Why are we so inconsistent? What the marshmallow test didn't tell you

We’re inconsistent because the brain was shaped to value what’s in our hands right now more than what might come later. Planning, waiting and picturing the future take effort, and the brain only signs up for that effort when it believes the reward will actually show up. And here’s the part few people talk about: to wait, we need to trust whoever made the promise, even when that person is us.

Almost everyone knows the cycle. The gym starts on Monday, the course on the first of the month, the diet after the holidays. The first few weeks go fine. Then comes a busy day, some tiredness, a good excuse, and before you notice, you’ve dropped it again.

What is the marshmallow test?

In the late 1960s and early 1970s, researchers at Stanford ran an experiment that became famous around the world.¹ A child sat in front of a treat, like a marshmallow, and got a simple offer: eat it now, or wait for the adult to come back and get two.

Some ate it in seconds. Others sang, covered their eyes, turned their chairs around, did anything to hold on. Years later, the experiment turned into a life lesson of sorts: those who can wait do better, and those who can’t have some kind of factory defect.

That version, though, left something out.

The part of the experiment nobody mentions

In 2013, a group of researchers decided to change one detail.² Before the marshmallow, each child got some art supplies and was told by the adult that, if they waited, they’d get better ones. For half the children, the adult came back with what was promised. For the other half, the adult came back empty-handed and just apologized.

Then came the treat test. The children who had seen the promise kept waited, on average, about four times longer than the others. The difference wasn’t in their willpower, it was in what they had just learned about that adult.

It makes perfect sense. If the adult has already let you down once, eating the marshmallow now is the smartest choice. One sure treat is worth more than two imaginary ones.

Some years later, another study repeated the research with many more children from more varied backgrounds, and found a much smaller effect than people used to claim.³ A good part of what looked like “willpower” had to do with the environment the child grew up in: whether promises around them were usually kept, whether there would be food tomorrow, whether waiting was worth it.

Why does the brain prefer right now?

For a very long time, surviving meant solving today. Food in front of you was food to eat now, because tomorrow was uncertain. The brain we inherited was shaped in that world, which is why an instant reward weighs so much. It’s concrete and already on the table. The future is an idea that has to be built inside your head, and that costs energy.

That’s why your phone beats the task and the couch beats the workout. We covered this when we talked about the quick reward trap. Nobody is lazy for choosing what the brain was wired to choose. But you can teach it something else.

Do you trust yourself?

Now go back to the experiment and swap the roles. In most of our goals, the adult who makes the promise and the child who has to wait are the same person.

Every time you say “I’ll start Monday” and don’t, a part of you takes note. Every plan dropped halfway is an adult coming back empty-handed. After a few rounds, your brain learns the same lesson those children learned: this person doesn’t deliver, so better grab the treat now.

Seen this way, inconsistency looks less like a personality flaw and more like distrust that has slowly piled up.

How do you become more consistent?

If the problem is trust, the way out is the same one that works with anyone: keep small promises, many times, until trust comes back. Promise yourself something smaller than you could handle, and do it. Ten minutes of reading instead of an hour. A short walk instead of the perfect workout plan. Every promise kept is an adult coming back with the supplies in hand.

The other part is training the brain to see the future. Research shows that when people vividly imagine a concrete scene of what lies ahead, they become more willing to wait for bigger rewards.⁴ The future stops being a blur and gets a face, a smell and a place. Letting go of today’s treat gets easier when you can actually see the second one.

Yes, it takes work. But it’s the kind of work that gets lighter, because trust grows along with it. As philosopher Will Durant put it, summing up Aristotle, we are what we repeatedly do. The big things in life tend to be built over many days of ordinary effort, far more than in one heroic day.

So here’s a question to take with you. If you were the child in front of the marshmallow, and the adult who promised the second treat were you, would you wait?

To keep building that trust, here are three more reads:

• If you tend to wait for the urge to show up before starting, here’s why motivation comes after action.
• Every postponed promise charges interest, and time always sends the bill for putting things off.
• When you drop a lot of things halfway, it’s easy to feel like everyone else has already made it, and we talk here about the feeling of being behind in life.

Editor’s note: Everyone has that friend who has been starting a diet every Monday since 2019 and knows every personal trainer in the neighborhood by name. Send them this article. Maybe they just need to start with one marshmallow at a time.

Footnotes

1. Mischel, W., Ebbesen, E. B. & Zeiss, A. R. (1972). Cognitive and attentional mechanisms in delay of gratification. Journal of Personality and Social Psychology, 21(2), 204-218. ↩ source

2. Kidd, C., Palmeri, H. & Aslin, R. N. (2013). Rational snacking: young children’s decision-making on the marshmallow task is moderated by beliefs about environmental reliability. Cognition, 126(1), 109-114. ↩ source

3. Watts, T. W., Duncan, G. J. & Quan, H. (2018). Revisiting the marshmallow test: a conceptual replication investigating links between early delay of gratification and later outcomes. Psychological Science, 29(7), 1159-1177. ↩ source

4. Peters, J. & Büchel, C. (2010). Episodic future thinking reduces reward delay discounting through an enhancement of prefrontal-mediotemporal interactions. Neuron, 66(1), 138-148. ↩ source