People who walk more each day tend to score better on several brain tests

By Vijay Kumar MalesuReviewed by Susha Cheriyedath, M.Sc.Oct 14 2026

Using years of Fitbit activity and blood pressure data, researchers investigated how daily movement and vascular health relate to attention, cognitive control, and decision-making.

Study: Physical activity and blood pressure as predictors of cognitive function in the all of us research program. Image Credit: Branislav Nenin / Shutterstock

In a recent study published in the journal Scientific Reports, researchers found positive associations between physical activity (PA) and cognitive performance among United States (US) adults, with greater PA linked to higher scores on selected cognitive tasks.

The study found that higher systolic blood pressure (SBP) was associated with poorer performance on the Delay Discounting Task (DD-Task), and SBP statistically accounted for a small proportion of the association between PA and performance on this task. Because cognition was measured only once, the findings cannot establish whether greater activity prevents cognitive decline or whether lowering blood pressure improves cognition.

Cognitive skills related to memory, attention, language, planning, and visuospatial processing are crucial for performing everyday tasks and maintaining quality of life. Cognitive performance can decline with age. Many researchers are investigating strategies to prevent or delay cognitive impairment and dementia among older adults.

Recent studies suggest that physical activity may positively influence cognitive outcomes, whereas high blood pressure (BP) has been associated with negative effects. Many previous studies assessed single exposures over shorter periods and relied on self-reported activity measures, highlighting the need for more objective evaluation to clarify the findings in larger and more diverse populations.

About the study

In the present study, the researchers explored PA and BP as cognitive health predictors for All of Us (AoU) program participants aged between 18 and 93 years.

Participants wore Fitbit trackers to record daily step counts. The available physical activity and blood pressure records covered a 14-year period from 2009 to 2023.

The team evaluated cognitive performance among 3,093 AoU participants once, from September 2023 onward, using the Flanker-Task, DD-Task, Gradual Onset Continuous Performance Task (Grad-CPT), and Multiracial Emotional Recognition Task (Emorecog-Task).

The tests assessed sustained attention, cognitive control, decisions involving immediate versus delayed rewards, and facial emotion recognition. The analysis excluded participants with diastolic blood pressure (DBP) below 30 mm Hg or above 140 mm Hg.

The team also accounted for age, educational attainment, marital status, income, and country of origin. Using linear mixed-effects (LME) models, they characterized changes in PA, SBP, and DBP over time.

The team then used participant-specific estimates from these models to examine how PA and BP related to cognitive performance through linear regression analyses. In post-hoc analysis, they tested whether SBP statistically mediated the association between PA and cognitive performance.

Results and discussion

Among the study participants, the mean age was 58 years; most were aged ≥60 years (57%), White (75%), female (68%), married (59%), US-born (92%), had health insurance (98%), and had some college education or higher (92%).

The team found positive associations of baseline PA with Flanker-Task, Grad-CPT, and DD-Task, whereas baseline SBP showed negative associations with the DD-Task. DBP was not significantly associated with any of the four cognitive tests.

The team grouped participants into three clusters based on their step-count patterns over time. Cluster A averaged 5,108 steps, cluster B 8,875 steps, and cluster C 14,161 steps.

Participants in the latter two clusters recorded higher scores on the Grad-CPT, Flanker-Task, and DD-Task than those from the first cluster, suggesting a link between greater physical activity and better cognitive performance. Cluster C had the highest Grad-CPT and DD-Task scores, while Cluster B had the highest Flanker-Task scores.

After adjusting for other factors, the researchers found no significant associations between BP and performance on the emotional recognition, Flanker, or Grad-CPT tasks.

The authors suggested that differences in study methods and strict statistical criteria may explain these findings. Unlike some earlier studies that compared people with high and normal BP, this study examined BP across a continuous range.

Participants with estimated daily activity levels 1,000 steps higher had Grad-CPT and Flanker-Task scores that were 0.31 and 0.26 points higher, respectively, and also performed better on the DD-Task. Each 1 mm Hg increase in SBP was associated with poorer DD-Task performance. Higher PA was associated with lower SBP, and both were independently associated with higher DD-Task scores. Changes in PA and SBP over time were not significantly associated with cognitive performance. The observed effect sizes were small.

The mediation analysis suggested that lower SBP could statistically account for a small part of the link between physical activity and DD-Task performance. The researchers estimated that SBP accounted for 4.8% of this association, while the relationship between activity and DD-Task performance persisted after adjusting for SBP.

The observational data cannot establish that lowering blood pressure improves cognition. Mechanisms involving reduced systemic inflammation, improved glucose metabolism, and greater neuroplasticity may help explain the associations, but these possibilities require further investigation.

Conclusion

The findings suggest that higher daily step counts and lower SBP are associated with better performance on selected cognitive tasks, especially those linked to sustained attention, reward valuation, and cognitive control. Because people with stronger pre-existing cognitive function may have been more likely to remain physically active, reverse causation cannot be excluded.

Future research should include repeated cognitive assessments to determine whether physical activity is associated with changes in cognition over time or a lower risk of dementia.

Accounting for factors such as obesity and smoking in further research could also help clarify how physical activity and BP relate to cognitive function in older adults. The sample's high educational attainment and insurance coverage may also limit how broadly these findings apply.