Where's Wendy? Toddlers with ASD Exhibit Limited Attentional Capture by Faces

Background: Many studies have demonstrated that children with autism spectrum disorders (ASD) attend to socially relevant aspects of their visual environment differently from their typically developing (TD) and developmentally delayed (DD) peers. Before attending to a face, one must first detect it in an otherwise complex visual environment. While deficits in attentional engagement with faces (Chawarska et al., 2003; 2010), as well as scanning and recognition, (Chawarska & Shic, 2009; Bradshaw et al., inpress) have been documented, it is not clear if faces capture the attention of toddlers with ASD as rapidly as in typical controls.   

Objectives: To examine attentional capture by faces in toddlers with ASD as compared to age-matched TD and DD controls. We evaluated (1) whether scenes containing faces elicit less attention from toddlers with ASD compared to controls; (2) where children with autism focus their attention within a naturalistic scene; and (3) whether children with autism take longer to orient towards a face in a complex scene.

Methods: Eye-tracking data were collected from toddlers diagnosed with ASD (N=28, M=2.30yrs) as well as TD (N=25, M=2.11yrs) and DD (N=16, M=2.07yrs) control groups. Participants were shown eight static images of women in naturalistic settings (e.g. in an office, living room) for 5s each. The faces of the women were positioned in random locations 7 visual degrees from a central fixation target which preceded each trial. Each image was parsed into discrete regions of interest (ROI). We also examined overall valid looking time and duration of looking at specific ROIs as well as latency to fixate the face for the first time after the stimulus onset.

Results: We found no significant differences between the groups in terms of overall amount of valid looking time recorded by the eye-tracker. However, children with ASD and DD spent more time looking away from the screen compared to TD children (p<.01; p<.05, respectively). Additionally, when children with ASD attend to the screen, they spend more time looking at the area surrounding the stimulus compared both to TD children (p < .05) and DD peers (p = .064). Children with ASD also oriented to the face later than both their TD (p < .05) and DD peers (p < .05).

Conclusions: Results show that both ASD and DD groups spent more time looking away from the screen compared to typically developing peers, which suggests difficulty in regulating attention, likely associated with presence of cognitive delays in both groups. However, when looking at the screen, toddlers with ASD were less likely to examine the complex scene containing the person than both contrast groups.  This may be a function of lesser salience of scenes containing relevant social stimuli for toddlers wtih ASD.  Finally, when children with ASD directed attention to the scene, they took longer to localize the face embedded into the scene. Thus, in addition to other attentional, processing, and recognition deficits associated with faces, toddlers with ASD also show limited attentional capture by faces.