Toward early markers for Autism Spectrum Disorder using eye tracking

Abstract

The goal of this dissertation was to explore two possible early markers of Autism Spectrum Disorder (ASD): visual search superiority, and gaze behavior during face perception. These possible markers were explored as they pertain to both the social deficits in ASD (gaze during face perception) and visual peculiarities (visual search superiority), thereby tackling both diagnostic criteria for ASD. Moreover, visual search superiority and gaze during face perception have been widely studied with older children and adults with ASD. The main reasons for investigating early markers of ASD is to get a better understanding of the atypicalities in ASD as they emerge during development, and to enable earlier diagnosis and intervention. Although exploring possible early markers of ASD was the main focus at the outset of the research conducted in this dissertation, it quickly became clear that the available methodology was critically lacking. Both early markers for ASD described above are dependent on eye-tracking technology. However, as infants cannot be instructed and are difficult to restrict in movement, adult eye-tracking methodology with regard to experimental setup and data processing might not suffice for infant participants. Methodological innovations were introduced that allow a better argumentation for the choice of eye tracker, achieving high data quality, and eye-tracking data analysis in infant research. These innovations are crucial in order to push the developmental eye-tracking field forward. Investigating early markers for ASD not only depends on adequate methodology, but also on knowledge of typical development. With respect to the first of the possible early markers, visual search superiority, relatively little knowledge on typical development of visual search behavior was available. Saccadic search behavior in typically-developing infants was investigated: Infants searched for discrepant items in the absence of instructions, and saccadic search behavior was similar to that observed in adults. These findings serve as a benchmark for the identification of search patterns of infants at-risk for ASD. The conclusions drawn from studies investigating the second possible early marker for ASD, gaze behavior during face perception, have been highly inconsistent. Therefore, a novel method for investigating gaze behavior during face perception was developed. A setup was developed through which people could see and hear each other (akin to a Skype call), while their eye movement were being recorded. First, gaze behavior to faces in this interaction setup revealed important corroborations of research using non-responsive stimuli (e.g. pictures). Second, gaze behavior in interaction revealed intriguing links to ASD traits. In conclusion, the present dissertation demonstrates important methodological advances that are crucial in order to grasp typical and atypical gaze behavior both early in development and in adults. Moreover, a novel setup was introduced for the investigation of gaze behavior during interaction, which has already revealed intriguing links between gaze behavior and traits of ASD. Our setup opens up entirely new avenues for the investigation of (a)typical gaze behavior in social interaction, and may help us grasp the core social deficits in ASD.