Altered Prefrontal Activation and Connectivity During Implicit Emotion Judgment in Autism Spectrum Disorders

Background:  Judging others’ emotions accurately is a critical skill for successful social interaction. Many individuals with autism spectrum disorders (ASD) struggle with everyday social interactions, largely due to difficulty reading others’ emotions from faces (Harms et al., 2010; Monk et al., 2010) and from their body postures (Grezes et al., 2009; Hadjikhani et al., 2009). Such difficulty in individuals with ASD may manifest more in implicit assessment of emotions than in explicit situations of emotion attribution. This is relatively under-examined in ASD. This study uses dynamic action scenarios to examine the neural mechanisms, especially the role of inferior frontal cortex (IFC, a key node involved in simulating actions), in mediating implicit and explicit emotion judgment in ASD.

Objectives:  The primary objective of this functional MRI study was to investigate the nature of brain responses associated with explicit and implicit perception of emotional information in high-functioning adults with ASD.

Methods: fMRI data were collected from fourteen high-functioning adults with ASD and thirteen typically developing control participants. Participants watched a series of short videos (average 10 seconds long) of characters involved in emotional action scenarios. Participants were told to judge the emotion expressed by the actor (happy, sad, angry, or afraid) or identify a masked object in each video. In control condition, participants watched neutral videos of characters and made a perceptual judgment. The stimuli were presented in an event-related design, and the data were acquired using a Siemens 3T scanner. Data analysis was performed using SPM8.

Results:  1) Behavioral data showed that both groups performed significantly above chance on all video types (p < .05), and there were no significant group differences in accuracy or reaction time; 2) There was robust bilateral extrastriate body area (EBA) activation (MNI coordinates: x = -46, y = -68, z = 0; x = 46, y = -66, z = 0) in both ASD and typical control participants for both emotion and object tasks; 3) While there was no difference in activation in explicit emotion condition, the participants with ASD showed significantly reduced response in medial prefrontal cortex (MPFC) and right IFC during implicit emotion judgment (p = 0.005; cluster size = 80mm³); and 4) Functional connectivity analysis revealed significant underconnectivity in autism between left IFC and right inferior parietal lobule (IPL) [t(25) = 2.36; p < 0.05], and between right IFC and left IPL [t(25) = 2.22; p < 0.05].

Conclusions:  Robust EBA activation seen in both groups across conditions suggests intact general perceptual explicit and implicit brain responses to emotional action observation in ASD. However, the limited response in MPFC, and IFC, as well as underconnectivity between IFC and IPL in ASD participants during implicit emotion judgment may suggest their difficulty in reading emotions automatically. It is possible that implicit information may not be associated to its valence in participants with ASD. The activation and functional connectivity data from this study provide valuable insight into understanding bodily emotions in autism.