Are there more words that start with the letters R or K than words with R or K as their third letters?
Most of us would guess there are more words that start with R or K, and if we made lists from memory those lists would seem to prove it. In fact there are about twice as many words with R or K as their third letters, but we don't mentally sort words by their third letters so it's harder to think of those examples.
It sounds like a silly example of language geekitude, but it's an example of a common way we misunderstand our experience ... including our experience of groups, discussions, and extremes.
More below the fold....
Groups, Discussions, and Extremes, Part II (Non-Cynical Saturday)
This week Morning Feature has looked at group polarization. I primed readers' thinking about that on Thursday by ranting about a very polarized group: the Tea Party GOP. Yesterday we defined group polarization and offered three contributing factors: (1) we tend to follow those who argue with confidence, and those arguments are less likely to recognize nuance and uncertainty; (2) most of us value group inclusion over our own views, more than we care to admit, so as a group position emerges we tend to bandwagon and adopt that position as our own; and, (3) groups tend to coalesce around positions with the greatest contrast to the positions they attribute to an opposing group. Put it all together, and group discussions tend to coalesce around extreme rather than moderate positions.
We can all think of plenty of examples of groups settling on extreme and (in our opinion) wrong positions. As noted above, I primed that response Thursday with the rant about the Tea Party GOP. So I wasn't surprised when many of yesterday's comments featured examples of groups reaching extreme and bad decisions. The implication, noted by Cass Sunstein in his article on group polarization (pdf file), is that group discussion might not be the best way to reach decisions.
Availability bias.
As we saw in the introduction, "we can think of plenty of examples" does not always mean we're thinking of the most common events. If asked that question about words beginning with the letters R or K vs. words with R or K as their third letters, and if we didn't already know the answer, most people will say there are more words that start with the letters R or K ... because examples of those spring more easily to mind. In psych-speak, those examples are more available because we sort and access words by their first letters more easily.
In estimating how often something happens, we tend to overvalue examples that spring easily to mind. That's called availability bias. Ironically, I found that concept by searching for other research by Cass Sunstein. He and Timur Kuran combined the concept of availability bias with the concept of information cascades - bandwagoning - into the concept of availability cascades. These are situations where groups or societies coalesce around a new idea that seems to explain a complex topic in a simple and intuitive way. The idea gains currency because of its simplicity, and as it does more people are more likely to adopt it because it seems to be emerging as the prevailing group or society position.
Another reason to distrust groups?
At first it seems ironic to offer availability bias in rebuttal to criticisms of group polarization. Haven't I just described yet another reason to distrust groups and group decisions? As you read the previous paragraph, you may have come up with your own examples of availability cascades ... most of them negative. Indeed much of the study of group dynamics in social psychology grew out of academic focus on a specific, horrifically dysfunctional group: Nazi Germany. Stanley Milgram's study on obedience to authority and Phillip Zimbardo's simulated prison study both sought to understand how ordinary people come to commit brutal, inhumane acts.
In fact, most studies in group dynamics focus on groups making mistakes, such as the example I gave yesterday where groups looked at a diagram with two lines and were led to agree on an obviously false statement of which line was longer. Those designing and conducting the studies do that in part to highlight the impact of the group dynamics they're examining. It's hard to learn much from watching a group of college students (the most common research sample) agree that two plus two does, in fact, equal four. But if you can get them to agree that two plus two equals chocolate ... there were probably some interesting group dynamics involved.
And it's not just what researchers study. We're more likely to focus on and remember things that go wrong than things that go right. That's not because we're "negative;" it's sound survival impulse. If you're in a burning building, it makes sense to forget about all of the other things that are going well in your life and focus on what's going badly ... the burning building. So we tend to discount the cases of groups that made extremely good decisions and focus on examples of groups that made extremely bad decisions.
"Extreme" can be good.
Indeed our first response when we hear that groups in discussion tend to coalesce around "extreme positions" is to imagine extremely bad positions. Much of the research and discussion of group polarization reflects a bias for moderation, tacitly or explicitly describing "extreme" positions as less reliable or useful.
But that is itself a bias. Most Beneficial, Least Costly, Most Possible, and Least Hopeless all describe "extremes." In fact, the idea that a group should choose an objectively moderate position - "This isn't as good as some of our alternatives, but it isn't as bad as some others, so let's take that" - is absurd on its face. Most of us want groups to choose the Best Possible Option, by definition an "extreme," and not the Most Moderate.
Groups' tendencies to coalesce around extreme positions often helps us in another, less obvious way. Our default solution to most problems is to Do Nothing: ignore them, endure them, or work around them. Often we can't Prevent or even Solve the problem, at least not as individuals, so Do Nothing is the only choice available. But get a group of us together to discuss the problem and we'll often work out a plan to Do Something. As compared to Doing Nothing ... Doing Something will be "extreme."
So yes, groups tend to polarize toward extreme positions, and individuals in groups tend to adopt those extreme positions as our own. But that's not always bad, and I suggest it's more often good. As I've written many times, our ability to communicate and cooperate - to make decisions and work together in groups - is how our social species moved from Tree Ape to Plains Ape without becoming Pootie Poop. We may idealize individuals, but we're a lot more effective in groups.
+++++
Happy Saturday!