Social Attention in 12- and 18-Month-Old Infants Later Diagnosed with Autism

Background:  Deficits in social attention and affective presentation are recognized features of autism spectrum disorder (ASD). However, the emergence of these social affective symptoms in infants and toddlers is poorly understood. Much of the recent research on these features in infancy comes from studies of infant siblings of children with ASD. There is limited understanding of the emergence of ASD symptoms in the presence of other non-specific risk factors.

Objectives:   To examine differences in social attention and affect in a cohort of infants later diagnosed with autism in comparison to a longitudinal cohort of babies matched on the presence of pre- and perinatal risk factors.

Methods:   Subjects were recruited from the Maternal Lifestyle Study (MLS), a longitudinal investigation of the impact of prenatal substance exposure and other pre- and perinatal risk factors on infant development, including prenatal drug exposure.  From the MLS (n = 1388), 11 children with ASD were identified and diagnoses were confirmed by standard methods. The ASD group was compared to a case control group matched on gestational age, presence/absence of prenatal cocaine exposure and other substances (alcohol, tobacco, marijuana; n=507). At 12 and 18 months corrected age, infants were videotaped during a standardized interaction with toys designed to assess attention, persistence, and affect in infants. Videotapes were coded for infant positive and negative affect, as well as for coordinated social attention (eye contact with the examiner) and other social communication acts defined as joint attention and behavioral requests.

Results: At 18 months, the case control group was more likely to hold up toys to show for the purpose of joint attention than was the ASD group (p < 0.01).  Also at 18 months, the ASD group was less likely to make behavioral requests (p < 0.01). There were no group differences on the social attention variables between the ASD and case control groups at 12 months. There were no differences in affect displays between these groups at either age point.

Conclusions:   Past research has reported diminished social attention by 12 months of age in infant siblings of children with ASD. In this study, which included infants with pre- and perinatal risk factors, we found diminished social communication at 18 months but not at the 12-month observation in children later diagnosed with autism. These findings suggest that social attention and social communication symptoms may not be reliably detected in the presence of confounding risk factors until later points in development. An alternative explanation is that delayed development of JA in a high-risk population may diminish the specificity of this indicator in some populations. Studies using infant sibling designs require replication with alternative ascertainment methods in order to determine whether findings will generalize to the population of infants at risk for ASD.  More sensitive and specific measurement tools may be needed to detect early precursor signs of autism in the context of multiple risk factors.