Working memory and high-level cognition in children: An analysis of timing and accuracy in complex span tasks.

This study examined working memory (WM) using complex span tasks (CSTs) to improve theoretical understanding of the relationship between WM and high-level cognition (HLC) in children. A total of 92 children aged 7 and 8 years were tested on three computer-paced CSTs and measures of nonverbal reasoning, reading, and mathematics. Processing times in the CSTs were restricted based on individually titrated processing speeds, and performance was compared with participant-led tasks with no time restrictions. Storage, processing accuracy, and both processing and recall times within the CSTs were used as performance indices to understand the effects of time restrictions at a granular level. Restricting processing times did not impair storage, challenging models that argue for a role of maintenance in WM. A task-switching account best explained the effect of time restrictions on performance indices and their interrelationships. Principal component analysis showed that a single factor with all performance indices from just one CST (counting span) was the best predictor of HLC. Storage in both the participant-led and computer-paced versions of this task explained unique and shared variance in HLC. However, the latter accounted for more variance in HLC when contributions from processing time were included in the model. Processing time in this condition also explained variance above and beyond storage. This suggests that faster processing is important to keep information active in WM; however, this is evident only when time restrictions are placed on the task and important when WM performance is applied in broader contexts that rely on this resource.