WISC-IV Vs. WISC-III: Cognitive Profile in Autistic, Asperger and Typically Developing Children

Background:  The 3^rd edition of the Wechsler Intelligence Scale for Children (WISC-III) has revealed consistent differences in the cognitive profile of autistic versus Asperger children. These two subtypes of the autistic spectrum mainly differ on the presence or absence of speech delay and visuospatial strengths. Does the 4^th edition of WISC yield the same cognitive profiles in autism spectrum children as the previous one, despite many changes in the structure of the test?

Objectives:  To compare WISC-III and WISC-IV cognitive profiles in Asperger, autistic, and typically developing children.

Methods:  22 autistic, 15 Asperger and 16 typically developing children (6-15 years; IQ 80-120) completed the WISC-IV. They were individually matched on age and Full-Scale IQ to 22 autistic, 15 Asperger and 16 typical children who completed the WISC-III. Full-Scale IQ differed between the typical group (M 106.6, SD 9.4) and the autistic (M 97.1, SD 9.8) and Asperger (M 98.2, SD 10.1) groups. Two sets of analyses of variance assessed (1) discrepancies across Wechsler indexes; (2) strengths and weaknesses, by comparing a participant’s performance on a given subtest with his mean performance on all subtests.

Results:  Profiles obtained on WISC-IV were consistent with those obtained on WISC-III. First, on WISC-IV Asperger children obtained a significantly higher Verbal Comprehension Index (VCI 110) than Perceptual Reasoning Index score (PRI 101), whereas the autistic group presented the opposite profile (PRI 111 vs. VCI 90). For the autistic group, the gap between VCI and PRI has nearly tripled relative to WISC-III results. No difference between VCI and PRI was found in the typical group. The new indexes of Working Memory and Processing Speed were the lowest in the three groups, but the largest gap between Processing Speed and Full-Scale IQ was found in Asperger children.

At the subtest level, autistic children displayed a significant strength on Block Design in both WISC versions, with an additional strength on the new WISC-IV motor-free visual reasoning subtest, Matrix Reasoning. A significant weakness was revealed on the Comprehension subtest in both WISC editions. As for Asperger children, significant strength on the Vocabulary subtest was present in both WISC versions, while strength in Similarities reached significance only in WISC-IV. For both WISC editions, the main weakness of Asperger children was on the Code subtest. 

Conclusions:  WISC-IV cognitive profiles are consistent with those obtained with the WISC-III. Discrepancies between VCI and PRI are more representative of the autistic spectrum than of typical children. Also, cognitive profiles of autistic versus Asperger children might be more differentiated with the WISC-IV than they were with WISC-III, due to their respective Matrix Reasoning and Similarities additional strengths. The greater difference between VCI and PRI obtained on WISC-IV (compared to WISC-III) for autistic children could also reflect a diminution of motor demands for non-verbal tasks, thus better revealing their visuospatial strengths. With the upcoming DSM-V and associated changes in nomenclature, using the WISC-IV can help highlight different cognitive profiles within the autistic spectrum and target educative methods accordingly.