Summary: A new study used machine learning to pinpoint the lifestyle and health factors most strongly associated with cognitive performance across the lifespan. Among 374 adults aged 19 to 82, age, blood pressure, and BMI were the top predictors of success on a focus-and-speed-based attention test.
While diet and exercise played a smaller role, they were still associated with better outcomes, particularly in offsetting high BMI or blood pressure. This data-driven approach highlights how combining multiple factors provides a clearer picture of what supports brain health with age.
Key Facts:
- Top Predictors: Age, diastolic blood pressure, and BMI most strongly influenced cognitive performance.
- Diet + Exercise: Healthy eating and physical activity contributed modestly but positively to focus and reaction speed.
- Machine Learning Advantage: Advanced algorithms revealed nuanced relationships traditional statistics may miss.
Source: University of Illinois
A new study offers insight into the health and lifestyle indicators — including diet, physical activity and weight — that align most closely with healthy brain function across the lifespan.
The study used machine learning to determine which variables best predicted a person’s ability to quickly complete a task without becoming distracted.
Reported in The Journal of Nutrition, the study found that age, blood pressure and body mass index were the strongest predictors of success on a test called the flanker task, which requires participants to focus on a central object without becoming distracted by flanking information.
Diet and exercise also played a smaller but relevant role in performance on the test, the team found, sometimes appearing to offset the ill effects of a high BMI or other potentially detrimental factors.
“This study used machine learning to evaluate a host of variables at once to help identify those that align most closely with cognitive performance,” said Naiman Khan, a professor of health and kinesiology at the University of Illinois Urbana-Champaign who led the work with kinesiology Ph.D. student Shreya Verma.
“Standard statistical approaches cannot embrace this level of complexity all at once.”
To build the model, the team used data collected from 374 adults 19 to 82 years of age. The data included participant demographics, such as age, BMI, blood pressure and physical activity levels, along with dietary patterns and performance on a flanker test that measured their processing speed and accuracy in determining the orientation of a central arrow flanked by other arrows that pointed in the same or opposite direction.
“This is a well-established measure of cognitive function that assesses attention and inhibitory control,” Khan said.
Previous studies have found that several factors are implicated in the preservation of cognitive function across the lifespan, Khan said.
“Adherence to the healthy eating index, a measure of diet quality, has been linked to superior executive function and processing speed in older adults,” he said. “Other studies have found that diets that are rich in antioxidants, omega-3 fatty acids and vitamins are associated with better cognitive function.”
The Dietary Approaches to Stop Hypertension, or DASH diet, the Mediterranean diet, and a diet that combines the two, called the MIND diet, all “have been linked to protective effects against cognitive decline and dementia,” the researchers wrote. Physical factors, such as BMI and blood pressure, along with increased physical activity also are strong predictors of cognitive health, or decline, in aging.
“Clearly, cognitive health is driven by a host of factors, but which ones are most important?” Verma said. “We wanted to evaluate the relative strength of each of these factors in combination with all the others.”
Machine learning “offers a promising avenue for analyzing large datasets with multiple variables and identifying patterns that may not be apparent through conventional statistical approaches,” the researchers wrote.
The team tested various machine learning algorithms to see which one best weighed the various factors to predict the speed of accurate responses in the flanker test. The researchers tested the predictive ability of each algorithm, using a variety of approaches to validate those that appeared to perform the best.
They found that age was the most influential predictor of performance on the test, followed by diastolic blood pressure, BMI and systolic blood pressure. Adherence to the healthy eating index was less predictive of cognitive performance than blood pressure or BMI but also correlated with better performance on the test.
“Physical activity emerged as a moderate predictor of reaction time, with results suggesting it may interact with other lifestyle factors, such as diet and body weight, to influence cognitive performance,” Khan said.
“This study reveals how machine learning can bring precision and nuance to the field of nutritional neuroscience,” he said.
“By moving beyond traditional approaches, machine learning could help tailor strategies for aging populations, individuals with metabolic risks or those seeking to enhance cognitive function through lifestyle changes.”
The Personalized Nutrition Initiative and National Center for Supercomputing Applications at the U. of I. supported this research.
Khan is a dietitian and an affiliate faculty member of the Division of Nutritional Sciences, the Neuroscience Program and the Beckman Institute for Advanced Science and Technology at Illinois.
About this AI and brain health research news
Author: Diana Yates
Source: University of Illinois
Contact: Diana Yates – University of Illinois
Image: The image is credited to Neuroscience News
Original Research: Open access.
“Predicting cognitive outcome through nutrition and health markers using supervised machine learning” by Naiman Khan et al. Journal of Nutrition
Abstract
Predicting cognitive outcome through nutrition and health markers using supervised machine learning
Background
Machine learning (ML) use in health research is growing, yet its application to predict cognitive outcomes using diverse health indicators is underinvestigated.
Objectives
We used ML models 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.
Methods
Data from 374 adults aged 19–82 y (227 females) were used to develop ML models predicting cognitive performance (reaction time in milliseconds) on a modified Eriksen flanker task.
Features included demographics, anthropometric measures, dietary indices (Healthy Eating Index, Dietary Approaches to Stop Hypertension, Mediterranean, and Mediterranean–Dietary Approaches to Stop Hypertension Intervention for Neurodegenerative Delay), self-reported physical activity, and systolic and diastolic blood pressures. The data set was split (80:20) for training and testing.
Predictive models (decision trees, random forest, AdaBoost, XGBoost, gradient boosting, linear, ridge, and lasso regression) were used with hyperparameter tuning and crossvalidation. Feature importance was calculated using permutation importance, whereas performance using mean absolute error (MAE) and mean squared error.
Results
Random forest regressor exhibited the best performance, with the lowest MAE (training: 0.66 ms; testing: 0.78 ms) and mean squared error (training: 0.70 ms2; testing: 1.05 ms2). Age was the most significant feature (score: 0.208), followed by diastolic blood pressure (0.169), BMI (0.079), systolic blood pressure (0.069), and Healthy Eating Index (0.048). Ethnicity (0.005) and sex (0.003) had minimal predictive effect.
Conclusions
Age, blood pressure, and BMI show strong associations with cognitive performance, whereas diet quality has a subtler effect. These findings highlight the potential of ML models for developing personalized interventions and preventive strategies for cognitive decline.