For decades, a simple experiment with sweets was held up as proof that self-control in early childhood predicts success in later life. Freelance journalist Tom Reeve spoke to Dr Anna Kowalski, a developmental psychologist based in Bristol, about what the famous 'marshmallow test' really tells us.
Tom: Let's start with the basics. What actually happened in the original studies? Anna: In the late 1960s and early 1970s, the psychologist Walter Mischel and his colleagues ran a series of experiments at a nursery on the campus of Stanford University in California. A child of about four or five was offered a choice: one treat straight away, or two treats if they could wait while the researcher left the room. The treat wasn't always a marshmallow, incidentally – sometimes it was a biscuit or a pretzel. The child was then left alone with the single treat and a bell they could ring to call the researcher back. Tom: And the part that made headlines came later? Anna: Exactly. Years afterwards, Mischel's team followed up some of those children and found that the ones who had waited longer tended, on average, to have done better at school and to be described by their parents as better at coping with stress. It was a striking result, and it was soon seized upon by writers of self-help books as evidence that delayed gratification is the key to success.
Tom: So what's the problem? Anna: Several, really. The original sample was small, and the children came mostly from families connected to a prestigious university, so they were hardly typical of the wider population. More importantly, when researchers repeated the study with a much larger and more diverse group of children, the link between waiting and later achievement turned out to be considerably weaker. Once they took into account factors such as family income and the child's early language and thinking skills, much of the effect disappeared. Tom: Meaning that waiting for the second treat might be a symptom of a child's circumstances rather than a cause of later success? Anna: That's one plausible interpretation. A child from a comfortable home may simply be more used to the idea that promises are kept and that food will still be there later.
Tom: There's a well-known experiment on exactly that point, isn't there? Anna: Yes, and I find it rather lovely. Researchers first had an adult make a small promise to each child – to bring better art materials, for instance. For half the children, the adult kept the promise; for the other half, the adult came back empty-handed. Afterwards, when the children were given the marshmallow test, those who had experienced the reliable adult waited far longer. In other words, not eating the marshmallow wasn't purely a matter of willpower. It was also a rational judgement about whether the world could be trusted. Tom: Does that mean the original work was worthless? Anna: Not at all, and I think that would be an overreaction. Mischel's real interest was in how children managed to wait, and that part of his research has stood the test of time. The children who succeeded rarely just stared at the treat and gritted their teeth. They covered their eyes, sang songs, invented games or imagined the marshmallow was a fluffy cloud. Self-control, in that sense, is less a fixed trait than a set of strategies – and strategies can be learned.
Tom: What should parents take from all this? Anna: Mainly, I'd say, not to panic if their four-year-old eats the marshmallow in ten seconds! A single test on a single afternoon cannot seal anyone's fate. What children do need is a predictable environment in which waiting actually pays off, plus plenty of chances to practise little techniques for managing their impulses. And for researchers, the story is a useful reminder that a famous finding is only as strong as its replications.