SYMPTOMS SEVERITY and VISUAL Attention to the Eyes IN Children with AUTISM: A Correlational STUDY

Background:  The reduced visual attention toward human faces in individuals with Autism Spectrum Disorders (ASDs) has been described in a number of studies (see Jones et al., 2008; Grelotti, Klin, Volkmar et al., 2002). In particular, while eyes are a special kind of stimuli for typically developing individuals (Ristic et al. 2002), this seems not to be true for people with ASDs.  Poor attention to social stimuli, corresponds to a great loss of relevant social information and this is likely to have a negative impact on symptoms severity. However, the link between attention to the eyes and symptoms severity has been rarely explored in a systematic way. For the best of our knowledge, to date only one study, investigated whether attention to social elements in complex social scenes predicted symptoms severity in ASDs (Klin, Schults, Volkmar & Cohen, 2002). However, this study used very complex stimuli, in which hand and mouth movements accompanied by language were extremely salient and this might have lowered attention to the eyes.

Objectives:  Our study aimed to investigate the possible correlation between fixation behavior to social stimuli and symptoms severity, in children with ASDs, using very simple static stimuli.

Methods: A group of 20 children with ASDs (10 M), aged between 3 and 8 yrs (mean age=4.5 yrs), were evaluated in respect to visual attention for photographs of human faces using an eye-tracker Tobii T60. The stimuli were pictures, depicting a human face with the eyes directed laterally towards one of two coloured objects depicted each on one side of the face, at the eyes level. Time to First Fixation (TFF) and Fixation Count (FC) for the eyes and for the non social elements of the stimuli (i.e. the background) were measured, and correlated with ADOS language scores and with ADOS social interaction scores.  

Results: As found in previous studies, the results confirmed the eyes were less relevant compared to the non social stimuli, since the children looked less to them (mean FC eyes=156 (sd=110); mean FC non social stimuli = 347 (sd=253); t=-2.53; df=19; p=0.020) and less faster (mean TFF eyes=1.3 (sd=0.76); mean TFF non social stimuli = 0.76 (sd=0.64); t=2.67; df=19; p=0.015). However, while FC did not correlate  with symptoms severity, TFF to the non social stimuli correlated both with ADOS language scores (r= -0.58; p=0.017) and with ADOS social interaction scores (r= -0.60; p=0.013), meaning that the children with more severe symptoms tended to focus more rapidly to non social elements of the stimuli rather than to the social relevant ones. 

Conclusions: It seems that, increasingly with their inability to be attuned to the social world, children with ASD tend to be attracted more by the non social elements of the stimuli, rather than by those that might bear a social meaning. This abnormal tendency already thoroughly documented in the literature (Osterling et al., 2002), needs early intervention, since the low focus on social information reduces opportunities for social learning in ASDs, hampering social and communicative development.