Karolina Grabowska For Unsplash+
Study Links UPFs To Rising Dementia Risk Scores
In A Nutshell
- A study of 2,192 Australian adults found that eating more ultra-processed foods was linked to worse attention scores, independent of overall diet quality.
- For every 10% increase in ultra-processed food calories, dementia risk scores also ticked upward on a scale measuring changeable risk factors.
- Ultra-processed food intake showed no link to memory, only attention, which researchers say may be an early warning sign of broader cognitive decline.
- The findings held up even after accounting for Mediterranean diet adherence, suggesting the degree of food processing itself may play a distinct role in brain health.
Potato chips, frozen dinners, and diet soda don’t exactly have a reputation for being brain food. But a new study of more than 2,000 Australian adults suggests these foods may be chipping away at something far more concerning than a waistline: the brain’s ability to pay attention. And eating a salad on the side may not be enough to undo it.
For every 10% increase in the share of calories coming from ultra-processed sources, attention scores dropped by a small but measurable amount (about 0.05 points on the study’s scale), and a score used to estimate future dementia risk ticked upward. Both associations held up even after accounting for how closely participants followed a Mediterranean-style diet, widely considered the gold standard for brain-healthy eating. That detail matters because it suggests something about the processing itself may be driving the effect, not simply the absence of better food choices.
Published in Alzheimer’s & Dementia: Diagnosis, Assessment & Disease Monitoring, the study doesn’t prove that ultra-processed foods directly cause cognitive problems. It captured a single snapshot in time rather than tracking people over years.
What Are Ultra-Processed Foods, and Why Do They Matter for Brain Health?
Scientists use a classification system called Nova, which sorts foods by how much industrial processing they’ve undergone rather than by nutritional content. At one end are whole or minimally processed foods like fresh fruits, vegetables, and plain meats. At the other end sit ultra-processed foods, industrial creations made largely from refined ingredients and chemical additives. Soft drinks, packaged salty snacks, processed meats, frozen ready meals, and flavored dairy desserts all fall into this category.
In Australia, these products make up roughly 42% of total calories consumed. In the United States and United Kingdom, that figure climbs above 50%. Participants in this study were right in line with the Australian national average, getting about 41% of their daily calories from ultra-processed sources.
How the Study Linked Ultra-Processed Foods to Attention
Led by Barbara R. Cardoso at Monash University, the research team analyzed data from 2,192 adults enrolled in the Healthy Brain Project, an online platform recruiting dementia-free people between the ages of 40 and 70. Most participants had a close family member with dementia, placing them at higher genetic risk for cognitive decline.
Participants filled out a detailed 130-item food questionnaire covering their eating habits over the previous 12 months, with every item classified using the Nova system. Cognitive function was measured using an established online test assessing processing speed, visual attention, visual recognition memory, and working memory, combined into two summary scores: one for attention, one for memory. Dementia risk was estimated using a tool called CAIDE, which weighs factors like age, cholesterol history, blood pressure, physical activity, and body mass index to predict 20-year dementia risk. The researchers also used a modified version of the CAIDE score that focuses only on changeable risk factors, though that version has not been independently validated as a dementia prediction tool.
Ultra-Processed Foods Linked to Worse Attention, Higher Dementia Risk Score
Higher consumption of ultra-processed foods was associated with worse attention scores, following a general trend where higher intake tracked with poorer performance. The effect remained across multiple rounds of analysis, even after accounting for Mediterranean diet adherence and body weight.
Ultra-processed food intake showed no connection to memory scores, a finding the authors noted aligns with previous research. Attention is foundational to learning and problem-solving, and disruptions there may surface before broader cognitive problems become apparent, a pattern documented in early stages of brain disease.
On the dementia risk front, each 10% increase in ultra-processed food consumption was linked to a 0.24-point rise on the modified CAIDE score, and that association held up after accounting for diet quality. The link with the original CAIDE score was weaker and lost statistical significance once Mediterranean diet adherence was factored in.
Salads Alone May Not Be Enough to Counter Ultra-Processed Food Risks
People who eat lots of ultra-processed food also tend to eat fewer fruits, vegetables, and other whole foods, so researchers have long struggled to separate the harm of junk food from the simple absence of better options. By adjusting for Mediterranean diet adherence, this team tried to isolate the processing question. The fact that the associations held up suggests something about the processing itself may matter.
The authors suggest several possible explanations, though none were directly tested in this study. Ultra-processing can strip away beneficial plant compounds, vitamins, and minerals while introducing potentially harmful substances from packaging materials, high-heat industrial cooking, and additives like emulsifiers and preservatives. Animal studies have shown some of these substances can disrupt gut microbes that communicate with the brain, potentially triggering inflammation and impairing nerve cell signaling. Ultra-processed foods are also strongly tied to obesity, diabetes, and high blood pressure, all of which damage blood vessels that feed the brain, and the researchers noted that attention is particularly sensitive to that kind of vascular damage.
Among the most commonly consumed ultra-processed foods in the study were dairy-based desserts, soft drinks, packaged salty snacks, processed meats, and ready meals. In a world where these products account for nearly half of all calories consumed in wealthy nations, even a modest link to diminished brain function in middle age deserves serious attention.
Disclaimer: The findings of this study are observational and do not establish a direct cause-and-effect relationship between ultra-processed food consumption and cognitive decline or dementia. This article is intended for informational purposes only and should not be taken as medical or dietary advice. Readers with concerns about their diet or cognitive health are encouraged to consult a qualified healthcare professional.
Paper Notes
Limitations
Several factors limit how broadly these findings can be applied. Because the study captured dietary habits and cognitive performance at a single point in time, it cannot establish whether ultra-processed food consumption preceded or contributed to cognitive changes, only that the two were associated. All dietary data came from self-reported questionnaires, which are subject to recall error, and the food frequency questionnaire used was originally developed for a British population and has not been validated for Australian dietary patterns. The sample skewed heavily female (about 75%), was predominantly white, and had higher-than-average education and socioeconomic status compared to the broader Australian population. Most participants also had a family history of dementia, meaning they were not representative of the general public. Finally, the modified CAIDE dementia risk score used as a secondary outcome has not been independently validated as a tool for predicting dementia, limiting the weight that can be given to those particular findings.
Funding and Disclosures
The Healthy Brain Project, from which study data were drawn, was funded by the Australian National Health and Medical Research Council, the Alzheimer’s Association, the Dementia Australia Research Foundation, the Bethlehem Griffiths Research Foundation, the Yulgilbar Alzheimer’s Research Program, the National Heart Foundation of Australia, and the Charleston Conference for Alzheimer’s Disease. Individual researchers received support from FAPESP and NHMRC fellowship and leadership grants. The authors declared no conflicts of interest. Open-access publication was facilitated by Monash University through the Wiley-Monash University agreement.
Publication Details
The study was authored by Barbara R. Cardoso, Euridice Martinez Steele, Barbara Brayner, Xinyi Yuan, Lisa Bransby, Hannah Cummins, Yen Ying Lim, and Priscila Machado, representing Monash University, Deakin University, and the University of São Paulo. It was published in Alzheimer’s & Dementia: Diagnosis, Assessment & Disease Monitoring in 2026 under the title “Ultra-processed food intake, cognitive function, and dementia risk: A cross-sectional study of middle-aged and older Australian adults.” DOI: 10.1002/dad2.70335.
