Pessimism of the Intellect, Optimism of the Will     Archive   Favorite posts   Twitter: @steve_hsu

Sunday, May 17, 2009

The marshmallow experiment

A simple test administered to four year olds has impressive predictive power for later life outcomes. Are we talking about an IQ test? No, a test of self-discipline, or the ability to delay gratification.

As described in Wikipedia:

In the 1960s, a group of four year olds were given a marshmallow and promised another if they could wait 20 minutes before eating the first one. Some children could wait and others could not. The researchers then followed the progress of each child into adolescence, and demonstrated that those with the ability to wait were better adjusted and more dependable (determined via surveys of their parents and teachers), and scored an average of 210 points higher on the Scholastic Aptitude Test.




See also this update on the research in the New Yorker.

...Footage of these experiments, which were conducted over several years, is poignant, as the kids struggle to delay gratification for just a little bit longer. Some cover their eyes with their hands or turn around so that they can’t see the tray. Others start kicking the desk, or tug on their pigtails, or stroke the marshmallow as if it were a tiny stuffed animal.

...Mischel and his colleagues continued to track the subjects into their late thirties—Ozlem Ayduk, an assistant professor of psychology at the University of California at Berkeley, found that low-delaying adults have a significantly higher body-mass index and are more likely to have had problems with drugs—but it was frustrating to have to rely on self-reports. “There’s often a gap between what people are willing to tell you and how they behave in the real world,” he explains. And so, last year, Mischel, who is now a professor at Columbia, and a team of collaborators began asking the original Bing subjects to travel to Stanford for a few days of experiments in an fMRI machine. ...The scientists are hoping to identify the particular brain regions that allow some people to delay gratification and control their temper. They’re also conducting a variety of genetic tests, as they search for the hereditary characteristics that influence the ability to wait for a second marshmallow.

It seems to me that the ability to delay gratification (future orientation), as tested in the experiment, is only moderately correlated with IQ. We all know high IQ slackers and diligent people of average intelligence. The data also show that IQ is highly correlated with income but much less so with accumulated wealth.

Both diligence and intelligence are necessary for success in life, which is why the tried and true inputs used in college admissions are high school grades and SAT score. The former are somewhat more correlated with diligence than the latter, so between the two we get a metric which is sensitive to both psychological factors.


Here are the original papers on the marshmallow study, kindly provided to me some time ago by the Shoda Lab:

Mischel, Shoda and Peake (1988)
Mischel, Shoda and Rodriguez (1989)
Shoda, Mischel and Peake (1990)


Time orientation (like other personality factors) and IQ are heritable traits. The dreaded issue of group differences is mentioned briefly in the New Yorker piece.

...Mischel soon grew interested in a different project. He lived in a part of the island that was evenly split between people of East Indian and of African descent; he noticed that each group defined the other in broad stereotypes. “The East Indians would describe the Africans as impulsive hedonists, who were always living for the moment and never thought about the future,” he says. “The Africans, meanwhile, would say that the East Indians didn’t know how to live and would stuff money in their mattress and never enjoy themselves.”

Mischel took young children from both ethnic groups and offered them a simple choice: they could have a miniature chocolate bar right away or, if they waited a few days, they could get a much bigger chocolate bar. Mischel’s results failed to justify the stereotypes—other variables, such as whether or not the children lived with their father, turned out to be much more important—but they did get him interested in the question of delayed gratification.

blog comments powered by Disqus

Blog Archive

Labels

Web Statistics