From big bowls of salad to platters of meatballs and spaghetti sauce, Srisatta noshed his way through a nutrition study aimed at understanding the health effects of ultraprocessed foods, the controversial fare that now accounts for more than 70% of the U.S. food supply. He allowed The Associated Press to tag along for a day.

“Today my lunch was chicken nuggets, some chips, some ketchup,” said Srisatta, one of three dozen participants paid $5,000 each to devote 28 days of their lives to science. “It was pretty fulfilling.”

Examining exactly what made those nuggets so satisfying is the goal of the widely anticipated research led by National Institutes of Health nutrition researcher Kevin Hall.

“What we hope to do is figure out what those mechanisms are so that we can better understand that process,” Hall said.

Hall’s study relies on 24/7 measurements of patients, rather than self-reported data, to investigate whether ultraprocessed foods cause people to eat more calories and gain weight, potentially leading to obesity and other well-documented health problems. And, if they do, how?

At a time when Health Secretary Robert F. Kennedy Jr. has made nutrition and chronic disease a key priority, the answers can’t come soon enough.

Kennedy has repeatedly targeted processed foods as the primary culprit behind a range of diseases that afflict Americans, particularly children. He vowed in a Senate confirmation hearing to focus on removing such foods from school lunches for kids because they’re “making them sick.”

Ultraprocessed foods have exploded in the U.S. and elsewhere in recent decades, just as rates of obesity and other diet-related diseases also rise.

The foods, which are often high in fat, sodium and sugar, are typically cheap, mass-produced and contain added colors and chemicals not found in a home kitchen. Think sugary cereals and potato chips, frozen pizzas, sodas and ice cream.

Studies have linked ultraprocessed foods to negative health effects, but whether it’s the actual processing of the foods — rather than the nutrients they contain or something else — remains uncertain.

A small 2019 analysis by Hall and his colleagues found that ultraprocessed foods led participants to eat about 500 calories a day more than when they ate a matched diet of unprocessed foods.

The new study aims to replicate and expand that research — and to test new theories about the effects of ultraprocessed foods. One is that some of the foods contain irresistible combinations of ingredients — fat, sugar, sodium and carbohydrates — that trigger people to eat more. The other is that the foods contain more calories per bite, making it possible to consume more without realizing it.

Teasing out those answers requires the willingness of volunteers like Srisatta and the know-how of health and diet experts who identify, gather and analyze the data behind the estimated multimillion-dollar study.

During his month at NIH, Srisatta sported monitors on his wrist, ankle and waist to track his every movement, and regularly gave up to 14 vials of blood. Once a week, he spent 24 hours inside a metabolic chamber, a tiny room outfitted with sensors to measure how his body was using food, water and air. He was allowed to go outside, but only with supervision to prevent any wayward snacks.

“It doesn’t really feel that bad,” Srisatta said.

He could eat as much or as little as he liked. The meals wheeled to his room three times a day were crafted to meet the precise requirements of the study, said Sara Turner, the NIH dietitian who designed the food plan. In the basement of the NIH building, a team carefully measured, weighed, sliced and cooked foods before sending them to Srisatta and other participants.

“The challenge is getting all the nutrients to work, but it still needs to be appetizing and look good,” Turner said.

Results from the trial are expected later this year, but preliminary results are intriguing. At a scientific conference in November, Hall reported that the first 18 trial participants ate about 1,000 calories a day more of an ultraprocessed diet that was particularly hyperpalatable and energy dense than those who ate minimally processed foods, leading to weight gain.

When those qualities were modified, consumption went down, even if the foods were considered ultraprocessed, Hall said. Data is still being collected from remaining participants and must be completed, analyzed and published in a peer-reviewed journal.

Still, the early results suggest that “you can almost normalize” energy intake, “despite the fact that they’re still eating a diet that is more than 80% of calories from ultraprocessed food,” Hall told the audience.

Not everyone agrees with Hall’s methods, or the implications of his research.

Dr. David Ludwig, an endocrinologist and researcher at Boston Children’s Hospital, criticized Hall’s 2019 study as “fundamentally flawed by its short duration” — about a month. Scientists have long known that it’s possible to get people to eat more or less for brief periods of time, but those effects quickly wane, he said.

“If they were persistent, we would have the answer to obesity,” said Ludwig, who has argued for years that consumption of highly processed carbohydrates is the “prime dietary culprit” and focusing on the processing of the foods is “distracting.”

He called for larger, better-designed studies lasting a minimum of two months, with “washout” periods separating the effects of one diet from the next. Otherwise, “we waste our energy, we mislead the science,” Ludwig said.

Concerns about the short length of the studies may be valid, said Marion Nestle, a nutritionist and food policy expert.

“To resolve that, Hall needs funding to conduct longer studies with more people,” she said in an email.

The NIH spends about $2 billion a year, about 5% of its total budget, on nutrition research, according to Senate documents.

At the same time, the agency cut the capacity of the metabolic unit where investigators conduct such studies, reducing the number of beds that must be shared among researchers. The two participants enrolled now at the center and the two planned for next month are the most Hall can study at any one time, adding months to the research process.

Read more from Fortune

Srisatta, the Florida volunteer who hopes to become an emergency room physician, said participating in the trial left him eager to know more about how processed foods affect human health.

“I mean, I think everyone knows it’s better to not eat processed foods, right?” he said. “But having the evidence to back that up in ways that the public can easily digest,” is important, he said.

HHS officials didn’t respond to questions about Kennedy’s intentions regarding nutrition research at NIH. The agency, like many others in the federal government, is being buffeted by the wave of cost cuts being directed by President Donald Trump and his billionaire aide Elon Musk.

Jerold Mande, a former federal food policy advisor in three administrations, said he supports Kennedy’s goals of addressing diet-related diseases. He has pushed a proposal for a 50-bed facility where government nutrition scientists could house and feed enough study volunteers like Srisatta to rigorously determine how specific diets affect human health.

“If you’re going to make America healthy again and you’re going to address chronic disease, we need better science to do it,” Mande said.