The Influence of Cognitive Load On Working Memory in Children with Autism

Background: Children with Autism Spectrum Disorder (ASD) exhibit various cognitive and executive deficits, including working memory (WM), which rely largely on the frontal lobes of the brain.  Abnormal maturation of the frontal lobes has been documented in ASD, yet little is known about the link between frontal lobe function and memory impairments in ASD.  We are not aware of any neuroimaging studies of WM in autism; however, recent studies of adults with and without ASD suggest differences in neural activation of WM networks (Luna et al., 2002; Koshino et al., 2008).  Studying the development of WM is extremely important as it may provide insight into the underlying cause of autistic symptomatology, and predict functional and academic outcomes. 

Objectives: To examine the effects of difficulty level on WM capacity and identify how the neural systems underlying WM differ in children with and without ASD using functional neuroimaging.

Methods: Measures of brain activity were acquired using functional magnetic resonance imaging (fMRI) in 24 children with high functioning ASD (7-13 years; IQ scores of 70 or above) and 24 age matched controls using a visuospatial working memory capacity task.  Diagnosis of ASD was confirmed by the Autism Diagnostic Observation Schedule-Generic and Autism Diagnostic Interview-Revised (Lord et al., 2004).  The WM task measures the number of items that can be held in mind across six levels of difficulty.  Neural activation was assessed between groups as a function of cognitive load (task difficulty level). 

Results: Preliminary results show activation in regions associated with WM (dorsolateral prefrontal cortex) in both ASD and control groups.  The autism group exhibited more activity than the control group in the frontal regions of the brain, such as the superior frontal gyrus and medial frontal gyrus, across levels of difficulty.  Children with autism showed more activation than controls in the anterior cingulate cortex (ACC); however the ASD group did not show change in activity as difficulty varied, whereas controls modulated ACC activity with difficulty.  Group differences were also apparent in the precuneus activation.

Conclusions: Patterns of neural activity in children with and without ASD suggest differences in working memory systems between the two groups. Brain activity varied with difficulty level in both groups; however, cognitive demand may only influence selective brain areas in children with ASD.  Findings will increase our knowledge of the neural and cognitive abnormalities of autism, and their links to ASD-related symptoms.  These data will allow us to identify the nature of atypical development, which is critical in establishing age-appropriate interventions that can effectively target working memory functions.