A study of EEG and eye tracking in children with autism

Abstract

Autism spectrum disorder (ASD) is a complex brain developmental disorder characterized by troubles with social interaction and communication, and by restricted interest and repetitive stereotypical behaviors. Childhood is a critical development period in a human life associated with physical, cognitive and social-emotional development, and has also been considered as a key time window for early diagnosis and intervention of autism. In this article, we focus on investigating the development rules of autistic children and typically developing (TD) children using electroencephalography (EEG) signal analysis and eye tracking analysis by conducting the following two independent experiments, respectively. In the first experiment, the resting-state EEG data of 351 children aged ranging from 3 to 9 years (80 children with autism, 271 TD children) are collected. We calculated the relative power spectrum of EEG using the well-known Pwelch’s method and then established a linear model to explore the main effects of age and diagnosis group at each frequency band. The results showed that the relative power spectrum in the slow-wave band decreased with age both in children with autism and TD children. Notably, autistic children demonstrated that significant differences in the relative power in the alpha band from children in TD group. In the second experiment, a total of 293 children aged ranging from 3 to 9 years (104 children with autism, 189 TD children) were recruited to acquire the eye gaze data when watching a dynamic social scene video. Different areas of interest, including eye, mouth, body, background and joint attention, were considered, and the total fixation time of each area of interest was statistically computed for children with autism and TD children with an interval of one year. The results demonstrated that the proportion of fixation time of the joint attention and the mouth in children with autism was significantly reduced compared to TD children. Specifically, autistic children were the common attention deficit, while paying more attention to the background and the body. In summary, two experiments above have been performed to explore the potential objective biomarkers and behavior markers from different perspectives of neural oscillation and statistical features of eye-tracking fixation patterns. Both results have proven that EEG oscillation activity and eye gaze behavior can provide an objective evaluation of brain development for children with autism, and furthermore can reveal the developmental rules of EEG and eye gaze behavior patterns with age, which is helpful for clinical diagnosis and early intervention of ASD.