I Ate Ultra-Processed Food for 30 Days. The Results Shocked Me!

Van Tulleken describes his own experiment: 'Around week three I had a chat with a colleague in Brazil... and she kept saying a thing — it's not food, Chris. It's an industrially processed edible substance. I sat down that evening to eat my favorite fried chicken brand. I got a little bucket of it and I was so looking forward to it and I couldn't finish it. She had made it disgusting.' A nutritionist/psychologist patient of a colleague achieved the same result by sitting in Burger King and taking 20 chews per bite: 'He came back and he was like, this was gross. I couldn't eat it after I did that.' Hyman confirms his own perceptual shift: 'It doesn't look like food to me. It looks like a rock or a piece of wood.' Van Tulleken also recommends the Pringles experiment: crush them into powder in a bowl and eat with a spoon — discovering whether you love the chip or love the hyperbolic paraboloid shape plus tube branding.

Van Tulleken: 'Understanding the tobacco industry massively helps smokers and understanding the food industry really helps people who are addicted to UPF.' He draws a direct parallel between the tobacco cessation literature — where understanding that cigarettes were engineered to trap you reduces blame and increases agency — and UPF. Key industry facts he documents: Unilever's head of neuroscience openly discusses putting people in brain scanners to watch them eat ice cream; companies internally track 'stomach share' (how much and how fast people eat); the bliss point was industrialized by Howard Moskowitz in the 1970s designing tomato sauce; Philip Morris/RJ Reynolds bought General Foods and Crisco in the mid-1980s.

Van Tulleken's description of his experimental state: 'I was never full. I was never hungry. I was just sort of eating all day because that's what the food allows you to do.' The food is designed to be eaten continuously — it does not produce the satiety signal that stops eating of minimally processed food. Removing the home supply breaks the automatic grazing default.

The key insight: the satiety blunting is not specific to processed food — it affects the hormonal response to any food. A standardized meal that would have produced normal satiety signals before the experiment produced a blunted response after 4 weeks of UPF. Someone switching from UPF to whole foods may not feel full from whole foods initially — not because whole foods are inadequate, but because the hormonal system has been recalibrated.

Van Tulleken criticizes the US FDA's proposed front-of-pack labeling: 'The low, you get low for added sugars if it's 5% sugar. Now, WHO says you shouldn't eat more than 5% of your calories from free sugar at all. You'll get health benefits all the way down to 5% if you can get it down. So saying that 5% is low when that's at the upper edge of what WHO recommend for a daily intake feels lax.' The South/Central American black-octagon system — a stop-sign warning that appears once any single threshold is exceeded — is his preferred model.

Van Tulleken: 'The food is so salty that you eat it in the evening and then you have to drink all this fluid. So then if you're a man in your 40s, you're kind of up all night peeing as well and then you're exhausted and you get constipated. Your life starts and you're angry and tired and stressed and so you get up in the middle of the night and then you're eating more because you're tired and your cortisol level goes up.' The UPF-sodium-nocturia-cortisol-overeating loop is self-reinforcing: poor sleep from excess sodium drives more UPF consumption because fatigue and high cortisol both increase appetite.

The specific fMRI pattern was connectivity between the cerebellum — which Van Tulleken describes as the 'automatic behavior bits at the back of the brain' — and the reward and addiction circuitry in the middle of the brain. The pattern is consistent with how addiction researchers characterize behavioral addictions: the goal-directed prefrontal system loses influence while the habit and reward systems become hyperconnected. Van Tulleken was in his mid-forties when he did the experiment, and he notes that this raises grave questions about children eating similar diets over years. The full clinical trial will test whether these changes replicate across 30+ participants and whether they are a function of UPF per se versus its nutrient composition.

Van Tulleken was never forced-fed and was not intentionally overeating — he simply ate to appetite with the rule that 80% of calories came from UPF. The result was constant eating throughout the day without ever feeling genuinely full or genuinely hungry: 'I was never full. I was never hungry. I was just sort of eating all day because that's what the food allows you to do.' The hormonal measurement was designed precisely to bypass subjective reporting — because UPF eating patterns can confound subjective experience. The finding that the same standardized meal elicits a blunted fullness response after weeks of UPF exposure suggests a lasting hormonal recalibration, not merely a property of the food in the moment.

Van Tulleken traces the 'corporate playbook' — funding academic research, seeding conflicts of interest in regulatory committees, creating patient advocacy front groups, funding nutrition charities (the British Nutrition Foundation's 'Healthy Eating Week' was sponsored by Coca-Cola), and intimidating critics. After his book was published, McDonald's sent him an offer to become their ambassador. He describes 65% of members of the UK Government's Scientific Advisory Committee on Nutrition as having financial relationships with food companies including Coke and Nestlé. He also describes Unilever's head of neuroscience openly discussing putting people in brain scanners to watch ice cream 'light up the orbital frontal cortex like a furnace' for product development.

Van Tulleken summarizes: 'The ultraprocessing — which includes the marketing, the coloring, the texture effects, the physical processing, thermal, chemical processing — that is the stuff that enables and allows you to eat so much of the sugar, salt, and fat, because no one eats sugar, salt, and fat from the bowl on the table.' He also cites the food matrix distinction: 1,000 calories of almonds versus 1,000 calories of Coke interact with your microbiome, hormones, immune system, and neurotransmitters completely differently. 'A calorie burned is a calorie burn, but a calorie eaten is not a calorie eaten' — quoting his colleague Robert Lustig.

The mechanism he proposes: UPF is not primarily a chemical addiction — it is a sensory and mythological addiction enabled by the processing (the Pringles hyperbolic paraboloid shape is part of the product), the packaging (tube branding), and the environment (eating quickly, in a car, from packaging). Slow down the intake and remove the environmental context and the food fails to hold up to scrutiny. Hyman confirms the same experience: processed food 'doesn't look like food to me anymore — it looks like a rock or a piece of wood.' Van Tulleken's within-48-hours recovery feeling is the opposite of withdrawal — the food was making everything worse (insomnia, constipation, anger, fatigue from high sodium) and removing it produced rapid improvement.

During the UPF diet, the excess sodium led to high fluid intake, nocturia, exhaustion, elevated cortisol from sleep disruption, and constipation — a cascade that in turn drove more eating (tired, cortisol-elevated individuals eat more). All of these resolved rapidly. But the fat mass deposited required fasting and 2 years to resolve. Van Tulleken notes the profound injustice: he has every possible advantage (physician, knowledge, resources, social support) and it was 'really, really, really hard' — pointing to how impossible it is for people in food-insecure or time-pressured environments.