DAVID EPSTEIN: One of the researchers I spoke with was himself an innovator and was picked by R&D Magazine as the Innovator of the Year. But he also studied innovation. And what he found in studying patent databases is maybe a change in the importance of specialists and generalists over time. And he defined this by looking at people's work histories.

So, creators who are very specialized did all their work in one or a small number of areas of technology as classified by the US Patent and Trademark Office. Creators who were more broad spread their work across a large number of technology classes, sometimes merging many into one project. And since about the 1990s, the explosion of the knowledge economy, what he found was that these more generalist inventors, or sometimes they were even polymaths with some depth and breadth, were making larger and larger and more and more important contributions whereas the specialists, who were still very important, were often making less impactful contributions.

And he thinks that this is part of the rise of rapid communication technologies, that the information created in many cases by specialists is so rapidly and thoroughly disseminated that there are many more opportunities than ever before to invent something new by taking things that aren't new and combining them in new ways. And one of my favorite examples of this is a Japanese man by the name of Gunpei Yokoi, who didn't score well in his electronics exams so he had to settle for a job in Kyoto as a machine maintenance worker at a playing card factory while a lot of his peers went off to big companies in Tokyo.

He realized that he wasn't equipped to work at the cutting edge, but that so much information was easily available that specialists were overlooking that he could just combine older and well understood technologies in ways that specialists couldn't see because they didn't have a broad enough view. In doing that, he started a toy and game operation at that playing card company -- that playing card company is called Nintendo -- and he continued combining old technologies for his magnum opus, the Game Boy. All of the technology was long out of date by the time it appeared and yet it became the best-selling video game console of the 20th century.

Yokoi called his creative philosophy 'lateral thinking with withered technology'. What he meant by lateral thinking was taking information from one area that may not be new, but just bringing it somewhere else where suddenly it's new to that area, combining technologies in ways that other people hadn't. By withered technology, he meant this older, well understood, often cheaper technology so he didn't have to worry about competing at the cutting edge.

And that's a nice story, but I think it also fits with multiple studies of patent research that show in many cases the biggest impacts come not from the people who drilled the deepest into a technological class, but those who spread their work across a large number of technological classes. And incidentally, there are analogous findings in other industries. In a really interesting study of comic book creators, researchers guessed at what would make comic book creators make commercially valuable comics, and also what would make them more likely to make a blockbuster comic. And it was a great study because they could track the value of comic books both up and down, it didn't suffer from the survivor bias that a lot of studies of excellence do. And they posited pretty intuitively that the resources of a publisher would make a creator better, or their years of experience would, or the number of comics they'd made previously. And they were wrong, wrong, and wrong. The most important factor was the number of different genres that a creator had worked in. The genres ranged from comedy, and crime, to fantasy, adult, horror, non-fiction.

And it was true that you could make a team and combine teams of genre specialists to get some of that diversity. But that was actually pretty limited. So if you had an individual who'd worked in two genres, you were better off having a team of three who had worked in one genre each. But after four genres, then an individual who had worked in more than four genres did better than a team who had the same genre experienced by platoon. So you could not re-create the diverse experience of an individual entirely with a team of specialists.

So these researchers named their paper 'Superman or the Fantastic Four'. They said, if you can find a Superman who has worked in a very diverse array of genres, do it. And if not, then create a fantastic team with diverse genre experience by platoon.