We have embarked upon a series of studies that involve the measurement of brain function and structure using MRI. In these studies, we examine how children with autism process information from faces, particularly those conveying emotion. We are also interested in the ability of children with autism to integrate information from multiple sensory modalities (e.g., visual, auditory) to make inferences about the emotional states of others. In the studies we conduct, we utilize sophisticated eye tracking technology that enables us to know precisely where children are looking when we present them with facial stimuli. In this way, we can track simultaneously where individuals are looking and what is occurring in their brains as they perform such tasks. Our initial findings indicate that children with autism do indeed spend less time fixating around the eyes of faces and, moreover, that the amount of time fixating around the eyes is strongly associated with their ability to make inferences about the emotion conveyed by the face. The measures of brain function indicate that when children with autism avert their gaze, they actually reduce activation levels in the amygdala and in other brain regions that are critical for activating emotion. These findings suggest that in children with autism, social stimuli like faces may elicit uncomfortable levels of emotional arousal that are partially attenuated by averting gaze.

Our studies have also been examining how these functional differences in the brain might be subserved by structural differences. We are finding significant differences in the volume of the amygdala in children with autism. We also observe that variations in amygdala volume are associated with our eye tracking measures and with other indices of social behavior. Our current and future studies include the assessment of siblings of children with autism who are themselves at increased risk for the development of the disorder. In addition, we are examining the extent to which the deficits in extracting information from faces generalizes to vocal expressions of emotion. We are also using information from brain imaging to help differentiate among subtypes of autism. Once we have better, brain-based phenotypes of the disorder, we can then make more rapid progress in the identification of the molecular genetic substrates of these subtypes by directly examining relations between the brain imaging findings and data from molecular genetics.

The Opportunity: We have an extraordinary opportunity to make rapid progress in understanding and then treating this devastating disorder. We critically need to increase the rate at which individuals are tested. We need to test large samples of children with autism in order to better characterize the different phenotypes. To do this, we will need to add personnel and resources to complement those we already have. There are also a number of novel studies that we wish to undertake that will test new behavioral treatments based upon our understanding of the brain mechanisms that underlie different subtypes of the disorder. If we can add to our already significant infrastructure, we can become one of the leading centers in the world for research on the brain mechanisms of autism.