Eye-Gaze Pattern Analysis As a Key to Understanding Co-Occurring Social Anxiety within Autism Spectrum Disorder

Background: Emerging research suggests that many adolescents and adults with Autism Spectrum Disorder (ASD) experience impairing social anxiety (Kreiser & White, 2013), yet there is little guidance or agreement about how to best assess social anxiety in this population. Eye-tracking methodology may prove useful in identifying and characterizing co-occurring social anxiety in adults with ASD. Social anxiety in non-ASD samples has been associated with a vigilance-avoidance gaze pattern (Garner et al., 2006), but this type of eye-tracking analysis has not been applied to individuals with ASD and co-occurring social anxiety.

Objectives: The primary objective is to explore the influence of co-occurring social anxiety on gaze patterns in individuals with ASD. We provide a developmental perspective by combining data from an adolescent sample and an adult sample.  

Methods: Data are drawn from two separate eye-tracking studies, both of which used the NimStim Set of Facial Expressions (Tottenham et al., 2009). Each stimulus presentation contained a pair of photographs of the same person, with one photo depicting an emotional expression and the other depicting a neutral or different emotional expression. All participants with ASD had a confirmed diagnosis based on the Autism Diagnostic Observation Schedule (ADOS; Lord et al., 2002). In the adolescent study, a sample of 33 participants (12-17 years old; 15 with ASD) completed the eye-tracking task. The adolescents and their parents also completed measures of ASD symptom severity and social anxiety. The adult study involves a three-group design, with 25 participants (age 16-45) per group: individuals with ASD, individuals with Social Anxiety Disorder (as supported by the Anxiety Disorders Interview Schedule [ADIS]; Brown et al., 1994), and non-socially anxious individuals without ASD. All participants completed the eye-tracking task, various questionnaires, a brief cognitive assessment and clinical interview, and the ADIS social anxiety module.  

Results: Preliminary analyses from the adolescent study show that greater self-reported fear of negative evaluation is associated with longer gaze duration to socially threatening cues (i.e., human faces expressing disgust) in the participants with ASD (r = .75, p = .031). This relationship between self-reported social anxiety and gaze duration toward faces displaying disgust was not seen in the group of adolescents without ASD. The ASD group, with relatively high levels of social anxiety, was also quicker to disengage attention from disgust faces, relative to the non-ASD group. Data collection is projected to be complete for the adult study by November 2013, and results from this well-characterized sample will be presented.   

Conclusions: Preliminary results highlight the importance of considering co-occurring social anxiety when conducting eye-tracking research with ASD. If a vigilance-avoidance gaze pattern (similar to that seen in non-ASD individuals with social anxiety) is found in the participants with ASD and social anxiety, then eye-tracking methodology could provide support for social anxiety being a separable phenomenon in ASD. The exploration of gaze pattern data to assess psychiatric comorbidity in individuals with ASD is novel and will require more research in the future. Clinical and scientific implications related to eye-tracking and ASD will be discussed.