Associations in mother-infant response latency vary across infants at high- and low-neurogenetic risks.

• In Event: Language and Communication Across Development

Abstract

Maternal responsivity has been shown to be positively associated with language skill in children with and without neurodevelopmental disorders (NDD; McFadden & Tamis-LeMonda, 2013; Warren et al., 2010). It is argued that this association is driven by a bidirectional feedback loop between responsive maternal and child vocalizations (e.g., Warlaumont et al., 2014). However, the temporal windows required for these associations are unclear, and based on past research, may vary with child factors such as development and diagnosis (e.g., Bergeson et al., 2006; Newport et al., 1977). Understanding these associations is an essential first step to designing personalized treatment approaches that are sensitive to the developmental needs of diverse groups, including children with neurogenetic syndromes who are known to be at high-risk for abnormal language outcomes from the time of diagnosis. The objective of the current study is to examine how child and maternal temporal responsivity are associated in dyads with infants at low- and high-risk (NDD) for atypical development. We hypothesize that child and maternal temporal responsivity will be positively associated, but may differ with diagnosis.

Methods: We used the LENA system to collect and process daylong audio recordings from 89 dyads when infants were 6-24 months old. Preliminary analyses were conducted using a subset of the full sample (Table 1): 10 low-risk infants (LR), 10 infants with Angelman syndrome (AS), 10 infants with Down syndrome (DS), and 10 children with Fragile X syndrome (FXS). Using LENA recordings and corresponding .its files, we used the automated chattr R package (Casillas, in prep) to extract the temporal latency of both maternal and child responsive utterances during mother-child dyadic interactions. Maternal and child responses were considered responsive if they occurred within 2 seconds of the initial vocalization prompt (for either infant or adult).

Results: Preliminary results demonstrated a significant positive correlation between the average maternal response latency and average infant response latency in infants with AS (ρ=0.65, p=0.05). Further inspection revealed that the temporal latency of responses for both mother and child decrease as a function of age in AS dyads (Figure 1), producing a positive association between average mother and child response. However, the relationship between maternal and child responsivity did not reach significance for all other groups: LR (ρ=0.18, p=0.63), DS (ρ=0.32, p=0.37), and FXS (ρ=0.04, p=0.92). Final results will probe these associations in an expanded sample (n=89; LR=44, AS=11, DS=24, FXS=10).

Discussion: The current analyses reveal that maternal temporal responsivity is positively associated with child temporal responsivity in AS dyads, and that response latency for both mother and child decreases with the age of the infant in these dyads. Our planned full sample analysis will use mixed models to investigate the associations between maternal and child response latency during individual interactional blocks and will explore the effect of age on response latency for all groups with increased power. These results may inform how to best coach maternal responses to act as a potentially modifiable factor that could be targeted in parent-mediated interventions for children at high risk for atypical language development.

Authors

• Katiana Adela Estrada, Purdue University, West Lafayette
  Presenting Author

• Amanda Seidl, Purdue University

• Bridgette Kelleher, Purdue University, West Lafayette