Twenty minutes after you smoke, buttered toast stops being toast. The crisp edge gives way under your teeth, the butter has gone down into the bread instead of sitting on top of it, and you find yourself paying real attention to something you’ve eaten several thousand times without once thinking about it.
The chemistry behind that part isn’t mysterious. THC switches on CB1 receptors, which are part of how the body manages appetite.
What happens between the receptor and the toast is where it gets complicated, and the better studies have spent less time on the chemistry than on the eating itself: what people reach for, how much they put away, and whether any of it actually tastes different.
Seventeen adults took either oral THC or a placebo for a 2019 paper in the American Journal of Clinical Nutrition, then rated photographs of high-calorie food. The THC group wanted it more and liked it more. Then everyone was given a chocolate milkshake, and “participants consumed significantly more milkshake after THC than after placebo.”
When the researchers delivered the same milkshake directly into people’s stomachs through a tube, the effect disappeared.
All of that points at the mouth. Flavor gets assembled from several things at once: chewing pushes aromas up the back of your throat into your nose, your tongue reports sweet and salty and sour, your nerves supply temperature and texture, and the whole thing arrives as one impression you call taste, according to the National Institute on Deafness and Other Communication Disorders. Cold cheddar out of the fridge and the same cheddar pulled apart in a grilled sandwich are chemically almost identical.
Cannabinoids do reach into that system. CB1 receptors “promote food intake in fasted mice by increasing odor detection,” according to a 2014 study in Nature Neuroscience that connected cannabinoid activity in the olfactory circuits to how much the animals could smell and then how much they ate.
In humans it went the other direction. Fifteen people given 20 milligrams of oral THC in a placebo-controlled trial got worse at smelling, including at telling two odors apart.
So the bloodhound-nose claim is running well out ahead of the evidence.
The largest recent attempt at this is a 2025 PNAS study that sat 82 adults down with snacks after they’d vaporized either cannabis or a placebo. The cannabis group ate more over the first thirty minutes. They did not report that anything tasted better, and the authors were direct about it: “Vaporized cannabis did not influence taste ratings.”
Rats in the same paper worked harder to get food and kept finding it rewarding well after they’d eaten their fill, which the authors flagged as needing confirmation in people. It’s a plausible account of a snack turning urgent without one thing on the plate tasting stronger.
Eating more, wanting more, and tasting more are three separate measurements, and these studies keep finding one without the others. Doses are all over the place, some of this is swallowed and some is inhaled, and lining the results up next to each other is generous.
What holds up across all of it is that the eating is doing the work, which is an argument for taking it a little more seriously than most of us do at eleven at night. Pick the thing you actually want instead of the thing within reach. Use a plate. Put your phone in another room. Pay attention to the butter, or to a pickle cutting through something rich.
You don’t have to earn it and it doesn’t even have to mean anything! The science can catch up on its own time.
