Theses and Dissertations

Date of Award


Document Type


Degree Name

Doctor of Psychology (PhD)


Center for Psychological Studies

First Advisor

Charles Golden

Second Advisor

Ed Simco

Third Advisor

Barry Schneider


Intelligence testing, Neuropsychological assessment, Neuropsychology measures


The current study examined the WAIS-IV and how the changes to the test may impact the measure's usefulness in neuropsychological evaluations. It was hypothesized that the WAIS-IV would be a significantly better predictor of performance on the neuropsychological measures of the Category Test, Finger Tapping Test, Trail Making Test, and Wisconsin Card Sorting Test over the WAIS-III. The mixed clinical sample came from an archival database of volunteer research participants and individuals clinically referred to a university outpatient facility. A total of 91 participants were administered the WAIS-III and WAIS-IV as part of a larger neuropsychological battery. The results of the current study found that both the WAIS-III and the WAIS-IV were able to account for a significant amount of the variance in performance on the neuropsychological measures, with the exception of the FTT dominant and non-dominant hands, where only the WAIS-IV was able to significantly account for the variance in performance on the measures. Using the Alf and Graf (1999) model, there were no significant R2 differences between the WAIS-III subtests and WAIS-IV at the .01 significance level. Thus, the WAIS-IV did not provide a better model for predicting performance on any of the neuropsychological measures. It should be noted that the small sample size of the current study may have inflated the R2, particularly in the WAIS-III models, which could have masked greater R2 differences between the two models. While the publishers endeavored to make the WAIS-IV a better measure of processing speed, working memory, and fluid reasoning, these goals were largely unmet. The analyses of the WAIS-IV working memory subtests, showed that the sequencing component added to the Digit Span subtest did not add to the relationship with neuropsychological measures with working memory components. The analyses of the WAIS-IV processing speed subtests showed that the Coding subtest of the WAIS-IV was a better measure of processing speed than the WAIS-III version, but this was not found for the PSI as a whole. Changes to Symbol Search did not show any improvement in the relationship to neuropsychological measures. One interesting finding was that the new subtest of Visual Puzzles does appear to add to the relationship with neuropsychological measures over the other subtests of the WAIS-IV. Visual Puzzles was consistently the highest correlated PRI subtest with the neuropsychological measures, with the exception of the WCST. The subtest appears to assess a wide range of abilities outside of the spatial reasoning skills purported by the test publishers. Specifically, the subtest was correlated with measures of processing speed, executive skills, and motor speed/reaction time. Thus, clinicians should use caution and examine all possible options when evaluating poor performance on this new subtest. Since none of the WAIS-IV models were able to significantly predict performance on any of the neuropsychological measures over the WAIS-III models, it would appear that the WAIS-IV as a whole is not a better neuropsychological measure than its predecessor. Despite being the gold standard for intellectual assessment, the WAIS-IV appears to add little to clinical utility over the WAIS-III outside of shorter administrative time. Clinicians are advised to continue using neuropsychological measures to assess processing speed, working memory, and higher order cognitive skills in conjunction with the WAIS.

  Link to NovaCat

Included in

Psychology Commons