An abundance of previous research has linked less-than-healthy foods like sugary snacks and deep-fried anything with poorer brain health compared to those who opt for healthier choices. But a new study in the journal Neurology suggests that what foods you eat together can be important when it comes to your risk of dementia.

Researchers asked 1,522 participants to complete a comprehensive dietary survey in 2002, including a qualitative food frequency questionnaire. At a followup 12 years later, they looked at 209 participants who had developed dementia, as well as 418 people who did not.

They used the data to create “food networks,” which identified what type of foods were eaten most in combination, and whether those groups of foods were substantially different between those who had dementia and those who didn’t.

Researchers found that the people who developed dementia were more likely to combine highly processed meats—such as sausages and cured meats—with starchy foods like potatoes, alcohol, and sugary snacks such as cookies and cakes.

“Processed meat appeared very central in their diet, which means it was connected with many foods,” lead author Cecilia Samieri, Ph.D., an epidemiologist at the University of Bordeaux, told Bicycling. “Worse eating habits toward processed meats and snacking were evident years before dementia diagnosis. In contrast, diverse and healthy diets appear to decrease the risk of developing dementia.”

In other words, it wasn’t the processed meat on its own that seemed problematic, she said, but the way its consumption was connected with so many other items that are considered less healthy, such as pasta, jam, and potatoes. (It’s worth noting, however, that athletes like cyclists can benefit from using these foods as fuel when fast-acting carbs and sugars are necessary to prevent from bonking on a long ride.)

[Gravel! prepares you with everything you need to know to crush it, including the best gear, how to train, and much more!]

The study does have some limitations though. Most notably, it required participants to recall what they ate rather than have their consumption monitored by researchers. Also, it did not track diet patterns over time, to see if changes made any difference.

Even with those caveats, the study does provide good evidence about the importance of diet diversity, according to the study’s coauthor, Abhijeet Sonawane, Ph.D., a research fellow in medicine at Harvard Medical School.

“Diversity is likely protective because it provides a combination of healthy nutrients, including vitamins, polyphenols, and carotenoids from plant foods, as well as good fats and protein,” he told Bicycling. “People in our study who did not develop dementia were more likely to have a lot of diversity in their diets, and that included healthier foods.”