Abstract

Objective: We investigated the construct validity of the Benton Serial Digit Learning Test total raw score (SDL8-TOT) with demographic and WAIS-III component variables. Method: SDL8-TOT was submitted to a principal component analysis in combination with age and educational level as performance level variables and the four factorial components of the WAIS-III as pattern of performance variables. Results: A four factor solution was derived that explained 83.33% of the variance. The first component showed loadings on Verbal Comprehension (.876), education (.792), and SDL8-TOT (.361). The second component had loadings on Working Memory (.930) and SDL8-TOT (.774). The third component showed loadings on Processing Speed (.941) and age (−.654). The fourth component showed loadings on age (-.607) and Perceptual Organization (.956). The SDL8-TOT score was primarily associated with attention span and working memory variables, and secondarily with verbal reasoning and educational level, but not with age, even though the SDL8 task is normed for age. Conclusion(s): Interpretation of SDL8-TOT scores appear to be enhanced by use of factorial measures that include intellectual and demographic variables other than age, which is the standard measure that is correlated with SDL8-TOT performance level.