A Behavioural Profile of Parent-Child Interaction Characteristics during the First Year of Life in High- and Low-Risk Infants

Background: During the first year of life the interaction between parent and child forms a key aspect of child’s social environment. Previous studies found that parent-child interaction (PCI) in children at high risk (HR) of autism spectrum disorder (ASD) is disrupted. This may be due to either parent or child behaviors that in an interaction between the child’s risk status and the social environment leads to an increasingly atypical development. More knowledge is needed about how early PCI differences arise in ASD and how they change over time. Answering these questions may help to strengthen PCI as a promising avenue for early intervention.

Objectives: (1) Investigating the inter-rater reliability of a newly developed coding scheme to rate PCI at 5 and 10 months of age; (2) Obtaining a detailed behavioural profile of the interaction between HR and LR infants and their caregiver, and to investigate differences between these groups.

Methods: As part of the Eurosibs Autism Research network (www.eurosibs.eu), the Parent-Infant/Toddler Coding of Interaction (PInTCI) was developed, including characteristics of PCI that predicted subsequent child development in ASD research. This global coding scheme consists of 5 child constructs (attentiveness – initiations – sharing of affect – positive affect – negative affect), 5 parent constructs (sensitive responsiveness – negative control – scaffolding – positive affect – negative affect) and 1 dyadic construct (dyadic reciprocity), rated on a 1-7 scale. Videotaped parent-child dyads were observed at 5 months (63 HR, 47 LR) and 10 months (93 HR, 73 LR). Coders were blind to risk status.

Results: (1) Intra-class correlations (ICC) showed excellent inter-rater reliability for all PCI constructs at 10 months (ICCs ranged from 0.79 to 0.96, all p ≤ 0.001). At 5 months, reliability ranged from good to excellent reliability (ICCs ranged from 0.68 to 0.95, all p < 0.05), except for a poor to fair reliability for infant initiations (ICC=0.47, p > 0.05) and parental negative affect (ICC=0.23, p > 0.05) (which were removed from further analyses). (2) Analyses adjusted for age, sex and site showed no differences at 5 months between HR and LR dyads (all p > 0.05). At 10 months, results revealed that HR siblings showed less initiations towards their parents as compared to LR siblings (p < 0.01). Differences in parent or dyadic behaviors were not found at 10 months (all p > 0.05). Examination of PCI characteristics across 5 and 10 months showed no significant effects of time (all p > 0.05).

Conclusions: Our results indicate that the PInTCI, based on its inter-rater reliability, seems to be a promising coding measure of PCI that can be used during the first year of life. Application of the PInTCI showed that HR and LR dyads did not differ at 5 months of age, but by the end of the first year dyads started to differ based on the child’s behaviour. Consistent with previous literature, our findings suggest that atypicalities start to emerge around the first birthday. Explanations for findings and recommendations for future research will be discussed.