Rethinking the marshmallow test helps me better frame my experience with entrepreneurship and software hacking, both of which taught me that there are times to jump in and just do it, times to persevere, and times to be patient.
In particular what I’d add to your marshmallow critique is that the real quality of value is the ability to delay when needed. But the test inadvertently evaluates another quality: how much you obey and follow the values and structures around you. Even harder than being able to delay gratification is knowing to tell when that willpower is needed.
I didn’t do very well at the marshmallow tests in school because I was stubbornly self-directed. I read the dictionary for fun, finished a whole set of encyclopedias, and taught myself to program in elementary school without any help — in fact despite resistance from teachers and parents who didn’t respect my priorities. But my grades were often terrible, because I didn’t care about them. Often I’d do well on tests, not because I studied for them, but because I happened to have read about the topic independently, and sometimes years before it came up in class. If an assignment interested me I’d do well; otherwise I might not do it at all. I produced some work that got teachers really excited, right alongside a lot of work that never even got started.
Maturing for me has been about learning to see when fitting into the system can also be valuable, rather than being a maverick all of the time.
That’s the other side of the marshmallow.