Marshmallow Test Redux

You may have heard about the Stanford Marshmallow Test. In the late 1960s Walter Mischel, a researcher at Stanford, was interested in how self-control worked. He took 4-to-6 year old children from the university’s lab preschool and presented them with a test. Alone in a room, he said, “Here’s a treat. [The kids had a choice of several, but people always remember the marshmallow.] You can eat it any time you want. But if you wait until I get back, you can have two treats.” And then he left for 15 minutes, and observed which kids ate the treat and which waited for his return. You can watch painful/amusing recreations of the test on YouTube—and a number of people have (ill-advisedly) tried this test on their own children and put the videos online. So what’s the big whoop? Some kids wait, some don’t, who cares?

Actually, it turns out it was a really big whoop. When they went back and followed up with these kids years and even decades later, they found out that the marshmallow task was an incredible predictor of success later in life, in ways Mischel and company never expected. Kids who waited for the marshmallow at age 4 had higher grades at age 12, higher SAT scores at age 18, higher college graduation rates at age 22. And not just school stuff—the marshmallow waiters, as adults, had significantly higher salaries, were less likely to have been addicted to drugs, and were rated much higher on measures of things like “social competence” and “planning ahead” and “coping with stress.” Why?

Waiting for the second marshmallow is a skill called “delay of gratification.” It’s one of a group of skills called “executive function” or “self-regulation” or, more loosely, “self-control.” Other skills in this group include planning ahead, flexibility, and attention. The idea is that the kids who wait for the second marshmallow are, generally speaking, stronger on all the executive function skills, and that executive function is crucial for success in school and in life—more important, this research would argue, than intelligence or subject knowledge or any of the other things our culture has traditionally focused on in school.

It’s important to remember that while children may be born predisposed to stronger or weaker executive function, these skills are both learnable and teachable. Which is where new research out of the University of Rochester comes in. The researchers there said, “What if children’s performance on this test depends on how dependable their environment is? If they don’t think the researcher is actually coming back with marshmallow #2, it makes complete sense to eat the first one.” They set up an experiment where grown-ups were shown to be more or less dependable, by promising to bring markers and then following through or not, and THEN doing the marshmallow test. Turns out? When kids were previously shown that grown-ups were as good as their word, they were way more likely to succeed at the marshmallow test; when grown-ups didn’t come through the first time, kids were way more likely to just eat the marshmallow.

Get it? Kids’ ability to employ and develop executive function skills is dependent upon the environment where they grow up. If your environment doesn’t set you up to believe in long-term outcomes, it wouldn’t make sense to build your planning and delay-of-gratification skills.

This is a huge contributor to the academic achievement gap between middle-class and poor kids. Middle class kids are taught study skills and stuff by their parents, sure, but perhaps more importantly their whole life rewards them for planning ahead and delaying gratification. For children to learn to self-regulate, they need to be taught, and they need to be in an environment where self-regulation is a successful tactic. There are a lot of reasons children—especially poor children—might learn that their environment is undependable. Maybe there’s not always enough food. Maybe you have a different caregiver every day because your parent work all the time. Maybe your family has to move a lot.

Even if all of a child’s needs for physical safety and environmental consistency are met, there are a lot of things families can do to build executive function—mostly having to do with letting children exercise their self-control, like a muscle. There are a lot of both serious and playful activities that give opportunities for this. And socio-dramatic play (i.e. pretend play with others) is a veritable aerobic workout for self-regulation. But all that’s another story.

For further reading, check out this article by Paul Tough and this one by Jonah Lehrer.

If you’re interested in how precisely grown-ups can help children build self-regulation skills, I give workshops on the topic for families and teachers.

2 thoughts on “Marshmallow Test Redux

  1. […] But remember: smart kids will stay smart. You don’t need to worry about that. Contrastingly, self-regulation skills need to be learned and taught, and a preschool classroom with sensitive, attentive teachers is one of the best places in the world to practice those skills. They’re the hardest skills to learn later on, so better to spend more time on it now, while you can. (See my blog post about executive function.) […]

  2. […] on This American Life and Planet Money and Morning Edition. Here’s a related piece I wrote on the famous Marshmallow Executive Function Test. And two of my favorite articles ever on EF—one by Paul Tough in the NYTimes, and one by Jonah […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: