Traditional approaches to intelligence and cognitive assessment have been criticized because they assess an individual’s prior knowledge rather than their aptitude to learn. Dynamic assessment has been proposed to address this limitation by integrating forms of learning into the assessment process. Dynamic assessment has been applied to many different arenas including intelligence, speech and language, and areas of achievement, and has generally been found to predict additional variance in the criterion measure beyond that which is predicted by the static measures. However, the variables that potentially moderate this additional variance have not been clearly explored. The purpose of this study is to evaluate at least three moderating variables (age, modality, type of dynamic measure) that may interact with the contribution of dynamic assessment of working memory performance in the predictions of reading and math performance. Three research questions are proposed which ask (1) does dynamic assessment of working memory contribute unique variance in predictions of reading and math performance above that of the static assessment, (2) does age moderate the effectiveness of dynamic assessment of a domain general construct of working memory in predictions of achievement measures, and (3) does performance on visual and verbal measures of working memory interact with age when predicting reading and math performance. Results of the study find support that dynamic assessment contributes unique variance in predictions of reading and math achievement. However, these results were qualified since age x dynamic interactions emerged when analyzing working memory as both a domain general and domain specific construct. The majority of these interactions were limited to predictions of math achievement. Implications of these findings and limitations of the study are discussed.