Sponsored By

Losing weight on a UPF diet? New research unpacks the causes of overeatingLosing weight on a UPF diet? New research unpacks the causes of overeating

Scientists are investigating the role of texture, calorie density, and hyperpalatability to shed more light on the causes of ultra-processed food (UPF) overconsumption.

Rik Moors, Content Editor

January 20, 2025

7 Min Read
Losing weight on a UPF diet? New research unpacks the causes of overeating
iStock/energyy

Ultra-processed foods have been repeatedly linked to increased calorie consumption and weight gain. However, the mechanisms explaining why people tend to overeat these foods remain under investigation. New scientific findings, including preliminary data from Dr Kevin Hall’s latest US National Institutes of Health (NIH) controlled feeding study (not yet published) and recent results from the RESTRUCTURE project’s analysis of bread consumption provide insights into the drivers of calorie overconsumption in UPFs.

Hall’s early results suggest that when energy density and hyperpalatability – the presence of nutrient combinations like fat, salt, and sugar – are reduced, even in UPFs, calorie intake can be moderated. The RESTRUCTURE project finds that modifying food texture to slow eating rate could be a practical strategy for reducing calorie intake.

Investigating UPF consumption: Kevin Hall’s latest controlled study

Kevin Hall, a senior investigator at the NIH, presented preliminary results from his ongoing trial during Imperial College London’s Ultra-processed food: The scope for Government action conference in November 2024. His previous study on UPFs and overeating, which was published in 2019, gained a lot of attention for demonstrating a highly increased energy intake when participants consumed an ultra-processed diet compared to a minimally processed diet. The participants in that study switched between two diets matched in calories, energy density, macronutrients, sugar, sodium, and fibre – and when allowed to eat until they felt satiated, they ate on average 500 calories more per day on the UPF-heavy diet compared to the minimally processed diet.

While that earlier research confirmed a clear link between UPF consumption and increased energy intake, the reasons for this pattern remained mostly speculative. Hall’s new, ongoing controlled feeding study is designed to investigate the mechanisms more closely by manipulating specific properties of the foods served to participants in a controlled environment.

Similar to the 2019 study, the ongoing research takes the form of a randomised, crossover feeding trial conducted under tightly controlled conditions at the NIH Clinical Center. Each participant follows four different diets for one week each, with all food prepared and measured to ensure precise monitoring of calorie intake, body weight, and metabolic responses. The diets tested are:

  • Minimally processed (0% UPF), low energy density (0.99 kcal per gram), low hyper-palatability 

  • Ultra-processed (88% of energy from UPFs), high energy density (1.95 kcal per gram), high hyper-palatability (closely matching the most common UPF diets)

  • Ultra-processed (80% of energy), high energy density (1.9 kcal per gram), low hyper-palatability

  • Ultra-processed (81% of energy), low energy density (1.01 kcal per gram), low hyper-palatability 

Carbohydrate, fat, sodium, fibre, and sugar content is nearly identical in each diet – and unlike in the 2019 study, protein content was also standardised at between 18 and 20% of energy intake for each diet. Participants are instructed to eat until comfortably full without specific calorie targets, and the quantity of food consumed is measured to determine ad libitum (freely available) calorie intake.

Appetite ratings for the different diets – measured by hunger, fullness, satisfaction, pleasantness, and eating capacity in surveys – show no significant differences between any of the four types. Eating rates also do not significantly differ between the groups.

Losing weight, but not fat on a UPF diet?

Preliminary results from the first half of the study participants suggest that both energy density and hyperpalatability significantly influence calorie intake, with energy density in particular accounting for a significant difference. Participants consuming the high energy density and high hyperpalatability UPF diet consumed roughly 1,000 more calories per day than those on the minimally processed diet – twice the difference found during the previous study.

In contrast, those following the reformulated UPF diet, where both energy density and hyper-palatability were reduced, consumed just 170 additional calories compared to the minimally processed group. The diet high in energy density but low in hyperpalatability resulted in an additional 800 calories per day compared to the MPF diet. This suggests that calorie overconsumption may be linked more closely to these specific food properties – and energy density in particular – than to ultra-processing alone.

“If you can reduce both the energy density of the overall dietary pattern and the percentage of calories coming from hyperpalatable foods you can almost normalise people’s ad libidum energy intake, despite the fact that they're still eating a diet that is more than 80% of calories from ultra-processed foods,” said Hall.

Weight gain was observed during the consumption of the two high-energy density UPF diets (1kg in a week for the high-high diet, and slightly less for the high-low diet), while the minimally processed and the low-density, low-hyperpalatability UPF diet both led to weight loss of around 0.5 kg in a week – leading Hall to conclude that “weight gain is not a necessary component of a highly ultra-processed diet.”

However, on closer examination, the study found that only the minimally processed diet group lost body fat, with the UPF weight loss coming from fat-free mass. “[T]his suggests that there's something going on with digestibility and ... energy partitioning in the body,” said Hall, who emphasised that these results are interim and the study remains ongoing. He hopes that future analyses can explore the role of gut microbiota and immune system function in response to different food patterns to explain the different types of weight loss observed.

Bread texture and eating rate: Findings from the RESTRUCTURE project

While Hall’s research focuses on whole diets, the RESTRUCTURE project, led by Ciaran Forde at Wageningen University, investigates the relationship between food texture and calorie intake through a more focused lens. For the last four years, studies have tried to obtain more precise data on how UPFs may impact bodyweight and overall metabolic health. A recent study from the project examined bread, a staple food often consumed rapidly, to assess how variations in texture influence eating speed and, by extension, calorie consumption.

The study, The eating rate of bread predicted from its sensory texture and physical properties, included 36 participants consuming nine commercially available bread types ranging from wholemeal bread slices and ciabatta to soft white buns and croissants. Each bread sample was tested under tightly controlled conditions, with eating rates measured using video recordings, while both sensory texture and objective physical measurements of crumb hardness, crust cohesiveness, and moisture content were assessed.

The study found that the texture of bread significantly influenced the speed at which participants ate. The ‘slowest’ bread type, a wholemeal bread slice, was consumed 40% more slowly than the fastest, a soft white bun. Softer, more adhesive breads were consumed faster, while those with greater hardness and dryness slowed the eating rate due to increased chewing and oro-sensory exposure.

The researchers conclude that since “texture-induced reductions of [the eating rate] by 20% can lead to a 10–15% reduction in ad libitum energy intake … switching to breads with low eating rates could be an effective strategy to modify energy intake.”

The researchers developed predictive models for eating rates based on both sensory texture ratings and objective physical measurements. Physical measurements, such as crumb hardness and crust cohesiveness, were slightly more reliable predictors of eating rate than subjective sensory evaluations.

Mechanisms driving UPF overconsumption: Connecting the dots

The evolving body of evidence highlights the importance of both food structure and nutrient density in shaping dietary patterns, offering new pathways for reformulation strategies aimed at supporting healthier eating habits.

Food texture, energy density, and hyper-palatability emerge as interconnected factors driving overconsumption. Softer, calorie-dense foods encourage faster eating, which can delay satiety signals, while hyper-palatable combinations of sugar, fat, and salt amplify the appeal and consumption of UPFs.

Reformulating ultra-processed foods by modifying their energy density and texture may offer a viable strategy for reducing their contribution to calorie overconsumption without compromising consumer acceptance. In bread products, for example, small adjustments such as increasing crumb hardness or crust thickness could slow eating rates and support better portion control. For more complex food products, reducing the calorie concentration per gram while maintaining flavour balance could help mitigate overeating.

About the Author

Rik Moors

Content Editor, Informa Markets

Rik Moors is a Content Editor for Fi Global Insights and Ingredients Network, writing about the latest trends and developments in food ingredients and nutrition. As a researcher and journalist, Rik has previously covered topics such as economic development and agriculture for various organisations in Africa and the Middle East.

SUBSCRIBE TO OUR FREE NEWSLETTERS
Get the latest food ingredient innovations, R&D breakthroughs, & sustainable sourcing strategies sent straight to your inbox.