The marshmallow test, redone in 2018: it was never really about a child's willpower

A child who grabs everything immediately, can't wait their turn, and falls apart the moment they can't have something right now — sooner or later, someone in the family brings up the famous marshmallow test: the idea that four-year-olds who could wait for a second marshmallow instead of eating the first one right away grew up more successful, while the ones who couldn't didn't. The conclusion seems obvious: the child is short on willpower, and that willpower needs to be built somehow. The problem is that half a century after the original study, someone redid it on a much larger, far more varied sample, and the headline conclusion didn't hold up.

The original test: a small sample that started out alike

The classic marshmallow test was run by psychologist Walter Mischel in the late 1960s and early 1970s at a preschool at Stanford University. A child was given one marshmallow and told they'd get a second if they could hold out 15-20 minutes without eating the first one alone. The study genuinely found that children who waited longer went on, years later, to score better on the SAT on average and have fewer behavioral problems. For decades, that became the textbook argument for willpower as a key predictor of success. But the original had a weak point that rarely comes up: fewer than a hundred children, almost all from the families of Stanford staff and students — meaning a similarly high level of income and parental education across the board.

The marshmallow test that changed everything

In 2018, Psychological Science published a paper by Tyler Watts, Greg Duncan, and Haonan Quan, "Revisiting the Marshmallow Test" — essentially the same test, but run again on an incomparably larger and far more diverse group: over 900 children from the long-running NICHD SECCYD study, where families varied by income, ethnicity, and maternal education. The waiting threshold was shorter this time, 7 minutes instead of 20, but the logic of the test stayed the same.

The result turned out to be far more modest than the popular version of the story suggests. The link between waiting for the marshmallow and later academic and behavioral outcomes came out roughly half as strong as in the original, and once the authors statistically accounted for family income, home environment, and maternal education, that link nearly vanished for most children. For children whose mothers had a college degree, there was no link at all between the test result and later outcomes, and the ability to wait didn't predict later behavioral or personality measures either. The authors' conclusion is direct: a child's isolated willpower isn't the main lever worth pulling first. What shapes later outcomes far more is the environment and everyday conditions the child grows up in.

To be fair, the science isn't fully settled here: a separate 2020 paper by Falk and Kosse partly pushes back on some of Watts et al.'s conclusions. But even accounting for that debate, the core finding held up: treating the ability to wait as an isolated character trait to be built outside the context of the family is an oversimplification that the replication doesn't support.

What this means in practice

If a child can't wait, grabs things, and falls apart when told no, the issue usually isn't a shortage of some innate willpower that can just be trained through lectures or punishing impatience. Self-control in children develops as a skill, and it develops mainly through the environment a child lives in every day: how predictable the routine is, how consistent adults' reactions are, and whether the child has actually experienced that waiting leads somewhere. That's exactly what parent coaching in applied behavior analysis (ABA) is built on: instead of a one-off office visit, a specialist teaches the parent themselves to build that environment and those reactions at home, where the child actually spends their time. If you want to see how this works in practice, we have a separate piece on it — "Parent coaching in ABA therapy".

What you can try today

• Make waiting predictable instead of abstract. Instead of "wait a bit," give something concrete: "once I finish washing this plate, I'll give you the cracker," with a visible end point to the wait.
• Start with very short intervals. A few seconds of successful waiting followed by praise works better than asking for five minutes of patience right away.
• Reinforce the act of waiting itself, not just what the child gets afterward. A quick "you waited, nice job" the moment patience pays off cements the skill itself, not just the reward that followed it.

If it's not clicking on your own at home

Sometimes it's genuinely hard for a parent to know where to start building that kind of environment, especially if trouble with waiting and impulsivity is pronounced and getting in the way at daycare, at home, or during activities. In cases like that, a board-certified behavior analyst (BCBA) can look at your specific situation and put together a step-by-step plan that fits into your everyday life, without needing a dedicated office or a scheduled appointment slot. You can start with a free consultation.