Estimating memory function: Disparity of Wechsler Memory Scale-Revised and California Verbal Learning Test indices in clinical and normal samples

Although a high degree of convergence has been reported between the California Verbal Learning Test (CVLT) and the Wechsler Memory Scale-Revised (3WMS-R), standard scores from these tests may not provide commensurate global estimations of memory function. Data were reviewed from 161 subjects who had taken both tests, including schizophrenic (n=33) and temporal-lobe epilepsy (n=31) patients and 97 normal volunteers. The patient group performed approximately one standard deviation below the normal mean on the WMS-R indexes, but three standard deviations below the normal mean on CVLT indices. The normal volunteers, who were well-educated as a group, performed approximately one standard deviation above the population-based mean on the WMS-R (commensurate with their IQ level), but at the mean of the CVLT reference sample. These findings suggest that the standardization samples for these two tests are not representative of the same population. The CVLT normative reference sample appears to have been higher functioning than the WMS-R sample, and may also have been characterized by a narrower range of memory function. The implications of these findings with respect to the clinical evaluation of memory are discussed.