringly important.

Ariely’s book makes economics and the strange happenings of the human mind fun, but it could do a little more to explore the bigger meaning of his research. Then again, maybe it’s the first step toward drawing in a crowd that might otherwise never have thought about economics and behavior. And that would be quite an achievement.