Anatomical Connectivity Abnormalities and Social Perception Deficits in Children with ASD: A MRI-DTI and Eye-Tracking Study

Friday, May 12, 2017: 12:00 PM-1:40 PM
Golden Gate Ballroom (Marriott Marquis Hotel)
A. Vincon-Leite1, A. Saitovitch1, H. Lemaître2, J. M. Tacchella1, E. Rechtman1, E. Douard1, N. Chabane3, A. Philippe4, D. Grevent1, R. Calmon1, F. Brunelle1, N. Boddaert1 and M. Zilbovicius1, (1)INSERM U1000, Institut Imagine, Paris, France, (2)INSERM U1000, Institut Imagine, Université Paris Sud, Paris, France, (3)INSERM U1000, Paris, France, (4)UMR 1163, Institut Imagine, Paris, France
Background: Autism spectrum disorder (ASD) is highly heterogeneous. Nevertheless, a core deficit in social interaction, notably a difficulty to establish a direct eye-contact, constitutes a frequent characteristic across the autism spectrum. Eye-gaze behaviour, crucial for human interaction, can be measured objectively with eye-tracking methodology. Moreover, it is now admitted that these social human abilities, disrupted in ASD, have a neural substrate. In a network point of view, recent advances in brain imaging as diffusion tensor imaging (DTI) permit to explore white matter (WM) microstructure mediating anatomical connectivity.

Objectives: In this context, after demonstrating that in our cohort, children with ASD met atypical eye-tracking pattern compared to typical developed (TD) children, the current study sought to investigate (i) whether children with ASD exhibit differences in WM microstructure approached by fractional anisotropy (FA) compared to TD children (ii) whether WM microstructure correlates with individual tendency to drive more or less interest to the eye region, which can be viewed as a social perception indicator.

Methods: Twenty-eight children with ASD (age = 8.4 ± 4) and twenty-five TD children (age = 10.4 ± 3) participated in this study. ASD diagnosis was based on DSM IV-R and ADI-R criteria. Tobii-T120 eye-tracker was used to measure eye-gaze processing during passive visualization of social movies, displaying two characters engaged in peer to peer social interactions. WM integrity was voxel-wise assessed over the whole brain diffusion tensor imaging (DTI). Diffusion data were acquired on a GE-Signa 1.5T using an echoplanar sequence (40 directions, TE=70 ms; TR=8800 ms; 2*1.8*1.8 mm3; b=1500 s/mm2). We used a whole-brain Tract-Based Spatial Statistics method. Firstly, whole brain FA values were compared between ASD and TD children controlling for age. Subsequently, whole brain correlation analysis was performed between normalized DTI images and the number of fixations to the eyes. For both analyses general linear model framework within FSL was used.

Results: Compared with the TD group, the ASD group had significantly reduced FA values (p < 0,05, corrected) in the fronto-temporo-parietal circuit, in particular in the right arcuate fasciculus, which connects frontal to temporal regions, known to be important for social perception and cognition. Moreover, the results for ASD children showed a significant positive correlation between FA values and number of fixations in the eyes in widespread clusters including the entire bilateral arcuate fasciculus. Interestingly, we found a significant interaction between groups and number of fixations in the eyes (p <0,05, uncorrected), localised in the right anterior temporal pole. In this anterior temporal region, ASD FA values strongly correlate with the number of fixations in the eyes whereas TD children FA values seem to be constant independently of the eye fixation.

Conclusions: The present study showed differences in WM connectivity between ASD and TD children, particularly in temporal regions. In addition, for the first time to our knowledge, we describe a significant correlation between the anatomical brain connectivity within temporal regions and an objective measure of visual social perception. Interestingly, taken together these results point to common temporal abnormalities across the autism spectrum.