2. Unpredictability of Maternal Sensory Signals and Child Internalizing Behaviors in Middle Childhood

• In Event: Novel Approaches to Characterize Observed Maternal Behavior in Various Cultural Contexts

Abstract

Background: Exposure to unpredictable caregiving is a form of early life adversity and conveys risk for poor mental health among offspring (Glynn, 2018; 2019). However, we know far less about how patterns of specific maternal behaviors, including sensory signals, shape infant development. Patterns of sensory inputs such as auditory and visual, are important for the maturation of sensory neural circuits during infancy (Espinosa & Stryker, 2012; Khazipov et al., 2004). Recent work indicates cognitive and emotional circuits may be impacted as well (Birnie & Baram, 2022; Granger et al., 2021). Studies with children show that greater unpredictability of maternal sensory signals during mother-child interactions is associated with poorer cognitive performance and altered stress response among offspring (Davis et al., 2017; 2019; Noroña-Zhou et al., 2020). The relation between unpredictable maternal sensory signals and child socioemotional development is unknown. The current study addresses these gaps in the literature by evaluating how unpredictable maternal sensory signals in infancy relate to internalizing behaviors at age 9.
Method: Participants were 129 women (0.8% African American/Black, 9.3% Asian, 31.8% Hispanic/Latinx White, 11.6% Multi-Ethnic, 46.5% Non-Hispanic/Latinx White) and their children (43.4% female). The unpredictability of maternal sensory signals was evaluated when the children were 6 (M=6.1, SD=0.4) and 12 months old (M=12.1, SD=0.4). Sensory signals were coded from video-recordings of parent-child free-play interactions. For each mother, a transitional probability matrix was created reflecting the transitions among auditory, tactile and visual inputs they provided to their infant. Entropy rate was computed as a metric indicating the degree of unpredictability in which mothers transitioned between sensory states. Entropy rate at 6 and 12 months was moderately correlated and averaged to create a measure of unpredictability during the first year. When children were nine years old (M=9.6, SD=0.7), they provided self-report ratings of anxiety and depression symptoms. Global ratings of maternal sensitivity in infancy, income, child sex at birth and maternal postpartum depression symptoms were covaried in the analysis.
Results: High entropy rate, reflecting greater unpredictable sensory signals, was associated with higher child anxiety symptoms (β=.23, p=.009). Associations persisted after covarying for maternal sensitivity (β=.21, p=.024), and other covariates (β=.21, p=.033). Maternal unpredictable sensory signals were positively associated with child depression symptoms when tested independently (β=.16, p=.068). This association did not persist after inclusion of maternal sensitivity and other covariates (β=.08, p=.389).
Conclusion: The results suggest that exposure to unpredictable sensory signals in infancy may contribute to the development of anxiety symptoms more strongly than depression symptoms. Importantly, the link between unpredictable maternal sensory signals and anxiety symptoms among children persisted above and beyond global ratings of maternal sensitivity. The association between unpredictable maternal sensory signals and depression symptoms did not reach statistical significance. This may be due to the young age of our sample, or the use of a global measure of depressed mood. Our study shows that unpredictable maternal sensory signals is an important measure to consider for child developmental outcomes and specifically anxiety symptoms.

Authors

• Özlü Aran, University of Denver
  Presenting Author

• Danielle Swales, University of North Carolina at Chapel Hill

• Curt Sandman, University of California, Irvine

• Laura Glynn, Chapman University

• Hal Stern, University of California, Irvine

• Tallie Z Baram, University of California-Irvine

• Elysia Poggi Davis, University of Denver