Amygdala Activation in Response to Configural and Featural Facial Changes in ASD

Background: Autism Spectrum Disorder (ASD) is characterized by deficits in social communication including an impaired ability to extract emotional and perceptual information from faces. Non-diagnosed siblings of individuals with ASD (ASD-sibs) share some less severe characteristics of autism including mild deficits in face perception. Whether these face processing deficits are due to a bias to process faces in a piecemeal, locally oriented fashion or due to impaired ability to process global information in faces remains in dispute. At present, the neural substrates of impaired face processing in ASD are not fully understood. For example, the fusiform face area (FFA) does not show strong activation to unfamiliar faces, but does show activation when face stimuli are relevant (Pierce et al., 2008). Amygdala abnormalities are widely reported in ASD during tasks involving emotional face discrimination (Golarai et al., 2006; Schultz, 2005). In typically developing (TD) individuals, the amygdala responds to configural and featural aspects of fearful faces (Morris, 2002). However, a possible role of the amygdala for featural and configural face processing in ASD remains unexplored.
Objectives: Using fMRI, we explore whether amygdala function for featural and configural face processing differs between ASD and ASD-sibs.
Methods: Participants enrolled thus far include five ASD children (mean age = 12) and five ASD-sibs (mean age = 13.5). During a block design fMRI task, subjects indicated via button press whether two images (either houses or faces) presented side by side were the same or different. Functional runs consisted of two blocks each of featural changes to faces (FF) or houses (HF) or configural changes to faces (FC) or houses (HC). Within each block two same and four different pairs were presented. fMRI data were analyzed using FMRIB’s FSL package. After standard preprocessing and individual-subject level statistics, preliminary analyses of each group (ASD and ASD-sibs) were conducted. The main contrasts of interest were: FF vs. HF, FC vs. HC, and Faces vs. Houses.                                                                                                                                                                                                Results: Behavioral results indicate that both groups of children performed the task well. Preliminary fMRI results indicate that the ASD group showed significantly greater activation of the right amygdala for configural face versus configural house processing, but greater left amygdala activation for featural face versus featural house processing. ASD-sibs did not show this differential response of the amygdala to faces. However, ASD-sibs activated a region consistent with the right fusiform face area (FFA) during face versus house blocks but the ASD group did not.
Conclusions: Preliminary results are consistent with other studies showing either reduced or no right FFA activation for unfamiliar faces in ASD individuals across a variety of face processing tasks. The present findings of greater amygdala activation in ASD subjects compared to ASD-sibs suggest that this region may play a role in the development of configural face processing. Moreover, whereas Morris et al. (2002) found left amygdala recruitment for configural aspects of faces in TD individuals, we find the right amygdala is responsive to configural processing in ASD. Future work will include comparisons with age-matched typically developing controls.