Individual Submission Summary
Share...

Direct link:

Poster #145 - Modeling the Dynamics of Mother-Child Interactions Across Middle Childhood

Fri, March 22, 12:45 to 2:00pm, Baltimore Convention Center, Floor: Level 1, Exhibit Hall B

Integrative Statement

Mothers and children fight; how they fight is related to multiple aspects of children’s development (e.g., Burt, Krueger, McGue, & Iacono, 2003; Gerard, Krishnakumar, & Buehler, 2006). For this reason, developmental researchers have spent years studying the dynamics of mother-child conflict interactions. This field of research has yielded many valuable insights, however, it has been limited by the scope of existing statistical methods. Dynamic structural equation modeling (DSEM) is a Bayesian, multilevel extension of the autoregressive model that merges structural equation, multilevel, time-series, and time-varying effect modeling concepts. This novel statistical method provides substantive advances over existing methods, as it can be used to estimate fixed and random effects for auto-regressions, cross-lagged effects and residual variances in a dynamic system, after accounting for an individual’s typical behaviour. Since DSEM uses multi-level modeling, it is also possible to test correlations between parameters and regress individual parameters on various predictors using this model.

In this study, we used DSEM to model the dynamics of five-minute conflict discussions between mothers and their younger (ages 5-9, n=217) and older (ages 7-13, n=220) children. Results showed that there are notable differences in children’s behaviour across development, and in maternal behaviour with children at different developmental stages. Young children tend to be the least constructive, the most likely to get ‘stuck’ in a certain affective state, and have the highest residual variances, whereas older children tend to exhibit more constructive regulated behaviours. Mothers are even more constructive, regulated and predictable than older children. Examination of cross-lag effects reveals that while children do not tend to change their behaviour in response to that of their mothers, mothers are responsive to children, but the direction of their responsiveness changes as a function of child age. With very young children, mothers exhibit complementary behaviour; with older children, mothers exhibit synchronicity. Furthermore, among older children, mothers seem to be most responsive to those who are rigid (higher auto-regression, lower residual variance). These results help us to better isolate the causal effect of children’s behaviours on that of their mothers and understand the dynamics of mother-child conflict interactions across development. Further research using DSEM has the potential to greatly expand our understanding of real-time dynamics of dyadic interactions and how they relate to other developmental factors.

Authors