Less Sensitive Face-Selective Responses in ASD Measured with Fast Periodic Visual Stimulation EEG

Friday, May 12, 2017: 5:00 PM-6:30 PM
Golden Gate Ballroom (Marriott Marquis Hotel)
S. Vettori1, S. Van der Donck2, M. Dzhelyova3, B. Rossion3 and B. Boets4, (1)Onderzoeksgroep Psychiatrie UZ Herestraat 49 - bus 7003 64, K U Leuven, Leuven, Belgium, (2)KU Leuven, 3000 Leuven, Belgium, (3)Psychological Sciences Research Institute and Institute of Neuroscience, UCL, Louvain-la-neuve, Belgium, (4)Katholieke Universiteit Leuven, Leuven, BELGIUM

Fluently recognizing faces and facial expressions is highly important for our social interactions. Impaired and atypical face processing have often been postulated as a key deficit in autism spectrum disorders (ASD). Despite the great amount of research on face identity and facial expression recognition in ASD, results are mixed. This is partly due to the widely used tasks tapping explicit face processing, which may give an incomplete estimate of face processing abilities in ASD.


Therefore, we wish to examine these face processing impairments in ASD with an innovative EEG approach. This method combines fast periodic visual stimulation (FPVS) with scalp electroencephalography (EEG). The main advantage of this new and highly versatile FPVS EEG approach is that it offers an objective, quantifiable and robust index of implicit face processing abilities, reliable at the individual level, within a few minutes of time and without any complex data analysis. The core idea of FPVS is that the periodicity of the electrophysiological response on the human scalp corresponds exactly with the periodicity (frequency) of the visual stimulation. Hence, it can be used for efficiently measuring categorization responses of complex visual stimuli in the human brain.

Methods:  In this study, the ability to rapid categorization of natural face images was assessed. 25 high-functioning young boys with ASD (age 8-12) and 25 typically developing (TD) boys without any psychiatric disorder completed the FPVS EEG paradigm. In this paradigm, images of objects in their natural background are presented at a baseline frequency rate of 6 Hz. Every fifth image, widely variable face images appear in the sequence. If participants detect the periodic appearance of the face stimuli, a face-selective response is observed in the EEG at exactly 1.2 Hz (6/5). Further, participants completed computerized versions of the Benton Facial Recognition Test (BFRT) and the Cambridge Face Memory Test (CFMT). In addition to verbal, performal and total IQ, clinical assessments included questionnaires assessing quantitative autism characteristics (Social Responsiveness Scale), symptoms of depression (Child Depression Inventory), symptoms of anxiety (SCARED), symptoms of congenital prosopagnosia (adapted version of the 20-item prosopagnosia index), and the Child Behavior Checklist.


Results show that the paradigm seems to be able to discriminate between children with and without ASD. We observe that boys with ASD are at least as sensitive as TD boys to the periodic flickering of the stimuli at the base rate (6 Hz), indicating that they pay equal attention to the images presented on the screen. However, they are far less sensitive to the periodic appearance of the face stimuli at oddball rate (1.2 Hz). We will further expand these results by showing correlations between the FPVS paradigm and symptom severity and IQ scores.


These results suggest that for boys with ASD, their neural system is less tuned to the socially relevant information on the screen, as their face-selective response is less sensitive compared to TD boys. Furthermore, the results indicate the strength of our technique as a powerful and sensitive tool to measure socio-communicative difficulties in ASD.