16117
Atypical Neural Response to Perceptual Saliency in Children with ASD

Friday, May 16, 2014
Atrium Ballroom (Marriott Marquis Atlanta)
C. J. Vaidya1,2, X. You1, M. Norr1, E. R. Murphy1, W. D. Gaillard2 and L. Kenworthy2, (1)Psychology, Georgetown University, Washington, DC, (2)Children’s Research Institute, Children's National Medical Center, Washington, DC
Background: Atypical top-down control and bottom-up perceptual processing has been posited to contribute to socio-emotional and cognitive deficits in children with Autism Spectrum Disorder (ASD). The top-down or goal-driven control of attention engages the dorsal attention network comprising superior frontal and dorsal parietal regions whereas the bottom-up or stimulus-driven capture of attention engages the ventral attention network comprising temporo-parietal junction and inferior frontal cortex. 

Objectives: To examine the two attention networks, functional magnetic resonance imaging (fMRI) was performed while ASD and control children performed an ongoing task, in the context of novel or familiar distractors which appeared unexpectedly (bottom-up condition) or under monitoring/response instructions (top-down condition). 

Methods: Eighteen 8-14 year old ASD and 22 control children with same age, gender and IQ distribution underwent two fMRI runs during a sustained attention task in which they pressed a button with their right hand for a triangle (target stimuli) among serially presented squares, circles, and rectangles.  In both runs, distracters (small flickering shape) were flashed in the periphery, which were either Familiar (high frequency) or Novel (rare).  In the bottom-up run, distractors were unexpected as children were not forewarned, whereas in the top-down run, they had to monitor for a target distractor and respond with their left hand while performing the central task with their right hand.

Images were slicetime and motion corrected, normalized to EPI template and resliced to 3mm, smoothed with 8mm FWHM in SPM8.  Contrasts of interest were Novel vs. Null, and Familiar vs. Null. The effects of any residual micromovements were removed by including six motion parameters and framewise displacement as regressors in the single subject-level analysis.  For each run separately, subject-specific contrast maps were entered into separate ANOVAs in SPM8 with Group and Condition as categorical variables and age and FIQ as covariates of no interest, with interaction examined at p<.005, 10 voxels within an anatomical mask including regions of the ventral and dorsal attention system.

Results: During bottom-up attention, ventral attention regions in left inferior frontal cortex (BA 47 and 46), were more responsive to Familiar than Novel distractors in ASD children whereas they were more responsive to Novel than Familiar distractors in control children.  Further, dorsal attention regions in parietal cortex (inferior and superior, BA 7) were also more responsive to Familiar distractors in ASD children, whereas they were equally responsive to both Novel and Familiar distractors in control children.  During top-down attention, a dorsal attention region in frontal cortex (BA 6), was more responsive to Familiar than Novel distractors in ASD children.

Conclusions: The ventral and dorsal attention networks were engaged qualitatively differently in ASD and control children such that they were driven by familiarity rather than by novelty.