Enhanced Fine-Form Perception Does Not Contribute to Gestalt Face Perception in Autism Spectrum Disorder

Abstract

Individuals with autism spectrum disorder (ASD) show superior performance in processing fine detail, but often exhibit impaired gestalt face perception. The ventral visual stream from the primary visual cortex (V1) to the fusiform gyrus (V4) plays an important role in form (including faces) and color perception. The aim of this study was to investigate how the ventral stream is functionally altered in ASD. Visual evoked potentials were recorded in high-functioning ASD adults (n = 14) and typically developing (TD) adults (n = 14). We used three types of visual stimuli as follows: isoluminant chromatic (red/green, RG) gratings, high-contrast achromatic (black/white, BW) gratings with high spatial frequency (HSF, 5.3 cycles/degree), and face (neutral, happy, and angry faces) stimuli. Compared with TD controls, ASD adults exhibited longer N1 latency for RG, shorter N1 latency for BW, and shorter P1 latency, but prolonged N170 latency, for face stimuli. Moreover, a greater difference in latency between P1 and N170, or between N1 for BW and N170 (i.e., the prolongation of cortico-cortical conduction time between V1 and V4) was observed in ASD adults. These findings indicate that ASD adults have enhanced fine-form (local HSF) processing, but impaired color processing at V1. In addition, they exhibit impaired gestalt face processing due to deficits in integration of multiple local HSF facial information at V4. Thus, altered ventral stream function may contribute to abnormal social processing in ASD.

Fig 1. VEPs for RG and BW stimuli at the occipital ROI in each group.

Grand-averaged waveforms show that N1 was a major component in both stimuli. Scalp topography shows N1 localized to the occipital area. VEPs, visual evoked potentials; RG, red/green; BW, black/white; ROI, region of interest.

Fig 2. VEPs for face stimuli at the right occipital ROI in each group.

Grand-averaged waveforms illustrate that P1 was evoked as a major component for all facial expressions. Scalp topography shows that P1 was distributed over the bilateral occipital area with right dominance.

Fig 3. VEPs for face stimuli at the right occipitotemporal ROI in each group.

Grand-averaged waveforms illustrate that N170 was evoked as a major component for all facial expressions. Scalp topography shows that N170 was distributed over the bilateral occipitotemporal area with right dominance.

Fig 4. ROIs for each VEP measurement.

We defined three ROIs: occipital ROI for N1 in response to RG and BW stimuli, and right occipital ROI for P1 and right occipitotemporal ROI for N170 in response to face stimuli.

Fig 5. Source analysis for VEPs with RG and BW stimuli in each group.

sLORETA activities were mapped onto the MNI space. N1 components for RG (A) and BW (B) stimuli were located at around occipital areas including V1 in both groups. sLORETA, standardized low resolution brain electromagnetic tomography; MNI, Montreal Neurological Institute.

Fig 6. Source analysis for VEPs with face stimuli in each group.

sLORETA activities were mapped onto the MNI space. P1 (A) and N170 (B) components for neutral face stimuli were located at around occipital areas including V1 and occipitotemporal areas including V4, respectively, in both groups.