The Key To Innovation: Making Smart Analogies

The following is the transcript of the closing keynote speach at Sandbox Summit, a fantastic annual conference that brings together all kinds of people interested in kids, technology and learning.

I want to talk today about how we build bridges between old ideas and new ideas, between concepts that are close to home intellectually and those that are far afield. Because it is these bridges, these connections, that produce innovation.

There’s a popular notion that innovation arrives like a bolt out of the blue, as a radical departure from previous knowledge—when really, most new ideas are extensions, twists, variations on what’s come before. The skill of generating innovations is largely the skill of putting old things together in a new way, or looking at a familiar idea from a novel perspective, or using what we know already to understand something new.

I say “skill” because we don’t have to leave these encounters up to chance. We don’t have wait for lightning to strike. We can, right now, hone our ability to deploy analogies. Analogies—comparing one entity to another, apparently different entity—is one of the most powerful tools humans have for understanding our world and for generating new knowledge.

In their book Mental Leaps: Analogy in Creative Thought, cognitive scientists Keith Holyoak and Paul Thagard point out how many intellectual advances through the ages have been built upon analogies:

The first-century Roman architect Vitruvius compared the sound of actors’ voices in an amphitheater to the movement of water in a pool, the first of many thinkers to compare sound waves to water waves.

The seventeenth-century scientist William Gilbert compared the earth to a magnet, advancing knowledge of the earth’s gravitational force.

The eighteenth-century chemist Antoine Lavoisier compared respiration to combustion, clarifying how breathing turns oxygen into carbon dioxide.

Even the great nineteenth-century biologist, Charles Darwin, built his theory of evolution on an analogy between artificial selection—the deliberate mating of animals by breeders—to the natural selection that goes on in the wild.

Analogies are still frequently used by scientists working today, as University of Maryland professor Kevin Dunbar discovered when he observed firsthand scientists working in four microbiology labs. Dunbar found that the scientists used as many as fifteen analogies in a one-hour laboratory meeting, and that the more successful labs employed more analogies in discussing their work.

Let’s look more closely at what is happening, conceptually, when we make an analogy. “The essential requirement for analogical thinking,” Holyoak and Thagard write, “is the ability to look at specific situations and pull out abstract patterns that may also be found in superficially different situations.” That’s important, so I’ll say it again in a slightly different way: A useful analogy reveals the deep commonalities beneath superficial differences.

We can think of analogies as having two parts: the base and the target. The base is the thing you know about. The target is the thing that’s new. Analogies are created by elaborating the similarities and the differences between the base and the target. When we use an analogy, we take what we know about the base and move some of it over to the target. Northwestern University psychologist Dedre Genter calls this process “bootstrapping the mind”—elevating ourselves into the realm of new knowledge, using the knowledge we have already to pull ourselves up.

What does this allow us to do? The scientists Kevin Dunbar studied used analogies, first, to formulate hypotheses that they could then test. Their thought process went something like this: If we know that X does Y when Z, is it possible that A does Y when Z, too? Let’s find out. That’s often how innovations get their start, in the lab and elsewhere: by taking a familiar starting point and using it as a launch pad to explore new territory.

Psychologist and author Steven Pinker, who has composed many an illuminating analogy himself, notes that “carefully interpreted, analogies are not just alluring frames but actual theories, which make testable predictions and can prompt new discoveries.”

Dunbar’s scientists found analogies useful on another kind of occasion as well, employing them as sense-making tools when they encountered anomalies or unexpected findings in the course of their work. This was also the case for another group of scientists whose use of analogy became the subject of study: scientists working on the Mars Rover mission.

A team led by University of Pittsburgh professor Christian Schunn analyzed transcripts of the discussions the Mars Rover scientists carried on while the mission was underway. Schunn and his colleagues found that the Mars Rover scientists used analogies more frequently when they felt confused or unsure. The appearance in the transcript of words indicating uncertainty, such as “maybe,” “I don’t know,” and “I don’t understand,” was often followed by an attempt to draw an analogy—to compare the ambiguous situation to a situation with which the scientists were familiar.

At such moments, the scientists were employing analogies as different sort of bridge—a conceptual catwalk that provides just enough space to move forward and keep searching for solutions. As Schunn writes: “Scientists and engineers do not always seek to completely eliminate uncertainty (and indeed, sometimes it is not possible to do so) but often drive problem solving with the aim of converting it into approximate ranges sufficient to continue problem solving.”

We’ve now seen two good reasons to employ analogies: They can provide us with new theories to test, and they can help us make sense of unanticipated results so that we can keep moving forward. So: how do we go about generating a useful analogy?

The first step is to choose the base and the target carefully. They should be similar enough to offer the possibility of building a serviceable bridge. If the target is too wildly different from the base, there won’t be any productive similarities to draw. In addition, these similarities must be deep and not merely superficial. Giovanni Gavetti and Jan Rivkin, both Harvard Business School professors, have found that management teams can easily be seduced by an analogy with similarities on the surface but differences at a deeper level (the opposite of what innovators should be looking for).

At the same time, the base and a target have to be different enough to offer a revealing insight. In a study led by researcher Diane Halpern of the University of Louisville, students who compared the movement of lymphatic fluid through the lymph system to the movement of blood through veins gained little from the comparison. The target and base were too close, and the analogy didn’t shed much light. Comparing the lymph system to the movement of water through spaces in a sponge, however, was a bigger mental leap, and it produced more learning in the students Halpern studied.

To aid in finding just the right analogy, it helps to have a deep pool of potential targets. The Boston Strategy Group, a consulting firm, has created an online gallery of sources of analogical inspiration for its consultants and their clients to use. We can do this, too—bookmarking or pinning websites that inspire connections, keeping a folder of ripped-out articles or pictures from newspapers and magazines. A class or a workplace team can create a shared repository of analogical targets.

The second step in using an analogy is to thoroughly draw out the similarities between base and target. Just posing the comparison isn’t enough; we need to think carefully about how and why the base and target are alike.

Third, we must define what doesn’t carry over from base to target—the places where the analogy “breaks down.” This is important, because it’s easy for us to allow prominent similarities to overshadow real differences. For example, education professor Rand Spiro of Michigan State University found that medical students—and even doctors—were sometimes misled by the common comparison of blood flowing through blood vessels to water flowing through pipes.

The students and physicians erred by attending to the similarities between blood vessels and pipes while failing to factor in the differences: the flow in blood vessels is affected not only by the diameter of the blood vessel, but also by the flexible contractions of the vessel and by the heart’s beating. It’s not that the medical students thought that blood vessels actually were rigid, like pipes. It’s that, focused on the similarities in their analogy, they weren’t even thinking about the differences.