A 9-WORD DEMENTIA VERSION OF THE CALIFORNIA VERBAL-LEARNING TEST

The validity of an experimental 9-word version of the California Verba l Learning Test (CVLT) was investigated. The construct validity of the 9-word CVLT was assessed with a principal component analysis (PCA) on a sample of nondemented subjects. A three - factor solution, related to immediate free recall, delayed recall and recognition, and intrusio n errors was produced. Construct validity was also assessed by compari ng subjects with probable Alzheimer's disease (AD), probable ischemic vascular dementia (IVD), and a normal control (NC) group. AD subjects evidenced little learning on immediate free recall test conditions, ma de no improvement on cued recall or recognition test conditions, and p roduced copious amounts of intrusion errors. IVD subjects performed si milarly to AD subjects on immediate free recall test conditions, but i mproved on all cued recall and recognition test conditions, and made s ignificantly fewer intrusion errors. The NC group was vastly superior to both groups of demented subjects on all CVLT indices. These results indicate that the 9-word CVLT has good psychometric properties, and t hat it may be a reasonable alternative to the standard 16-word CVLT am ong subjects who are obviously impaired.