Know thyself (through others)
Next time you walk into a restaurant, don’t even look at the menu. Ask one of the regulars what he thinks the best dish is–you’re far more likely to be pleased with your meal.
This is the basic conclusion of a study led by Harvard psychologist Daniel Gilbert. They used a speed-dating scenario to test the accuracy of “affective forecasting”–people’s projections about how happy something will make them. The results indicate that when it comes to helpful knowledge about future experiences, less really is more: Women who were given more information about a man they were about to speed-date tended to make less accurate forecasts about how enjoyable the date would turn out. Women with only a tiny bit of info–the rating of the man by his previous date–were significantly better at estimating how satisfied they would walk away feeling. As Gilbert puts it, “If you want to know how much you will enjoy an experience, you are better off knowing how much someone else enjoyed it than knowing anything about the experience itself.” (You can read about the study here and here.) Is this good news, or does this spell big trouble for freedom and liberty?
If you think you know yourself better than anyone, Gilbert’s study should unsettle you at least a bit. An even more bizarre conclusion of the experiment was that participants largely preferred to be “informed” predictors–getting all the information available–even after they learned that it would make them less accurate at predicting their own happiness. (In a certain sense, the information makes them less informed!) They hold that deeply universal belief: “I know what is best for me.” It is a fundamental tenet of liberty, one of Western society’s core principles. People ought to be free to do what they want, our implicit logic goes, because they tend to want the sorts of things that enable them to lead happy, fulfilling lives.
So should we be “outsourcing” our decisions to the wisdom of the crowds? If so, what sort of decisions should we trust to others? Where to eat? What to read? Whom to love? (Hint: you probably outsource far more decisions than you realize.)
–Roland Nadler
March 25th, 2009 at 8:39 am
Here’s another way of motivating the problem: a philosophy professor of mine, in the first lecture for Ethical Theory, remarked that it would seem “terribly untoward” for someone facing an ethically challenging life decision to simply seek out and consult a “moral expert” and simply accept her recommendation without further deliberation. It makes sense: we would be likely to say that this person is being lazy, or doing something dangerous, or maybe even diminishing his personhood somehow.
The same thing seems to be true of decisions about personal happiness. It strikes us as a bad thing when someone basically says “tell me what will make me happy.” Is this just an obsolete psychological bias?
March 26th, 2009 at 8:38 am
I just read Gilbert’s original 20-March paper in Science:
http://www.sciencemag.org/cgi/content/abstract/sci;323/5921/1617
He sums up the implications for our lives like this:
“Surrogation is by definition superior to simulation when individual differences are relatively small and simulations errors are relatively large, and it is inferior to simulation when the opposite is true. Although there is no way to know which of these is more typical in everyday life, the situations we studied—dating and peer-evaluation—are by no means exotic.”
So there’s the (obvious) rub. You should eschew information other than another person’s enjoyment of an experience to help you decide whether to have the experience yourself (“surrogation”) if it’s the sort that doesn’t vary wildly between people (love-it-or-hate-it experiences) AND it’s the sort that you’re unlikely to accurately imagine (“simulation”) yourself doing. So the question is, what sort of experiences are those? Apparently, speed-dating and getting judged by peers. What else? How far does it go?
OK, one experience that is NOT amenable to surrogation is high school. I pretty much hated it. Many people have fond memories. Doing a PhD? The reverse. Perhaps there’s just too much variation in enjoyment for surrogation to be useful.
–John Bohannon
March 26th, 2009 at 10:38 am
It’s funny, because you would think a speed dating experience would be particularly unfriendly to surrogation. Maybe this is my Y chromosome talking, but tastes in potential mate vary widely from person to person. Maybe women are actually more likely to agree on what they want in a man, but I can say with confidence despite the small sample size that my three blockmates and I have completely divergent tastes in women and as such surrogation would be a disaster.
March 26th, 2009 at 10:42 am
But then, all that was measured was the enjoyment of the speed-date itself, not of actually dating the guys. It’s more of a surrogate measure of people’s ability to be good conversationalists than mates. Right?