Integrated Study of Joint Attention in Autism Spectrum Disorders By High-Density EEG and Eye-Tracking

Background: Autism Spectrum Disorders (ASD) are characterized by a strong impairment in joint attention (JA) development. This deficit is associated with difficulties in social learning and has consequences on the subsequent behavioral and neural development of children with ASD. For this reason the implementation of protocols for the study of brain-behavior correlation of JA are crucial in understanding and treating ASD. Electrophysiology (EEG) can provide cues about the behavioral correlates of JA with an excellent temporal resolution. To our knowledge no study in literature have explored yet the neurophysiological correlates of JA in ASD using EEG. Moreover eye-tracking can provide additional information about the direction of the gaze during the task.

Objectives: The aim of this study was to describe the neurophysiological and gaze correlates of JA in children with ASD during responding and initiating joint attention tasks. A further aim was to use this quantitative assessment to evaluate the eventual modifications in EEG and eye-tracking measures due to the treatment.

Methods: Data were acquired in 10 children with ASD at T0 and after six months of treatment (T1). Stimuli consisted in video sequences, with an actor between two identical toys placed on the floor in front. The experiment involved two conditions: the responding to joint attention (RJA) and the initiating joint attention (IJA). A control condition without objects was also added. EEG data were acquired using a high-density 128-channels system (HCGSN 128; Electrical Geodesics Inc) while children’ gaze was recorded with SMI Eye Tracking device by SensoMotoric Instruments (Germany). A switch connected to the stimulus PC allowed sending the events simultaneously to the two systems. EEG data epochs, in which the child attended the screen, were pre-processed to remove artifacts and analyzed using quantitative methods (QEEG) to obtain power, asymmetry and coherence. For eye-tracking data analysis three regions of interests (ROIs) were selected: model’s face, target object and non-target object. Gaze duration within each ROI and transitions between different ROIs were computed. 

Results: In the IJA condition treatment induced an increased in frontal gamma activity and a decrease in alpha and delta activity in frontal, temporal and occipital areas. Coherence in gamma band increased in frontal and occipital areas while delta and alpha coherence decreased in occipital areas. In the RJA condition power decreased in frontal regions in theta and alpha bands, in right temporal area in alpha and in left occipital area in delta. Coherence increased in delta band between occipital and temporal areas and in parietal area in theta and alpha bands. Eye-tracking analysis revealed an increased attention at face and a decreased attention at objects at T1. In addition in the IJA task transitions from both target and non-target object to face increased compared with T0. Significant correlations were observed among QEEG measures and eye-tracking parameters.

Conclusions: The integrated protocol realized allowed to explore neurophysiological and gaze correlates of the different components of JA. The results of this study suggest that IJA and RJA subtend two different neural circuitries, which can be modified by treatment.