A study in the 1970s shows that people in groups tend to make riskier decisions and take more extreme views than they would alone. Here's how to defuse that group dynamic when it's not on your side.
In 1970, psychologists David Myers and George Bishop rounded up several groups of students. They wanted to see what happened to a group of individuals when they discussed the topics of the day, but the groups they chose were not created to foster balanced debate. They were divided by their attitudes on race. The students were given a test called "The Multifactor Racial Attitude Inventory (MRAI)", which was created in 1967 and includes such questions as, "I would not take a Negro to eat with me in a restaurant where I was well-known." Based on their scores, the psychologists separated the students into groups of people with highly racist views, ones without racist views, and ones who took the middle ground.
The groups discussed questions such as school busing, desegregation, and housing covenants. (Housing covenants are provisions on a deed to a house which state that the house cannot be sold or rented to a certain group of people. If the owner does sell or rent to that group, they forfeit the property.) After the discussions the students took the MRAI again. As the researchers predicted, the students evinced more extreme attitudes than they had before. Those who were progressive were more so after they discussed the issue with other progressive people. Those who were racist were more so after their discussions. The "risky" opinions that they had held going into the discussion were validated and made to look like the norm, and so they felt comfortable taking even more risks.
What's interesting about the study is how the authors end it. Free discussion of issues, they conclude, might do more harm than good. People aren't likely to have breakthroughs or epiphanies. They're not likely to be repulsed by a look at the extremes to which people with their general opinions can go. They're more likely to intensify their existing views to match those extremes. The psychologists recommend that to avoid this, groups limit discussion until after educational programs that discourage certain views are shown. At the very least, arbitrators should start discussions "with remarks which make it likely that the desired comments will predominate."
Like all techniques, it can be used for good or for evil. No one can argue that restrictive housing covenants or segregation are morally acceptable. Still, this shows how to stifle free conversation and the airing of views without appearing to do so. Even brief introductory remarks let people know that, if they express a different opinion they are taking a risk. That knowledge will discourage them from taking further risks.