Working memory predictors of mathematics across the middle primary school years

Abstract

Background: Work surrounding the relationship between visuospatial working memory (WM) and mathematics performance is gaining significant traction as a result of a focus on improving academic attainment.

Aims: This study examined the relative contributions of verbal and visuospatial simple and complex WM measures to mathematics in primary school children aged 6-10 years.

Sample: A sample of 111 children in years 2-5 were assessed (M_age = 100.06 months, SD = 14.47).

Method: Children were tested individually on all memory measures, followed by a separate mathematics testing session as a class group in the same assessment wave.

Results and conclusions: Results revealed an age-dependent relationship, with a move towards visuospatial influence in older children. Further analyses demonstrated that backward word span and backward matrices contributed unique portions of variance of mathematics, regardless of the regression model specified. We discuss possible explanations for our preliminary findings in relation to the existing literature alongside their implications for educators and further research.

Keywords: children; mathematics; verbal; visuospatial.

Figure 1

Instructions for the verbal dual task.

Figure 2

Variance decomposition. WM‐V = verbal working memory, WM‐VS = visuospatial WM.

Figure 3

Loadings and correlations for Model 1.

Figure 4

Loadings and correlations for Model 2. Correlations of 1 or higher indicate that factors are not empirically distinguishable.

Figure 5

Loadings and correlations for Model 3. Correlations of 1 or higher indicate that factors are not empirically distinguishable.

Figure 6

Fisher's Z transformed correlations for the relationship between mathematics and dual task spatial as function of age. Error bars represent 95% confidence intervals, while grey diamonds represent predicted effects at each age.