Kevin Lewis of the Boston Globe shares some surprising insights from the social sciences, including this not-so-surprising insight on countering groupthink:
One of the most famous phenomena in social psychology is groupthink, the tendency of a group to converge on a consensus without much critical evaluation, even if the consensus is wrong. Various remedies have been proposed over the years, but some management researchers are presenting an interesting new angle on it. They invited a couple hundred members of fraternities and sororities to participate in a problem-solving experiment. The students were given 20 minutes to read a murder mystery and deduce the most likely perpetrator out of three suspects. Individually, only 44 percent of the students got it right, which is slightly better than chance. The students were then sorted into groups of three, all from the same fraternity or sorority, and were given 20 minutes to come to agreement on the most likely suspect. After a few minutes, a fourth member was added to the group — sometimes from the same fraternity or sorority, sometimes from a different one. If the new member came from a different fraternity or sorority, the group performed objectively better then the totally homogeneous groups (75 percent vs. 54 percent correct), and members with incorrect guesses were much more likely to change their minds. Nevertheless, the homogeneous groups perceived themselves as having more confidence, consensus, and effective interaction.
The original research comes from Phillips, K. et al., “Is the Pain Worth the Gain? The Advantages and Liabilities of Agreeing With Socially Distinct Newcomers,” Personality and Social Psychology Bulletin (forthcoming).