Test The Marshmallow

One of our favorite activities at the d.school is Tom Wujec’s famous Spaghetti Marshmallow Tower Challenge. As a Fellow at Autodesk, Wujec has conducted the challenge with thousands of people, and gathered some shocking data along the way. At the d.school, we’ve done the same, and our results are very similar to his, if not worse… (If you’ve never experienced the challenge with a team, you really should. Don’t watch Wujec’s video until after trying it out. Here are the instructions. I’ll try not to give away too many secrets of the activity here, so that you can still be surprised at the results.)

The thing that shocks most people is to discover that the group you’d think would be great — MBA’s, who’ve likely visited the Eiffel Tower, and have a fair grasp on concepts like gravity — would be so categorically outperformed by children. Having run the activity scores, if not hundreds of times, the simple way I’d describe the difference in mindset is, “Kindergartners know they don’t know the answer; MBA’s assume they do.” That simple mindset difference has profound implications for how teams approach the challenge.

What’s striking is that “the MBA mindset” has been empirically demonstrated to be challenged in this same domain by another finding. A fascinating study on the application of valuable risk-reducing methods by some of my colleagues at Stanford last fall concluded, “Specifically, MBAs resist the use of the method despite being in a strong position to leverage it…” Apparently, “Formal training in learning-by-thinking methods thus appears to limit the (application) of learning-by-doing methods.”

At the d.school, I sometimes joke that I’m a “recovering MBA.” This is mostly a tongue in cheek way to play along with the needling of my engineering colleagues, but there’s some truth to it: especially as it pertains to doing something new, deep familiarity with what’s worked in the past, and an ability to lay out a plan based on past experience, is increasingly irrelevant. In fact, it’s a liability. The challenge is, due to our incredibly ability to pattern match and rationalize our matches, we often don’t know when we’re doing something new. And so we approach a new problem as if it’s somewhat like the old problem, and trigger a bunch of wrong assumptions about how to approach the work.

All the while, building a tower that’s going to collapse under the weight of a marshmallow.

One approach I recommend is to cultivate an environment where “kindergarten-minded individuals” (novices, new folks) are comfortable asking “dumb questions.”

Click here to subscribe to Paint & Pipette, the weekly digest of these daily posts.