Gratification Delay and Adult Outcomes: Does the Marshmallow Test predict adult functioning?

• In Event: Early Self-Regulation and Long-Term Outcomes: Why Does Childhood Self-Regulation Predict Adult Success?

Abstract

The seminal Marshmallow Test is a measure of gratification delay that assesses a child’s ability to resist temptation in favor of a future reward (Mischel, 2014). Delay of gratification performance has been shown to predict a host of positive outcomes, including higher SAT scores (Shoda et al., 1990), better coping abilities (Mischel et al., 1988), and lower body mass index in adulthood (BMI; Schlam et al., 2013). However, a recent conceptual replication challenges this predictive validity: Watts et al. (2018) found that the association between Marshmallow Test performance and adolescent outcomes was heavily attenuated by controls for early life factors, challenging the utility of delay of gratification as both a predictive screener and potential target for intervention. However, other longitudinal work has found important relations between early measures of self-control and a host of adult outcomes (Moffitt et al., 2011), raising the possibility that early gratification delay may also predict adult functioning. In an extension of Watts et al. (2018), we examined whether Marshmallow Test performance at age-4 predicts a host of adult outcomes at age-26.
We used data from the NICHD Study of Early Child Care and Youth Development (SECCYD; NICHD Early Child Care Research Network, 2002). 1,364 children (86% White) were recruited during infancy and followed into adulthood, with the most recent assessment occurring at age-26. At 54 months, children participated in a version of the Marshmallow Test. Children were left in a lab room with a treat and given a choice: they could either eat the treat now, or wait for an experimenter’s return to be given an additional reward. The amount of time waited (up to 7 minutes) was extracted as the delay of gratification measure. Data was also collected regarding children’s cognitive abilities, early home life, and demographic characteristics across early childhood.
At age-26, participants reported on their symptoms of depression, BMI, police contact, educational attainment, impulse control, annual earnings, substance use, risk taking behaviors, and debt. These outcomes were chosen as they are theoretically aligned with delay of gratification ability based on the available literature, and because they are important indicators of success in early adulthood. Only participants who completed the age-26 assessment and had a valid measure of delay of gratification (n=702) were included in analyses. All outcomes and analyses were preregistered before viewing results, though authors did have prior experience with the study data.
When controlling for children’s early cognitive abilities, home life, and demographic characteristics, we found no significant associations between Marshmallow Test performance and any of the preregistered adult outcomes. Given that gratification delay did not predict any adult outcome, these findings call into question the predictive validity of the test. Furthermore, this evidence challenges delay of gratification as a useful point for intervention. Limitations of the present study include the limited diversity of the sample.

Authors

• Jessica Sperber, Columbia University
  Presenting Author

• Deborah Lowe Vandell, University of California - Irvine

• Greg J Duncan, University of California, Irvine

• Tyler Watts, Teachers College, Columbia University