Sunstein Cass R - Wiser : getting beyond groupthink to make groups smarter

Did you recently finish a project a lot later than you expected? You are subject to the planning fallacy, which means that you expect to complete tasks far more quickly than you do.

Are you confident that your friend will like that movie you recently loved? You might well be vulnerable to egocentric bias, thinking that if you like something, other people are going to like it too.

People tend to ignore the long term; to be unduly afraid of losses; to display unrealistic optimism; to make self-serving judgments (whats fair is whats best for me!); and to deal poorly with risks. For these reasons, and many others, each of usas individualscan get into a lot of trouble.

The last few decades have seen pathbreaking research on how people think, act, and sometimes blunder. Behavioral scientists have shown that human beings make all sorts of mistakes. By itself, this observation is not exactly news. Whats important is in the detailsthe specification of what mistakes we make, and why.

Social scientists, most notably Daniel Kahneman in his masterful book on the topic, have distinguished between two ways of thinking: fast and slow. Fast thinking, rooted in what behavioral scientists often call System 1, is rapid, automatic, emotional, and intuitive. Slow thinking, rooted in what is often called System 2, is slow, calculating, and deliberative. If we want to know what to do, especially in complex situations, System 1 is not reliable (even though it is good at escaping from trouble in a hurry). If System 2 is working well, slow thinking is an important safeguard, because it runs the numbers and usually comes up with the right solution. Unfortunately, System 1 is often in charge, and it is responsible for many of the errors that individuals make.

But what about groups? In this book, we begin with a simple question: Do groups usually correct individual mistakes? Our simple answer is that they do not. Far too often, groups actually amplify those mistakes. With respect to the planning fallacy, for example, groups turn out to be even worse than individuals arewhich is a clue to a lot of failures in business, government, and daily life. Within groups, System 1 has a great deal of power.

We also ask a second question: Can groups correct individual mistakes? Our simple answer is that they can. We aim to explain how.

Sometimes groups get wiser with the help of some easy, informal methods or tactics. A simple tactic: group leaders often do best if they shut up and let other people speak. A slightly less simple method: managers are frequently more effective if they assign specific roles to group members, thus making it more likely that groups will get the information they need. Wise groups sometimes adopt red teams that are designed to identify vulnerabilities in existing practices or coming decisions. You can also increase wisdom within a group or a firm by cultivating certain social norms that redefine what it means to be a team playernot to go along with the group, not to engage in happy talk, not to show unqualified enthusiasm for the bosss demonstrable brilliance, but to add new information .

Sometimes groups do best when they adopt more formal approaches. Groups can enlist modern technologies that provide astounding opportunities both to capture and to improve collective wisdomthrough tournaments, prediction markets, and public comment processes. By the way, we didnt come up with the title of this book; it was produced by a collection of complete strangers, working on a website that we didnt know about until we decided to look for help with our title. Well tell that tale in due course.

We also apply a central distinction from biology and modern computer science. The distinction can help people to improve group processes of all kinds by dividing decisions or problem-solving tasks into a creative, divergent-thinking stage and a critical, solution-integration stage. We will offer many specific suggestions to managers of all stripes, who are frequently faced with group processes that just arent working, or arent working well enough, and who want to know how things can be set right.