Journal Article 1 Mention
Delay of gratification and adult outcomes: The Marshmallow Test does not reliably predict adult functioning
Jessica F. Sperber2024
Deborah Lowe VandellGreg J. Duncan
Top 9% · 91st Percentile
8 citations · Genetics
Open Access

TLDR

A preschool child’s waiting time for a bigger treat did not reliably forecast education, health, or behavior at age 26. Broader early-life experiences and abilities mattered more than this single waiting task.

Summary

1 Study Aim

The study examined the long-term predictive validity (how well a measure forecasts later outcomes) of the Marshmallow Test, a waiting task involving smaller immediate and larger delayed treats. The authors tested whether performance at 54 months predicted achievement, health, and behavior at age 26. They expected simple links to weaken after adjusting for confounding characteristics (other factors affecting both waiting and adult outcomes). The researchers asked whether one preschool waiting task can meaningfully forecast how adults function later in life.

2 Study Design

The study analyzed 702 participants from the NICHD Study of Early Child Care and Youth Development, recruited across 10 United States sites. The sample was 83% White, 46% male, and mainly middle-class, so it was not nationally representative. At 54 months, children chose between waiting seven minutes for more food or ending the task for less food. At age 26, surveys measured education, earnings, debt, depression, body mass index (BMI, a height-and-weight measure), substance use, police contact, impulsive behavior, and risk-taking. Preregistered regressions (statistical models estimating links while accounting for other factors) added demographics, early home conditions, and childhood cognitive and behavioral measures. Multiple imputation with chained equations (a method for filling in missing data) addressed missing values. The researchers compared simple links with increasingly adjusted estimates using childhood and adult data.

3 Findings

The study found modest unadjusted links between longer waiting and higher educational attainment, and between longer waiting and lower BMI. These links were measured as correlations (statistical relationships), with both values near .17 in size. Nearly all regression-adjusted coefficients became nonsignificant, meaning they no longer provided reliable evidence of a relationship. Waiting the full seven minutes did not consistently distinguish adults from those waiting less than 20 seconds. No clear moderation (a difference in the relationship across groups) appeared by socioeconomic status (SES, access to social and economic resources) or sex. The findings challenge the Marshmallow Test’s predictive validity and construct validity (whether it measures the specific ability it claims to measure). The authors caution against using it as a screener or narrowly targeted intervention measure, recommending broader attention to children’s abilities and environments. The BMI finding requires further experimental testing because it came from an exploratory analysis without adjustment for multiple comparisons. One brief preschool task did not reliably forecast adult success, health, or behavior.

Abstract

This study extends the analytic approach conducted by Watts et al. (2018) to examine the long-term predictive validity of delay of gratification. Participants (n = 702; 83% White, 46% male) completed the Marshmallow Test at 54 months (1995-1996) and survey measures at age 26 (2017-2018). Using a preregistered analysis, Marshmallow Test performance was not strongly predictive of adult achievement, health, or behavior. Although modest bivariate associations were detected with educational attainment (r = .17) and body mass index (r = -.17), almost all regression-adjusted coefficients were nonsignificant. No clear pattern of moderation was detected between delay of gratification and either socioeconomic status or sex. Results indicate that Marshmallow Test performance does not reliably predict adult outcomes. The predictive and construct validity of the ability to delay of gratification are discussed.

Referenced In