Wrong lessons from the Marshmellow test?

For decades, the marshmellow test occupied the popular imagination in demonstrating the effect of delayed gratification in our lives. The lesson was simple as it was powerful – more willpower = more success in every aspect of our lives.

But, perhaps we got it wrong?

Researchers from the University of Rochester ran a variation of the test with 28 children in 2012 that involved a first step in which half the kids were exposed to an adult who promised to bring them supplies for an activity and didn’t. When this half was exposed to the marshmellow test, they did far worse.

The research hypothesis was that a child’s ability to wait wasn’t just about the amount of willpower they were born with. It was also about how much they trusted the word of the adults who said they’d come back and give them more. Willpower is still important in enabling kids to be successful. But, it is likely more important for kids to grow up in an environment where they trust the adults around them.

While the test was conducted with a a small sample, the conclusion is thought provoking.

The meta learning, one that Nassim Taleb vocally advocates, is to carefully consider the results of the latest and greatest social science experiment. Far too often, factors outside the lab affect the behavior of the subjects in the lab.