Tuesday, April 06, 2010

Too smart for our own good?

An experiment that shows that too much smartness could be counter-productive! I think this is also the reason that good businesses are not built solely by numbers and analytics, but by that indefinable thing called the instinct or gut-feel or intuition.

Sourced from here:
The Trouble with Humans: Why rats and pigeons might make better investors than people do

"Psychologists have long known that if rats or pigeons knew what the NASDAQ is, they might be better investors than most humans are. That's because, in some ways, animals are better than people at predicting random events. If, for instance, you set up two lights in a laboratory and flash them in a random sequence, humans will persistently try to predict which of the two lights will flash next. Stranger still, they'll keep trying even when you tell them that the flashing of the lights is purely random. Let's say you flash a green light 80% of the time and a red one 20% of the time but keep the exact sequences random. (A run of 20 flashes could look something like this: GGGGRGGGGGGGRRGGGGGR.) In guessing which light will flash next, the best strategy is simply to predict green every time, since you stand an 80% chance of being right. That's what rats or pigeons generally do in a similar experiment that rewards them with a crumb of food whenever they correctly guess the next outcome".


But humans are apparently convinced that they're smart enough to predict each upcoming result even in a process they've been told is random. On average, this misguided confidence leads people to get the right answer in this experiment on only 68% of their tries. In other words, it's precisely our higher intelligence that leads us to score lower on this kind of task than rats and pigeons do.