A child sits alone with a marshmallow. Eat it now, and the reward is gone. Wait, and a second treat arrives. For decades, that simple scene became one of psychology’s most famous lessons about self-control, success, and what a 4-year-old’s patience might reveal about the future.
However, a larger 2018 study found a more complicated story. When researchers looked at a bigger and more economically diverse group of children, the marshmallow test still said something about later achievement, but far less than many people assumed.
Once researchers considered family background, early cognitive skills, and home environment, much of the famous link faded.
A famous test gets a second look
The original marshmallow studies grew out of work by Walter Mischel and colleagues at Stanford University. In the best-known version, Mischel and colleagues offered preschool children one treat right away or two treats if they could wait.
Later follow-ups suggested that children who waited longer tended to do better as teenagers, including in school and in social measures. It was a neat story. Maybe too neat.
The 2018 research, led by Tyler W. Watts, Greg J. Duncan, and Haonan Quan, did not try to recreate every detail of the original test. Instead, the authors called their work a “conceptual replication,” meaning they tested the same broad idea using a different and larger data set.
10 times more children
The researchers used data from the National Institute of Child Health and Human Development (NICHD) Study of Early Child Care and Youth Development, a long-running U.S. study that followed children from early childhood into adolescence.
Their analysis included 918 children with valid delay-of-gratification measures at about 4.5 years old, along with academic and behavioral data at age 15.
Much of the analysis focused on 552 children whose mothers had not completed college. That subgroup was about 10 times larger than the sample used in the 1990 follow-up study, and it gave researchers a wider view of family circumstances than the original Stanford-based group.
Why does that matter? Because a child’s ability to wait for a sweet may depend on more than personality. It can also reflect whether the child has learned that promises are reliable, whether the home feels stable, and whether basic needs are usually met.
The link became much weaker
Among children whose mothers had not completed college, each extra minute of waiting at age 4 predicted about one tenth of a standard deviation in achievement at age 15. That may sound technical, but effectively, the effect was modest.
The authors also found that the original-looking relationship was only about half as large as earlier reports. After they added controls for family background, early cognitive ability, and home environment, the association dropped by about two thirds.
So, was patience meaningless? Not quite. The test still captured something, but it no longer looked like a magic window into a child’s destiny.
The first 20 seconds mattered most
One of the most striking findings involved the beginning of the waiting period. Most of the variation in later achievement came from whether a child could wait at least 20 seconds, not whether the child heroically made it through the entire waiting period.
That changes the picture. The popular version asks us to admire the child who sits for minutes, staring down temptation like a tiny monk. The data suggest that, after a brief threshold, longer waiting did not clearly separate children in the same dramatic way.
Anyone who has watched a child negotiate dessert knows how much can happen in 20 seconds. A glance at the door. A nervous smile. One tiny hand moving closer to the plate.
Behavior was the weakest part
The bigger surprise came from behavior. The famous story often implies that a patient preschooler becomes a calmer, better-adjusted teenager. The 2018 replication did not strongly support that part.
In the study, links between delay time and later behavioral outcomes at age 15 were much smaller and rarely statistically significant. The authors also reported virtually no relationship between delay of gratification and a composite measure of mother-reported internalizing and externalizing behavior in their key subgroup.
That does not mean self-control is unimportant. It means a single snack-based task should not carry the weight of explaining a child’s teenage years.
Circumstances shape the score
The deeper lesson is not that willpower is fake. It is that willpower grows inside a context.
The researchers added controls for things such as child background, home environment, and cognitive and behavioral skills already measured in early childhood. Their goal was to estimate how much of the simple marshmallow-to-success link might be tangled up with the child’s surroundings.
That is where the story becomes more human. A child who waits may be showing self-control, yes, but may also be showing the effects of a more predictable home, stronger early learning, or adults who have kept promises before.
What parents should take from it
For parents, the takeaway is not to panic if a child grabs the marshmallow. Nor is it to treat waiting as a tiny entrance exam for future success.
A better reading is that patience can be taught, supported, and practiced, but it is not separate from the world around the child. A calm routine, trustworthy adults, sleep, food security, and early learning may matter more than one dramatic moment at a table.
Watts put the point carefully in a summary from the Association for Psychological Science, saying that an intervention focused only on making a child delay, without improving broader cognitive and behavioral capacities, would “probably have very small effects on later outcomes.”
The marshmallow lesson changed
The marshmallow test survived because it offered a simple moral. Wait now, win later. It felt clean, memorable, and easy to repeat at kitchen tables and in classrooms.
The larger study does not erase that lesson, but it softens it. A 4-year-old’s patience may tell us something, but much of what it tells us is about the child’s environment, not just inner grit.
At the end of the day, the marshmallow was never just a marshmallow. It was also a small measure of trust, stability, and opportunity.
The study was published in Psychological Science.
