The Nature of Working Memory Impairments In Children and Adolescents with Autism Spectrum Disorder

Background:   

Executive functioning, involving attention shifting, inhibition, working memory (WM), and planning, is thought to be impaired in autism spectrum disorders (ASD; Eigsti, in press). However, because executive function encompasses such a wide range of processes, the exact nature of these impairments remains unclear. There is controversy regarding the presence or specificity of WM impairments in ASD, with some studies finding deficits (Bennetto, et al, 1996; Williams, et al., 2005) and others reporting adequate performance (Dawson, et al., 2002). The type of WM assessed (verbal vs. spatial) and task complexity likely amplify these discrepancies. This study aims to better understand WM abilities in children and adolescents with ASD in a large, carefully-matched, high-functioning sample.

Objectives:    

The current study examined WM abilities in ASD, utilizing standardized assessments of verbal and spatial WM abilities, which varied in complexity. We hypothesized that children with ASD will display WM deficits in relation to their typically-developing (TD) peers, particularly for spatial WM as consistent with previous research (e.g., Williams, et al. , 2006), and that deficits would be most marked for the most complex skills.

Methods:  

Children with ASD (n = 19) and TD controls (n = 24) matched on age (range 9-16; mean 12.9 years), gender, IQ, and receptive vocabulary completed four WM tasks. 1) Spatial short term memory was examined using the Finger Windows subtest from the Wide Range Assessment of Memory and Learning (WRAML); 2-3) Verbal WM was examined using Letter-Number Sequencing from the Wechsler Intelligence Scale for Children-IV Edition (WISC-IV), and with a modified version of the Competing Language Processing Task (CLPT), which required participants to respond to true/false statements and remember the last word from each; 4) Verbal short term memory was examined via non-word repetition. A repeated-measures ANOVA on four WM measures, with appropriate post-hocs, examined performance by group, and correlational analyses examined relationships between WM performance and standardized measures of language (Clinical Evaluation of Language Fundamentals, 4^th Edition; CELF-4) and parent-report measures of executive function (Behavior Rating Inventory of Executive Function; BRIEF), behavior (Child Behavior Checklist; CBCL), and ASD symptom severity (Social Responsiveness Scale; SRS).

Results:  

There was a main effect of Group on WM measures, p < .001; specifically, the ASD group had lower scores across verbal and spatial WM tasks. For both groups, better language skills were associated with higher scores on verbal, p = .001, and spatial, p = .04, WM, and behavioral regulation (as measured by the BRIEF) was associated with verbal WM, p = .03. For the ASD group, there were trends for poorer spatial WM to associate with symptom severity (SRS), p = .10, and attentional problems (CBCL), p = .06.

Conclusions:   

Individuals with ASD displayed notable impairments in WM across verbal and spatial domains. Behavioral regulation and language ability for both groups, and inattention and symptom severity for the ASD group, were related to WM performance. Findings highlight the need to incorporate WM goals into interventions for individuals with ASD.