The Marshmallow Test

The Marshmallow Test: Why Self-Control is the Engine of Success, a book written by Walter Mischel. A book that doesn’t match the hype but is a good book, nonetheless. 

The Marshmallow test is an extremely famous self-help book about a popular series of experiments done inside psychology labs. The experiments attempted to measure the tricks and strategies used by children to delay gratification, to say no to an immediate reward and to wait for a long-term, but also bigger, reward. Basically, give the children one marshmallow now or two marshmallows 15 minutes from now.

For most people, the ideas inside the book will be rather obvious, but some of the correlations found with different performance levels are less obvious and quite interesting. And arguably, the most exciting part of the book is when the author discusses those correlations. With that said, the author also explains some strategies and tricks which can help increase our own levels of self-control, which I found quite helpful. Let’s have a look at both.

Interesting Findings From The Marshmallow Test

Right from the beginning the studies had found some controversial data, and this data is perhaps why the book became so famous. Children who delayed their reward, that is, the children who waited for the full 15 minutes to get a bigger reward, also had better grades, earned more money, and had lower BMIs than the children who did not wait. Put otherwise, what we might hastily conclude is that your child’s performance on the marshmallow test will predict a ton of significant things about the remainder of their lives.

Imagine, your child is 10 years old and now you believe, because they ate a marshmallow too soon, that they will be failures doomed to poverty and poor performance in school.

Furthermore, what sparked even more controversy is that the original study was done on upper class children; meaning, it was a no brainer that children in upper classes could succeed if they just applied themselves.

These two controversies, however, are not the complete picture. What the researchers found has important implications.

Firstly, people can be selective in their effort to delay gratification. For example, Bill Clinton could delay gratification enough to become a Rhodes scholar and a president, but he clearly did not care to resist temptation in his romantic life. So, some of the children will have simply decided that waiting was not worth it. And to add to that point, levels of self-control – as determined by ability to delay gratification – varies across the lifespan. Some people become high in self-control after 18, some before 18, and some never.  Hence, performance on the marshmallow test at age 10 doesn’t determine your entire life course.

Secondly, the results from the marshmallow test were replicated across socio-economic statuses (SES). Children who delayed gratification in low SES positions went on to achieve better grades, earn more money, and maintain lower BMIs; whereas, children in low SES who failed to delay gratification, on average, performed worse than their peers.

To put those two findings into context, the marshmallow tests are just averages. They are not deterministic about each individual. So, a child can perform poorly yet still become successful. Unfortunately, the book doesn’t mention much about other factors that drive success, which I think is a missed opportunity.

Some other interesting findings are that households without fathers resulted in lower performance on delayed gratification tests, that controlling mothers influence performance on delayed gratification tests, that those who performed worse on the test were also bad at regulating their emotions, and that someone’s performance on the test can predict their willingness to cheat in games.

Indians. Curious as to why, the author looked at household situations. He soon discovered rates of fatherlessness were higher for Africans from Trinidad. When he then controlled for rates of fatherlessness, the disparity between ethnicities vanished, which led him to conclude that fatherlessness can impact performance. The explanation given was that children without fathers could not rely on the promises made by strangers since they experienced more cases where men did not keep their promises.

For controlling mothers, another interesting finding was made. When children had controlling mothers, the children would only perform well if they were both high in self-control skills already and also distant from their mothers. Whereas, when children had relaxed mothers, the children would only perform well if the mother was closer to them. Meaning, children with self-control need distance from their controlling mothers to perform well, and children with controlling mothers but low self-control perform poorly. Those who have relaxed mothers require their mothers to be close by to perform well.

The last two interesting findings related to cheating and emotional regulation. Children who performed poorly on the marshmallow test were more likely to cheat in other experiments where children were made to play games. The implication from the research being delayed gratification has a link to moral behavior. And the other finding was that those who perform well also regulate their emotions better. That is because the areas in the brain responsible for delaying gratification are the same areas which regulate emotions.

So, to summarize, the interesting findings from the marshmallow test are:

  1. Delayed gratification correlates with life outcomes like: education, income, and BMI
  2. Delayed gratification predicts willingness to cheat
  3. Fatherlessness can hinder willingness to wait for future rewards 
  4. Controlling mothers can hinder performance, and relaxed mothers are needed for performance
  5. Those who delay gratification can also regulate their own emotions better

Tips and Tricks to Delay Gratification

Now let’s talk about some of the tips and tricks to delay gratification. Here are all the ones I thought were worth remembering:

  1. Focusing on long-term consequences lessens cravings
  2. Identifying with your future self can increase delayed gratification
  3. Belief in self-control can lead to greater levels of delayed gratification
  4. Exposure to realistic images of a reward can lead to longer levels of delayed gratification.
  5. Exposure to actual rewards leads to lower levels of delayed gratification
  6. Hidden rewards can lead to higher levels of delayed gratification.

To start, let’s explore 4,5, and 6. When the author wanted to see what would happen if the conditions of the experiment were changed, he found the results mentioned in 4, 5, and 6.

Counter intuitively, when children were exposed to hyper realistic images of a reward, they delayed gratification. The explanation given was that the children focused on the cool, cognitive features of the reward rather than the hot, impulsive features of the reward: i.e. that a marshmallow is white and like a cylinder: cognitive features; as opposed to hedonic details like the marshmallow being chewy, soft, and incredibly delicious.

When children were exposed to the actual reward, by the same logic, they would wait less time. And indeed they did. Children exposed to the actual reward had significantly lower waits times than those who were exposed to realistic images of the reward. To add onto that, when the reward was hidden entirely, children waited longer. However, they did not wait as long as those who were exposed to the realistic images. Why?

Because the realistic images diverted attention to the cognitive features of the marshmallow whereas simply hiding the marshmallow still left children with the task of “do not think about the marshmallow”. What can be done with that data is nearly endless.

Perhaps blocking your nose from the delicious smells of food can help resist the hot features of food, hiding junk food from sight can ensure you only eat it on the weekends, or adding hyper-realistic images of the food items on the pantry can help you be more cognitive about your choices. Applying those three conditions to your own life successfully just boils down to your creativity!

Now to address points 1, 2, and 3. What the author also found, in conjunction with research done by his colleagues, was that belief in self-control increased impulse control, identifying with your future self increased impulse control, and as did focusing on long-term consequences.

When people were asked how much they identify with their future selves, those who identified with their future selves more strongly than others also had higher incomes and greater amounts of assets. In addition to that, the book also found interesting FMRI data to back this point up, but we won’t go over that here.

Moreover, the author also cites research done by his colleague who found those who believe in self-control, that is, that they have control of their choices and actions, were more likely to complete their goals, change their behaviors and start new habits. They were also more able to adapt to challenges: i.e. high- schoolers who reported higher rates of belief in their self-control were performing better and experienced less stress in adapting to their new environment.

The last point regards long term consequences. When children in the marshmallow test were asked to focus on the long term consequences of eating the marshmallow too early, they were much more likely to delay gratification and get the bigger reward.

Which means, we should implement some of these things in our everyday lives! We should more frequently focus on the long term consequences of our choices and actions, we should imagine the connection between our current self and future self more often, and we should, despite what any philosopher might say, believe in self-control: that we can control our actions and choices.

Critiques and Conclusions

There are a few critiques I have of the book. When I picked the book up, I was expecting more literature on delayed gratification itself. The book, although discusses delayed gratification, becomes far more about self-control, self-development, and theories of human nature as it progresses. From pages 72 and onward the book becomes more about topics slightly outside of delayed gratification; for example, there is an entire chapter on human nature which I did not care much for.

Another critique I have is that the book is very basic. But that is likely by design. The reason why I avoid most self-help books is because they are usually not that groundbreaking. Much of what was said in this book probably could have been explained in a single ted talk, and since I had previously studied psychology, much of what was said in this book wasn’t very new to me; albeit, there were a few counter-intuitive results which absolutely were new to me. But this second critique might be legitimate for only a small group of people.

Overall, the book is still good, and even more so if you have no previous background in psychology. So I do recommend people read the book. Just be cautious to not mistake the book as “the only reason why people are successful is because of self-control”. After all, even the correlations drawn from the original studies are mixed, sometimes being only 0.3 and other times being higher.

Email Notifications


Ideasinhat is a business development analyst and longtime reader of academic literature. He writes books and essays on science and philosophy, and posts them to this website. The essays, as with the books, cover topics from psychology, philosophy, and cognitive science to economics, politics, and law.

Leave a Reply