Identifying Mechanisms Underlying Neurodevelopmental Disorders: Family-Wide and Inherited Influences on Early Social Attention Skills

Background: To understand mechanisms underlying the emergence of Autism Spectrum Disorder (ASD) and co-occurring conditions, we need to identify processes that mediate between genetic risk and later symptoms. One candidate process is attention, which gates what the child can learn. In fact, attention to people in the first years is highly heritable and atypical in infants who later receive diagnosis of ASD. However, this could be related to both social features of ASD, and co-occurring attention difficulties like Attention Deficit/Hyperactivity Disorder (ADHD). Here we test whether disruptions in attention towards faces could lie between genetic risk and outcome by looking at i) familial risk-group difference; ii) its association with quantitative familial risk for ASD and ADHD; iii) its association with polygenic scores.

Objectives: To investigate contribution of familial likelihood of early social and attention problems, wecombined genotypic, phenotypic and eye-tracking data from 334 14-month-old infants who participated in the British Autism Study of Infant Siblings (BASIS: http://www.basisnetwork.org/) and their older siblings.

Methods: i) For risk-group analysis, mean duration of the longest look at the face stimulus (henceforth, peak-look duration) during a face pop-out task was calculated for 222 infants with an older sibling with ASD and/or ADHD (high-risk), compared to 112 infants without family history of ASD (low-risk); ii) For relation to quantitative familial load, the Social Communication Questionnaire (SCQ) and Strengths and Difficulties Questionnaire (SDQ) were used to measure social and attentional problems in a subset of the infants’ older siblings with ASD (N=114); iii) To understand to what extent the aggregate effect of common genetic variants explains atypicalities in early social attention behavior, polygenic risk scores (PGS) for two neurodevelopmental disorders were obtained for 198 infants (173 high-risk, 62 low-risk). Summary statistics by the Integrative Psychiatric Research and the Psychiatric Genomics Consortium were used to compute PGS for ASD and ADHD.

Results: i) Peak-look to faces was longer in high-risk vs. low-risk infants (F(1,327)=19.4, p<0.001, Figure 1). ii) Longer peak-look duration was associated with more parent-reported social (b=0.02, p=0.009) and attention (b=0.08, p=0.013) difficulties in the older sibling with ASD. This confirmed that early differences in this measure of visual attention to social stimuli might be associated with familial burden for both ASD and ADHD. iii) Peak-look duration was not significantly predicted by ASD-PGS (Nagelgerke’s R²= 0.017, p=0.072) after controlling for possible design-related confounders such as recruitment phase, number of valid trials and age in months (b=945.97, p=0.23). ADHD-PGS significantly predicted peak-look duration (Nagelgerke’s R²= 0.22, p=0.038), even when controlling for design-related covariates (b=160.27, p=0.023, Figure 2).

Conclusions:

Disruptions in early social attention in high-risk infants quantitatively vary with ASD and ADHD symptoms in their older sibling. Moreover, a modest but significant contribution of genetic risk for ADHD might influence infants’ looking behavior when paying attention to people, having cascading effects on learning. Further work will examine the longitudinal relation to ASD and ADHD symptoms in toddlerhood. Taken together, our work highlights a promising new approach to understanding developmental mechanisms that underlie the emergence of neurodevelopmental disorders.