The research team, led by Xinyu Fan and Jing Liao of Sun Yat-sen University together with colleagues, analyzed data from 2,516 US adults aged 60 and older who participated in the 2011–2014 cycles of the National Health and Nutrition Examination Survey, or NHANES. This sample represents approximately 45.7 million older US adults. Unlike most previous studies, which have treated sleep, sedentary behavior, light physical activity, and moderate-to-vigorous physical activity as independent predictors, the investigators employed compositional data analysis, a statistical framework designed for variables that are inherently co-dependent because they must sum to a fixed total—in this case, the 1,440 minutes of every day.

The technical logic of compositional data analysis is central to the study’s power. Because time spent in one behavior necessarily displaces time in another, standard regression models can produce misleading estimates. The researchers instead transformed the daily time-use composition using isometric log-ratio coordinates, allowing each behavior to be evaluated relative to the geometric average of all the others. Physical activity was measured objectively using wrist-worn ActiGraph accelerometers worn for seven consecutive days, with raw acceleration processed into Monitor-Independent Motion Summary units. Sleep and wake periods were identified with validated algorithms, and wake minutes were classified as sedentary behavior, light physical activity, or moderate-to-vigorous physical activity using age- and sex-specific thresholds that account for declining physical capacity in older adults.

The study’s most innovative methodological step was disaggregating total sedentary time into mentally active and mentally inactive components. Because accelerometers cannot distinguish the cognitive content of sitting, the researchers used proportional weights derived from the Global Physical Activity Questionnaire, in which participants reported their daily hours of television viewing and computer use. Television watching served as the proxy for mentally inactive sedentary behavior, while computer use—which in this older cohort typically involves managing finances, reading news online, or playing digital games—served as the proxy for mentally active sedentary behavior. The remaining sedentary time was classified as an unmeasured residual category encompassing activities such as reading, socializing, dining, and passive travel.

Cognitive function was assessed across three domains using well-validated instruments: memory through the Consortium to Establish a Registry for Alzheimer’s Disease Word Learning test, language through the Animal Fluency Test, and executive function through the Digit Symbol Substitution Test. Standardized scores from the three tests were combined into a global cognition score. In the initial four-component model, greater relative time in moderate-to-vigorous physical activity and total sedentary behavior were both positively associated with global cognition, while relative time in sleep and light physical activity showed negative associations—a pattern the authors attribute to the compositional trade-offs inherent in a fixed 24-hour budget.

The picture transformed when sedentary time was split into its subtypes. Computer use emerged as positively associated with all cognitive domains, with a coefficient of 0.045 for global cognition, while television watching was significantly negatively associated with both language and global cognition. Remarkably, neither the residual other sedentary category nor sleep remained significantly associated with cognition in this expanded model, suggesting that the apparent cognitive effects previously attributed to sleep and total sitting were largely absorbed by the divergent signals of the specific sedentary subtypes. Moderate-to-vigorous physical activity retained its positive association across all domains, accounting for only about 1.1 percent of daily time—roughly 16 minutes—underscoring how small doses of intense movement carry outsized cognitive weight.

The isotemporal substitution analyses added a sobering dimension of asymmetry. Reallocating 10 minutes per day from sleep to computer use or moderate-to-vigorous physical activity was associated with predicted increases in global cognition of 0.034 and 0.059, respectively. But the reverse change told a steeper story: removing 10 minutes of moderate-to-vigorous physical activity was linked to a predicted decline of 0.103 in global cognition—nearly twice the magnitude of the corresponding gain. Because the baseline geometric means for these behaviors were so small, the models predicted that even modest reductions would produce disproportionately large cognitive costs. The authors emphasize that these cross-sectional estimates represent hypothetical, associational differences rather than proven causal effects.

The biological interpretation offered by the researchers is grounded in established neuroscience. Moderate-to-vigorous physical activity promotes cardiovascular health, enhances cerebral blood flow, and stimulates neurobiological pathways including the upregulation of brain-derived neurotrophic factor and hippocampal neurogenesis—mechanisms not sufficiently activated by light activity. For mentally active sedentary behavior, the authors invoke cognitive disuse atrophy theory: interactive tasks such as computer use engage repetitive cognitive processing that supports cognitive maintenance, and cognitively demanding activities increase cerebral blood flow and metabolism. Television watching, by contrast, delivers a predominantly one-way flow of information that lacks the interactive engagement needed to activate language-processing brain regions, offering a plausible explanation for its negative association with verbal fluency.

The public health implications are pragmatic and potentially transformative. The authors propose a hierarchy of recommendations: increasing moderate-to-vigorous physical activity should remain the primary target for cognitive health at all ages. But for older adults with mobility limitations who cannot increase physical activity, substituting mentally inactive sedentary time with cognitively engaging activities such as computer use represents a viable complementary strategy. The researchers caution that their cross-sectional design cannot exclude reverse causality—people experiencing early cognitive decline may withdraw from demanding activities and drift toward passive screen time—and that self-reported sedentary subtypes carry measurement imprecision. Still, the core message stands: sedentary behavior is not a homogeneous construct, and its cognitive nature profoundly matters. Preserving even short bouts of brisk activity and swapping passive screen time for reading or interactive tasks may be among the most accessible tools available for protecting the aging brain.

Subject of Research: Associations between 24-hour movement behavior compositions, sedentary behavior types, and cognitive function in older adults

Article Title: Sedentary behavior type matters: compositional analysis of 24-hour movement behaviors and cognitive function in older adults

Article References: Sedentary behavior type matters: compositional analysis of 24-hour movement behaviors and cognitive function in older adults. (n.d.). https://doi.org/10.1186/s44167-026-00105-2

Image Credits: AI Generated

DOI: 10.1186/s44167-026-00105-2

Keywords: sedentary behavior, cognitive function, older adults, 24-hour movement behaviors, compositional data analysis, physical activity, television watching, computer use, NHANES, accelerometry, cognitive aging, isotemporal substitution