An examination of the Wechsler Adult Intelligence Scale (WAIS) subtests from a neuopsychological perspective
The primary purpose of this study was to determine the kind of neuropsychological information that can be obtained from an investigation of the Wechsler Adult Intelligence Scale (WAIS) subtests. Additionally, there was an examination of the shared variance between the WAIS and the Halstead-Reitan Neuropsychological Battery (HRNB). The archival data collected from the files of '220 females and 188 males from a large midwestern medical center were used. They had been originally diagnosed with objective research criteria.All eleven subtests of the WAIS and the HalsteadReitan Neuropsychological Battery were administered to the subjects in the years between 1981 and 1983. Thirteen scores were obtained from the HRNB measures. Statistical analyses of the results made use of the techniques of multiple regression and canonical correlation.The individual WAIS subtests were examined for the neuropsychological information they provided. Globally, three HRNB measures, APHASIA, RHY, AND CAT-TOT contributed significantly to a majority of the regression equations for the WAIS subtests. Their presence suggested that language skills, an auditory attention factor, and a general intellective factor were being tapped (Dean, 1985a).A canonical correlation was computed. The results yielded one significant correlation between the linear components of the WAIS and the HRNB tests. Only canonical variates with weights of +/- .2 were considered large enough for interpretation. The WAIS subtests meeting the .2 criteria included Block Design, Digit Symbol, and Similarities, while the HRNB measures meeting criteria were APHASIA and CAT-TOT. Therefore, it would appear that the significant variables measured the general (g) factor as in Spearman's research (1927). According to the Stewart and Love formula (cited in Pedhazur, 1982), the variability of the WAIS did overlap with the HRNB, and their relationship was symmetrical.This research demonstrated that the measured tasks from the WAIS were a complex of underlying constructs. The verbal portion of the WAIS was shown to be less highly related to the HRNB variables than the performance portion of the scale. The WAIS and HRNB do offer nonredundant information concerning the impaired and unimpaired adult's cognitive functioning.