Abstract
Previous research has assessed the relationship between maternal sensitivity and infant hypothalamic-pituitary-adrenocortical [HPA] axis function, yet neglected additional stress systems. Using a multi-system method (HPA measured via cortisol and sympathetic nervous system via salivary alpha-amylase; sAA), we assessed the relationship between maternal sensitivity and infant stress system coordination and flexibility in response to acute stress. A community sample of 125 mother-infant dyads participated in a toy frustration (age 15 months) and separation procedure (age 16 months). Maternal sensitivity was measured via naturalistic observation. Multilevel-modeling analyses found that maternal sensitivity moderates the relationship between infant sAA and cortisol basal activity and reactivity, such that systems were coordinated at higher, but not lower, levels of sensitivity. SAA output was greater in response to separation compared to frustration, though sensitivity did not moderate this variability. Findings suggest that the quality of early caregiving relationships is important for the development of coordinated stress physiology.
Highlights
Previous research has assessed the relationship between maternal sensitivity and infant hypothalamic-pituitary-adrenocortical [HPA] axis function, yet neglected additional stress systems
The current study further examines the relationship between maternal sensitivity and infant stress responsivity in a community sample of mother-infant dyads, with a particular focus on sympathetic nervous system (SNS) and hypothalamic-pituitary-adrenocortical (HPA) axis coordination and flexibility
To test the multiple hypotheses put forward, the full model of interest was inputted into multilevel modeling software (HLM-7, Scientific Software International): Salivary alpha-amylase (sAA) was selected as the outcome variable; cortisol and time were entered as level-1 predictors; stressor challenge as a level-2 predictor; and at the individual level, maternal sensitivity was grand mean centered
Summary
Following approval from the Research Ethics Boards at the Centre for Addiction and Mental Health and Ryerson University, a low-risk community sample of 297 mother-infant dyads were recruited from in-person visits and flyers posted at community centres, activity centres and infant-related events to participate in our laboratory’s original longitudinal study (Atkinson et al, 2013; Khoury et al, 2016; Ludmer et al, 2015; Pereira et al, 2012). To test the multiple hypotheses put forward, the full model of interest was inputted into multilevel modeling software (HLM-7, Scientific Software International): sAA was selected as the outcome variable (as opposed to cortisol which was the focus of previous work on this sample; Atkinson et al, 2013); cortisol (group mean centered) and time (coded as baseline = 0, peak = 1 and recovery = 2 ; uncentered) were entered as level-1 predictors; stressor challenge (uncentered, dummy-coded as toy frustration = 0, strange situation = 1) as a level-2 predictor; and at the individual level (level-3), maternal sensitivity was grand mean centered. It uses covariance and variance input from the final multilevel model and requires the selection of conditional values for each predictor of interest Where applicable, these conditional values were selected as the mean and ± 1 standard deviation values for cortisol and maternal sensitivity, or for sampling time, as the integer representing either baseline (0), peak (1), or recovery (2).
Sign-in/Register to view highlights & summary 
Submitted Version (
Free)
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call