The marshmallow test, developed by Walter Mischel and colleagues at Stanford University beginning in the late 1960s, is one of psychology's most iconic experiments. A child (typically 4–6 years old) is placed in a room with a single marshmallow (or other treat) and told that if they can wait until the experimenter returns (typically 15 minutes), they will receive two treats instead. Some children eat the marshmallow immediately, others wait varying lengths of time, and some successfully delay for the full period. The test measures the capacity for delayed gratification — the ability to resist an immediate reward in favor of a larger future reward.
Longitudinal Findings
Mischel's follow-up studies revealed remarkable correlations between seconds of delay at age 4 and outcomes decades later. Children who waited longer scored higher on the SAT as adolescents (by approximately 200 points), were rated as more socially competent, coped better with stress and frustration, and showed better executive function. Further follow-ups into adulthood found associations with educational achievement, body mass index, and substance use. These correlations — spanning over 40 years — generated enormous interest in self-control as a foundational capacity for life success.
Strategies for Delay
Mischel's research identified specific cognitive strategies that successful delayers used. "Hot" strategies that focused on the reward's appealing qualities (its taste, smell, texture) undermined delay. "Cool" strategies that reframed the reward abstractly (thinking of the marshmallow as a cloud or a cotton ball) extended delay dramatically. Distraction — looking away, singing, playing — was also highly effective. These findings showed that delay of gratification is not simply a trait — it is a skill that depends on cognitive strategies for transforming the psychological situation.
Recent replication studies have qualified the original findings. Watts et al. (2018) found that the marshmallow test's predictive power was substantially reduced when controlling for family socioeconomic status, home environment, and early cognitive ability. Children from disadvantaged backgrounds may rationally choose the immediate reward based on experience that promised future rewards often don't materialize. These findings suggest that the marshmallow test measures not just self-control capacity but also trust in the environment, rational adaptation to unstable conditions, and the cognitive and emotional resources that socioeconomic advantage provides.