© chameleonseye/iStockphotoBreaking the stranglehold of conformity only takes a few outliers.
Even when people try to be different from each other, they may end up conforming with the majority around them, a new model suggests.Breaking the stranglehold of conformity only depends on a few extreme outliers, say the authors of a new paper published today in the Royal Society journal
Open Science.
"Conformity is 'doubly hard' to beat -- it can obviously happen when people imitate one another, but it can also set in even when folks are trying to be distinct," says mathematical social scientist Professor Joshua Epstein of
Johns Hopkins University.
"You have to be creative if you want to resist it, reverse it or interrupt it."
Epstein says previous attempts to model conformity in social systems have focused on how it results from some kind of imitation.
In their model, he and colleague Dr Paul Smaldino of the
University of California, Davis did something different. They allowed for people wanting to be a bit different from the average.
Each individual in the model had a starting position -- it could be a political view or a fashion statement -- that was measured in standard deviations from the mean.
And they had a preferred position, which might be different from this starting position. As the model ran, an individual would move in the direction of their preferred position (towards or away from the mean). But as each individual moved, this changed the mean and also changed the movement of all other individuals.
"There's this constant feedback between people's positions and their objectives."
The amazing thing was that the end point of this process was that all positions collapsed in to the average position."If all they [people] care about is moving in the direction of the goal, they all end up in the same darn position," says Epstein.
"It is a very counter-intuitive result ... Nobody had really thought that you could get conformity out of people trying to be distinct."
Epstein says this could help explain phenomenon such as fashion cycles.
Comment: