Contextual Influences on Eccentric Viewing in Young Children with ASD

Background:  Children with Autism Spectrum Disorder (ASD) show atypical visual behaviors either during object exploration (Mottron et al. 2007; Ozonoff et al. 2008; Hellendoorn et al. 2014) and social interactions (Noris et al. 2012). In general, they tend to produce eccentric gazes (EG), which seem mainly to serve the function to filter details during visual exploration of objects. A subtype of EG, downcast gaze (DG), are probably functional to shift attention away from over-arousing visual stimuli (Noris et al. 2012; Bogdashina 2003). However, despite the relevance of contextual factors as possible triggers of EG (namely the presence of objects or over-arousing stimuli), previous studies observed children in a single context, not comparing the frequencies of EG under different contextual conditions.

Objectives:  This study aimed to investigate whether contextual factors might influence the frequency of EG in young children with ASD, and specifically the occurrence of DG.

Methods:  20 preschoolers with ASD mean age 4,8 years (3,7 to 6,4 years, SD= 10 months),  mean non verbal IQ 106 (Leiter-VR range 87-131, SD= 12.86) were included in the study. All the children were diagnosed by expert clinicians according to DSM-IV criteria for ASD and standardized diagnostic instruments (ADOS and ADI-R). We observed the children in two video-recorded conditions: COND-1 (free play with toys); COND-2 (structured social interaction and free play with toys). Frequency of EG performed by the children towards 8 possible directions (up, up-left, up-right, lateral-left, lateral-right, down, down-left, down-right) were coded from the videos. The mean frequency of DG and of EG per minute in each condition were considered in the analysis.

Results:  The results indicated that, in general, the mean frequency of EG performed by the children in COND-2 was significantly lower than in COND-1 (COND-1=6.33 [SD=3.66]; COND-2=9.12 [SD=4.11]; t=3.77, df=19, p=0.001). Specifically, DG were more frequent in COND-2 than in COND-1 (COND-1=0.98 [SD=0.90]; COND-2=2.47 [SD=1.34]; t=5.1, df=19, p<0.001). 

Conclusions:  The results indicated a significant effect of the context on the mean frequency of EG. In general, EG were significantly more frequent in the most stimulating condition, in which the children were playing with toys and interacting with the experimenter. Specifically, such a difference was mainly accounted by a higher frequency of DG, which were rare when the children were only playing with toys. These findings seem to confirm the possible function of DG in reducing sensory overload.