Parent ratings of working memory are uniquely related to performance-based measures of secondary memory but not primary memory

Introduction: Existing evidence suggests that performance- and rating-based measures of working memory (WM) correlate poorly. Although some researchers have interpreted this evidence as suggesting that these measures may be assessing distinct cognitive constructs, another possibility is that rating-based measures are related to some but not all theoretically motivated performance-based measures. The current study distinguished between performance-based measures of primary memory (PM) and secondary memory (SM), and examined the relation between each of these components of WM and parent-ratings on the WM subscale of the Behavior Rating Inventory of Executive Function (BRIEF-WM). Because SM and BRIEF-WM scores have both been associated with group differences in attention-deficit/hyperactivity disorder (ADHD), it was hypothesized that SM scores would be uniquely related to parent-rated BRIEFWM scores. Method: Participants were a sample of 77 adolescents with and without an ADHD diagnosis, aged 11 to 15 years, from a midwestern school district. Participant scores on verbal and spatial immediate free recall tasks were used to estimate both PM and SM capacities. Partial correlation analyses were used to evaluate the extent to which estimates of PM and SM were uniquely related parent-rated BRIEF-WM scores. Results: Both verbal and spatial SM scores were significantly related to parent-rated BRIEF-WM scores, when corresponding PM scores were controlled. Higher verbal and spatial SM scores were associated with less frequent parent-report of WM-related failures in their child's everyday life. However, neither verbal nor spatial PM scores significantly related to parent-rated BRIEF-WM scores, when corresponding SM scores were controlled. Conclusion: The current study suggested that previously observed low correlations between performance- and rating-based measures of WM may result from use of performance-based WM measures that do not capture the unique contributions of PM and SM components of WM.