Schooling duration rather than chronological age predicts working memory between 6 and 7 years: memory maestros study

Abstract

Objective: Low working memory (WM) is strongly linked with poor academic outcomes. WM capacity increases across childhood but how exposure to school is associated with WM development is not known. We aimed to determine extent to which chronological age and schooling duration are associated with WM at the population level. Methods: In 2012, children in Grade 1 (the second year of formal schooling in Victoria, Australia) from 44 schools in metropolitan Melbourne were recruited. Assessments occurred over the entire school year, with schools quasi-randomly allocated to one of the 4 school terms. WM (primary outcome) was measured using 2 subtests from the computerized Automated Working Memory Assessment: Backwards Digit Recall (verbal) and Mister X (visuospatial). Linear regression was used to examine relationships of WM with time in school and age. Results: Of the 1765 who provided consent, 1727 children (97.9%) had WM assessed throughout the 2012 school year. WM scores became steadily higher over the course of the year. Thus, scores were .77 and .53 SDs higher in Term 4 than Term 1 for verbal and visuospatial WM, respectively (p values for trend for both scores <.001); conclusions were unchanged when adjusted for age and potential confounders. Conversely, age associations attenuated fully once adjusted for school duration. Conclusions: Our results demonstrate, for the first time, that the developmental increases in WM are strongly associated with time spent in the classroom, above and beyond chronological age.