Executive Functions in Asperger's Disorder: An Empirical Investigation of Verbal and Nonverbal Skills

Background: Researchers have investigated the specific abilities of individuals with Asperger’s Disorder (AD).  One construct used in this effort is Executive Functioning (EF).  While research has shown that this population performs more poorly than typically-developing matched controls on many EF tasks, there is a lack of consistency in these results (Calhoun, 2006; Hill, 2004).  This is likely due to the use of inconsistent diagnostic criteria leading to incomparable studies (Klin, Pauls, Schultz & Volkmar, 2005).  This study was designed to investigate if individuals with AD demonstrate a performance bias in favour of verbal Executive Functioning.

Objectives: To empirically determine if individuals with Asperger's Disorder demonstrate a performance bias in favour of verbal Executive Functioning.

Methods: The present study investigated EF in AD using a bottom-up method whereby several EF tasks were administered to 35 adolescents and young adults aged 16-21 with AD and 35 age- and gender-matched controls.  VIQ and PIQ cutoffs of 85 as measured by the Wechsler Abbreviated Scale of Intelligence (WASI) were used to ensure cognitive abilities of all participants fell within the average or above average range.  Two-step cluster analysis was used to derive subgroups of participants based upon performance on several subtests of the Delis-Kaplan Executive Functioning System (DKEFS).  Diagnostic composition of these subgroups was examined to provide empirical evidence of a performance bias towards verbal EF functioning for the AD group.

Results: The participants did not differ in terms of VIQ, PIQ, or FSIQ.  The two-step cluster analysis indicated that the data was best described by a two cluster solution.  The first cluster was primarily comprised by individuals with AD (23 individuals with AsD, 8 controls) while the second primarily contained typically developing controls (9 individuals with AD, 24 controls).  Performance on the EF measures between these groups was investigated by T-tests.  Cluster 2's performance was significantly higher than cluster 1 on Trail Making 4, Verbal Fluency 3, Design Fluency 3, Color Word Identification 3, Word Context, and Tower.  Performance did not differ on the Proverb subtest.

Conclusions: The performance of individuals with AD and typically developing controls on seven specific measures from the DKEFS was empirically examined.  A cluster of participants consisting primarily of individuals with AD and a cluster consisting primarily of typically-developing control participants were derived on the basis of this performance.  Although the performance of the individuals in cluster 2 was significantly above that of cluster 1 on six of the seven subtest components, the hypothesized verbal EF performance bias was not found.  Rather, the clusters were differentiated on the basis of better performance on both verbal and visual subtests except for the Proverb subtest where performance did not differ between the groups.  These results lend continued support for an overall deficit in EF abilities in individuals with AD.