Only a few experiments drew as much attention as the famous "marshmallow experiments" by Walter Mischel. To test the ability of self-control, Mischel & Co. confronted hundreds of children with a difficult decision: "You can either eat a small amount of sweets (e.g., one marshmallow) right now or you wait for a larger amount of sweets (e.g. two marshmallows)." Many years later the researchers observed an amazing correlation (Mischel et al., 1988; Shoda et al., 1990): Those children who had been able to delay the gratification, were rated higher in social competence, frustration tolerance and academic success. In addition, they even showed better performance in the Scholastic Aptitude Test (SAT). However, especially the calculated correlations with the SAT scores were based on a very small sample (N = 35). Therefore the recently published replication by Duckworth, Tsukayama and Kirby (2013) is highly important to back up the results of the original marshmallow experiments.
Keywords: Walter Mischel - Marshmallow Experiment - Delay of Gratification - Delay Time at age of four - correlation - social competence, academic competence, frustration tolerance - SAT Scores - small sample size - replications - Duckworth, Tsukayama, Kirby - 966 children - smaller correlation
Hello and welcome back to the world of psychology.
Delay of gratification or the famous marshmallow experiments by Walter Mischel.
We live in an age of temptations and instant gratifications.
You can go to the supermarket and get chocolate, cigarettes, alcohol for very little money.
But you can also play video games for which is known that they stimulate the same reward systems of the brain just like drugs.
Or you can, and this is one of the biggest temptations, or you can look for sexual entertainment.
So all the time we have to decide whether we go for the instant gratification or whether we delay the gratification to get a usually much bigger reward in the future.
So, for example, students have to decide, am I going to learn for the test tomorrow or do I prefer to stalk my friends on Facebook?
Well, stalking is so much fun, so let's better stalk.
No, disclaimer, don't stalk.
But also in work life you have to decide, well, am I going to prepare the presentation for tomorrow,
which will improve my career chances in the long run, or do I prefer to watch my favorite soap opera or whatever on the telly?
So there is no doubt that the ability to delay gratifications is extremely important.
And how important it really is was first shown by the famous delay of gratification experiments by Walter Mischel in the public better known as the marshmallow experiments.
In the late 60s and early 70s of the last century, Mischel and colleagues conducted one and the same experiment with some variations with hundreds of four-year-old children.
The experimental design was very simple.
All they did was to lead every single four-year-old in a room with little distractions.
So there was not much more than a table and a chair.
And first of all, also to give the child a chance to get used to the new situation, first of all, they played a little bit.
The next step was they asked the children, well, here you see some sweets or sometimes there were also toys like a poker chip.
Here you see pretzels, marshmallows or some other sweets.
Just tell me, which would you prefer?
I prefer the marshmallow.
Okay, that's a good decision.
And if you had to choose between one marshmallow and two marshmallows, which one would you prefer?
Two?
Well, in fact, all children decided that two marshmallows are definitely better than one.
But you have to ask this question, because maybe some children were influenced by their parents like, well, sweets are bad, don't eat sweets at all.
And so they might have preferred one sweet in comparison to two.
So the experimenter had to make this clear, because otherwise the results would have been influenced by this preference.
The next step was that the experimenter told the child, well, I have to do something else right now.
I'll leave this one marshmallow here and you can eat it if you want.
All you have to do is to ring the bell, so they were given a little bell, and then I'll come and you can eat the marshmallow.
But if you can wait until I come back, you will get two marshmallows.
You can do whatever you want. It's your free decision.
Then the experimenter leaves the room and the children are observed unobtrusively in order to record their delay time.
And one reason for the popularity of the marshmallow experiment is the funny struggle of the children when they try to not eat the marshmallow.
They try to distract themselves by looking away, turning their back to the marshmallow, trying to think of something else or singing songs.
But after some time, their eyes begin to wander back to the marshmallow.
And they begin to smell the marshmallow and they take the marshmallow.
I need this marshmallow.
And it's like a drug addict. It's really funny to watch.
I linked some YouTube videos on the website or just put in marshmallow experiment on YouTube.
It's really great to watch.
The other reason for the popularity of the marshmallow experiment is what was able to be predicted by the delay time of the children.
Ten years later, the children were now in the 8th or 9th grade.
Michel and colleagues asked the parents of those children, well, how would you rate the academic competence of your child?
Or what about the social competence?
The problem frequency or also very important, the coping competence, the ability to get along with difficult situations.
And what they found was really astonishing because they found significant correlations between the delay time and, for example, the academic competence, the social competence or the coping competence.
And that ten years later.
I mean, there are so many things that happen in the time of ten years.
And nevertheless, children that were able to delay gratification at age four in this simple, really simple test were rated better in very important competences.
But that's not all.
Michel and colleagues did not only report the correlation between the parents' impression of competences and the delay time.
They also, and that was even more astonishing, they also reported the correlation between delay time and SAT scores.
Because four years later, for some students, unfortunately, it were just 35.
That's a really big problem.
We will cover this in a minute.
For some students, the SAT scores were available.
And the correlation they reported was sensational.
For the SAT verbal, they reported a correlation of 0.42.
And for the SAT quantitative, they reported a correlation of 0.57.
And this is absolutely crazy.
If you keep in mind that the delay of gratification task took place 14 years before that.
But as I said, it was based on just 35 participants.
And this is highly problematic.
If you study psychology, your professor will tell you, hey, don't trust a correlation that is based on such a small number of participants.
You mustn't do that.
And how dangerous it is, you can see on another result, Michel and colleagues reported.
Because in some variations of their delay experiment, the rewards were not visible.
So the marshmallow was not visible to the child.
And if you correlate the delay time for those children with the SAT scores, you get a negative correlation.
It's not significant, but it's negative.
And this doesn't make sense.
For what reason should there be a negative correlation?
So you see, it's dangerous to trust such correlations that are based on so small numbers of participants.
So what we really need are replications of the marshmallow experiments with bigger number of participants.
And at best, from another research team.
That would be something we could trust.
And I'm very happy to present you a recently published study from Duckworth, Tsukayama, and Kirby.
Published in 2013.
In which 966 4-year-old children took part in the marshmallow experiment.
It was not an exact replication of the Michel studies.
For example, the experimenter waited just 7 minutes until he came back in the room.
In Michel's experiments, the experimenter usually came back after 15 or 20 minutes.
So if the results should differ from those of Michel, there might be some other reasons at play as well.
Again, 10 years later, children were now in grade 8 or 9.
They found a significant correlation between delay time and grade point average of 0.21.
And they found a bigger correlation between delay time and standardized achievement test scores of 0.27.
And in respect to the question, is it possible to replicate the SAT correlation from the Michel studies,
this correlation is very important because standardized achievement test scores are better comparable to SAT scores than the grade point average.
And what we see, just from the numbers, you don't have to be an expert,
the correlation is 0.27, which is obviously far smaller than the correlation from the Michel studies.
Nevertheless, it's still surprising because there were 10 years in between and still there is such a correlation.
For me, this is still surprising.
Last but not least, let's take a look at what you have to do to be able to delay gratification,
to be able not to eat the marshmallow.
And that's what Michel and colleagues tried to figure out during the last 50 years and they're still working at it.
They suppose it's all about attention and attention control.
And if you take a look at the behavior of successful children in the delay of gratification task, it really makes sense.
Those children that look away, that are able to distract themselves, those show higher levels of perseverance.
To confirm the importance of attention control, Michel and colleagues conducted several studies in which they changed small aspects of the experimental setting.
For example, as you already heard, in some conditions the gratification, for example the marshmallow, was not visible.
And if it wasn't visible, if the child wasn't able to smell at it, children showed much longer delay times.
And this was even true, and I think this has some strong implications for our everyday life,
and this was even true when the bigger reward, when the two marshmallows were presented in front of the kid.
So, originally, Michel and colleagues expected to see higher delay times in this condition,
because now the child sees, well, if I wait, I will get these two marshmallows.
But this was not the case.
Knowing that they will get only one marshmallow if they ring the bell, they rang the bell,
because, probably because, they just couldn't resist the temptation.
So, the everyday life implication is, if you should have alcohol problems, or have a smoking addiction, or a serious drug addiction,
the worst thing you can do is to say, well, I'll put the cigarettes right here on the table,
and I will work three hours at my essay, or whatever you have to do, and then I will go and smoke a cigarette.
Probably, you will have trouble to resist.
The best thing you can do, and it's a really simple, but powerful intervention, the best thing you can do is,
if you have alcohol problems, don't have alcohol in your house.
If you want to quit smoking, don't have cigarettes in your house.
On the other hand, the things you really want to do, those should be in your sight.
So, if you want to read a book, put it on the table, put it somewhere where you can see it.
This is basic stuff, but it's really powerful.
Another interesting strategy, and according to the results of Michel's experiment, a very powerful strategy,
is to think of something else, and to change the modality of your thoughts.
So, in some experiments, Michel and colleagues told the four-year-olds, well, it might be helpful,
you can try this if you want, it might be helpful for you to think of the pretzel sticks as long, thin, brown logs.
And children who thought about the rewards in this neutral way were able to delay the gratification for a much longer time
than children who were told, well, think of the arousal aspects of the pretzel sticks.
Think of the crunchy, salty taste of the pretzel sticks.
Of course, those children, they had no chance.
But the biggest effect occurred when children were told, think about the positive aspects of something else.
So, if they were presented a marshmallow, they were, for example, told, think of the crunchy and salty taste of a pretzel.
And in this condition, children waited for almost 17 minutes on average.
So, this seemed to be the most effective strategy.
What can we learn from that?
Well, if you have the urge to smoke a cigarette, maybe it helps to think of the positive feeling you get when you are running,
in case that you like running.
So, maybe this could help.
Well, we will cover more of that in the next episodes.
Especially in the next episode, we will take a look whether it is really self-control,
which is important in the delay of gratification task, or maybe it's intelligence or social trust.
Do I believe that the experimenter will come back and give me the marshmallow?
So, maybe there are some other factors at play.