Despite considerable evidence that working memory is involved in children’s mathematical performance (Gathercole and Pickering, 2000; Swanson, 2004), there has been little attempt to investigate possible mathematical benefits of working memory training. Forty-five children in Year 5 (9-10 years old) were given tests of visual-spatial working memory, central executive working memory and a test of addition to assess mathematical performance. Half of the group was then given a six-week programme of working memory training focusing on the central executive. The participants were selected using a matched-pairs design with participants matched as closely as possible for working memory and mathematical performance. Following the working memory training, all the participating children were re-tested on working memory and mathematical performance. The effectiveness of the working memory training to improve working memory functioning and possibly mathematical performance was assessed. Possible implications for the classroom are also discussed.

Working memory is defined as ‘a multi-component, limited capacity system responsible for the concurrent processing and storage of information during cognitive tasks’ (Baddeley & Hitch, 1974; Baddeley, 1986; 2000). The system comprises four components: a phonological store, a visual-spatial store, a central executive and an ‘episodic buffer’ responsible for integrating different types of information There is a considerable body of evidence (e.g. Gathercole and Pickering, 2000; Holmes and Adams, 2006) that working memory is linked to mathematical performance at a range of ages. There is less consensus as to the precise components of the working memory model that are implicated in mathematical functioning. Despite this confusion, the central executive has been shown in a number of studies (e.g. McLean & Hitch, 1999; Bull and Scerif, 2001; D'Amico and Guarnera, 2005) to be important in a range of mathematical situations. The central executive is thought to be important in regulating the flow of information in working memory, for inhibiting unwanted information from working memory (Passolunghi & Siegel, 2001 and 2004) and for other executive functions such as updating the contents of working memory (Passolunghi and Pazzaglia, 2004).

There is, however, relatively little research on the possibility of improving working memory performance through practice. The majority of studies looking at this have been done with specific groups of adults such as those with schizophrenia (e.g. Bell et al, 2003; Wexler et al, 2000). The majority of studies done with children have looked at those with specific attentional difficulties, such as children with ADHD (Klingberg et al, 2002; 2005) rather than average children in an average classroom. Even fewer studies have looked at the possibility that that improvements in working memory could lead to improvements in scholastic performance. Those that have been done suggest that working memory intervention may help with early literacy skills (Maridaki-Kassotaki, K., 2002). There do not appear to be any published studies looking at the possible mathematical benefits of working memory training.

The study sought to answer two research questions:

1. What are the effects on working memory performance of explicit practice and training of working memory tasks?


2. Do any improvements in working memory functioning have benefits for mathematical performance.

The participants in this study were selected from four primary schools in the South West of England. Forty-five children took part, and were given pre-intervention tests of visual-spatial working memory (visual patterns test), central executive working functioning (backward digit recall) and mathematical functioning. In order to avoid possible confounding factors of using a general measure of mathematical performance, the children’s performance in addition was measured.

The visual-spatial working memory task was adapted from the visual-patterns task (Della Sala, 1999) and administered using a computer. The backward digit recall task involves participants in repeating back strings of digits of increasing length in reverse order. The mathematical task consisted of 20 addition questions presented on a PC. The response time was measured as well as the number of errors. The questions were a mixture of single digit + single digit, three single digit numbers, double-digit + single digit (including carrying) and double digit + double digit (including carrying). The children’s predominant calculation strategy was also noted.

The participants were then put into matched pairs within their schools to minimise the impact of possible differences in teaching over the weeks of the intervention. Within each school, the children were matched as closely as possible for central executive function and for performance on the addition task (time and accuracy). They were not necessarily matched for gender, as there appeared to be no significant differences between the working memory or mathematical performance of the boys and girls in the original pre-intervention measures.

Given the importance of the central executive in a number of mathematical tasks (e.g. Bull and Scerif, 2001), the decision was made to focus on this aspect of working memory for the intervention. The working memory intervention took place over six weeks and included the following tasks and objectives:

Week 1 – Using visual strategies to remember lists of objects and strings of digits. This was linked to practice on a concurrent processing and recall task.

Week 2 – Using phonological repetition strategies to remember lists of objects and digits. This was linked to practice on a concurrent processing and recall task.

Week 3 – Updating information. Children were given increasingly long lists of objects and asked to recall the 3, 4 or 5 smallest objects in the list.

Week 4 – Inhibition practice. The children were shown pictures of objects surrounded by ‘distracter’ pictures and encouraged to focus attention on the ‘target’ items.

Week 5 – Concurrent processing and recall. The children played a counting recall game in which they had to count objects and remember the outcome of the counts.

Week 6 – Recap and summary. Further practice on selected tasks.  

Following the intervention period, all the children were re-tested. The stimuli were altered slightly from those of the pre-intervention tasks, but the body of all the tasks was left unaltered. During the intervention period, careful notes were made concerning the children’s memory performance. Their own comments were also noted as they made progress with the backward-digit recall task.

Results.

At the time of writing, the study is in its final week. Observations and notes made during the training period suggest that many of the children have made gains in their working memory skills. It is too early to comment on possible benefits for mathematical performance. It is hoped that the findings of the study will provide some initial evidence about the possibilities of working memory training and whether such training might have benefits in the classroom.