Speaker: HOU Xiaohui

Time：13:30-14:30 (Tue), December 9, 2014

Venue: Meeting Room Level 5, South Building

Host: ZUO Xinian (Professor)

Abstract:
Autism Spectrum Disorders (ASD) has been increasingly recognized as a brain disorder with aberrant activity measured by functional magnetic resonance imaging (FMRI). Previous findings on both functional activation and connectivity suggested large-scale neuronal dysfunctions at a system level in ASD. Here, we conducted a series of comprehensive meta-analyses of 101 fMRI studies on ASD (1748 patients versus 1762 healthy controls) and linked the observations to a functional parcellation of connectomes derived from 1000 healthy participants. This approach offers us an opportunity of mapping the ASD-related functional changes at a system network level in detail. We found that the default, ventral attention and control networks exhibited both hyper- and hypoactivations whereas the somatomotor network mainly presented hypo-activation and the dorsal attention network primarily showed hyper-activation. Hypo- and hyper-activations were more detectable in adulthood samples for cognitive control network and visual network, respectivelywhereas the abnormal activations were mainly presented for limbic network and dorsal attention network in neurodevelopmental samples. Interestingly, default network demonstrated remarkable reduction of proportion of abnormal activations from childhood to adulthood. Specifically, this age-related change of default network contains a system level shift that aberrant activations dominated by its dorsal medial prefrontal cortex (dMPFC) subsystem were replaced with abnormal activations within its hub subsystem and its medial temporal lobe (MTL) subsystem. These findings extend early neural pathophysiological models of ASD to a whole brain system-level model by proposing a miswired connectome profile during the neurodevelopment, which will serve as a target for basic research, clinical tests and brain-based biomarkers of ASD in future.