Cortisol reactivity, maternal sensitivity, and infant preference for mother's familiar face and rhyme in 6‐month‐old infants

Abstract

This study investigated how cortisol (stress) reactivity and mothers' behavioral sensitivity affect familiarity preferences in 6-month-old infants. Relations between sensitivity and stress were explored using saliva samples taken from mothers and infants before, and 20-min after, two preferential looking experiments. Photographs and voice recordings from infants' mothers were incorporated into standard visual preference tasks. Sensitivity was assessed by determining the degree of behavioral synchrony between mother and infant from a 10-min interaction period preceding the preferential looking experiments. Results showed that decreasing infant cortisol reactivity and greater maternal sensitivity were associated with familiarity preferences for mother's face stimuli. For the experiment with voice stimuli, a sex difference was obtained in the relationship between the directionality of cortisol reactivity and familiarity preferences. Results are related to a parallel study with 3-monthold infants (Thompson & Trevathan, 2008), and issues are discussed in terms of infants' developing emotional independence from mother. KeywordsCognitive; Infant; Social Interaction; Mothers; Quantitative Infants' ability to recognize caregivers' faces and voices is important for the development of attachment (Bowlby, 1969). Numerous studies have investigated the newborn infant's preferences for, and ability to discriminate, their mother's face and voice. For example, research has shown that newborn infants look for longer periods of time at their mother's face compared to a stranger's face (Bushnell, Sai, & Mullin, 1989;Pascalis et al., 1995;Walton, Bower, & Bower, 1992). In the auditory recognition domain, newborn infant's ability to recognize their mother's voice is a well-established finding (Spence & Freeman, 1996). Newborns can also discern the native tongue of their mother's voice (Mehler et al., 1988), and, at just 3 days of age, discriminate between unfamiliar, and familiar, story passages that were spoken by their mothers during the last trimester of prenatal development (e.g., DeCasper & Spence, 1986). At four months, event-related potential (ERP) studies show enhanced arousal and a clear memory template for mother's voice compared to an unfamiliar voice (Purhonen et al., 2005). By eight months, infants are able to preserve information about talker identity when exposed to the same talker over a 2-week time interval (Juscyzk, Hohne, Jusczyk & Redanz, 1993). Address correspondence to: Laura A. Thompson Department of Psychology MSC 3452/Box 30001 New Mexico State University Las Cruces, NM 88003 Phone: 575-646-4024 Fax: 575-646-6212 Email: thompson@nmsu.edu. 1 To establish whether or not there was a difference in cortisol reactivity between those infants with usable looking time data and those without, two-tailed paired samples t-tests were conducted between log 10 T1 and T2 values for infants whose looking time data could not be analyzed, separately for the faces and rhymes experiments. As was the case in the larger sam...

Do different data analytic approaches generate discrepant findings when measuring mother–infant HPA axis attunement?

“…We next examined the extent to which the change in the baby's cortisol correlated with the change in the mother's cortisol, a strategy used by a number of past researchers (Sethre‐Hofstad et al, ; Thompson & Trevathan, ; Van Bakel & Riksen‐Walraven, ). This type of correlation between signed difference scores has been called a dynamic correlation coefficient (Vroom, ).…”

“…Further complicating this issue is that studies of physiological attunement have used a number of different data analytic approaches, which themselves have different assumptions about the nature of the phenomenon being studied (see Butler, ). Studies have used: (1) simple correlations between maternal and child cortisol levels at the same time point (Middlemiss, Granger, Goldberg, & Nathans, ; Thompson & Trevathan, ), suggesting that attunement is an association between mother and child at individual time points; (2) simple correlations between mothers and children in their “difference” or “change” scores across two time points (Sethre‐Hofstad et al, ; Thompson & Trevathan, ; Van Bakel & Riksen‐Walraven, ), suggesting that attunement is an association between how mother and child change from one time point to the next; (3) linear regression (Van Bakel & Riksen‐Walraven, ), suggesting that attunement exists when one person's cortisol level can predict the other person's cortisol level; and (4) more complex multilevel models (e.g., growth models, mixed model ANOVAs, dual trajectory models; Atkinson et al, ; Crockett et al, ; Hibel et al, , 2015; Laurent et al, , 2012), suggesting that attunement is when mothers and infants change in the same way over time. For the multilevel approaches, depending on the analysis, attunement can be modeled as either reciprocal or nonreciprocal, and either concurrent or lagged.…”

“…However, there is notable variability in the findings (Table 1). Indeed, while numerous studies found clear evidence of attunement (e.g., Atkinson et al, 2013;Hibel, Granger, Blair, & Finegood, 2015;Laurent et al, 2012), others have not (e.g., Crockett, Holmes, Granger, & Lyons-Ruth, 2013; Thompson & Trevathan, 2009), and yet others found attunement only after accounting for a psychosocial risk variable such as maternal sensitivity, postpartum depression, household conflict or violence, or restrictive or punitive parenting (Atkinson et al, 2013;Field, Healy, Goldstein, & Guthertz, 1990;Granger et al, 1998;Hibel et al, 2009;Laurent, Ablow, & Measelle, 2011;Sethre-Hofstad et al, 2002;Van Bakel & Riksen-Walraven, 2008).…”

“…(1) simple correlations between maternal and child cortisol levels at the same time point (Middlemiss, Granger, Goldberg, & Nathans, 2012;Thompson & Trevathan, 2009), suggesting that attunement is an association between mother and child at individual time points;…”

Maternal Self-Reported Depressive Symptoms and Maternal Cortisol Levels Interact to Predict Infant Cortisol Levels

“…All procedures occurred between 9 and 10 a.m. Morning cortisol collection is common‐practice for infants (e.g., Azar et al., ; Castral et al., ; Dougherty et al., ; O'Connor, Bergman, Sarkar, & Glover, ; Thompson & Trevathan, ), as this is often the most convenient time for families and minimizes confounding cortisol levels with daytime changes in eating and napping routines (Goldberg et al., ; Gunnar & White, ; Thompson & Trevathan, ; van Bakel & Riksen‐Walraven, ). However, afternoon collection is preferable for adults to avoid the influence of the cortisol awakening response (CAR; Lovallo, Farag, & Vincent, ).…”