Dr. Axe Wire · Study decoded · nutrition

Eating Disorder Traits Shape How Addictive Processed Foods Feel

A survey of 1,664 U.S. adults links eating-behavior traits, not body weight, to how addictive people perceive ultra-processed foods to be.

By PJ Howland · November 14, 2026 · 1 source · Watch

Photo: Pexels

At a glance

What they found: Addictiveness ratings rose most steeply for Nova 4 and hyperpalatable foods as eating-disorder and food-addiction scores increased.
Who was studied: 1,664 U.S. adults over 18, surveyed in 2024 through the Prolific research platform.

A survey of 1,664 U.S. adults, published November 11, 2026 in the Journal of the Academy of Nutrition and Dietetics, found that people with more eating-disorder symptoms and food-addiction traits rated ultra-processed and hyperpalatable foods as far more addictive than minimally processed foods, while body mass index showed no such pattern.

1,664US adults surveyed

2024survey data

Researchers surveyed 1,664 U.S. adults recruited through the Prolific research platform in 2024, collecting demographics, height, weight, and scores on two validated questionnaires: the Eating Disorder Examination Questionnaire Short and the Modified Yale Food Addiction Scale 2.0. Each participant rated how problematic 99 randomly assigned foods, drawn from a pool of 297, felt on a 100-point scale. Foods were sorted by Nova processing category and standardized hyperpalatable-food criteria. The gap in addictiveness ratings between processed and minimally processed foods widened as eating-disorder and food-addiction scores rose, steepest for Nova category 4 (ultra-processed) foods; the same pattern held for hyperpalatable versus non-hyperpalatable foods. Body mass index showed no such interaction. The authors concluded that eating-behavior traits, not body weight, shape perceived food addictiveness, with ultra-processed and hyperpalatable foods rated substantially more addictive among people with food addiction and eating-disorder symptoms.

In general, foods combining refined carbohydrates, fats, sugar and salt can engage the brain's reward circuitry more strongly than whole foods. This survey did not test that mechanism directly, but its results fit that framework: people already prone to disordered eating or food-addiction symptoms were linked to the strongest response to the most processed, palatable products.

Foods Rated in the Survey

Journal of the Academy of Nutrition and Dietetics, 2026, Ahrens et al.

This is a cross-sectional snapshot from one year and cannot show ultra-processed foods caused food-addiction symptoms, only that the two were linked. The sample came from one online platform, ratings were self-reported, and no actual intake was measured. Larger studies that track the same people over time are the next step, to test whether this pattern predicts future eating problems.

Dr. Axe’s Take

People most vulnerable to food addiction feel the strongest pull toward ultra-processed, hyperpalatable foods, and that gap has nothing to do with their weight. I'd build meals around whole foods — vegetables, quality protein, healthy fats — so the brain's reward system has less engineered bait to respond to. If you can stop easily after berries or roasted vegetables but can't stop after one serving of a packaged snack, that difference tells you which foods to keep out of your kitchen. This is a link, not proof that processed foods cause addictive eating, but it's a reason to audit your pantry this week.

— Dr. Axe

Take the Cellular Health Quiz →

Study summary

Who: 1,664 U.S. adults over 18, surveyed in 2024 via the Prolific research platform.
Design: Nationally representative cross-sectional survey using linear mixed-effects models.
Dose: Each participant rated 99 randomly selected foods out of a pool of 297.
Primary result: Addictiveness ratings rose most steeply for Nova 4 (ultra-processed) and hyperpalatable foods as eating-disorder and food-addiction scores increased; BMI showed no such interaction.
Secondary: Significant interactions found for processing category with both EDE-QS and mYFAS 2.0 scores, and separately for hyperpalatable classification with both scores (size not reported).
Funding / conflicts: Funder listed in the journal record: National Institute of Diabetes and Digestive and Kidney Diseases. Conflicts not reported in the abstract.
Date: Nov 11, 2026