I Think A Lot About The Marshmallow Test. Why?

A thing about me is that I often think about the marshmallow test. The marshmallow test is a test of delayed gratification abilities in which little kids are given a small treat, like a marshmallow, and told that if they wait a bit, they can have extra. One marshmallow now, or two later.

The test is famous because researchers said there were correlations between choosing “two marshmallows later” and, years after, getting the good things of modern capitalism like career success, academic achievement, and better SAT scores. Endless variations on the main study have been undertaken since the 1930s; the standard conclusion is that having the self-control to delay gratification is a useful and virtuous.

I’ve always been a skeptic about the marshmallow test. I was a shy, somewhat nervous kid, eager to please the adults around me. If I’d been given the test, I expect I would have been more motivated by social pressure than actual marshmallows. As I wrote on this blog in 2009, the original test set up was that kids who wanted the first marshmallow right away would have to ring a bell to summon a researcher. Are you kidding?! There is no way at four years old I would have rung a bell to summon a strange adult, even if 50 marshmallows had been on the line.

Also, is it even obvious that more later is better than less now? The thing about now is that it’s now: if you have your treat immediately, you are virtually guaranteed satisfaction: you’re having it at the moment that it looks delicious, there’s no risk of some diabolical behavior or random obstacles blocking your treat, and you can move on with your life rather than sitting there in the painful condition of "waiting for a treat."

Color me unsurprised, therefore, that as the years went on, studies showed the marshmallow test was more complicated than it may have appeared. As with the young Patricia, researchers found that children engage in “reputation management,” and were more likely to delay if a teacher knew their choices.

Furthermore, kids from wealthier socio-economic backgrounds were found to do better on the test. It was pointed out that one reason for that could be that if you’re from a richer family, the likelihood of “more later” was more likely to actually materialize. If the adults around you have more resources, they are more able to provide in a stable and predictable way, and to prevent unexpected diabolical behavior or random obstacles from interrupting your treat.

The socio-economic explanation obviously leads to a hypothesis almost diametrically opposed to the original one: that it’s not your inner character that matters, it’s your environment. The potential significance is huge. A few years ago the New Yorker had an article about the test, including how whole schools were being designed to promote learning self-control and delaying gratification. If it’s more the external environment than that intervention is a major investment in a completely wrong direction.

Since this is a blog post and not a philosophy article I will leave aside correlation versus causation and how-do-we-really-know-anything, and get straight to the personal: Patricia, why do you think so often about the marshmallow test?

I think one reason is that I am often amazed by the contrast class between things I can do and things I cannot get myself to do, which makes me wonder: if self-control makes me able to do the first things, why can’t I do the second things?

I do a fair number of things that appear to require self-control: I work by myself on large unstructured research projects with no deadlines; I go to the gym even when it’s freezing outside and cozy at home; in the course of my life, I have quit smoking, Diet Coke, and a range of other things we won’t get into here.

On the other hand, the things I cannot get myself to do is mystifying. I have never been able to prepare lunch at home to eat later during the day — now I can afford to buy lunch, but even when I had no money, I would eat like one donut or just skip lunch. I am trying to learn Italian, and I found this great site with Easy Italian News for practice: listening to it is reasonably fun, but am I doing it? No. I’ve been trying to form a new habit of bringing my own silverware from my office to the lunch place on campus, so I can avoid using all that plastic. Easy, but my success rate? Just reaching toward 20 percent.

I’m forced into the conclusion that for me, it's less like there is a self-control part that I direct at one activity or another and more like some habits take and some don't. It’s a confusing mix why. Partly, some things are engaging despite being difficult and some things are just boring and annoying. Partly, a habit is different from self-control. Partly, behavior is social not individual, which is why nagging the people you love is actually an important thing to do.

So I guess that’s why I think about the marshmallow test so often. It seems to test for a quality I feel I don’t really have. I don’t know if y’all have a similar experience, but that’s what’s going on with me.