Groupthink: many lessons for think tanks

1 March 2012

Not often does one find a an article that covers so many issues of great relevance and interest. Jonah Lehrer’s article in The New Yorker, Groupthink: The brainstorming myth, is one. The main argument he makes is that brainstorming is not all it is made to be. You are more likely to get more and better ideas out of working on your own. But the detail in his argument is rather interesting.

First, it is not so much the concept of getting a bunch of people together to brainstorm what is wrong, but that the ‘rules’ we tend to adopt when this happens are responsible for curtailing innovation and creativity. One of those rules is that there are no bad ideas in a brainstorming session: quantity rather than quality. Lehrer reports on research that shows that when you introduce debate rule into a brainstorming session -you are allowed to shoot down bad ideas- the quantity and quality of the ideas increases. Not only that but people are more likely to keep thinking about ideas well after the session is over.

Debating rather than just accepting everything that comes our way, it seems, makes us think -who would have known.

Debating also works best because it is likely to lead to dissent. Being exposed to alternative views (alternative to the obvious and accepted truths), even if these views are wrong, expands our creative potential. Most think tanks public events tend to feature a panel of people who all agree with each other. Few are true debates. But unless think tanks are willing to listed to opposing views they are unlikely to reach their creative potential. Dissent may not be enjoyable for all, but it is certainly more productive.

The third idea that Lehrer puts forward has to do with the composition of the group. Groupthink tends to fail because the group is not often the most appropriate one. He quotes the work Brian Uzzi who looked at Broadway productions to see whether the density of the connections between their staff had any effect on the outcomes of the plays. In other words, he wanted to know if a play produced by people who knew each other would be better than one by people who had never met before. This is an interesting question for think tanks. The think tank community all over the world is rather small. In some cases it is impossible to go from one centre to another without running into someones husband wife, brother, mother, cousin, etc. In countries with few universities it is not uncommon to find that all economists studied together. In the U.K. the think tank community is also quite small -imagine how tiny is the international development one?

Well, Uzzi found that when connections are low (very low) and when connections are high the quality of the plays are low. ‘The best plays were produced by teams with an intermediate level of social intimacy’. So when putting together teams (within think tanks or across them) we should consider the relationships between their members. The right mix of familiarity and surprise will yield the best results.

Which brings us to the best idea of all. Much is said about face to face meetings and work in the development industry. Lately, a great deal of attention has been given to online tools such as webinars and, pressured to guarantee value for money, donors and think tanks are replacing face to face interaction with Skype calls. Lehrer refers to Isaac Kohane’s work on the subject. Kohane found that the ‘best research is consistently produced when scientists were working together within ten meters of each other.’ Even in today’s world when it is possible to get in touch with anyone in the world with a couple of clicks of the mouse working together is better than not.

This flies in the face of efforts to do research together-apart that are supported by donors everywhere (regional and global networks of organisations which have entirely different ways of working, pressures, cultures, etc.). The effort required to produce high quality research in these circumstances, with researchers based in different countries and regions, is much higher than that required when they work together. This is common sense. Like most things we do not do… it is common sense. I wonder if any provisions are made for this, though? Based on the number of times I have been told that face to face meetings are too expensive for a research programme consortium I would say that no, provisions are not made.

But the idea of proximity gets even more interesting. Last year I helped Grupo FARO, a think tank in Ecuador, to choose a design for its new headquarters in Quito. Design (or lack of) matters. One of the few things I miss about leaving ODI is that I don’t get to talk to as many people as I did during the day -although I am able to avoid admin discussions, which is a huge plus. The best place at ODI to talk to people was the kitchen -and the kitchen (or the water cooler or the cafeteria) is probably the best place to talk to people in every organisation. Lehrer talk about the history of Building 20 at MIT. Building 20 had been an emergency building, meant to be torn down after the Second World War. But demand for space meant that the building became home to an eclectic community of researchers, housing centres as diverse as a laboratory for nuclear science, the linguistics department, a cell-culture lab, a piano repair facility, etc.

Because Building 20 was a temporary construction, staff felt free to adapt it to fit their own interests. They would tear down walls, open new spaces, decorate following their own styles, etc. This flexibility meant that the building adapted to the researchers’ own learning. If they needed more space for a lab the walls came down. If they needed to bring in extra help, new desks were added, etc. But most interesting, the diversity in the background and focus of the researchers meant that the right mix of familiarity and surprise was reached. It reminds me of the think tank hub idea. Noam Chomsky’s work on linguistics drew from biology, psychology, and computer science -all from the researchers working within Building 20.

This is a fantastic story. It confirms that while great facilitators can do great things, nothing matches having the right people int the right space. ‘Group dynamic can take care of itself.’