New research finds beliefs and perceptions drive overeating more than food’s UPF label

Ultra-processed foods (UPFs) have been cast as the villains of modern nutrition. From obesity to dementia, critics blame factory-made products – crisps, ready meals, fizzy drinks and packaged snacks – for fuelling an epidemic of poor health. Campaigners argue these foods are designed to hijack our brains, forcing us to eat more than we need.

Policymakers have responded with proposals ranging from warning labels and taxes to outright bans near schools. Yet new research suggests the picture may be far more complicated.

A team of UK scientists studied the eating habits of more than 3,000 adults, analysing their responses to over 400 common foods. Participants rated each food by how much they liked it and how likely they were to overeat it. The foods spanned a typical shopping basket: apples, noodles, jacket potatoes, custard creams, cottage pie.

Researchers then compared those responses with nutritional content – fat, sugar, fibre, energy density – as well as how the foods were perceived: sweet, fatty, processed, or healthy. The results challenge the idea that UPFs alone are driving our overconsumption.

Taste vs. overeating

One key finding was that liking a food and overeating it are not the same thing. Many people like porridge, for example, but rarely binge on it. Chocolate, biscuits and ice cream, however, topped both lists: highly liked, and highly “bingeable”.

The team found calorie-dense foods were more likely to lead to overeating. But perceptions played a crucial role. Foods believed to be sweet, fatty or highly processed were more likely to be overeaten, regardless of their actual nutritional makeup. In contrast, foods perceived as bitter or high in fibre were far less tempting.

In fact, when researchers combined nutrient data with perceptions, they could predict nearly 80% of variation in overeating. Nutritional factors explained about 41%, while beliefs and sensory impressions added another 38%.

The UPF label under scrutiny

Where did UPFs fit into this? Surprisingly, the Nova system of classification – which sorts foods by how much they are processed – explained less than 2% of variation in liking and only 4% in overeating.

This suggests the UPF label itself is a blunt tool. It sweeps together soft drinks with fortified cereals, protein bars with vegan meat substitutes. While many UPFs are calorie-dense and easy to overconsume, others can serve useful roles, particularly for older adults with low appetite, people on special diets, or those needing accessible nutrition.

The blanket message that “all UPFs are bad” oversimplifies and risks steering people away from foods that can be beneficial.

Rethinking solutions

So what should be done? The researchers argue that policy must focus less on demonising entire categories of food and more on understanding the psychology of eating.

They recommend:

• Boosting food literacy: teaching people what makes food satisfying, what drives cravings, and how to spot personal triggers for overeating.
  
• Reformulating food with purpose: designing meals and snacks that are enjoyable yet filling, rather than simply creating bland diet versions or hyper-palatable treats.
  
• Acknowledging motivations: recognising that people eat for comfort, connection and pleasure as well as hunger, and helping them build healthier habits around those needs.
  

Beyond the blame game

Ultra-processed foods do deserve some scrutiny. They are often cheap, heavily marketed and sold in oversized portions. But blaming processing alone misses the bigger picture. Eating behaviour is shaped not just by nutrients but by perception, psychology and context.

Ultimately, what matters most is how food tastes, how it makes us feel and how it fits into our lives. Demonising everything that comes in a packet risks creating more confusion than clarity. If we want healthier diets, we need to understand not just what people eat – but why