Circumscribed Attention and Social Salience in Infants at Risk for Developing ASD

Background: Individuals with autism spectrum disorders (ASD) orient and respond to social stimuli abnormally, which may be indicative of important differences in the distribution and prioritization of attentional resources. Previous work has shown that school-aged children with ASD exhibit restrictive and perseverative patterns of attention when presented with complex visual arrays matched for social and nonsocial content (Sasson et al, 2008), though this pattern is driven primarily by disproportionate attention to objects related to circumscribed interests (CI; e.g., trains). Attention to social stimuli (i.e., faces) is reduced in the presence of CI objects, but appears normal in the presence of other object types, suggesting that heightened salience of CI objects in ASD may result in attentional prioritization that reduces attention to social information. Whether this pattern extends to very young children with ASD is of significant clinical relevance, as early emerging biases toward nonsocial information may affect the development of neural specialization (Johnson, 2000), including neural circuitry supporting abilities related to social information processing (Sasson, 2006; Schultz, 2005).

Objectives: To determine whether infants at risk for developing ASD exhibit reduced social attention in the presence of CI objects relative to non-CI objects.

Methods: Twenty-six infants between the ages of six and twenty-four months at risk for developing ASD (i.e., siblings of children with an ASD diagnosis) were recruited through the Infant Brain Imaging Study. Each infant was eye-tracked while passively viewing the Social vs. Object task, a novel free-viewing visual preference task consisting of twenty slides of paired face and object images. The twenty pairs included ten object images found to be common targets of CI for individuals with ASD (South, et al., 2005) that disproportionately capture attention in school-aged children with ASD (Sasson, et al., 2008), and ten objects not typically related to CI.

Results: An analysis of fixation time on stimulus categories revealed a three-way interaction between age (6 vs. 12 vs. 24 months), stimulus type (face vs. object) and object type (CI vs. non-CI) (F (2, 26) = 5.61, p < .01). Post-hoc Tukey analyses indicated that this interaction was driven by an increase with age in fixations to CI objects, resulting in an age-related decrease in fixations on concurrently paired social stimuli. In contrast, fixations to faces remained stable with age when non-CI objects were paired with faces. Several non-significant mean differences also emerged with age, suggesting that older children may be more likely to orient to CI objects relative to-non CI objects in the context of social stimuli, a trend that will be followed up with a larger sample prior to the meeting.

Conclusions: Findings suggest that between six and twenty four months, infants at risk for developing ASD become increasingly attentive to CI, but not non-CI, objects, a pattern that results in a decrease in social attention when these CI objects are present. Data collection is ongoing with a typically-developing comparative sample of infants to determine whether this pattern is specific to at-risk infants. These data should be available in time for the conference.