Objectives: This study examines, through eye-tracking, the differences between the visual scanning patterns of toddlers diagnosed with ASD (age: M=21 (SD=3) months; gender: 22M, 4F; N=26) and typically-developing controls (age: 20(3) months; gender: 20M, 6F; N=26) as they watched a video of a child-adult play interaction.
Methods: Subjects were presented with a 30 second video in which a female adult was directing a male child in the assembly of a child’s puzzle toy. The scene took place in a typical office with other toys strewn about the floor. The subjects’ viewing patterns were tracked and the time spent examining each of the following areas was recorded: 1) activity area (area of joint focus of the adult and the child, including the puzzle); 2) actors (the adult and the child); 3) toys (not including the puzzle activity); and 4) background (furniture, carpet, etc.). The proportion of time spent looking at these regions was analyzed with a MANOVA with diagnosis (ASD or TD) as a factor.
Results: ASD toddlers monitored the activity area significantly less than TD toddlers (ASD: .43(.2); TD: .56(.15); F(1,50)=6.6, p<.05) and attended more to the other toys in the scene (ASD: .12(.09); TD: .07(.05); F(1,50)=8.0, p<.01). No differences were found between ASD and TD toddlers in the proportion of time spent examining the actors or background features. However, in ASD toddlers, nonverbal cognitive scores on the Mullen Scales was positively correlated with activity monitoring (r =.44, p<.05) and negatively correlated with attention to the background (r=-.44, p<.05). Attention to the actors was negatively correlated with socialization scores (r=-.53, p<.01), social-communicative impairment (r=-.42, p<.05), and repetitive behavior/restricted interests (r=-.60, p<.01) on the ADOS.
Conclusions: Toddlers with ASD monitor the activities of others in an atypical fashion, attending less to the area of joint focus and shared activity and more to background distracters such as toys. Patterns of abnormal viewing were associated with social and cognitive measures of functioning. These results suggest that, by 20 months, the pathogenic factors involved in ASD are affecting the ability of ASD toddlers to attend to socially relevant areas of scenes. It is likely that, as the ASD toddlers grow older, the cumulative effects of this atypical monitoring will further impair their imitative learning and their ability to acquire knowledge regarding the rules of social interaction and play.