A new study published last week in the journal Frontiers in Psychology has found that "delay discounting" is more predictive of how much people earn than "age, ethnicity, or height".
The study from Temple University recruited over 2,500 participants from the crowdsourcing marketplace Amazon Mechanical Turk to participate in a delay discounting trial.
Delay discounting is the perceived decline in the value of an object when it is received at a later date, as opposed to receiving it in the present.
Delay discounting focuses on monetary rewards and it's a simplified version of something you may have heard of before: delayed gratification.
In the Frontiers in Psychology study, participants were offered a small sum of money to be collected immediately, or a larger sum later ($500 now or an additional $1,000 offered at six increasing time intervals, bringing the total to $6500 at one year).
The participant sample was important – it was extremely broad compared to the Western, Educated, Industrialised, Rich, and Democratic (W.E.I.R.D) first-year college students that are typically recruited for psychology studies.
The participants varied greatly in age (25 to 65 years old), education level, annual income and ethnicity.
Using a modelling system, the researchers found that occupation, education, gender, and postcode code were the most prominent predictors for income level.
However, the fifth most important variable was how well the participant performed in delay discounting, which trumped race, ethnicity, height, and even age.
Dr William Hampton, a researcher from Temple University and the study's lead author, told German research platform Morressier that these results could have significant implications for how we raise and educate children.
"This research may spur a greater interest in how we could train people, especially children, to be better at delaying gratification.
"That is, if you want your child to grow up to earn a good salary you might consider teaching them the importance of passing on smaller immediate rewards in favour of larger ones that they have to wait for."
What exactly is delayed gratification and what are the claims here?
Delayed gratification is a developmental psychology term that was defined through the famous Stanford marshmallow experiments.
In the 1960s Walter Mischel and his co-authors seated children in a laboratory in front of a single marshmallow on a plate and told them that they would reap a bigger reward (two marshmallows) if they did not eat the confectionery before the researcher returned to the room.
The children were all of preschool age and attended a preschool on the Stanford University campus (this will become important later).
In a 1988 follow-up study of the Stanford children, Mischel found that successfully delaying the gratification and holding out for the two marshmallows at 4 years-old was associated with better outcomes in later cognitive and social development, health and even brain structure in their early college years.
The marshmallow experiment has since become a classic laboratory study and you can see a cute video of children doing a bad job of self-regulating in this video of the study.
Delaying gratification, delay discounting, and self-regulation are all variations on the same concept — self-control. The ability to put aside short-term gain for long-term goals makes us more successful.
People who tend to think and act with an eye on the future naturally have a greater ability to delay their gratification and reap the larger reward.
Children and older adults have been shown to see the perceived value of a reward dropping off more sharply than other age groups when they're forced to wait.
Children who can value the long-term gain and hold out on eating the marshmallow and delay their gratification have been shown to have superior working (short-term) memory and motor inhibition (regulation of physical impulses) relative to those who have failed their gratification delay and ate the treats.
The children who are more successful on the marshmallow test have also been found to have better academic results in college as well as better emotional regulation when facing stress.
And a 1999 study showed that among 56 people addicted to heroin and 60 people who weren't addicted, the people with addictions showed much sharper delay discounting (valuing the monetary reward far less when it was handed over in the future).
The researchers concluded that their results lent validity to the idea that delay discounting was a measure of "impulsiveness, a characteristic associated with substance abuse".
There have been a number of interventions designed to try and improve children's scores on the marshmallow test.
Parenting blogs and websites espouse the value of teaching children to be "waiters" and suggest instilling distraction techniques to equip kids with skills that appear so predictive of lifetime success.
One 2015 study found that employing a ritualised session of games and instructions for children in school improved children's ability to delay gratification.
Another study in 2016 used attention training techniques to "turn one marshmallow into two" by playing recordings of sounds such as birds singing, traffic, and ticking clocks with instructions for children, to direct attention in specific ways. This training significantly improved children's ability to delay gratification.
What's the problem here?
Earlier this year, researchers from the University of California (UC) and New York University (NYU), decided to revisit the famed marshmallow test focusing on a large sample of children born to mothers who had either not completed college or who had.
This was an important difference from the children used in the original study, whose mothers would have invariably been college-educated, based on their choice of preschool.
Mischel's original study used fewer than 90 children, while assistant professor Tyler Watts and his research team in the new study observed 918 participants who had completed a gratification delay (marshmallow) test before four and a half years old.
The research team then observed the children's test scores at 15 years old, including mathematics, reading comprehension, vocabulary and word recognition.
The UQ and NYU researchers were replicating elements of Mischel's marshmallow study and found that there were still statistically significant associations between the ability to delay eating the marshmallow and later achievement in life.
However, when the researchers accounted for variables in the lives of these children, the marshmallow effect was negligible, and the researchers concluded that perhaps the marshmallow test wasn't capturing what Mischel and his co-authors first believed.
Watts told BuzzFeed News that while there was still a small correlation between delaying gratification and success, that correlation was tempered when factors such as race, ethnicity, socioeconomic status, and gender were included.
"So the main takeaway from our paper is that Mischel wasn't wrong, there was an association between delayed gratification early on and later achievement, but there's kind of an extra nuance, there's a caveat to that.
"What's making delayed gratification predictive of later stuff in life are probably these other large factors that are present in these kids' lives, and other cognitive ability."
Watts notes that while the Temple University study is interesting, it still does not prove delayed gratification as a singular cause of higher income, just that it's an associated factor.
Therefore, Watts says, interventions designed to simply help children delay gratification and improve their self control might not actually be that helpful.
"What we're probably saying is that you can find interventions that would maybe just teach a kid strategies to get through the marshmallow test. I'm a lot more sceptical that that sort of thing alone would have much benefit."
Watts says that any of the early intervention strategies that solely teach children to strive for the second marshmallow probably won't have "really big effects later on in life".
Wealth might be playing into the marshmallow story in a big way.
In the UQ and NYU study, children whose mothers had obtained a college degree managed to wait out the full seven minutes of the trial at a rate of 68%, while only 45% of children whose mothers did not complete college managed to hit the end of the seven-minute bracket without consuming the snack.
Assistant professor Jessica McCrory Calarco from Indiana University theorised in an article for The Atlantic that the marshmallow test was actually just biased by the tendencies of children who come from less privileged backgrounds.
"The failed replication of the marshmallow test does more than just debunk the earlier notion; it suggests other possible explanations for why poorer kids would be less motivated to wait for that second marshmallow," said McCrory Calarco.
"For them, daily life holds fewer guarantees: There might be food in the pantry today, but there might not be tomorrow, so there is a risk that comes with waiting."
Watts notes that this is a valid sociological interpretation and Calarco is "taking the data points that we laid out and trying to connect them from her perspective", but says the study wasn't specifically designed to prove the points Calarco raised.
So, should we still care about the marshmallow test at all?
Watts says that the test still has its place in developmental psychology.
"We still want to understand how people think and how they behave and how they feel ... the ability to delay gratification and self control, I think we still have a lot of reason to believe that those are life skills and describe meaningful differences between people and how they interact with the world."
Watts believes there are benefits implied by the marshmallow test that we need to apply in our day-to-day lives for basic success and people still "need to be able to delay gratification to get through the working day".