Researchers at Virginia Tech’s Fralin Biomedical Research Institute investigated whether the level of food processing influences metabolism and brain activity independently of basic nutrients. They recruited healthy adults aged 18 to 45 and prepared two meals that were matched within 1 percent for carbohydrates, fat, protein, water, calories, salt, and weight but differed in processing level according to the NOVA classification system. One meal consisted of minimally processed items such as sliced banana, dried cranberries, cheese, and egg; the other was highly processed and included a peanut butter and jelly sandwich, deli turkey, veggie chips, a cookie, cereal, and instant mashed potatoes.
This image illustrates the meal comparison central to the experiment, where nutrient content was matched but processing level differed.
After participants ate each meal in separate sessions, metabolic testing showed that the ultraprocessed meal prompted a significantly higher insulin response and caused blood sugar to remain elevated longer compared to the minimally processed meal. Additionally, energy expenditure was greater and carbohydrate oxidation was lower following the ultraprocessed meal, indicating that the body burned less carbohydrate for fuel even though the meals provided the same amount of carbohydrates.
Functional MRI scans conducted while participants viewed pictures of the foods revealed distinct brain responses in the ventral striatum and nucleus accumbens, regions associated with motivation, learning, and reward. The differences in brain activity were correlated with how participants burned carbohydrates for energy after each meal, suggesting that metabolic differences produced by food processing are linked to the brain’s reward response to food cues, even though participants did not report a willingness to pay more for either meal type.
This photo depicts the imaging setup used to measure brain responses to food cues, linking neural activity to metabolic differences.
The study acknowledges several limitations: only 57 adults were recruited and 32 completed both the metabolic and fMRI sessions, providing a relatively small sample size; the research examined just one set of 300‑calorie meals, and while protein content was matched, the protein sources differed between meals, which could have influenced outcomes; and the study did not measure specific additives or the physical and chemical structure of the foods. The researchers note that future work should explore different populations, longer durations, larger meals, and various food pairings to isolate the effects of particular processing steps or commercial additives.