Accepted answer
Food noise is a lay term for something with a reasonable scientific analogue: incentive salience of food cues, sometimes framed as hedonic hunger or cue reactivity. It is separable from homeostatic hunger both experientially and anatomically, which is why your description of it being independent of fullness is not a quirk of introspection but an accurate report.
Is it measured?
Yes, imperfectly, and the semaglutide programme did measure it. The instrument used was the Control of Eating Questionnaire, which scores craving control, craving for sweet and savoury foods, positive mood and hunger as separate domains. In the trial analyses using it, semaglutide improved craving control and reduced craving intensity separately from its effect on hunger [1]. Other instruments in the wider literature include the Power of Food Scale and various food-craving questionnaires. None of them are called "food noise" and none capture the intrusive, involuntary-thought quality you describe particularly well, which is a genuine gap.
How a peptide that size reaches the brain
It largely does not cross the blood-brain barrier by diffusion. It does not need to. There are three routes and all three matter:
- Circumventricular organs, which lack a tight barrier by design because their job is to sample the blood. The area postrema and the subfornical organ are directly accessible to circulating peptide.
- Vagal afferents, whose terminals in the gut wall express the receptor and project to the nucleus tractus solitarius. This is an indirect, neural route rather than a chemical one.
- Limited direct access to circumscribed brain regions. Rodent work tracing labelled semaglutide found uptake in specific hypothalamic and brainstem regions rather than diffusely, with downstream activation propagating to areas the drug itself never reached [2].
That third point is the key architectural fact: the drug binds in a few accessible places, and the effect spreads through existing neural circuitry. So you should not expect the receptor map and the effect map to coincide.
The regions and what each one does
| Region | Receptor population | Direct access to circulating drug | Functional contribution | Subjective correlate |
|---|
| Hypothalamic arcuate nucleus | GLP-1R on POMC/CART neurons; AgRP/NPY neurons inhibited indirectly via local GABAergic signalling | Partial, via median eminence proximity | Primary hypothalamic mediator of the weight effect; this population was shown to be necessary for the effect of a GLP-1 agonist in rodents [3] | Reduced drive to eat; earlier and more durable satiation |
| Area postrema | Dense GLP-1R | Yes, circumventricular | Chemoreceptive trigger zone for emesis; also contributes to meal termination | Nausea, vomiting, food aversion |
| Nucleus tractus solitarius | GLP-1R plus the brain's own preproglucagon neurons, which are the endogenous central source of GLP-1 | Partial; also receives vagal input | Integrates gut signals with hypothalamic and forebrain output | Fullness, meal-size limitation |
| Paraventricular nucleus | GLP-1R | Indirect | Autonomic and neuroendocrine output; stress-eating interactions | Reduced stress-driven intake in some reports |
| Ventral tegmental area and nucleus accumbens | GLP-1R, at lower density | Largely indirect | Mesolimbic dopamine signalling; assigns incentive salience to cues | This is the food-noise circuit. Reduced wanting, reduced cue-triggered seeking |
| Amygdala and insula | GLP-1R and downstream activation | Indirect | Affective valuation of food; interoception | Food becomes less emotionally loaded |
| Subfornical organ | GLP-1R | Yes, circumventricular | Fluid balance and thirst regulation | Altered thirst reported by some |
| Vagal afferent terminals (peripheral) | GLP-1R on nodose ganglion neurons | Yes, peripheral | Gut-to-brain signalling; also mediates delayed gastric emptying | Fullness, early satiation, reflux |
Answering your third question directly: separate pathways
Yes, and this is the most practically important thing in the whole topic. Human imaging work with GLP-1 receptor activation showed changes in responses to food cues in reward-related regions including the insula, amygdala, orbitofrontal cortex and putamen, and those changes were separable from the gastrointestinal effects [4].
Three effects, three substrates:
- Reduced incentive salience (the food-noise effect): mesolimbic, arcuate-mediated downstream. Persistent, dose-dependent, and it does not require any GI symptom.
- Delayed gastric emptying and early satiation: peripheral vagal plus brainstem. Substantially tachyphylactic, meaning it diminishes over weeks to months of continued exposure.
- Nausea and aversion: area postrema. Also tachyphylactic, which is why side effects fade.
The fact that the second and third diminish over time while the first largely does not explains a great deal of what people report, including the question two answers below about whether returning food noise means tolerance. If someone's "quiet" was mostly aversion, it goes away on the normal side-effect timeline. If it was mostly reduced salience, it tends to persist.
On your eleven-day onset
That timing fits the pharmacokinetics well. With a roughly one-week half-life you approach steady state over four to five weeks, but you cross the threshold for a noticeable central effect much earlier, and eleven days is around two half-lives, i.e. about 75% of the way to your first steady state. The abruptness people report, the sense that it switched off rather than faded, is consistent with a threshold effect in a circuit rather than a linear dose-response in a symptom.
edited 16 Dec 2024 by lane_transit — removed a claim I could not source
The three-pathways framing is the most useful thing I have read on this, particularly the point that two of the three are tachyphylactic and one is not. – kirsi_lahtinen 3 months ago Worth stressing that the arcuate work was done in rodents with receptor knockdowns. The human evidence is imaging and inference. – tri_gly_ala 2 months ago 3The "kitchen became furniture" description in the question is a better phenomenological account than most of the instruments capture. – mz_4113 8 days ago add a comment