Study Finds Age, Blood Pressure, and BMI Are Top Predictors of Cognitive Performance

Why do some people stay quick-thinking and focused throughout life, while others notice their attention and reaction speed change as they get older? Can today’s artificial intelligence help us discover what actually influences our mental sharpness? A new study from the University of Illinois Urbana-Champaign set out to answer these questions, using cutting-edge machine learning to analyze diet, physical activity, blood pressure, body weight, and other health markers in nearly 400 adults. The results point to age, blood pressure, and BMI (Body Mass Index). as the strongest predictors of how well people perform on cognitive tests, and show how new data tools can find connections that traditional methods often miss.

Study Finds Age, Blood Pressure, and BMI Are Top Predictors of Cognitive Performance. Image by Pexels

Note: This article is intended for general information and educational purposes. It summarizes scientific research in accessible language for a broad audience and is not an official scientific press release.

Understanding what helps us maintain mental clarity and attention as we age is an important question for both scientists and the public. Many people wonder how their lifestyle choices today — such as what they eat, how active they are, and how they manage their weight and blood pressure — may impact their ability to concentrate, solve problems, or react quickly. Researchers have long studied how individual factors like diet or exercise influence brain function. But real life is complex: multiple health, lifestyle, and demographic factors are always interacting, and their combined influence on cognitive abilities can be hard to untangle.

To address this challenge, a team from the University of Illinois Urbana-Champaign, led by Shreya Verma and colleagues, designed a new study published in The Journal of Nutrition. The study took advantage of machine learning, a branch of artificial intelligence that can sift through large, complex datasets and spot patterns that might go unnoticed by standard statistical methods. The researchers’ aim was to “predict cognitive performance based on a set of health and behavioral factors, aiming to identify key contributors to cognitive function for insights into potential personalized interventions.”

What Did the Researchers Study?

The team combined data from seven previous studies to build a diverse group of 374 adults, aged 19 to 82 years, who were living in Illinois. Participants came from a variety of backgrounds and were recruited over several years. Each person provided a wide range of information, including their age, sex, ethnicity, weight, height, waist circumference, and income. Participants also shared details about their dietary habits and completed a questionnaire on their physical activity. Importantly, everyone took a computer-based cognitive test known as the modified Eriksen flanker task — a test that measures how quickly and accurately someone can focus and respond when faced with distractions.

Diet quality was scored using several widely recognized tools, including the Healthy Eating Index (HEI-2020), the DASH diet, the Mediterranean diet, and the MIND diet. These scoring systems allowed the researchers to compare not just individual foods, but overall dietary patterns. Physical activity was measured by asking participants how often and how vigorously they exercised during their free time.

Blood pressure, both systolic and diastolic, was measured in a standardized way, and body mass index (BMI) was calculated from measured height and weight. By combining all these data points, the research team could look at the big picture and see how each factor related to people’s attention and reaction speed.

How Was the Study Conducted?

After ensuring all data were complete and consistent, the researchers analyzed the information from 374 adults. Participants who had major missing data were not included, to ensure the accuracy of the models. The group was fairly balanced in gender and included people from different age ranges and backgrounds.

To analyze which health and lifestyle factors had the greatest impact on cognitive performance, the team used several supervised machine learning models. These included decision trees, random forest, AdaBoost, XGBoost, and different types of regression models. Machine learning is particularly good at looking at many variables at once and finding patterns or relationships that might be hidden in the data. In this study, all the information about health, lifestyle, and demographics was entered into the models, and the main outcome measured was how quickly people responded on the attention test.

The researchers then compared the different models to see which gave the most accurate predictions. Once they identified the best-performing model, they used a method called permutation importance to determine which factors were most strongly connected to cognitive performance.

What Makes This Study Different?

While many previous studies have looked at the role of individual factors — like exercise, blood pressure, or a specific diet—this study stands out for its comprehensive approach. The authors highlight that “the complex interplay of these factors and their relative importance in predicting cognitive outcomes remains incompletely understood.” By using machine learning, the researchers could examine all the variables together and “identify patterns that may not be apparent through conventional statistical approaches.” This allowed them to not only find the strongest predictors, but also to see how combinations of factors —such as age and blood pressure, or diet and physical activity — might interact to affect cognitive performance.

What Did the Study Find?

1.Machine Learning Reveals Complex Interactions

Another key insight from the study is that machine learning was especially useful for detecting subtle connections and interactions among health factors. As the authors explain, “ML algorithms excel at analyzing large data sets with multiple variables, uncovering patterns and relationships that conventional approaches might overlook.” This made it possible to see, for example, that greater adherence to the HEI-2020 diet partially offset the negative effects of higher BMI on cognitive function, or that the benefits of a healthy diet were most pronounced in people with lower blood pressure. The study’s approach helped reveal how diet and lifestyle might interact with other health indicators, such as age or body composition, to influence cognitive abilities.

2. Age, Blood Pressure, and BMI Are the Strongest Predictors

The study’s most important finding was that age stood out as the top predictor of cognitive performance in the attention test, followed by diastolic blood pressure and body mass index (BMI). Systolic blood pressure and overall diet quality (measured by the Healthy Eating Index) were also found to be significant, but less so than age and diastolic blood pressure. As the authors report, “Age, blood pressure, and BMI show strong associations with cognitive performance, whereas diet quality has a subtler effect.” In practice, this means that being older, having higher blood pressure, or a higher BMI were each linked to slower reaction times and lower performance on the attention task.

3. Diet and Physical Activity Play a Supportive but Modest Role

While the impact of age, blood pressure, and BMI was most prominent, the study also found that healthier lifestyle habits—including a balanced diet and regular physical activity — were associated with better cognitive test results. People who reported being more physically active and who scored higher on healthy diet patterns (like the HEI-2020 or DASH diets) tended to perform better, even if these effects were not as large as those of age or blood pressure. The authors note, “physical activity was found to be an important predictor,” and add that the combination of activity and good diet “highlighted the importance of maintaining both an active lifestyle and a balanced diet for optimal cognitive function.” In some cases, the study found that a healthier diet could help offset some of the negative cognitive effects linked to higher BMI or elevated blood pressure. Specific diets such as the DASH, MIND, and Mediterranean also contributed, but their importance scores were lower compared to the overall Healthy Eating Index.

What Are the Limitations and Next Steps?

Like any research, this study has its limitations. The authors are careful to note that their study only looked at one specific measure of cognitive function (reaction time in a particular test). This means the results might not apply to other aspects of thinking, such as memory or reasoning. Additionally, because the study collected data at only one point in time, it cannot prove cause and effect — only associations. The researchers write that “longitudinal and experimental studies are needed to confirm these findings and explore the causal and synergistic relationships between the features and cognitive performance.”

The group of participants, while diverse, came from one region and may not fully represent all populations. The authors recommend that future studies should include more people from different backgrounds and add more health data, such as genetic information or detailed metabolic profiles, to improve our understanding of what shapes cognitive performance over time.

Conclusion

In summary, this study offers new insight into which health and lifestyle factors are most strongly linked to cognitive performance in adults. Age, blood pressure, and BMI stand out as the top predictors, while diet and physical activity also play a role, especially when combined. The use of machine learning allowed the research team to spot patterns and interactions that may be missed by traditional analysis. While more research is needed, especially to track changes over time and include broader measures of cognitive health, these findings underscore the importance of looking at the whole picture when it comes to supporting mental sharpness throughout adulthood.

The full study, “Predicting Cognitive Outcome Through Nutrition and Health Markers Using Supervised Machine Learning,” can be read at https://doi.org/10.1016/j.tjnut.2025.05.003.

The information in this article is provided for informational purposes only and is not medical advice. For medical advice, please consult your doctor.