Compensatory Strategies in Locating Referential Object in Children with ASD: The Contribute of An Eye-Tracker Paradigm

Background: Gaze direction detection plays an important role in socio-communicative development. Accurate eye gaze judgement appears problematic for individual with autism (Gepner et al., 1996; Sweetenham et al., 2001; Webster & Potter, 2008). However, the basic geometric understanding of gaze direction seems to be preserved in ASD (Leekam et al., 1997). Thus evidence concerning the eye gaze direction detection abilities in individual with Autism Spectrum Disorders (ASD) needs to be further investigated. Recently, researchers focused on tracking eye movements to better understand the strategies used to perform these tasks and detect possibly atypicity (Klin et al., 2002).

Objectives: Investigate if children with ASD are sensitive to the orientation of an adult’s eye gaze in relation to an object which is within the child’s field of view.

Methods: We compared on an eye-gaze detection task 30 children with ASD (20 M; 10 F), aged between 2-8yrs (mean=4; DS=1,7), IQ Leiter-R=73 (DS=22), and 30 TDs (12 M; 18 F), age-matched. The stimuli were presented with Tobii T60 Eye Tracker and were a modified version of the ones used in Hoehl, Reid, Mooney, and Striano (2008): an adult looking either toward or away from an object that was depicted on eye-level next to the adult’s head. There was a second object on the other side of the adult’s head. We defined 4 areas of interest for all the stimulus images: eyes, gaze target, no gaze target,  screen-others (that is the area of the screen external to the other AOI). In order to define the strategies used to perform the tasks we considered Fixation Count (FC: the number of times that the child fixate an AOI), Time of First Fixation (TFF: how long it took to the participants to fixate an AOI for the first time) and Observation Length (OL: the total time in seconds for every time a person has looked within an AOI).

Results: The results show that ASD children looked more times outside the AOI (FC screen-others=34%) than controls (FC screen-others=22%) and less to the eyes (FC eyes ASD=22%; FC eyes TD=40%), while the rate of FC to the gaze target (FC gaze target ASD= 22%; FC gaze target TD= 21%) was very similar ( chi square = 222; df=1; p=0.000). The two groups did not differ in TFF and OL meaning that both groups were able to locate the gaze target.

Conclusions: These findings shed new light on the processing of dyadic and triadic relations in ASD children  There were quantitative differences in the way that children with ASD process static eye-gaze direction (see also Klin et al., Spezio et al, 2007). They demonstrated fewer fixations to the eyes region of face compared to controls but equal fixations to the gaze target. These results indicate that children with ASD  might be successful in locating relevant objects in the environment due to a compensatory strategy involving over-exploration of the visual-field. Empirical evidences of gaze-direction detection in ASDs might be critical for developing effective treatment approaches aimed to optimized learning outcomes.