One of the two meals in this trial was a few slices of banana, a piece of cheese, some dried cranberries and 30 grams of hard-boiled egg. The other was deli turkey, instant mashed potatoes, a chocolate chip cookie, frosted wheat cereal, veggie chips, white bread and a bite of peanut butter and jelly. Both came to about 300 calories, both were assembled from a pool of 28 validated foods, and on paper they were almost impossible to tell apart. The bodies eating them responded differently anyway.
The work comes from Alexandra DiFeliceantonio’s lab at Virginia Tech’s Fralin Biomedical Research Institute, and it was published October 5 as a randomized crossover study in Nature Metabolism. The team matched the meals within roughly 1.6% on calories, carbohydrates, fat, protein, fiber, sodium, water and glycemic index. Zach Hutelin, the first author, put it this way: “If these meals had a nutritional label, they would be practically identical.”
Fifty-seven healthy-weight adults were enrolled, with a mean age of 26, and 32 of them ate both meals on separate days inside a whole-room metabolic chamber. Over the next three hours the researchers drew blood repeatedly and measured how much energy each person burned and from which fuel.
Blood sugar looked the same across the two meals, but insulin did not. Total glucose exposure over three hours showed no difference between the meals, and both peaked around 40 minutes. Insulin matched at the 5 and 20 minute marks, then ran higher after the ultra-processed meal at 40, 60, 90 and 120 minutes before settling back near baseline by the three-hour mark. The same meal also produced a larger rise in energy expenditure and more fat burning, while the abstract describes lower carbohydrate burning than after the minimally processed one.
A second group of 52 participants did a brain scan while bidding on pictures of foods. Their willingness to pay did not differ between the two categories, yet the size of the gap in carbohydrate burning tracked activity in regions tied to motivation and reward, including the caudate and ventral striatum. In other words, people did not consciously value the foods differently, but their brains appeared to respond to them differently depending on what the body had just done with the meal.
What none of this shows is that the extra insulin is harmful. Insulin release also responds to protein and amino acids, and the two meals drew their protein from different foods, which is one of the authors’ own listed limitations. The ultra-processed meal also carried more additives, so processing itself cannot be separated from ingredients here.
A few other boundaries matter when reading the result:
- Only one pair of 300-calorie meals was tested, which is much smaller than a typical dinner.
- Participants were young, healthy-weight adults, and only 31.6% were men.
- Just 29 of the 57 people had complete metabolic and brain data.
- Insulin sensitivity, overeating and long-term health outcomes were not measured.
The research was funded through a National Institutes of Health grant from the National Institute of Diabetes and Digestive and Kidney Diseases, plus a National Science Foundation graduate fellowship. DiFeliceantonio’s team says the next step is to isolate specific factors such as commercial additives or particular processing steps, which is the experiment that could say what is actually driving the difference.
The context is why this keeps getting studied. Ultra-processed foods, defined by the NOVA classification system as industrial formulations built from extracted ingredients and additives, supply more than half of the calories the average American eats. Earlier work on this site covered how heavier intake was tied to slower attention and focus, and this trial adds a mechanism-level hint from a different direction, since the body handled identical-looking nutrition in a measurably different way.
For anyone reading food labels to make sense of this, the practical point is modest. A matching nutrition panel does not guarantee a matching metabolic response, though one small trial of single meals cannot say what that means for diabetes, weight or heart disease over the years. Those answers will need longer studies in more varied groups, including people who already have insulin resistance.
