Summary: A variation of the classic marshmallow test shows that young children will delay gratification for much longer when they know a teacher will learn how long they waited.
Source: APS
The “marshmallow test” is one of psychology’s best-known experiments: a young child is offered a small reward immediately (one marshmallow) or a larger reward for waiting (two marshmallows). Prior research has linked the ability to wait with later positive outcomes, including academic performance years later. New research published in Psychological Science refines our understanding of what drives children’s willingness to wait.
In a series of experiments, researchers led by Gail Heyman at the University of California San Diego investigated whether concerns about reputation — specifically, who will learn about a child’s choice — influence delay-of-gratification behavior. The study found that children’s waiting times increased substantially when they were told others would know how long they had waited, and that teachers had a stronger effect than peers.
Heyman explains that while delay-of-gratification tasks have traditionally been viewed as measures of self-control, they may also reflect young children’s sensitivity to social signals about what others value. “Children are constantly observing cues about what adults and peers appreciate,” she said. “Our results suggest that a child’s decision to wait can reflect not only internal self-control but also an assessment of the social benefits associated with appearing patient.”
The study combined data from two experiments with a total of 273 children aged three to four in China, conducted by teams at UC San Diego and Zhejiang Sci-Tech University. Each child was given the standard choice: take a small treat immediately or wait to receive a larger treat. Participants were randomly assigned to one of three conditions. In the teacher condition, children were told their teacher would be told how long they waited. In the peer condition, they were told a classmate would learn their waiting time. In the standard condition, no one was specified as finding out.
Results showed that children waited longer in the reputation conditions than in the standard condition, indicating that the prospect of others knowing their choice influenced behavior. Furthermore, waiting times were longest in the teacher condition and shorter in the peer condition, with children in the teacher condition waiting roughly twice as long as those in the peer condition. The pattern suggests that the identity of the potential evaluator matters: teachers, as authority figures and important evaluators in early childhood contexts, produced a larger reputational effect than peers.
The researchers interpret these findings through a cost–benefit lens. When deciding whether to wait, children appear to weigh immediate material gains against potential social gains, such as a reputation boost. This reputational motivation can change the calculus in favor of waiting even when no explicit rewards or instructions promote patience. The study therefore extends the interpretation of delay-of-gratification tasks beyond pure self-control to include social-cognitive factors like reputation management and sensitivity to who will observe or hear about their behavior.
The findings surprised the team because conventional thinking holds that three- and four-year-olds are generally too young to worry about others’ opinions. Instead, the study provides evidence that even preschool-age children can make inferences about the values of people around them and adjust their behavior accordingly. Heyman notes that children’s sensitivity to reputational cues may arise from many everyday sources, such as parental remarks, teacher feedback, or broader social signals that indicate which behaviors are socially rewarded.
Co-authors on the paper include Fengling Ma, Dan Zeng, and Fen Xu from Zhejiang Sci-Tech University, and Brian J. Compton from the University of California San Diego. The research received support for Fengling Ma from the National Natural Science Foundation of China (grant 31400892), the Natural Science Foundation of Zhejiang Province (LY17C090010), and the China Scholarship Council.
About this psychology research article
Source:
APS
Contacts:
Charles Blue – APS
Image Source:
The image is in the public domain.
Original Research: Closed access. “Delay of Gratification as Reputation Management” by Gail Heyman et al., Psychological Science. DOI: 10.1177/0956797620939940
Abstract
Delay of Gratification as Reputation Management
Delay-of-gratification tasks have traditionally served as measures of self-control, but recent work suggests that rational decision processes also guide performance. This research tested whether children’s choices are shaped not only by task-internal costs and benefits but also by anticipated reputational consequences. Across two studies involving 273 three- and four-year-old children in China, participants were placed in either a standard delay-of-gratification condition or a reputation condition in which they were told their teacher or a peer would learn how long they waited. Children in the reputation conditions waited longer than those in the standard condition, and waiting was longer when the evaluator was a teacher compared with a peer. These results indicate that performance on delay-of-gratification tasks is sensitive to reputational concerns and to who might evaluate the child’s behavior.