Sources of Variable Functionality of the Execution/Observation Matching System in ASD

Thursday, May 17, 2012
Sheraton Hall (Sheraton Centre Toronto)
10:00 AM
R. Bernier1, B. Aaronson2 and J. McPartland3, (1)University of Washington, Seattle, WA, (2)University of Washington, Seattle, WA, United States, (3)Yale Child Study Center, New Haven, CT

Attenuation of the EEG mu rhythm is reliably observed at central scalp electrodes during both the execution and observation of human motion, reflecting activity in an execution/observation matching system. Disruptions to this mechanism are hypothesized to contribute to social deficits in autism. Discrepant findings have been reported, with some studies reporting atypical activation within the mu range in autism and others demonstrating intact activation. Phenotypic and developmental variability in samples of individuals with ASD studied in this work has been posited to account for these heterogeneous results.  


To (a) improve upon prior research by examining activity in the execution/observation matching system in a sample of well characterized children with ASD within a constrained age range and to (b) systematically examine the influence of additional phenotypic characteristics that may contribute to variability in mu attenuation. 


46 children (23 ASD, 23 TYP) between 5 and 9 years of age observed and executed simple grasping actions while continuous EEG was recorded with a high density EGI sensor net. All ASD children met ADOS, ADI, and clinical criteria. The TYP group consisted of age and gender matched typically developing children without a family history of ASD. EEG mu rhythm was characterized in two ways: 1) Ratio of power during observe and execute conditions over resting baseline in 8-13 Hz band from centrally located electrodes (Bernier et al, 2007); 2) Ratio of power in a 2 Hz band identified as the peak difference between execute and resting baseline for each participant (Muthukumaraswamy et al, 2004).  Groups were compared with separate ANOVAs for each mu calculation method, and, within the ASD group, regression analysis was applied to examine relations between phenotypic characteristics (IQ, ADOS scores, and ADI scores) and mu rhythm attenuation. Finally, children failing to demonstrate mu rhythm attenuation were compared to those who did using the above variables.


The ASD group showed reduced attenuation of the EEG mu rhythm during observation but not execution relative to the TYP group (p<.05) in terms of individualized mu rhythm frequencies but not the gross 8-13 Hz band approach. Both parent report (ADI, p<.05) and clinician rating (ADOS, p<.01) of increased repetitive behaviors predicted decreased mu rhythm attenuation, but IQ, social, and communication domain scores on the ADOS and ADI were unrelated. Attenuation was observed in all TYP children; however a portion of the ASD group (N=5) showed a complete absence of attenuation. 


Independent measures of repetitive behaviors, based on both parent report and clinical observation, were predictive of activation in the execution/observation matching system in children with ASD. Furthermore, while, in general, children with ASD showed relatively decreased activity in this system, a subset of children with ASD exhibited absolute absence of activity, a pattern not observed in TYP children. Findings offer insight into the heterogeneous results observed in prior work and clarify the importance of evaluating neural phenotypes in the context of well-characterized behavioral phenotypes. 

See more of: Neurophysiology II
See more of: Neurophysiology
See more of: Brain Structure & Function
| More