Ricochet is the best place on the internet to discuss the issues of the day, either through commenting on posts or writing your own for our active and dynamic community in a fully moderated environment. In addition, the Ricochet Audio Network offers over 50 original podcasts with new episodes released every day.
Un-Teaching Grit: The Marshmallow Test Revisited
James Pethokoulis recently asked how we can teach children the grit – the persistence, self-control, and conscientiousness – they need to climb the opportunity ladder. I don’t have a great answer to that question, but I do think there’s a likely answer to the question, “How do we un-teach grit?”
Many of us are probably acquainted with the Stanford marshmallow experiment. In this experiment, a young child is left alone in a room with a reward (in some cases, a marshmallow) with the promise that — if he refrains from eating the reward for a fixed length of time — he can have an even bigger reward once the time is up. It is considered a classic test of a child’s ability to delay gratification. When the researchers followed up on the children years later, they found that those able better to delay gratification had turned into more successful young adults: they had higher SAT scores and educational attainment, lower BMIs, were better able to cope with stress and frustration, and so on. Self-control wins! Grit wins! Conservatism wins! Go us!
Not so fast. The marshmallow test could also be considered a classic test of the child’s ability to trust the researcher.
The researchers who ran the original marshmallow experiment forgot to look at things from the child’s perspective: i.e., while the researchers knew that their promise of the greater reward would be fulfilled on time, the children had no good way of judging whether the researches could be trusted. As anyone who has ever been a kid should remember, sometimes even the most well-meaning adults break their promises to children. Children who have little reason to expect adults to keep their promises have little reason to expect that the researchers will actually return with the promised reward, and consequently less reason to wait
The marshmallow test has since been revisited with the child’s perspective in mind. In the new version, the children are divided up into two groups. The researchers interact reliably with the children in the first group, always keeping their promises. In the other group, the they are deliberately unreliable and consistently break their promises (though always with an apology). Only after being exposed to either reliable or unreliable interaction from the researchers, are the children given the original marshmallow test.
To the researcher’s astonishment — but not at all to mine — their behavior had a ginormous effect on how well the children do on the subsequent marshmallow test. Children whose prior experience with the researchers was unreliable wait only about three minutes, on average, before consuming their marshmallow. Children whose prior experience with the researchers was reliable, on the other hand, wait an average of about 12 minutes before consuming their marshmallow, four times as long!
Since, in prior studies, the average time a child would wait during the marshmallow test was about 6 minutes, this shows that reliably interacting with children doubles their apparent self-control, while interacting unreliably with children cuts their apparent self-control in half. Even “young children’s actions are… based on rational decisions about their environment”, the marshmallow-test revisitors conclude:
Being able to delay gratification – in this case to wait 15 difficult minutes to earn a second marshmallow – not only reflects a child’s capacity for self-control, it also reflects their belief about the practicality of waiting,” says Kidd. “Delaying gratification is only the rational choice if the child believes a second marshmallow is likely to be delivered after a reasonably short delay.
So, how do you un-teach children grit? Behave unreliably toward them and break your promises to them. Tell them they’ll get an allowance if they do their chores for the week, then — on the day when all their chores are actually done — inform them that their allowance money has already been spent on milk or gas. Tempt them with treats that never appear, and forget to follow through on your promised punishments.
This kind of unreliability is more likely to happen when you’re living beyond your means. Many parents aren’t heartless or foolish enough to spend their children’s allowance on frivolities, but will feel perfectly justified in spending it on necessities like milk or gas when family funds are running low. Similarly, many parents would feel justified spending money saved for promised Christmas gifts on an emergency car repair if they feel they have no alternate source of funds. From the adult perspective, this is eminently reasonable. Necessities like food and transportation must take precedence over frivolities like gifts and kids’ pocket money. But from the kid’s perspective, it is still a broken promise.
Even a child praised for his self-sacrifice in understanding his parents’ decision may nonetheless get the message that promised rewards aren’t worth waiting for. Heck, adult behavior is similarly influenced by an unreliable environment, so even a child “old enough to understand” that his college fund has disappeared due to a market crash or a family medical emergency may nonetheless find the disappearance of the promised funds demotivating. No wonder the advice “Don’t set up a college fund for your child. Make him understand from the beginning that he is to pay his own way,” is currently so popular in some circles. During tough economic times, it is a promise any parent can keep.
Poor people often find themselves in situations where it’s difficult to live within their means. Undoubtedly, many people are poor because they lack inherent self-control, but even poor people with good self-control find themselves unable to follow through as often as their better-off counterparts, as they’re more prone to changes in circumstance. Poor children, therefore, grow up in a more unreliable environment even when their family structure is stable. When economic fragility is paired with family instability — as it so often is among America’s poor — it’s no wonder that poor children disportionally mistrust adults and their promised rewards. Why should a teenager raised from infancy in an unreliable environment believe the adults who tell him that — if he works hard and stays in school, he can someday, perhaps years from now – escape the poverty he has known all his life? As Russ Roberts, whom Pethokoulis cited, notes, adults can tell kids lots of things, but that doesn’t mean they’ll believe them.
Paul Tough, author of How Children Succeed, admits “We don’t yet have a curriculum or a real methodology that’s going to say: Here’s how you boost conscientiousness scores.” He speculates that talking up the importance of good character traits and “how hard work and conscientiousness now is going to lead to clear paths later” makes an impression on children. He notes that the Knowledge Is Power Program (KIPP) has had great success fostering grit in kids, and speculates that KIPP’s strong group identity and its use of intense propaganda (“wall-to-wall messaging”) plays a role.
Unmentioned, though, is the role that the sheer reliability of the KIPP environment might play in fostering children’s apparent self-control. Perhaps what KIPP is really good at is giving children exactly what they were promised, maybe for the first time in their lives.
Published in General
The difference is that kids are MUCH smarter than dogs, and thus are far more manipulative.
Example: they used to use breath monitors on newborns, only to discover that some of them, within days of birth, would deliberately hold their breaths for the attention.
Kids have nothing better to do than raise their parents – and they have much more energy than we do. Which is why bad parents are usually very, very tired.
Quite.
Example: Kid does something bad. Parent gets mad.
Why did the kid do something bad? He wanted attention, and would settle for bad attention if that is the only option. Which means that parents who are actually smart will see that what the kid wants is NOT to do something bad and get away with it. The kid wants the outcome!
This is why I will not put down a quiet kid who wants to be held. And why I will not hold a screaming child. Most parents do this exactly backward: pick up the screamer, put him down when he is quiet. But: the kid is screaming because he wants to be held! Why punish good behavior and reward the negative behavior? Because parents are stupid.
In this case, parents are teaching their kids to scream. I teach mine to be quiet. Some of my kids stopped screaming to be picked up at the ripe old age of 8 days. Result? Happier home. Squeeze-toy children who are highly tactile and cuddly from 8 days to 18 years (my oldest).
Similarly, from the age of 2, our kids are welcome to scream all they want. But I don’t have to hear it: they can do it in a closed room.
Somehow, screaming is more work and less fun without an audience. Who knew?
Despite my step-children all being in their 20’s they know enough to NOT question the existence of Santa Claus. He is very real in our house….he likes motorcycles and spends too much time on Ricochet…and he is very reliable come the end of December.
Other than maybe slathering it with Peanut Butter first, I can’t imagine a worse punishment than being given a marshmellow.
Oh, agreed. Which is why if you can’t handle a dog, you’ll never be able to handle a child. And while children are more manipulative than dogs, a lot of the lessons learned scale up. (And in my experience, I’ve never seen a family with a well-behaved dog and bratty kids.)
My training in training was primarily horses instead of dogs, which adds that wonderful element of getting an animal that could easily kill you into behaving correctly, and I was doing it at an age when I couldn’t see over the horse’s back. You learn to take charge or you get hurt.
As a side note my dad was a genius at brainwashing our dog. Whenever I’d volunteer to take her for a walk, he’d sit down next to her, throw an arm around her an explain how “Tom didn’t want to take you on a walk, but daddy made him because he loves you and cares about his dog.”
She was just smart enough to get the gist of this without seeing through the subterfuge. No wonder she worshiped him.
This post put me in mind of the recent TV show “Legends” about an FBI agent played by Sean Bean, who was constantly making and breaking promises to his son, who seemed to be 10 years old or so, to do things together.
It was heartbreaking to watch the poor kid be disappointed each and every time. It didn’t matter to him that his father was needed at that moment to prevent someone from being murdered, or to catch the world’s most notorious terrorist. I mean to say, he understood that his father had to disappoint him, but it hurt nonetheless, and he learned the lesson not to trust his promises. Perhaps he would grow up not trusting much of anyone.
Amen! This is crucial when raising kids, and it goes for rewards and punishment.
On the issue of trust of the actual researchers: Actually life includes risk. Doing the right thing doesn’t always result in good things for the one doing the right thing.
It almost always results in good things over the long term, but adults often have their trust abused in small and large things. And a child growing up learns that too long before reaching adulthood.
And leaving out trust issues, just because you work hard on a project, doesn’t mean the project will turn out right. On the other hand, you’ll probably learn from a failure you worked hard on.
So a child gets disappointed by an adult. It happens. It’s not the end of the world, or the end of that child’s development towards being a productive adult.
Of course life includes risk, and it’s unrealistic to expect good behavior to be rewarded 100% of the time. But doesn’t it make sense that the more uncertain a reward, the less incentive there is to seek it?
For example, suppose your boss promises you that your salary this year will be $100,000. Is there a chance that the salary will be less than promised due to unforeseen circumstances, even if your boss is a very honest person? Yes. But if you suspected your boss would delay all paychecks till the end of the year, promising you a lump sum, then flip a coin to decide whether to pay you or not, suddenly that promise of $100,000 a year becomes much less attractive.
“That’s no way to treat an adult,” you might object. But that’s no way to treat a child, either.
@MFR#30: yes, but ordinary adults are not claiming to be elite behavioral experts experienced in conducting high cost research projects to produce results that whole displines will base their premises on. So when such people do research I would expect more knowledge in animal behavior much less human behavior than your average animal trainer. But it seems that I am wrong.
@MFR#42: you know my boss? You basically described several of my jobs over the years.
Wow. Is it too soon to talk about congratulations? (It also caught me by surprise. I had figured your great wisdom involved a few more years.)
Yes, a bit.
Thanks! Conservatives are natural-born curmudgeons.