Submission 424
Working Memory Development in Children: Using Large-Scale Adaptive Learning Data in the Mathematical Domain
SymposiumTalk-01
Presented by: Seyma Nur Ertekin
Working memory (WM) develops throughout childhood and is a strong predictor of mathematical ability. To better understand the trajectory and constraints of cognitive development, it is essential to examine how WM changes during childhood. To this end, we analyze data from over 100,000 Dutch primary school children (ages 6–12) who completed a spatial serial recall task within an adaptive learning platform, enabling the study of working memory in a regular educational context.
We first examine the development of WM capacity across primary school age and highlight difficulties with estimating WM capacity based on the set size and structure of the items played. Next, we investigate the relationship between WM and mathematical abilities using both cross-sectional and longitudinal data. Cross-sectional correlations show associations between WM and mathematical domains, with domain-specific differences across primary school years. Furthermore, longitudinal analyses examine the role of WM in mathematical learning over time. Together, these findings provide valuable insights into the development of WM processing in children and underscore the value of large-scale online educational data to better understand the relationships between different ability domains.