Why You Are Hungry an Hour After a Big Lunch
Two amnesic patients ate lunch three times in an hour, which is the clearest evidence anybody has that short-term hunger is a forecast rather than a reading
Key takeaways
- Paul Rozin and colleagues offered a second full lunch to two densely amnesic patients, ten to thirty minutes after they had finished the first. Both ate it, and usually started a third. Control participants refused. That was two patients, which makes it an existence proof rather than a measurement, and what it rules out is the idea that a meal ends because the body alone has registered enough food.
- The hunger hormone runs on a timetable. In a study of ten people on fixed meal schedules, ghrelin roughly doubled in the period just before each of their usual mealtimes and fell within an hour of eating, which makes it evidence of learning as much as of pure physiology.
- The blood-sugar story is wrong twice. Blood sugar actually falling below the normal range after a meal is uncommon in people without diabetes or prediabetes, and the large glucose-monitor study that does find a link points at a dip two to three hours after eating rather than at the earlier spike.
- That glucose finding is real, consistent across 1,070 people, and weak. It shifts the average of a large group and tells you very little about any one person's afternoon. It also comes from a company that sells personalised nutrition subscriptions built on it.
- Larger servings reliably produce larger meals, by a modest amount, across 61 studies and 6,711 participants. Smaller plates do not. The plate claim is popular, and the evidence under it is thin and partly traceable to a laboratory whose data cannot be trusted.
- Eating slowly does reduce how much is eaten at the meal, moderately, with the studies disagreeing a good deal. It has no measurable effect on hunger at the end of the meal or up to three and a half hours later, which is precisely the window this article is about.
Two patients. Three lunches, one after another, inside about an hour.
In a study published in 1998, Paul Rozin and his colleagues at the University of Pennsylvania gave a full lunch to two men with dense amnesia. Both had brain damage severe enough that almost nothing stayed in mind for longer than a minute. They ate. Between ten and thirty minutes after the plates were cleared, they were offered a second full lunch. They ate that too, and usually started on a third. The participants without amnesia refused.
Two patients is the whole sample, and small studies get treated as laws all the time, which this publication has written a whole article about. Two patients cannot tell you how common anything is, or how large; what they can do is rule something out. Both men had eaten, and their stomachs were as full as yours would have been. What was missing was the memory.
Gauge
Most of us carry the same picture of hunger, and it is a fuel gauge. Food goes in, the needle climbs, it falls across the afternoon, and when it drops far enough you go and find a biscuit. Over weeks and months that picture is roughly right. It explains almost nothing about three o'clock.
What the research supports is closer to a forecast than a reading. Your body predicts what it is about to need, using what it has learned from your days and your meals and what those meals have done for you before. A forecast built on an unfamiliar meal, eaten at an unusual hour in front of a screen, can be badly wrong, and it feels exactly as convincing as the accurate version. The easy way to overstate this is to say hunger has nothing to do with the body, and the founding study shows why that is wrong.
Clocks
In 1983 Harvey Weingarten trained rats to associate an arbitrary signal, a buzzer and a light, with food, then gave the signal to rats that had already eaten their fill. They sat down to a proper meal, roughly a fifth of what they normally ate in a day, because a learned cue said it was time.
The half nobody quotes is what happened afterwards. The animals compensated, eating less later on. Learning started the meal and the body squared the books over the day, which is the honest shape of the whole subject, and the shape of the argument about willpower too. Those were rats, so this is a mechanism rather than proof about people.
The human version turns up in the hormone everybody has heard of. In a study of ten people on fixed meal schedules, ghrelin, which the press calls the hunger hormone, roughly doubled just before each meal and fell within an hour of eating. The rise comes before the times those people habitually ate. The hormone keeps a diary, which is why the learned half and the bodily half cannot be cleanly separated.