Marshmallow[1] experiment (
English Stanford marshmallow experiment)
[2] — a series of studies on
delayed gratification, conducted in the late 1960s and early 1970s under the direction of psychologist
Walter Mischel, who later became a professor at
Stanford University. In these studies, children were offered a choice between one small reward provided immediately and a reward doubled in size if they could patiently wait for it for a short period (approximately 15 minutes), during which the experimenter left the room and returned after the waiting period. The rewards used were
marshmallows,
cookies or
a cracker. In subsequent studies, scientists found that children who were able to wait for the larger reward generally went on to lead more successful lives. These conclusions were based on their
SAT scores
[3], level of education
[4],
body mass index (BMI)
[5] and other quality-of-life indicators
[6].
Original experiment
[
edit |
edit source]
The experiment originated from an earlier study conducted in
Trinidad, where Mischel noticed that members of different ethnic groups living on the island held contrasting
stereotypes about their island neighbors. Members of different groups believed that they differed greatly in such traits as recklessness,
self-control and the ability to have fun
[7]. This small study (n = 53) focused on boys and girls aged 7 to 9 (35 West Indian Black children and 18 Indian Caribbeans (
English Indo-Caribbeans)) at a rural school on the island of Trinidad. The children were asked to report their choice: receive a 1¢ candy immediately or a 10¢ candy in a week. Mischel reported a significant difference between the ethnic groups
[8], since the Indian children demonstrated a much greater ability to delay gratification compared with the African pupils, as well as greater age differences, and that “a comparison of the ‘high’ and ‘low’ socioeconomic groups in terms of experimental choice revealed no significant difference.” The absence of a father was common in the African group (in the East Indian group, only one child had this circumstance), and this characteristic showed the strongest effect: children from intact families demonstrated superior ability to restrain their desires.
Stanford experiment
[
edit |
edit source]
The first “marshmallow” test was a study conducted by Walter Mischel and Ebbe B. Ebbesen at Stanford University in 1970
[9].
The purpose of the original study was to understand when children develop the ability to delay gratification—that is, the ability to wait in order to obtain what they want. The original experiment took place at
Bing Nursery School, located at Stanford University, with children aged four to six. The children were taken to a distraction-free room where a treat of their choice (an
Oreo cookie, marshmallow, or pretzel) was placed on a table
[2]. The researchers allowed the children to eat the treat immediately, but promised to double the reward if they could wait 15 minutes without giving in to temptation
[2]. Mischel observed that some children covered their eyes with their hands or turned away so they would not see the treat; others banged on the table, tugged at their hair, or played with the marshmallow as if it were a tiny stuffed animal, while others simply ate the treat as soon as the researchers left
[2].
Of the more than 600 children who took part in the experiment, most ate the sweet immediately. Approximately one third managed to wait long enough to receive the doubled treat
[2]. The determining factor in delayed gratification was age.
Method
[
edit |
edit source]
Participants
[
edit |
edit source]
The participants were 16 boys and 16 girls attending
Bing Nursery School at Stanford University. The results of three participants were excluded because they were unable to understand the instructions. The children were between 3 years and 6 months and 5 years and 8 months old (with a mean age of 4 years and 6 months). Eight researchers (four men and four women) were randomly assigned to each of the four experimental conditions. Under each condition, every experimenter tested two boys and two girls to avoid systematic bias based on the experimenters’ sex
[9].
Conditions
[
edit |
edit source]
- The immediate (less preferred) and delayed (more preferred) rewards were placed in front of the child and were visible[9].
- Both rewards were hidden from the child[9].
- The child could see only the doubled (delayed) reward[9].
- The child could see only the ordinary (immediate) reward[9].
Follow-up studies
[
edit |
edit source]
In subsequent studies, Mischel found unexpected correlations between the results of the marshmallow test and the children’s life achievements many years later
[6]. The first follow-up study, in 1988, showed that “preschool children capable of delaying gratification were described by their parents, more than 10 years later, as adolescents who were significantly more competent.”
A second follow-up study, in 1990, showed that the ability to delay gratification also correlated with higher
SAT scores
[6].
In 2011, a brain-imaging study was conducted on the experiment’s participants when they reached middle age. The main differences between individuals with longer and shorter delay times were found in two areas: the
prefrontal cortex (more active in participants with longer delay times) and the ventral striatum (the area “responsible” for cravings), when they tried to control their responses to tempting stimuli
[10][11].
In 2018, Tyler Watts and others published a study that refuted the experiment
[12]. The authors argued that the original experiment had been conducted on several dozen children of university employees, whereas Watts conducted it on 900 children, divided by race and income. By age twenty, the influence of “willpower” was close to statistical error, while a
common cause was identified: family wealth, which affected both “willpower” and success. For poor children, “later” might never come, while for wealthy children the treat in the experiment was an insignificant reward because a more substantial one was available to them at home.
In 2019, the experiment was repeated once again; it confirmed the conclusions of the first experiment and refuted the 2018 refutation
[13][14].
