Slowly eating less: how texture-based differences in eating rate moderate energy intake from ultra-processed foods

Numerous studies have shown associations between higher consumption of ultra-processed foods (UPFs) and negative health outcomes, including high energy intakes and obesity. The underlying mechanisms remain unclear. One of the proposed mechanisms underlying the higher energy intake of UPFs compared to unprocessed foods is eating rate (ER). ER is primarily driven by food texture and individual eating behaviour. Foods with slower texture-based ERs are consumed in smaller amounts compared to foods eaten at a faster ER. However, research to date has focused only on the acute impact of texture-based ER on intake. The main aim of this thesis was to determine whether the effect of ER of UPFs on food and energy intake is consistent across a wide variety of meals and sustained over time. The sub-aims were (1) to assess the effect of various textures on ER and (2) to assess how the ERs of meal components affect the overall ER and intake of multi-component meals.

Regarding sub-aim 1, we showed in Chapter 2 that models based on the texture and physical properties of bread were able to predict ER with high accuracy. In Chapter 3, we showed that soft penne and soft carrots were consumed faster than their hard variants and the addition of sauce increased the ER of both penne and carrots by approximately 30%. In Chapter 4, the ER of 54 meal components were strongly correlated to perceived hardness, dryness, and chewiness. Overall, we demonstrated that textures that reduced ER achieved this by reducing bite size, increasing the number of chews per bite, and prolonging oro-sensory exposure (Chapter 2 to Chapter 4).

Regarding sub-aim 2, we demonstrated in Chapter 3 that the ER of individual components cumulatively determined the ER of the composite pasta dish. In Chapter 4, we found that combining individual solid meal components led to a higher overall meal ER compared to the ER of the individual components. We also demonstrated that the ER of solid individual meal components was positively associated with the intake of the same meal components when consumed as part of a meal.

Regarding the main aim, we showed in Chapter 5 with twelve breakfast meals and twelve lunch meals that were all ultra-processed but differed in texture properties, that meals with a slower ER were consistently consumed with a reduced food intake, where an average decrease of 20% in ER produced an 11% decrease in food intake. In Chapter 6, participants consumed 369 kcal/day less when on the 14-day Slow UPF diet compared to the 14-day Fast UPF diet and the effect of texture on ER and energy intake was consistent over two weeks.

In conclusion, this thesis demonstrates that individuals consistently adapt their eating behaviour and intake in response to food texture and this effect is sustained over time. These findings highlight that texture modification is an effective approach to moderate ER and energy intake from UPFs, and provide valuable insights for food-based strategies aimed at reducing excessive energy consumption in environments that promote chronic energy surplus.