Neuro-Functional Networks Supporting Cross-Sensory Emotion Processing in Teens with Autism Spectrum Disorder

Background: The processing of social information involves a network of distributed brain regions, many of which appear to be aberrant in ASD. Because information from the different senses is typically complimentary, cross-sensory integration of sensory input provides information about the environment that is unobtainable from any one sense in isolation (O'Hare, 1991). It has been proposed that the organizational structure of the brain in ASD follows a pattern of high local connectivity and low long-range connectivity (Just et al., 2004). Failure to integrate multi-sensory social-emotional information could be expected to result in considerable social deficit. Objectives: To investigate the neural circuitry underlying the processing of socially important, affective multimodal cues in high functioning teenage boys with ASD. We focused on an age where maturational changes in cortical myelination support faster, superior integration, in order to identify newly established brain networks that support the processing of multimodal cues and detail functional regions that distinguish teens with ASD from healthy controls. Methods: High functioning teenage males with autism (13-19 years old) and matched male controls participated. Stimuli were constructed from simple emotions (happy, sad, angry and no emotion). In advance of the functional Magnetic Resonance Imaging scan, a behavioural test was conducted to establish individual face perception thresholds for visual facial emotion stimuli. Subsequently thresholded stimuli were used in a cross-sensory task that involved the matching of an emotion face and voice combination to an emotion label. This design permitted us to test how successful integration can facilitate the processing of emotion in difficult to detect situations. In the scanner, visual stimuli were projected onto an overhead visor and auditory stimuli were presented using MRI compatible sound isolation headphones. Responses were made via a hand-held response pad. Functional BOLD imaging was done using a gradient-echo planar imaging (EPI) sequence, with 36 axial contiguous slices (3-mm thick, no skip) encompassing the entire cerebrum (repetition time/ echo time [TR/TE] 2500/35 milliseconds, flip angle=90 degrees, field of view [FOV] 24 cm, matrix 64 x 64). An event related design was employed where emotion trials were presented with variable jittered interstimulus intervals. Results: Data analysis is underway, however, preliminary between group results identify greater recruitment of the insula, superior temporal and parietal cortices in typically developing teens. In contrast, teens with autism show greater recruitment of the superior prefrontal cortex. Conclusions: These results suggest, when typically developing teens are processing cross-sensory emotion cues they recruit areas involved in face and emotion processing, integration of emotional state and physical responses to arousal and social relevance. In comparison, teens with autism show greater engagement of regions associated with working memory and problem solving.