Maternal mood and contingency to infant distress in everyday settings

• In Event: Poster Session 05
  In Poster Session: PS 05 Topic - Parenting & Parent-Child Relationships

Abstract

Background:
In the current report we use home audio recordings captured by LENA to characterize parents’ contingent responses to infant distress. Parents’ responses to infant distress are thought to form a foundation for self-regulation (Bowlby, 1969, Leerkes et al, 2009). However, naturally occurring infant distress is notoriously difficult to observe in laboratory situations, meaning that researchers do not have much data to characterize this feature of caregiving (Hubbard & Van IJzendoorn, 1991). Additionally, rather than being a static feature of caregiving, parents’ contingency to infant distress is likely dynamic, changing as a function of parents’ mood or feelings of self-efficacy. In particular, while studies have shown that maternal depression is associated with reduced contingent responding to distress (Beebe et al, 2012, Field, 2009), it is possible that more subtle shifts in mood are associated with concomitant shifts in caregiving responses. Understanding the dynamics between mood and caregiving can help to reveal the processes by which well-known associations between maternal depression, reduced contingency and infant negative affectivity become established.

Aims/Hypotheses:
The aim of this study is to provide a descriptive analysis of caregivers’ contingency to infant distress in natural home settings, as well as explore maternal mood as a dynamic (ie time-varying) factor that might modulate contingent responding while caregiving over the course of a day. We hypothesize that on occasions where mothers report feeling more negative affect, they will show a lower proportion of contingent responses to infant distress.

Methods: Our sample includes 20 dyads each contributing a 24h LENA recording, randomly selected from a larger study (n=87) leveraging sensors to study daily activity. Complete 24-hour recordings have been annotated for onset and duration of each instance of infant cry (Cohen’s kappa = 0.77). Caregiver speech will be extracted from automated LENA outputs. For each episode of crying, we will calculate the caregivers’ contingent response, operationalized as a new vocal response occurring within three seconds of the onset of infant cry, consistent with previous studies (Van Egeren et al., 2001). Maternal negative affect is reported by mothers six times a day over the course of the recording period and will be synchronized with the LENA recordings.

Data Analysis Plan: All data have already been collected and all audio has been annotated for infant distress (fussing and crying). We will develop pipelines to extract LENA annotations for female vocal response and determine contingency for each episode of annotated distress by Dec 2020. We will then examine the most recent mood report (up to 3 hours prior to the episode of distress) as a time-varying predictor of contingency using multi-level models with individual episodes of distress nested within each dyad.

Authors

• Miriam Mikhelson, University of Texas at Austin
  Presenting Author

• Megan Micheletti, University of Texas at Austin
  Non-Presenting Author

• Xuewen Yao
  Non-Presenting Author

• Kaya Anna De Barbaro
  Non-Presenting Author