Think back to chapter 1, to the types of intuitive experts that Gary Klein studied in kind learning environments, like chess masters and firefighters. Rather than beginning by generating options, they leap to a decision based on pattern recognition of surface features. They may then evaluate it, if they have time, but often stick with it. This time will probably be like the last time, so extensive narrow experience works. Generating new ideas or facing novel problems with high uncertainty is nothing like that. Evaluating an array of options before letting intuition reign is a trick for the wicked world.
In another experiment, Lovallo and his collaborator Ferdinand Dubin asked 150 business students to generate strategies to help the fictitious Mickey Company, which was struggling with its computer mouse business in Australia and China. After business students learned about the company’s challenges, they were told to write down all the strategies they could think of to try to improve Mickey’s position.
Lovallo and Dubin gave some students one or more analogies in their instructions. (For example: “The profile of Nike Inc. and McDonald’s Corp. may be helpful to supplement your recommendations but should not limit them.”) Other students got none. The students prompted with one analogy came up with more strategies than those given no analogies, and students given multiple analogies came up with more strategies than those reminded only of one. And the more distant the analogy, the better it was for idea generation. Students who were pointed to Nike and McDonald’s generated more strategic options than their peers who were reminded of computer companies Apple and Dell. Just being reminded to analogize widely made the business students more creative. Unfortunately, students also said that if they were to use analogy companies at all, they believed the best way to generate strategic options would be to focus on a single example in the same field. Like the venture capitalists, their intuition was to use too few analogies, and to rely on those that were the most superficially similar. “That’s usually exactly the wrong way to go about it regardless of what you’re using analogy for,” Lovallo told me.
The good news is that it is easy to ride analogies from the intuitive inside view to the outside view. In 2001, the Boston Consulting Group, one of the most successful in the world, created an intranet site to provide consultants with collections of material to facilitate wide-ranging analogical thinking. The interactive “exhibits” were sorted by discipline (anthropology, psychology, history, and others), concept (change, logistics, productivity, and so on), and strategic theme (competition, cooperation, unions and alliances, and more). A consultant generating strategies for a post-merger integration might have perused the exhibit on how William the Conqueror “merged” England with the Norman Kingdom in the eleventh century. An exhibit that described Sherlock Holmes’s observational strategies could have provided ideas for learning from details that experienced professionals take for granted. And a consultant working with a rapidly expanding start-up might have gleaned ideas from the writing of a Prussian military strategist who studied the fragile equilibrium between maintaining momentum after a victory and overshooting a goal by so much that it turns into a defeat. If that all sounds incredibly remote from pressing business concerns, that is exactly the point.
* * *
• • •
Dedre Gentner wanted to find out if everyone can be a bit more like Kepler, capable of wielding distant analogies to understand problems. So she helped create the “Ambiguous Sorting Task.”
It consists of twenty-five cards, each one describing a real-world phenomenon, like how internet routers or economic bubbles work. Each card falls into two main categories, one for its domain (economics, biology, and so on) and one for its deep structure. Participants are asked to sort the cards into like categories.
For a deep structure example, you might put economic bubbles and melting polar ice caps together as positive-feedback loops. (In economic bubbles, consumers buy stocks or property with the idea that the price will increase; that buying causes the price to increase, which leads to more buying. When ice caps melt, they reflect less sunlight back to space, which warms the planet, causing more ice to melt.) Or perhaps you would put the act of sweating and actions of the Federal Reserve together as negative-feedback loops. (Sweating cools the body so that more sweating is no longer required. The Fed lowers interest rates to spur the economy; if the economy grows too quickly, the Fed raises rates to slow down the activity it launched.) The way gas prices lead to an increase in grocery prices and the steps needed for a message to traverse neurons in your brain are both examples of causal chains, where one event leads to another, which leads to another, in linear order.
Alternatively, you might group Federal Reserve rate changes, economic bubbles, and gas price changes together because they are all in the same domain: economics. And you might put sweating and neurotransmission together under biology.
Gentner and colleagues gave the Ambiguous Sorting Task to Northwestern University students from an array of majors and found that all of the students figured out how to group phenomena by domains. But fewer could come up with groupings based on causal structure. There was a group of students, however, who were particularly good at finding common deep structures: students who had taken classes in a range of domains, like those in the Integrated Science Program.
Northwestern’s website for the program features an alum’s description: “Think of the Integrated Science Program as a biology minor, chemistry minor, physics minor, and math minor combined into a single major. The primary intent of this program is to expose students to all fields of the natural and mathematical sciences so that they can see commonalities among different fields of the natural sciences. . . . The ISP major allows you to see connections across different disciplines.”
A professor I asked about the Integrated Science Program told me that specific academic departments are generally not big fans. They want students to take more specialized classes in a single department. They are concerned about the students falling behind. They would rather rush them to specialization than equip them with ideas from what Gentner referred to as a “variety of base domains,” which foster analogical thinking and conceptual connections that can help students categorize the type of problem they are facing. That is precisely a skill that sets the most adept problem solvers apart.
In one of the most cited studies of expert problem solving ever conducted, an interdisciplinary team of scientists came to a pretty simple conclusion: successful problem solvers are more able to determine the deep structure of a problem before they proceed to match a strategy to it. Less successful problem solvers are more like most students in the Ambiguous Sorting Task: they mentally classify problems only by superficial, overtly stated features, like the domain context. For the best performers, they wrote, problem solving “begins with the typing of the problem.”
As education pioneer John Dewey put it in Logic, The Theory of Inquiry, “a problem well put is half-solved.”
* * *
• • •