Someone opens the fridge at nine in the evening, not hungry in any way they would describe as hunger, and wants cheese. Not food. Cheese. The internet has an explanation ready: your body is telling you it needs calcium, or fat, or salt. Listen to it.
It is a nice theory. It has been tested, and it did not do well.
The short answer
A craving for a specific food is very unlikely to be a nutrient signal. The best experimental evidence points at sensory properties and learned association instead: what the food tastes and smells like, and how many times you have had it before. In a 1994 study in Physiology and Behavior, Michener and Rozin gave chocolate cravers cocoa capsules containing all of chocolate's pharmacological components and no taste at all. The effect on the craving was, in the authors' words, virtually none. White chocolate, which contains none of those compounds but does have the aroma, sweetness and texture, produced partial relief. The active ingredients were not the point. The experience was.
The experiment that tested the deficiency theory
That study is worth sitting with, because it is one of the rare cases where a folk explanation was put in a position to fail cleanly. Chocolate was chosen because it was the most commonly craved food in North America, and because it comes loaded with candidate compounds people had blamed for years. If the craving were pharmacological, capsules should have worked. If it were about calories, the caloric equivalent should have worked. Neither did. Adding the full set of cocoa compounds to white chocolate did not improve on white chocolate alone.
What that leaves is aroma, sweetness, fat and texture, arriving together in a form your brain has filed as worth wanting. The craving is a memory of an experience, playing back.
Why cheese specifically
Cheese lands in a combination the research treats as unusually rewarding. In a 2019 paper in Obesity, Fazzino and colleagues built a numeric definition of hyper-palatable food and found three recurring clusters. One of them is fat plus sodium: more than 25 percent of energy from fat, and at least 0.30 percent sodium by weight. Most hard cheeses sit inside that cluster comfortably.
The same paper applied the definition to the United States food system database and found that 62 percent of the foods in it, 4,795 out of 7,757, met the criteria for hyper-palatability. Including plenty of foods nobody thinks of that way. Vegetables cooked in cream or fat qualified. So did some items labelled reduced fat.
So the honest answer to "why cheese" is not that cheese is uniquely dangerous. It is that cheese is very good at being food, and you have learned that.
Bread, fruit and crunchy things follow the same rule
The same logic covers the other specific cravings people search for. Bread is the carbohydrate and sodium cluster. Fruit is sweetness with genuine novelty in texture, and craving fruit is often just a person eating well and wanting something bright. A craving for crunchy food is usually about texture and jaw feedback rather than the food itself, which is why celery sometimes works and sometimes does not.
None of these need a nutrient story to make sense. Once you stop looking for the deficiency, the pattern is easier to read: you crave what you have had, recently and often, in the states you tend to be in when you eat it.
Restricting a food makes you crave it more
This is the part that changes what you should do. In a 2012 study in Appetite, Massey and Hill followed 129 women for a week, distinguishing women dieting to lose weight from women simply watching their weight and from non-dieters. Across 393 recorded craving incidents, dieters had more cravings than non-dieters, and rated them stronger and harder to resist. Crucially, the foods they craved were the foods they were restricting.
Chocolate was the single most craved item at 37 percent of all incidents. About 70 percent of cravings led to eating, and that proportion did not differ between the groups. Everyone eats after a craving at roughly the same rate. Dieters just have more of them, and have a worse time.
Which means the standard advice, cut out the food you crave, is the one move most likely to make the craving louder. If cheese is on a forbidden list, cheese gets more interesting, not less.
What actually reduces specific cravings
Three things, none of them dramatic. Eat enough protein, because appetite that has been properly answered generates fewer of these episodes in the first place. Sleep, because short sleep reliably raises evening appetite. And stop putting the food in a special category, because the category is doing more work than the food.
The last one is the hardest and the most effective. A person who eats cheese sometimes, without ceremony, thinks about cheese less than a person who has banned it. We wrote about the wider version of that in what a healthy relationship with food actually looks like, and about the specific mechanisms in what causes sugar cravings. If you are still working out whether what you feel is a craving at all, food noise and hunger are separable, and the difference matters.
How OffRamp handles a craving
OffRamp has no forbidden list, and that is a deliberate product decision rather than a slogan. There is no food you log that comes back flagged, no red state for a bad day, no lecture. You log the cheese and it is a number, because the 2012 data says the moralising is the part that makes the next craving stronger.
What the app does have is a protein floor, calculated at 1.6 grams per kilogram of your goal weight and rounded to the nearest 5 grams, and a weight trend rather than a daily reading. Regain Radar looks at a 7-day rolling average against your baseline and stays quiet unless that average sits at or above baseline times 1.02 for seven consecutive days. One evening of cheese cannot move a 7-day average enough to matter. The app is built so it does not pretend otherwise.
