Contextualizing the Relation Between Child Inhibition and Maternal Anxiety Beliefs: Demographic Characteristics and Multi-Method Assessment

• In Event: Poster Session 08
  In Poster Session: PS 08 Topic - Developmental Psychopathology

Abstract

Parenting behavior is crucial in childhood development. Protective parenting, defined by excessively comforting behaviors, predicts child anxiety outcomes (Edwards et al., 2010). Child factors predict parenting, with higher child inhibited temperament, characterized by social withdrawal, predicting more maternal protective parenting (Edwards et al., 2010). The mechanism by which child inhibition predicts protective parenting remains unknown. Social information and emotion processing theory provides theoretical support for cognition serving as a mechanism (Lemerise & Arsenio, 2000), positing that how individuals think about and emotionally (e.g. temperamentally) respond to social environments impacts behavior. Research suggests that cognitions mediate the relation between emotion processes and behaviors (Leerkes et al., 2016). There is also evidence that demographic characteristics such as maternal education and child gender affect how mothers think about and respond to their children (Mills & Rubin, 1990; Hastings & Rubin, 1999), thus it is important to understand whether this mechanistic process occurs consistently across different demographics.
The current study investigated whether maternal beliefs about their children’s anxiety are one mechanism by which child inhibition longitudinally predicts maternal protective parenting, and if demographics are contextualizing factors. 195 mother-child dyads participated at three timepoints when children were ages 1, 2, and 3 years. Demographics were collected at 1-yr. Child inhibition was assessed at all timepoints via maternal report (Infant-Toddler Social and Emotional Assessment-Revised; Carter et al., 2000) and behavioral observation during a standardized procedure (Fox et al., 2001). Maternal anxiety beliefs were assessed at all timepoints via questionnaire (adapted from Anxiety Sensitivity Index, Taylor et al., 2007). Protective parenting was assessed at the 2 and 3-yr timepoints via behavioral observation during another standardized procedure (Nachmias et al., 1996; Kiel & Buss, 2012).
Moderated mediation analyses were run with FIML in MPlus. Both models’ indices indicated good fit to the data. In Model 1, the interaction between 1-yr observed inhibition and 1-yr maternal education was significant (b= -.532, SE=.174, t= -3.059, p= .002) with 1-yr inhibition predicting more 2-yr maternal beliefs that their children’s anxiety was harmful at low maternal education (X̄-1 SD, b= .713, SE= .254, t= 2.811, p= .005) but not at mean or high maternal education (X̄+1 SD). In Model 2, the interaction between 1-yr mother-reported inhibition and child gender was significant (b= -1.603, SE= .548, t= -2.927, p= .003) with 1-yr mother-reported inhibition predicting more 2-yr maternal beliefs that anxiety was harmful for males (b= 1.095, SE= .401, t= 2.729, p= .006) but not for females. In both models, 2-yr maternal beliefs did not predict 3-yr protective parenting controlling for 1-yr inhibition and 2-yr protective parenting and the conditional indirect effects were not significant.
Results suggest that the longitudinal path from child inhibition to maternal anxiety beliefs is contextualized differently dependent upon the assessment method, with significant prediction found in select demographic by method contexts. Results do not provide support for maternal beliefs mediating the relation between child inhibition and maternal protective parenting. Post-hoc analyses to be presented at the conference will investigate other parenting outcomes within these models.

Authors

• Elizabeth Aaron, Miami University - Ohio
  Presenting Author

• Elizabeth J Kiel, Miami University - Ohio
  Non-Presenting Author