Forget You Ate Lunch and You'll Eat Another One: Your Memory Center Helps Run Your Appetite

I have never once thought of lunch as something I’d need to remember. You eat it, the plate goes in the sink, and your stomach handles the rest. Then I read about two patients who finished a full lunch, watched it get cleared away, and 10 to 30 minutes later sat down and ate a second one.

Paul Rozin and colleagues ran that experiment in 1998 with two people whose amnesia was so severe they could barely recall anything from more than a minute earlier, often including the meal they’d just eaten. Both accepted the second lunch and ate it, on three separate occasions each, and when a third lunch arrived they usually started on that one too. Their stomachs worked fine and they’d just taken in a full meal. What they didn’t have was any record of having eaten it, and without that record the meal didn’t seem to count.

I went digging for that study after Scott Kanoski, a professor of human and evolutionary biology and neurobiology at USC, sat down with Andrew Huberman this week to talk about how food shapes memory and cravings. Kanoski has been making the same argument for a while; he told USC’s news office last year that “remembering what and when you ate is just as crucial for healthy eating as the food choices themselves.” I’ll admit I came in braced for a mindful-eating app pitch. What I found was a run of experiments tracing part of the circuit that keeps the meal log, plus evidence that the food itself can blunt it.

The rats that looked around between bites

In a 2025 paper in Nature Communications, Léa Décarie-Spain, Kanoski and colleagues watched neurons in the ventral hippocampus of male rats while they ate. The hippocampus is the brain’s classic memory structure, the place that maps where you parked and what happened at last week’s dinner. Activity in this set of neurons climbed between eating bouts rather than during them, in the little pauses when the rats stopped chewing and surveyed the room, and the stronger that between-bite activity, the better the rats later remembered where they’d eaten.

Wait, why would a memory structure care about the pause and not the bite? The researchers’ read is that the pauses are when the brain files the meal, building what Kanoski calls a meal engram. “An engram is the physical trace that a memory leaves behind in the brain,” he told USC, and he describes meal engrams as “sophisticated biological databases” holding what you ate, plus where and when.

When the team destroyed those meal-responsive neurons, the rats lost track of where they’d eaten. They still wanted food as much as before, and they still remembered escape locations in a separate spatial task, so the damage was narrower than a general memory wipe. The neurons send their output to the lateral hypothalamus, one of the brain’s main hunger regions, and they’re loaded with serotonin 2A receptors. Silencing that hippocampus-to-hypothalamus line, or blocking those receptors, knocked down meal-location memory and pushed calorie intake up. The rats didn’t get there by eating bigger meals. They came back to the food sooner, with shorter gaps between meals, which is about as close as a rat gets to saying yes to the second lunch.

Marise Parent’s group reached the same place by a different road. In a 2019 eNeuro study, they used light to switch off hippocampal neurons for 10 minutes right after a rat finished eating, the window when a memory of that meal would be settling in. By the next meal the neurons were working normally again, and the rats still started that meal sooner and ate more of it.

The hippocampus doesn’t keep this log on its own, either. In a 2018 study, Kanoski’s group cut most of the sensory signal running from rats’ guts up the vagus nerve, and the animals got worse at hippocampus-dependent memory tasks while their hippocampi lost BDNF, a growth factor that keeps neurons healthy. Their body weights didn’t budge. The team traced the route from gut to brainstem to medial septum to hippocampus. So the gut helps keep the meal-recording system tuned, and that system helps decide when the gut gets fed again.

One of the gut’s own messengers lands there too. GLP-1 is made in the far end of the small intestine and in the brainstem, and in a review, Kanoski and Harvey Grill describe rat experiments where switching on GLP-1 receptors directly in the ventral hippocampus cut 24-hour chow intake by about 40 percent and Western-diet intake by about 50 percent, mostly by shrinking meals. Blocking those receptors made rats eat more, which suggests the body’s own GLP-1 does this job every day. Nobody has shown that the weekly injections people now pay for work through this hippocampal route. Still, I can’t stop thinking about the fact that the molecule those drugs copy is one your gut already sends, for free, to the part of your brain that keeps the meal log.

A week of waffles

The loop gets less friendly from here. Kanoski and Terry Davidson laid out the uncomfortable version back in 2010: saturated fat and refined sugar, the backbone of a Western diet, impair hippocampus-dependent learning and memory. A weaker hippocampus makes food cues harder to resist, which means more of that diet, which weakens the hippocampus further. They called it a vicious cycle.

His lab’s sugar work pulled the gut into it. In a 2021 Translational Psychiatry study led by Emily Noble, adolescent male rats got free access to an 11 percent sugar solution, 65 percent fructose and 35 percent glucose, built to mimic soda. They didn’t gain weight or take in more total calories, because they simply ate less chow to make room. As adults, though, they did worse on a hippocampus-dependent memory test while a memory test run by a different brain region came out normal. Their guts carried more Parabacteroides bacteria, and the more a rat carried, the worse it remembered.

Then the team wiped out a separate group of rats’ gut bacteria with antibiotics and dosed them with two Parabacteroides species, and those rats developed memory problems without drinking any sugar at all. Kanoski told USC Dornsife the team was surprised they could reproduce the impairment “simply by enriching a single bacterial population in the gut.” Keep the scale in mind before you empty the fridge, though: the dosed rats ended up with Parabacteroides making up about 56 percent of their sampled gut bacteria, far beyond what sugar produced, in groups of seven or eight animals. The bacteria can do this. Whether soda does it to a kid is a different and still open question.

Rats supply the mechanism. What I wanted to know was whether any of it shows up in people, and fast. Richard Stevenson’s team at Macquarie University ran a randomized trial in 110 lean, healthy adults aged 17 to 35. Half kept eating normally, and the other half spent a week eating the way plenty of us do on a bad week: two Belgian waffles on four days, a fast-food meal on two, and a lab breakfast of a toasted sandwich and a milkshake at either end.

After eight days, the waffle group recalled 92 percent of what they’d learned on a delayed word-list test that leans on the hippocampus, against 97 percent for controls. That’s a modest gap in young, lean volunteers after seven days, and a memory test that doesn’t depend on the hippocampus showed no difference at all.

The appetite test was cleverer. Volunteers rated how much they liked and how much they wanted foods like Froot Loops and toast with Nutella, before and after a filling breakfast. Normally, once you’re full, wanting falls a lot further than liking: the Nutella still tastes great, you just don’t need it. After the week of waffles, that drop shrank, and within the waffle group the people whose appetite control slipped most tended to be the ones whose recall slipped most. Three weeks later the two groups no longer differed on the memory test, though the authors couldn’t say whether the waffle eaters had recovered or the controls had drifted down.

Eating with your eyes on the plate

If meals get filed in the pauses, what happens when you spend the pauses scrolling? Kanoski’s rats never saw a phone. The screen worry is his extrapolation, about attention pulled away during those encoding moments. People have been tested on the broader question, though. A 2013 meta-analysis pooled 24 experiments and found that eating while distracted raised intake at that meal with a standardized effect size of 0.39 and raised how much people ate later with an effect of 0.76, roughly double. Making people’s memory of what they’d just eaten more vivid cut how much they ate afterward.

So distraction does more damage to the next meal than to the one in front of you, which is exactly what you’d predict if the meal never got properly filed. Nobody is going to market that finding, because there’s nothing to patent in eating at a table. I’m doing it anyway: lunch at the table, phone in another room, and I’m leaving the pauses between bites empty so my hippocampus has time to write the meal down.

Sources

1. Nature Communications: Décarie-Spain, Kanoski et al., “Ventral hippocampus neurons encode meal-related memory” (2025)
2. Royal Society Open Science: Stevenson et al., one-week Western-style diet RCT on hippocampal memory and appetite control (2020)
3. Translational Psychiatry: Noble et al., gut microbial taxa elevated by dietary sugar disrupt memory function (2021)
4. Psychological Science: Rozin et al., multiple-meal eating in amnesic patients (1998)
5. American Journal of Clinical Nutrition: Robinson et al., eating attentively systematic review and meta-analysis (2013)
6. eNeuro: Hannapel, Parent et al., postmeal hippocampal inhibition increases future intake (2019)
7. Nature Communications: Suarez, Kanoski et al., gut vagal sensory signaling and hippocampal memory (2018)
8. Biological Psychiatry: Kanoski and Grill, hippocampal contributions to food intake control (review)
9. Physiology & Behavior: Kanoski and Davidson, Western diet, hippocampal dysfunction and obesity (2010)
10. USC Today: “Brain cell discovery may explain excessive hunger” (2025)
11. USC Dornsife: sugary diet early in life and later memory trouble (2021)
12. Huberman Lab: “How Food Choices Shape Memory, Cognition & Cravings” with Dr. Scott Kanoski (2026)