Study Finds Food Processing Itself, Not Just Nutrients, Alters Insulin and Brain Response

Two meals built to be nutritionally identical — same calories, same carbohydrates, fat, protein, sodium and water, down to within one percent — nonetheless sent healthy volunteers' bodies down different metabolic paths, according to a randomized, controlled crossover trial published Monday in Nature Metabolism. The only thing that differed between the two meals was how heavily processed the food was. That difference alone was enough to produce a sharper insulin response, altered energy expenditure and distinct activity in brain regions tied to motivation and reward.

The study, led by researchers at Virginia Tech's Fralin Biomedical Research Institute at VTC in Roanoke, set out to test a basic assumption in nutrition science: that a calorie, a gram of fat or a gram of protein behaves the same way in the body regardless of how the food containing it was made. The finding that it may not, researchers say, offers one of the clearer experimental hints yet at a biological mechanism connecting ultra-processed foods to the chronic disease rates associated with them in decades of population data.

The Numbers

Fifty-seven healthy adults between 18 and 45 enrolled in the trial; 32 completed both a metabolic-chamber session and a brain-imaging session. Each participant ate two different 300-calorie meals on separate days, at least three days apart, after an overnight fast. One meal was built from ultra-processed ingredients — a bite of peanut butter and jelly sandwich, deli turkey, veggie chips, a cookie, cereal and instant mashed potatoes. The other was minimally processed: sliced banana, dried cranberries, cheese and egg. Researchers used the Nova scale, a four-tier system catalogued by the National Cancer Institute that ranks foods by degree of industrial processing rather than by nutrient content, to select the two meal types.

Despite the meals' near-identical nutritional profiles, blood draws taken immediately after eating and six more times over the following three hours showed that the ultra-processed meal produced a significantly higher insulin response, with blood sugar remaining elevated somewhat longer even though total glucose exposure over the three hours was similar between the two conditions. Participants also burned relatively more fat and less carbohydrate for fuel after the ultra-processed meal, and took about 90 seconds longer, on average, to finish eating it. In a separate brain-imaging session, participants viewed photographs of the same matched foods while undergoing functional MRI; differences in how individuals metabolized carbohydrates after each meal type corresponded with differences in activity in the ventral striatum and nucleus accumbens — regions involved in reward and motivation — when they looked at pictures of food.

How Researchers Got Here

The study did not emerge from nowhere. Ultra-processed foods — products reformulated with industrial ingredients and additives rather than prepared from whole foods — now account for more than half of daily caloric intake in the United States. A federal data brief from the CDC's National Center for Health Statistics put the figure at 55 percent of calories for people age 1 and older during 2021–2023, with youth ages 1 to 18 drawing nearly 62 percent of their calories from such foods compared with 53 percent for adults. Epidemiological studies have repeatedly linked heavy consumption of these foods to obesity, cardiovascular disease, type 2 diabetes and, in some research, measures of mental health — associations the Virginia Tech team was not trying to relitigate. Their question was narrower: is something about processing itself driving those outcomes, independent of the fat, sugar, sodium and fiber content that ultra-processed foods typically carry in different proportions than whole foods?

"We know that the majority of ultraprocessed foods are high in fat, they're high in sugar, they're low in fiber, and they're low in protein," said Alex DiFeliceantonio, a Fralin Biomedical Research Institute associate professor whose lab led the study, in the university's announcement. "But if we artificially hold all of those things constant, is there something about the processing that leads to a different outcome?" Zach Hutelin, the study's first author and a doctoral researcher at the institute, put the design in blunter terms: "The two meals are matched within 1 percent of carbs, fat, proteins, calories, water, and salt. If these meals had a nutritional label, they would be practically identical."

Because ultra-processed foods make up roughly half the American diet, the population with some stake in the question is, functionally, most of the country. The study's direct subjects, though, were a narrow slice of it: young, healthy-weight adults eating small, single meals under tightly controlled laboratory conditions, not the general public eating ordinary-sized meals over months or years. The researchers were explicit that their results describe a short-term physiological signal, not a demonstrated pathway to obesity, diabetes or heart disease. "We were shocked when every single metabolic metric differed," Hutelin said, describing the insulin finding as the study's most unexpected result, but he and DiFeliceantonio both cautioned that chronically elevated insulin and blood sugar are mechanisms associated with later disease risk, not proof that a single processed meal causes it.

Reaction

Outside researchers who study food processing were cautiously positive about the study's design, which several noted is unusually rigorous for isolating processing from nutrient content — a persistent confound in ultra-processed food research. Carlos A. Monteiro, the University of São Paulo epidemiologist who developed the Nova classification system the study relied on, reviewed the findings for Nature Metabolism.

"A key debate in the literature is whether [ultraprocessed food] associations with poor health outcomes are driven by nutrient composition rather than processing," Monteiro told the journal. "This is a novel and important contribution, as the brain ultimately regulates eating behavior."

Coverage of the study by Newsweek and Neuroscience News emphasized the same point researchers made in their own release: the result does not overturn existing nutrition guidance so much as identify a specific, testable mechanism — insulin response tied to processing — that had previously been largely theoretical. News-Medical's summary of the paper noted that the brain-imaging results, while statistically linked to the metabolic differences, did not show participants were willing to pay more for the ultra-processed foods when asked to bid on them — complicating any simple story about processed food being more "rewarding" in a straightforward behavioral sense.

What Happens Next

The authors list several limitations they intend to address in follow-up work. The two test meals, while matched on calories, macronutrients, sodium and water, used different protein sources and the ultra-processed meal contained more additives — variables the team did not isolate individually. The study measured total dietary fiber but not the physical or chemical structure of the food matrices, a property some nutrition researchers believe affects digestion speed independent of processing category. And because the sample consisted of young, healthy-weight volunteers eating single 300-calorie meals, the findings may not generalize to older adults, people with obesity or diabetes, or to the larger, everyday-sized meals most people actually eat.

DiFeliceantonio said her lab next wants to test specific processing variables — individual additives or processing steps — in isolation, along with larger meals, longer observation windows and a broader range of participants, to determine whether the insulin and brain-activity differences observed here persist outside a single controlled sitting. The research was funded by a National Institutes of Health grant through the National Institute of Diabetes and Digestive and Kidney Diseases and a National Science Foundation graduate fellowship. For now, the authors describe their result as a candidate mechanism rather than a verdict — evidence that processing can matter biologically, even when a nutrition label says two meals should be the same.