The influence of environmental reliability in the marshmallow task: An extension study.
Level 2 - randomized trial
Controlled experimental trial allocating children to reliable versus unreliable pre-test conditions
PubMed 32169745 · doi:10.1016/j.jecp.2020.104821
What was done
This study replicated and extended an experiment investigating whether environmental reliability influences children's performance on the Marshmallow Task. Sixty children aged 3 to 5 years were tested in a familiar school setting rather than a laboratory. Before the delay-of-gratification task, children were exposed to either a reliable experimenter (who delivered on promised rewards) or an unreliable experimenter (who did not deliver promised rewards). Data were evaluated using Bayesian analyses.
What was found
The authors reported an effect of experimenter reliability on waiting time, noting that children in the unreliable condition did not wait as long, though the effect size was smaller than in the original 2013 study. They also found a statistically significant gender-by-condition interaction effect. The abstract reports no specific numerical values, point estimates, or credible intervals.
Why it matters
It reinforces evidence that children's delay of gratification is not solely a measure of self-control capacity, but can reflect a rational response to the perceived reliability of adults and their immediate environment.
Limits
The sample size is relatively small (60 children across conditions). The abstract omits exact numerical results, Bayesian effect sizes, and details on the direction or nature of the gender interaction. It was limited to children aged 3 to 5 years in a school setting.