Dan Ariely, professor of psychology and behavioral economics at Duke University, gave a fascinating talk at a conference I attended a few months ago. (The talk was particularly powerful because the previous speaker, David Freedman, gave an unfortunately weak presentation, far beneath the quality of his article "Lies, Damned Lies, and Medical Science".) Ariely discussed perception, bias, and decision-making, with comments based on his research and personal experience.

In high school Ariely was accidentally burned over ~70% of his body and spent three years in a hospital. He saw lots of mistakes there in patient care, and when he became a professor he did experiments on his students (!) including putting their fingers into vises, giving them electrical shocks, etc. Among the results of his measurements of pain and perception:

• pain intensity matters more than duration — for instance, yanking a bandage off and causing intense pain is worse than removing a bandage slowly, according to his volunteer-victims
• pain progression over time makes a difference — it's best to get the worst over with first and then have decreasing levels of pain
• taking breaks to recover helps

The nurses who treated Ariely in the hospital meant well, he said, but they generally yanked the bandages off his burns abruptly, started with his feet and worked their way up to his head where it hurt more, and didn't give him any time to rest up between the necessarily painful steps of the procedure. They didn't know any better. Hence, the importance of doing experiments and measuring what people actually prefer, which can be counterintuitive. This field, behavioral economics, became Ariely's career choice.

Prof. Ariely discussed the importance of defaults. Organ donation, for example, rates vary wildly between otherwise-similar nations in Europe. The entire difference is explained by the opt-out versus opt-in choice given to potential donors. "When we don't know what to do, we do nothing!" summarizes the situation, Ariely said, especially when the situation is complex. Ariely gave examples from medicine, from retirement planning, and from marketing. Offer people 6 different flavors of jam to taste, and most of them ll try and then buy something. Offer them 24 flavors and they're overwhelmed and generally don't buy any. Ariely's conclusion: "Defaults are neither good nor bad ... but as people who design environments, we should consider defaults carefully" — and help people make the best possible decisions by setting the default choice to something sensible.

In another analysis that Prof. Ariely described, folks who are asked to do something difficult become biased against the activity. Ask some people to come up with 3 reasons why they love their spouse, for instance, and ask others for 10 reasons. The ones who had to come up with 3 answers will then rate their spouses higher on a scale of appeal than the folks who had to struggle to produce 10.

Asking a question forces people to make up an answer — and once we have an answer we tend to become more certain. "When we don't know what we want, we focus on the small universe of choices given," Ariely observed. An economist who believed that humans are rational actors would have to conclude that "people prefer food to living", given the epidemic of obesity today.

Ariely then discussed cheating and how honor systems work, when they do. He talked about the "Ikea phenomenon", that we love things more when we've made them ourselves. The same principle applies to craft projects, to ideas, and to the love one has of one's own children. Conflicts of interest cause people to rationalize their behavior. One's culture can affect choices also. In a restaurant in the USA, people in a group feel they should order different things; in Japan, diners feel they should order the same dishes.

Dan Ariely is the author of Predictably Irrational and The Upside of Irrationality. Recently I began reading the first; much of its content overlaps Ariely's lecture. Review to follow ...

^z - 2011-02-10