Psychologist logo
Two children from Cameroon
Children, young people and families, Developmental, Race, ethnicity and culture

Rural Cameroonian pre-schoolers just aced Mischel’s iconic Marshmallow Test

Comparing the performance of 125 four-year-olds from urban middle-class Germany with the performance of dozens of four-year-olds from the Nso farming families of rural Cameroon.

07 July 2017

By Christian Jarrett

Walter Mischel’s Marshmallow Test of self-control is one of psychology’s iconic experimental set-ups. First conducted in the 1960s, Mischel told the kids he tested that if they managed to resist eating the marshmallow in front of them until he returned (usually about 15 minutes later), they would be rewarded with a second marshmallow.

The children varied greatly in their powers of restraint and those who performed better displayed some cute distraction strategies, such as singing to themselves and covering their eyes. Perhaps most important, those kids who performed well at the test tended to do well in later life too, in terms of their health, education and career success. Given the huge impact this research has had, it’s amazing that it’s never been exported to a non-Western setting. Until now.

In a new paper in Child Development, Bettina Lamm and her colleagues have compared the performance of 125 4-year-olds from urban middle-class Germany with the performance of dozens of 4-year-olds from the Nso farming families of rural Cameroon. The Cameroonian kids aced the test, performing much better than their German peers. What’s more, their success seemed to be tied to the traditional, strict, hierarchical culture in which they’d been raised. The results challenge Western assumptions about what constitutes an ideal parenting style, and they provide another powerful demonstration of the urgent need for psychology to conduct more research outside of its usual Western focus.

The task faced by the German and Cameroonian children was very similar to the original version of the Marshmallow Test except the source of temptation for the German kids was either a lollipop or a chocolate bar (no one treat is equally liked in Germany so the researchers used whichever was the more desirable for each child), and for the Cameroonians it was a puff-puff, which is a kind of doughnut. On first seeing whichever treat was used, the children’s initial outward behavior in both cultures made it clear that they found it extremely tempting.

The children sat at a table on which either the lollipop, chocolate bar or puff-puff was placed within their reach. Before leaving the room, the researcher told them that they could eat the treat now, or if they waited until the researcher came back (which he or she did 10 minutes later), they would get the first treat plus a second.

Nearly 70 per cent of the Cameroonian children succeeded in waiting for the researcher to return without eating the first treat, compared with fewer than 30 per cent of the German children. Furthermore, the emotions the children displayed and the strategies they used differed depending on their cultural background. Whereas the German children showed more frustration and used a lot of distraction techniques, such as turning away, talking and singing, the Cameroonians showed little emotion or activity, in fact eight of them fell asleep! Also several of the German kids who couldn’t wait actually left the room to go and look for the researcher, but only one Cameroonian kid did this.

The researchers believe the contrast in behavior between the children in the two cultures has to do with the very different social milieu and parenting styles they have experienced. The Cameroonian kids are raised in multi-generational family groups with many siblings and they are taught to contain their emotions, obey their elders and to recognise the importance of group solidarity and social responsibility. In contrast, and as in most rich Western countries, the German kids are mostly raised in small family groups and they are taught the importance of self-expression, individuality and freedom of choice.

Lamm and her team think this manifests in the way the kids from the two cultures handle the Marshmallow Test: the German kids sought to impose their will on the situation, such as by districting themselves or leaving the room to go look for the researcher. By contrast, the Cameroonian kids adopted what the researchers call “holistic self-regulation”, that is they seemed to be better able to control their own emotions from within, perhaps through all the practice they’d had in their upbringing.

A lot of this is conjecture, of course. There are other ways to interpret the findings – for instance, we don’t know for sure that the Cameroonian kids really were as tempted by the treats as the Germans. Perhaps they found the whole set-up more intimidating and they were more subdued (though there was no evidence that they displayed more negative emotion). As the researchers admitted, “it remains unclear whether [the Cameroonian kids’ success] is the result of effective emotion regulation in this specific situation or whether emotional socialisation [their strict upbringing] produces a completely different experience of the situation.”

To test their theory that children’s self-control develops differently across the cultures due to different social values and parenting styles, Lamm and her colleagues performed a follow up study using data they’d collected from a sub-set of the kids in both cultures when they were just 9-month-old babies. This involved videoing the babies and their mothers in a free play session and coding the mothers’ behavior. The mothers also completed questionnaires about the social values they most cherished.

As expected, the researchers found that the Cameroonian mothers placed more value on conservative goals (loyalty, obedience, solidarity etc) and they were more directive in their play, creating a highly structured play situation and continually engaging their infants. The German mothers said they valued encouraging psychological autonomy (freedom of preference, individuality etc) and in their play they were less hands on, tending to let their babies take the lead and letting them have more unstructured breaks.

Crucially, these distinctive parenting patterns correlated with the children’s Marshmallow test performance at age 4. Across groups, kids with mothers who particularly emphasized autononomy and freedom and allowed more unstructured play particularly struggled to resist the treat – this was mostly the German kids. In contrast, kids with mothers who particularly emphasized traditional hiercharchial goals and structured play tended to excel at showing self-restraint – this was mostly the Cameroonians.

These are group-level correlations so causal effects of parenting and values have not been established (among other things, the genetic inheritance of self-control is an important confound). But one credible interpretation of the findings is that children’s self-control is shaped by the culture of their upbringing, and that there are different ways to find success in the Marshmallow test – the less effective “external” approach of the Western kids, which is based on exerting control over the environment, versus the more social, self-disciplined approach of the Cameroonian kids quite likely based on their having learnt greater control of their emotions in the service of obedience and group solidarity.

Intriguingly, this picture clashes somewhat with popular Western findings and theory around parenting, which stated crudely have suggested that authoritative parenting (in which children are given boundaries but are also supported and encouraged to pursue their own interests and desires) is the ideal way to foster a conscientious child whereas authoritarian parenting (strict and limiting parenting that emphasises obedience, arguably similar in some ways to the Cameroonian approach) should be avoided.

A fascinating possibility is the developmental consequences of different parenting styles is culture-dependent: for instance, perhaps a more authoritarian style is beneficial to children raised in rural Cameroon in large families with many generations under one roof and strict social hierarchies. If nothing else, this research is a reminder that we must study a broader range of humanity else we might find our most cherished theories are built on foundations of Western sand.