The convergent effects of primary school physical activity, sleep, and recreational screen time on cognition and academic performance in grade 9

Abstract

Lab-based experiments and randomized controlled trials consistently demonstrate improvements in youth cognition following physical activity (PA), while cross-sectional studies suggest that sedentary behavior (especially recreational screen time [RST]) and poor sleep are inversely related to cognition. However, little is known about how these 24-h movement behaviors-sleep, PA, and sedentary behavior-converge to affect youth cognition. Therefore, the purpose of this study is to test the associations between childhood 24-h movement behaviors and adolescent cognition using a longitudinal design and examine moderating effects of each behavior. This study utilized structural equation modeling with data from the NICHD Study of Early Child Care and Youth Development (N = 1,364, 52% female, 80% White). Independent variables-sleep, RST, and PA-were collected in grade 5. Dependent variables of cognitive and academic performance were collected at grade 9, including the Stroop task, Woodcock-Johnson, and Tower of London. Grade 5 PA was inversely associated with grade 9 cognition, but this relationship was no longer significant once grade 5 cognition was controlled for in analyses. Grade 5 sleep was positively related to grade 9 cognition, whether baseline cognition was controlled for or not. Finally, grade 5 RST was inversely related to cognition and academic performance, regardless of whether baseline values were controlled. Moderation analyses showed the relationship between grade 5 RST and grade 9 cognition was moderated by grade 5 PA, while the relationship between grade 5 PA and grade 9 cognition was moderated by grade 5 sleep. In each case, more PA and sleep blunted the negative relationships. These findings extend evidence that greater sleep promotes cognition and greater RST impairs cognition, by affirming these relationships over a longer period. They extend the evidence by demonstrating that the longitudinal relationship between individual 24-h movement behavior and cognition is moderated by other behaviors.

Keywords: development; device assessed data; longitudinal design; movement behaviors; self-report data; structural equation modeling; youth.