Introduction
The adverse and protective consequences of intrusive and nurturing parenting behaviors on children’s adjustment are well documented (Hastings et al., Reference Hastings, Nuselovici, Rubin, Cheah, Rubin and Coplan2010; Mills-Koonce et al., Reference Mills-Koonce, Willoughby, Garrett-Peters, Wagner and Vernon-Feagans2016; Rubin et al., Reference Rubin, Coplan and Bowker2009). However, the effects of parenting behaviors may vary depending on children’s characteristics, including children’s physiological regulation (Belsky & Pluess, Reference Belsky and Pluess2009; Ellis et al., Reference Ellis, Boyce, Belsky, Bakermans-Kranenburg and Van IJzendoorn2011; Wagner et al., Reference Wagner, Propper, Gueron-Sela and Mills-Koonce2016). For example, a large body of work suggests that children’s parasympathetic activity, commonly indexed by RSA, is one biomarker of individual differences in sensitivity to caregiving contexts (Conradt et al., Reference Conradt, Measelle and Ablow2013; Keller et al., Reference Keller, Kouros, Erath, Dahl and El-Sheikh2014; McLaughlin et al., Reference McLaughlin, Alves and Sheridan2014). That is, some children are more likely to be influenced by their caregiving environment and develop mental health problems under adverse conditions (e.g., when experiencing intrusive parenting), whereas others seem to be less affected, depending on their parasympathetic functioning – either when at rest or in response to psychosocial challenges.
Recent work suggests that additional insight into individual differences in these developmental processes may be gained by considering patterns of physiological regulation across the dyad (i.e., concordance) (Lan et al., Reference Lan, Zhang, Lunkenheimer, Chang, Li and Wang2024; Oshri et al., Reference Oshri, Liu, Suveg, Caughy and Huffman2023). In the current study, we advance this literature by characterizing patterns of dyadic RSA concordance between young children and their primary caregivers across tasks purposefully designed to mimic children’s exposure to novel social interactions while parents observed their children as they navigated the tasks. Furthermore, we examined whether dyadic RSA concordance and parenting behaviors (i.e., nurturing and intrusive behaviors) may uniquely and interactively predict children’s internalizing and externalizing behavior problems, as reported by multiple informants (i.e., teachers and parents).
The current study advances the literature by considering these questions using pretreatment data from a sample of young children high on behavioral inhibition (BI). BI is a temperamental predisposition to experience negative affect and behaviorally withdraw in the face of novelty (Fox et al., Reference Fox, Buzzell, Morales, Valadez, Wilson and Henderson2021; Kagan et al., Reference Kagan, Reznick and Gibbons1989) that has been well documented as an early antecedent to social reticence and anxiety (Chronis-Tuscano et al., Reference Chronis-Tuscano, Degnan, Pine, Perez-Edgar, Henderson, Diaz, Raggi and Fox2009; Rubin et al., Reference Rubin, Coplan and Bowker2009). Children with BI tend to be extremely shy and avoid novel situations and social interactions, particularly with unfamiliar others (Rubin et al., Reference Rubin, Coplan and Bowker2009). The current study contributes to the literature by considering the moderating role of dyadic RSA concordance among children who are at greater risk for later psychopathology during novel laboratory tasks designed to elicit stress associated with social exposures and novelty in children high on BI and parents’ physiological regulation while observing their children navigate these stressful experiences (Kagan, Reference Kagan, Beauchaine and Hinshaw2008; Rubin et al., Reference Rubin, Coplan and Bowker2009).
Physiological concordance
The PNS, a branch of the autonomic nervous system, exerts inhibitory and chronotropic (i.e., frequency) cardiac influences, which supports the ongoing and dynamic maintenance of homeostasis and the allocation of regulatory resources to engage with the environment (Cacioppo et al., Reference Cacioppo, Tassinary and Berntson2007). RSA represents the variability in the heart rate associated with a respiration cycle and is a specific measure of PNS functioning (Beauchaine, Reference Beauchaine2001; Porges, Reference Porges2007). Resting RSA and patterns of reactivity (i.e., withdrawal of PNS influence typically characterized by lower RSA relative to the resting state) and recovery (e.g., return to tonic or basal levels of functioning following stressors) represent key neurophysiological substrates of individual differences in regulation (Porges, Reference Porges2007). They provide insight into the capacity for flexible self-regulation that underlies effective emotion regulation, social communication, and attentional orienting (Porges, Reference Porges2007; Shakiba et al., Reference Shakiba, Lynch, Propper, Mills-Koonce and Wagner2023; Wagner et al., Reference Wagner, Shakiba, Bui, Sem, Novick, Danko, Dougherty, Chronis-Tuscano and Rubin2023). In response to arousing or stressful stimuli or external stressors, children tend to show a dynamic pattern of parasympathetic regulation, such that their mean RSA values tend to decrease from baseline RSA (i.e., suppression of RSA or vagal withdrawal) and increase and return to basal levels of functioning (i.e., vagal augmentation or recovery) when the source of stress is removed. In addition, individual variability in patterns of PNS functioning provides insight into risk for psychopathology and emotion dysregulation (Beauchaine, Reference Beauchaine2001, Reference Beauchaine2015; Wagner & Waller, Reference Wagner and Waller2020). Typically, lower resting RSA (Beauchaine, Reference Beauchaine2001; Crowell et al., Reference Crowell, Beauchaine, McCauley, Smith, Stevens and Sylvers2005) or aberrant patterns of RSA reactivity and recovery, often characterized by either exaggerated or reduced patterns of RSA change to the stressors and/or slower patterns of recovery, reflect reduced regulatory capacity to actively cope with external demands (Wagner et al., Reference Wagner, Shakiba, Bui, Sem, Novick, Danko, Dougherty, Chronis-Tuscano and Rubin2023) and are associated with increased risk for internalizing and externalizing psychopathology (Beauchaine, Reference Beauchaine2015; Beauchaine & Thayer, Reference Beauchaine and Thayer2015; Porges, Reference Porges2007).
Existing research documents caregiver-child physiological concordance (also termed synchrony, coregulation, covariation, attunement, or linkage) as another important biomarker of risk and resilience for childhood psychosocial and behavioral adjustment (DePasquale, Reference DePasquale2020; Feldman, Reference Feldman2007, Reference Feldman2017; Lunkenheimer et al., Reference Lunkenheimer, Tiberio, Buss, Lucas-Thompson, Boker and Timpe2015). Consistent with previous research, the current study defines dyadic parent–child physiological concordance as the coordination or matching of biological states or similar patterns of physiological responses or stress reactivity patterns between the caregiver and the child (Feldman, Reference Feldman2012), and it promotes children’s regulatory development (Feldman, Reference Feldman2012; Morris et al., Reference Morris, Cui, Criss and Simmons2018). Indeed, in a recent meta-analytic report, Miller and colleagues (Reference Miller, Armstrong-Carter, Balter and Lorah2023) suggested that the extent to which a caregiver and her/his child exhibited concordant patterns of regulation may reflect dyadic processes of mutual responsivity and coregulation. Furthermore, researchers have suggested that parent–child physiological concordance helps to shape children’s neurobiological, social, and emotional development, and it is thought to influence children’s ability to self-regulate and engage positively with their environment (Bell, Reference Bell2020; Feldman, Reference Feldman2007, Reference Feldman2012).
Although there are some inconsistencies within the literature, overall, studies seem to report the positive concordance, or similar patterns of parent and child RSA responses, to be associated with an array of positive outcomes in children, including adaptive socioemotional development, more prosocial and empathetic behaviors, self-control, better school adjustment and socialization, and lower levels of child externalizing symptoms (Feldman, Reference Feldman2007; Lunkenheimer et al., Reference Lunkenheimer, Olson, Hollenstein, Sameroff and Winter2011; Moffitt et al., Reference Moffitt, Arseneault, Belsky, Dickson, Hancox, Harrington, Houts, Poulton, Roberts, Ross, Sears, Thomson and Caspi2011; Suveg et al., Reference Suveg, Shaffer and Davis2016). In contrast, negative concordance or dissimilar patterns of caregiver-child dyadic RSA responses (i.e., as one person’s RSA increases, the other person’s RSA decreases, and vice versa) seem to reflect disrupted biobehavioral regulatory processes in the dyad, which may impact children’s developing biobehavioral regulatory processes and, in the long term, confer risk for psychopathology (Fuchs et al., Reference Fuchs, Lunkenheimer and Lobo2021; Lunkenheimer et al., Reference Lunkenheimer, Tiberio, Buss, Lucas-Thompson, Boker and Timpe2015, Reference Lunkenheimer, Busuito, Brown and Skowron2018). However, it is important to note that whether positive or negative dyadic RSA concordance is associated with or serves as a risk or resilience factor for child’s adjustment and developmental outcomes depends, in part, on the context (i.e., acutely stressful or emotionally challenging vs. nonstressful and relaxing conditions) in which dyadic RSA concordance is measured, as well as how well-regulated or dysregulated the two members of the dyads are (DePasquale, Reference DePasquale2020; Lunkenheimer et al., Reference Lunkenheimer, Busuito, Brown and Skowron2018; Palumbo et al., Reference Palumbo, Marraccini, Weyandt, Wilder-Smith, McGee, Liu and Goodwin2017). For example, in the context of stress and emotionally evocative tasks, when the child exhibits suppressed parasympathetic activity – often considered a less typical or potentially maladaptive pattern of regulation – a parent who shows augmented parasympathetic activity – often considered a more typical and potentially adaptive pattern – may exert a buffering or soothing effect on the child’s physiological functioning. In this case, negative rather than positive concordance could be associated with better outcomes for the child. Therefore, physiological concordance may function differently and have distinct developmental and adjustment implications for children across various contexts, parent–child dyads, and developmental periods, during which the nature and function of the parent–child interaction may change (DePasquale, Reference DePasquale2020). In addition, studies often use different methodological approaches to model parent–child dyadic concordance and synchrony over time and across context, which may contribute to the existing diverse findings for the link between dyadic RSA concordance and children’s adjustment (Helm et al., Reference Helm, Miller, Kahle, Troxel and Hastings2018; Palumbo et al., Reference Palumbo, Marraccini, Weyandt, Wilder-Smith, McGee, Liu and Goodwin2017; Stallworthy et al., Reference Stallworthy, Elison and Berry2024).
The innovative paradigm design used in the current study was informed by Rubin’s developmental-transactional model of child BI and anxiety (Rubin et al., Reference Rubin, Coplan and Bowker2009) and was intended to provide insight into highly behaviorally inhibited children’s physiological reactivity to social stressors and parents’ physiological reactivity to observing their child navigate these stressors. Parents of BI children often perceive them as vulnerable and exhibit overly protective and controlling behaviors (e.g., accommodation) as a result (Kennedy et al., Reference Kennedy, Rubin, Hastings and Maisel2004; Hastings et al., Reference Hastings, Nuselovici, Rubin, Cheah, Rubin and Coplan2010; Root et al., Reference Root, Hastings and Rubin2016; Rubin et al., Reference Rubin, Coplan and Bowker2009). An important component of understanding the determinants of parenting is the neurophysiological processes that support them (Hastings et al., Reference Hastings, Rubin, Smith and Wagner2019; Kennedy et al., Reference Kennedy, Rubin, Hastings and Maisel2004; Root et al., Reference Root, Hastings and Rubin2016). When children face novelty, challenge, or threat in the environment, parents must draw on these physiological systems to self-regulate to adequately respond to the needs of their child. The current study intentionally uses the term “concordance” rather than “synchrony” (as in dyadic mutually reciprocal influences) to distinguish our focus on the coordination or matching of biological states or similar patterns of physiological responses or stress reactivity patterns between the caregiver and the child (Feldman, Reference Feldman2012). Within this context, concordance provides unique insight into how perceptions of threats or challenges in the environment operate at the level of the dyad, which advances our understanding of processes underlying the intergenerational transmission of risk for anxiety disorders among BI children (Hastings et al., Reference Hastings, Nuselovici, Rubin, Cheah, Rubin and Coplan2010; Rubin et al., Reference Rubin, Coplan and Bowker2009). Although the current study focuses on physiological concordance, the broader literature on dyadic synchrony during parent–child interactions provides a useful foundation for considering the aims of the current study.
Physiological concordance as a moderator of parenting influences on children’s adjustment problems
A preponderance of research highlights the important and enduring influences that parents have on every aspect of children’s development (McKee et al., Reference McKee, Colletti, Rakow, Jones and Forehand2008; Rankin Williams et al., Reference Rankin Williams, Degnan, Perez-Edgar, Henderson, Rubin, Pine and Fox2009). In particular, warm and nurturing parenting promotes children’s psychosocial adjustment and well-being, whereas harsh and intrusive parenting confers risk for adjustment problems, such as externalizing and internalizing psychopathology (Lupien et al., Reference Lupien, McEwen, Gunnar and Heim2009; Murray et al., Reference Murray, Dwyer, Rubin, Knighton-Wisor and Booth-LaForce2014; Van Zeijl et al., Reference van Zeijl, Mesman, Stolk, Alink, van IJzendoorn, Bakermans-Kranenburg, Juffer and Koot2007; Zheng et al., Reference Zheng, Pasalich, Oberth, McMahon and Pinderhughes2017). Parenting behaviors have also been shown to moderate the risk for social anxiety among behaviorally inhibited youth (Fox et al., Reference Fox, Zeytinoglu, Valadez, Buzzell, Morales and Henderson2023; Hastings et al., Reference Hastings, Nuselovici, Rubin, Cheah, Rubin and Coplan2010; Suarez et al., Reference Suarez, Morales, Miller, Penela, Chronis-Tuscano, Henderson and Fox2021). Parents of children with BI often perceive them as vulnerable in the face of novel, unfamiliar, and socially challenging situations (Hastings et al., Reference Hastings, Nuselovici, Rubin, Cheah, Rubin and Coplan2010; Rubin et al., Reference Rubin, Coplan and Bowker2009). Therefore, they are more likely to become physiologically aroused when they see their child becoming (or expect them to become) anxious or upset in the face of novelty and unfamiliar social interactions. These parents are also more likely to practice overly protective and controlling behaviors (Kennedy et al., Reference Kennedy, Rubin, Hastings and Maisel2004; Root et al., Reference Root, Hastings and Rubin2016), which may, in turn, predict more behavior problems in these children (Fox et al., Reference Fox, Zeytinoglu, Valadez, Buzzell, Morales and Henderson2023; Hastings et al., Reference Hastings, Nuselovici, Rubin, Cheah, Rubin and Coplan2010; Rubin et al., Reference Rubin, Coplan and Bowker2009).
Nevertheless, existing research documents a wide range of individual variability in the extent to which parenting behaviors, including both nurturing and harsh parenting behaviors, differentially impact children’s adjustment (Hastings et al., Reference Hastings, Rubin, Smith and Wagner2019; Oshri et al., Reference Oshri, Liu, Suveg, Caughy and Huffman2023; Shakiba et al., Reference Shakiba, Gao, Conradt, Terrell and Lester2022; Wagner et al., Reference Wagner, Propper, Gueron-Sela and Mills-Koonce2016). A promising line of research suggests that heterogeneity in children’s responses to their rearing experiences and subsequent adjustment outcomes could also be partially explained by the pattern of physiological concordance or coregulation between children and their caregivers (i.e., whether they show similar or different patterns of regulation). In a longitudinal, multilevel study of economically disadvantaged youths (M age = 10 years old) and their families, Oshri and colleagues (2023) reported the highest levels of externalizing and internalizing symptoms among youths who were exposed to negative parenting (e.g., parents’ use of coercion, intrusiveness, and negativity towards the child) and also showed high positive RSA synchrony patterns with their parents during a parent–child conflict task. Conversely, they observed the lowest levels of psychopathology symptoms among youths who also exhibited high positive dyadic RSA synchrony, but experienced positive parenting (e.g., parents’ use of reasoning, positivity towards the child, and respect for the child’s autonomy) during the conflict task (Oshri et al., Reference Oshri, Liu, Suveg, Caughy and Huffman2023). The authors further concluded that positive dyadic RSA synchrony could function as a neurobiological susceptibility factor that influences youths’ responses to positive and negative parenting behaviors in a “for better or for worse” manner.
In another longitudinal study of a rural, low-SES, community sample of parents and their 6-month-old infants, Lan and colleagues (2023), found evidence for the moderating role of parent-infant RSA synchrony in the association between maternal depressive symptoms and children’s internalizing problems. In particular, these researchers found greater internalizing problems among 24-month-old infants who showed low (but not high) levels of RSA synchrony with their parents during the free play task whose mothers also had greater depression when the child was 6 months of age. The authors further suggested that high mother–infant RSA synchrony or coregulation may be a resilience factor that mitigates the intergenerational transmission of internalizing psychopathology in community samples.
Much of the research examining individual differences in children’s level of susceptibility to caregiving conditions has been conducted with typically developing infants and young children (e.g., Conradt et al., Reference Conradt, Measelle and Ablow2013; Gueron-Sela et al., Reference Gueron-Sela, Wagner, Propper, Mills-Koonce, Moore and Cox2017; Somers et al., Reference Somers, Jewell, Hanna Ibrahim and Luecken2019), resulting in limited evidence for at-risk populations, particularly those who may have difficulties with regulating their own emotions, behaviors, and physiological responses in the face of novelty and socially challenging situations, as is the case for children high on BI (Davis et al., Reference Davis, West, Bilms, Morelen and Suveg2018). Furthermore, very few studies have examined dyadic physiological concordance as a moderator of the influences of early rearing experiences on children’s adjustment problems, particularly among at-risk children. The current study contributes to this literature by examining whether parent–child dyadic RSA concordance moderates the links between nurturing and/or intrusive parenting behaviors and behavioral adjustment (i.e., internalizing and externalizing problems) in a sample of young children with elevated BI. Our analytic approach allowed for the characterization of dyadic RSA concordance between children and their caregivers and for tests of potential moderated associations between parenting, dyadic concordance, and children’s adjustment problems. Although there is support for considering the role of RSA concordance in the links between parenting and child outcomes, to our knowledge, no studies have characterized dyadic physiological concordance among a sample of children with elevated BI who tend to experience high levels of anxiety and negative affect, are socially reticent, and behaviorally withdraw in the face of threat, novelty, and socially challenging situations (e.g., meeting unfamiliar peers; Fox et al., Reference Fox, Henderson, Marshall, Nichols and Ghera2005; Reference Fox, Zeytinoglu, Valadez, Buzzell, Morales and Henderson2023; Kagan et al., Reference Kagan, Reznick, Clarke, Snidman and Garcia-Coll1984).
The current study
First, we employed a multilevel modeling (MLM) approach to model parent–child parasympathetic regulation (i.e., the direction of parent–child RSA concordance) in a sample of highly behaviorally inhibited children and their primary caregivers. Unlike tasks used to measure dyadic physiological synchrony, we utilized a series of social stressor laboratory tasks designed to elicit physiological arousal in inhibited young children while their parents observed their children navigating these tasks. Second, we expanded the models to test whether nurturing and/or intrusive parenting behaviors and parent–child RSA concordance across tasks are associated with children’s adjustment problems, including internalizing and externalizing behavior problems. Third, we investigated whether parent–child RSA concordance across tasks moderates the effects of parenting behaviors on children’s adjustment problems. We took a multi-informant approach to gain a better understanding of children’s adjustment problems across different contexts. Both parents and teachers reported separately on children’s internalizing and externalizing problems. We defined and operationalized parent–child RSA concordance as within-dyad similarities or dissimilarities in the pattern of RSA change across socially challenging stressors in the laboratory. Specifically, we defined positive concordance as the positive prediction of the mother’s RSA by the child’s RSA (i.e., the caregiver’s and child’s RSA change in the same direction across the tasks at the same time; both the child’s and mother’s RSA increase or decrease simultaneously). By contrast, we defined a negative concordance as the negative prediction of the mother’s RSA by the child’s RSA (i.e., the caregiver’s and child’s RSA change in the opposite direction across tasks at the same time; as the child’s RSA increases, the mother’s RSA decreases simultaneously, and vice versa).
Given the limited body of research on cardiac parasympathetic regulation among samples of children with BI, we did not make any preliminary assumptions about the nature and influence of parent–child RSA concordance on children’s adjustment. However, a growing body of evidence supports our preliminary expectation that the effects of parenting behaviors on children’s adjustment would vary as a function of dyadic RSA concordance across socially challenging stressors.
Method
Participants
To test the study’s hypotheses, we used the pretreatment (i.e., baseline) data from a randomized controlled trial (ClinicalTrials.gov registration: NCT02308826) early intervention study for children with elevated BI (Chronis-Tuscano et al., Reference Chronis‐Tuscano, Novick, Danko, Smith, Wagner, Wang, Druskin, Dougherty and Rubin2022). The original intervention study has a longitudinal design which took place at different time points (i.e., pretreatment, mid-treatment, immediate posttreatment, and 1-year follow-up), but the pretreatment data used for the current study are cross-sectional. Participants (n = 151) were recruited from preschool and pre-kindergarten programs, childcare centers, pediatricians, and community organizations in the greater Washington D.C. area, metropolitan area. At the time of recruitment, eligibility criteria required that children: (1) were between ages 45 to 64 months old (M = 52.9 months, SD = 5.7 months) and attending a structured school; (2) scored within the top 15% on the parent-related Behavioral Inhibition Questionnaire (BIQ; Bishop et al., Reference Bishop, Spence and McDonald2003) based on conceptual models of risk associated with heightened BI in this range (Henderson et al., Reference Henderson, Marshall, Fox and Rubin2004; Kagan et al., Reference Kagan, Snidman, Kahn, Towsley, Steinberg and Fox2007); (3) were never diagnosed with an ASD; and had a score below the clinical cutoff (≤15) on the Social Communication Questionnaire (SCQ; Eaves et al., Reference Eaves, Wingert, Ho and Mickelson2006), a parent-report screen for social/communication deficits characteristic of ASD; (4) were not be receiving any anxiety treatments, and; (5) did not have selective mutism. In total, 151 families met the inclusion criteria. For more information about the original randomized controlled trial, see (Chronis-Tuscano et al., Reference Chronis‐Tuscano, Novick, Danko, Smith, Wagner, Wang, Druskin, Dougherty and Rubin2022). The current analyses involve only the pretreatment data (n = 107 parent–child dyads), which were collected at the first laboratory visit. At the pretreatment, teachers also reported on children’s behaviors at the preschool or daycare settings. `
At pretreatment, parents (n = 107) reported on children’s demographic information as 54% female, 47% European American, 13.5% Asian American, 11% African American, and 28.5% other racial groups. Parents reported 9% of children as Hispanic or Latinx. Participating primary caregivers were 84% female (n = 90), well educated (92.2% held a bachelor’s degree), racially diverse (6.6% Hispanic or Latino/a; 13% Asian American; 12.3% African American; 61.3% European American; 13.4% other/not reported), and reported annual household incomes ranging from > $24,999 to < $150,000 (M = $100,000 – $124,999).
Procedure
The study was approved by the University of Maryland Institutional Review Board. Interested parents were invited for a telephone screening to assess participants using the study’s inclusion criteria. Parent–child dyads who met the criteria provided informed consent and were invited to participate in a laboratory-based assessment, which involved diagnostic interviews, parent-report questionnaires, and social stressors tasks.
Laboratory protocol
The baseline laboratory visit started by having the child seated at a small desk to complete the novelty tasks that were presented on a computer screen. The child’s parent was seated behind the child during the socially challenging conditions, so the child could not see the parent, but the parent could see the tasks the child was presented with and their reactions. The laboratory tasks (Figure 1) consisted of three Calm-Video conditions, each lasting 3 minutes, during which the children were shown a Sesame Street music lullaby. Each of the three Calm-Video conditions was presented prior to each of the socially challenging conditions (Wagner et al., Reference Wagner, Shakiba, Bui, Sem, Novick, Danko, Dougherty, Chronis-Tuscano and Rubin2023) to establish and record the child’s physiological resting state. They also provided opportunities for the child’s physiological recovery following each of the socially challenging conditions. A Clown condition was the first socially challenging condition that was administered immediately following the Calm-Video 1. The task was designed to mimic the socially challenging condition of a child’s speaking to an unfamiliar and strange adult. During the Clown condition, a research assistant dressed as a clown entered the assessment room through a door that was immediately adjacent to the computer screen located in front of the child, waited 10 s while staring at the child with neutral affect, and then engaged in a structured conversation for approximately 60 s. The Clown condition is a well-validated paradigm commonly used in studies examining BI and social fear (Fox et al., Reference Fox, Henderson, Rubin, Calkins and Schmidt2001; Kagan et al., 1989; Rubin et al., Reference Rubin, Burgess and Hastings2002). The Calm-Video 2 condition was shown to the child immediately after the Clown condition. The second socially challenging condition, the Kids condition, was administered following the Calm-Video 2, for which the experimenter explained to the child that they wanted to introduce several children with whom the child might meet in the future. The Kids condition was designed to mimic the common experiences of getting introduced to unfamiliar, same-aged peers. This condition involved viewing images of 6 smiling children’s faces at 10-second intervals. Smiling images for the fictitious children were taken from the Child Affective Facial Expression set (LoBue & Thrasher, Reference LoBue and Thrasher2014). For the first 5 s, only the image of a fictitious child appeared on the screen. For the next 5 s, the image of the child was accompanied by clip art indicating an object or a place the fictitious child enjoys. The experimenter offered the fictitious child’s name and a description of the fictitious child’s interests during the task. In total, participating children were shown six fictitious children, which were counterbalanced on race and biological sex. The current study is the first to assess parents’ physiological responses to stress while observing the tasks and how their behaviorally inhibited children navigate these stressful experiences.
Figure 1. Laboratory tasks used to assess parent–child dyadic RSA concordance. These tasks were purposefully designed to mimic exposure to novel social interactions in behaviorally inhibited children.
Next, during the Introduction condition, the experimenter asked the child to introduce him- or herself to “the kids who you just learned about so that they can learn something about you” and directed the child’s attention to a camera mounted on a nearby wall. The Introduction condition was designed to mimic aspects of the Trier Social Stress Test (Kirschbaum et al., Reference Kirschbaum, Pirke and Hellhammer1993), a well-established validated task to elicit changes in the functioning of the autonomic nervous system and the hypothalamic-pituitary-adrenal axis (Kudielka et al., Reference Kudielka, Hellhammer, Kirschbaum, Harmon-Jones and Winkielman2010). The camera that the child was asked to speak into was mounted next to a mirror to reinforce self-conscious emotional states. If the child remained silent (which was common), the experimenter then offered standardized prompts such as, “You can tell them how old you are” or “You can describe something you like to do.” Children were given 90 s to complete the task. Finally, the Calm-Video 3 condition was presented right after the completion of the Introduction task. RSA reactivity during these tasks has been successfully modeled over time and has been shown to relate to children’s anxiety (Wagner et al., Reference Wagner, Shakiba, Bui, Sem, Novick, Danko, Dougherty, Chronis-Tuscano and Rubin2023).
Measures
Behavioral inhibition
The BIQ (Bishop et al., Reference Bishop, Spence and McDonald2003) was used to assess children’s BI and whether they meet the inclusion criteria for participation in the Study. The BIQ is a 30-item parent report that assesses the child’s BI across social novelty, situational novelty, and physical activity with a minor risk of injury. On a 7-point Likert scale ranging from 1 (hardly ever) to 7 (almost always), parents reported whether each behavior occurs for their child. Examples of items include “is shy when first meeting new children,” “dislikes being the center of attention,” and “takes many days to adjust to new situations.” The BIQ has shown good convergent and discriminant validity (Kim et al., Reference Kim, Klein, Olino, Dyson, Dougherty and Durbin2011) and internal consistency (α = 0.72 – 0.95; Bishop et al., Reference Bishop, Spence and McDonald2003; Kim et al., Reference Kim, Klein, Olino, Dyson, Dougherty and Durbin2011). In this sample, the Cronbach alphas ranged from 0.87 – 0.94.
Parent–child dyadic RSA concordance
Parents’ and children’s RSA were derived from data collected simultaneously using the Biopac MP150 data acquisition system at a sampling rate of 2000Hz. Both parents and children wore wireless Bio-Nomadix PPG-ED transmitters (“BIOPAC Systems, Inc.” 2005) throughout the series of lab tasks. In a control room adjacent to the assessment room, two separate video streams were recorded concurrently with the heart rate data, which were used by research assistants to identify timing designations for the four different conditions that occurred during each assessment: Calm-Video (repeated three times), Clown, Kids, and Introduction. The first video captured the child from a wall-mounted camera feed, while the second video captured the computer screen where all the videos and other visual cues were presented to the child.
For both parents and children, measures of RSA across all the tasks were derived from inter-beat interval (IBI) data, following the “Porges–Bohrer method” (Porges & Bohrer, Reference Porges and Bohrer1990) through which the IBI data were first derived from a peak identification algorithm (“BIOPAC Systems, Inc.” 2005), then visually inspected and edited for potential movement artifact (Porges, Reference Porges1985; Brain-Body Center for Psychophysiology and Bioengineering) by trained and reliable research assistants. Finally, RSA, the high-frequency component of heart rate variability, was calculated across 15-second epochs across all conditions (Calm Videos, Clown, Kids, Introduction) using methods by Porges and Byrne (Reference Porges and Byrne1992). The Porges-Bohrer method conforms to the assumptions of parametric analysis, is not moderated by respiration, is robust to nonstationarity, and is the most sensitive to vagal blockade (Lewis et al., Reference Lewis, Furman, McCool and Porges2012). No more than 5% of the included data required editing, and approximately 6% of average missingness across tasks was due to movement artifacts.
Parenting behaviors
Parents reported on their intrusive and nurturing parenting behaviors using the restrictiveness and nurturance subscales of the Child-Rearing Practice Report (CRPR; Rickel & Biasatti, Reference Rickel and Biasatti1982), respectively. Example items for the restrictiveness subscale include “I try to keep my child away from children or families who have different ideas or values from our own” and “I prefer my child not try things if there is a chance he/she might fail.” Example items for the nurturance subscale include “I express my affection by hugging, kissing, and holding my child” and “I encourage my child to be curious, to explore, and question things.” The CRPR has 36 items which are rated on a 6-point Likert scale (1 = strongly disagree, 6 = strongly agree) that assess parents’ child-rearing values and attitudes. Internal consistency for restrictiveness (α = 0.79) and nurturance (α = 0.68) subscales were adequate, and comparable to estimates reported in other studies with a sample of parents of primary and middle school children (α = 0.82, 0.85 respectively; Booth-LaForce et al., Reference Booth-LaForce, Oh, Kennedy, Rubin, Rose-Krasnor and Laursen2012).
Child’s adjustment problems
Teachers reported on children’s adjustment problems using the Teacher Report Form (TRF; Achenbach & Rescorla, Reference Achenbach and Rescorla2000), which is designed primarily to be completed by daycare providers, teachers, and other adults who work closely with children in a daycare or preschool setting. The TRF questionnaire includes 99 items and assesses children’s emotional and behavioral functioning. On a 3-point Likert scale ranging from 1 (not true) to 3 (very/often true), teachers reported the presence of externalizing (e.g., attention problems: fails to carry out assigned tasks, and aggressive behaviors: disturbs other children) and internalizing (e.g., emotionally reactive: sudden changes in mood or feelings, anxious/depressed: nervous, high-strung, or tense, somatic complaints: headaches without medical cause, and withdrawn: doesn’t answer when people talk to him/her) behavioral symptoms in the child over the past 2 months. We used the T scores in all analyses. Internal consistency was good for externalizing (α = .94) and internalizing behaviors (α = 88).
Parents also reported on their child’s adjustment problems using the Child Behavior Checklist (CBCL; Achenbach, Reference Achenbach1991), which, similar to the TRF measure, includes 99 items. This well-validated measure assesses children’s emotional and behavioral functioning relative to their same-aged and same-sex peers. On a 3-point Likert scale ranging from 1 (not true) to 3 (very/often true), parents reported the presence of externalizing (e.g., attention problems: quickly shifts from one activity to another and aggressive behaviors: destroys things belonging to his/her family or other children) and internalizing (e.g., emotionally reactive: upset by new people or situations, anxious/depressed: unhappy, sad, or depressed, somatic complaints: painful bowel movements without medical cause, and withdrawn: shows little affection toward people) behavioral symptoms over the past 6 months. We used the T scores in all analyses. Internal consistency was good for externalizing (α = 0.91) and internalizing (α = 0.84) behaviors.
Parent’s anxiety
The ADIS-V – Adult and Lifetime Version (ADIS-V-L; Brown & Barlow, Reference Brown and Barlow2014) is a gold-standard semi-structured diagnostic interview, used to assess parent’s current anxiety using the DSM-5 criteria. Clinician Severity Ratings (CSRs) were summed for separation, specific, social, and generalized anxiety modules to calculate total parent anxiety severity (Ginsburg et al., Reference Ginsburg, Drake, Tein, Teetsel and Riddle2015). CSRs ranged from 0 to 8, where scores ≥ 4 are clinically significant. Twenty percent of clinician interviews were coded for inter-rater reliability (ICC = .93).
Data analytic plan
We performed a series of multilevel models (MLM) to test the study’s aims. Using the MLM allowed us to account for the nested nature of the data (i.e., repeated assessment of RSA responses nested within parent/child and dyads across tasks). To conduct the MLM models, at the first level (i.e., within-dyad), we characterized parent–child dyadic RSA concordance across all the laboratory social stressors, in which cRSAi,t and pRSAi,t denoted the ith child’s and parent’s average of the 15-second epochs RSA values, respectively, at time t (the total of 42 epoch-level timepoints across all the social stressors). As shown in Equation 1 (Eq. 1 below), we modeled RSA concordance (denoted as Concori) as the effect of the ith child’s RSA (cRSAi,t) on the parent’s RSA (pRSAi,t). The paradigm employed in the study was purposefully designed to gain insight into children’s patterns of physiological regulation across a series of tasks that mimic social exposures and novelty, and parents’ physiological regulation while observing the tasks and how their children navigate these stressful experiences.
Eq. 1. Level 1 (within-dyad):
pRSA i, t = Intercept0 + Concori* cRSAi,t + Epochi,t + ri,t
Next, the model was expanded at level 2 (i.e., between-dyads), as shown in Equation 2 (Eq. 2 below), to examine the effects of parenting behaviors (i.e., nurturing and intrusive behaviors) and dyadic RSA concordance on children’s parent- and teacher-reported internalizing and externalizing behavior problems. For each set of the models, the outcomes (i.e., internalizing or externalizing problems) were first regressed on dyadic RSA concordance, parenting behaviors, and study covariates (i.e., the main effect model).
Eq. 2. Level 2 (between-dyad):
Intercept00 + γ01 (intrusive/nurturing parenting)i + γ02 (Concori)i + u0i + ri,t
Next, we examined whether and how dyadic RSA concordance moderates the associations between parenting and children’s adjustment problems (i.e., interaction model).
Finally, significant interactions were probed by estimating the simple slopes at negative (− 1SD) and positive (+ 1SD) dyadic RSA concordance and the regions of significance (RoS), using the analytic approach recommended by Roisman et al. (Reference Roisman, Newman, Fraley, Haltigan, Groh and Haydon2012). The RoS analyses, which identify the exact range of values of the moderator for which the independent variable significantly predicts the dependent variable, were used to determine at which levels of dyadic RSA concordance parenting behaviors predict children’s adjustment problems, but also for which levels of parenting behaviors these associations were significant. The analytic approach proposed by Roisman and colleagues is specifically recommended when investigators are not certain about the type and nature of interaction effects that they aim to examine. Parenting predictors were mean-centered to reduce multicollinearity and to aid the interpretation of simple slopes. Parent and child RSA levels were also grand-mean-centered at each level of the MLM. All the MLM analyses were conducted using Mplus version 8.6 (Muthen & Muthen, Reference Muthen and Muthen1998–Reference Muthen and Muthen2017). The exploratory analyses for probing the significant interactions were performed using a web-based application developed by R. Chris Fraley (https://www.yourpersonality.net/interaction/). This application specifically follows the analytic approach of Roisman et al. (Reference Roisman, Newman, Fraley, Haltigan, Groh and Haydon2012) and provides a range of tools for investigators to explore interactions in multiple regression, helping them better understand the implications for alternative models of risk and resilience. Missing data were handled using the full-information maximum likelihood estimation (Enders & Bandalos, Reference Enders and Bandalos2001).
Covariates
Covariates were selected based on their theoretical relations with the primary predictors and outcome measures to account for confounding contributions to the processes of interest. Parents reported on children’s biological sex (male = 0, female = 1), age, race (Non-White = 1, White = 2), and their educational attainment at the time of the baseline laboratory visit. The randomized controlled trial was conducted in 12 cohorts to ensure the feasibility of effective treatment allocation. Cohort membership was included as a covariate in multilevel models as a multinominal variable with 12 categories. We did not control for the treatment conditions because all data were collected before the start of treatment and were unaffected by randomization into treatment groups. Parents’ anxiety was also included as another covariate in the models predicting parent-report outcomes to address potential bias associated with parental anxiety and their responses to the parent-report measures.
Results
Descriptive statistics and zero-order bivariate correlations of study variables are presented in Table 1. The first level analyses (i.e., characterizing parent–child dyadic RSA concordance) indicated a significant residual variance (σ 2 = 1.31, p <0.001), suggesting that there is a variation in parents’ RSA levels that is not accounted for by children’s RSA levels. In addition, at this level (i.e., within-dyad), results indicated that, on average, the pattern of dyadic RSA concordance was not statistically significant, Mean = 0.082, p = 0.27, but there was individual variability in the dyadic RSA concordance, var = 0.48, p <0.001. In addition, the ICC for the parent–child dyadic RSA were 0.54, suggesting a significant amount of variability in parent–child dyadic RSA levels across different levels. The ICC of 0.54 indicates that 54% of the total variability observed in parents’ and children’s RSA values is due to their dyad’s membership. The ICCs also reflect a relatively strong degree of dependence on the data and the strength of the nesting design for applying a MLM approach. Tables 2 and 3 summarize the results of the multilevel models testing the study’s aims. Unstandardized coefficients as the estimates of the effect sizes and confidence intervals for the main and interaction effects models are reported separately in the Tables.
Table 1. Zero-order bivariate correlations between study variables
Note. RSA = respiratory sinus arrhythmia
p ≤ .05*, p ≤ .01**
Table 2. The multilevel model of the moderating effects of parent–child dyadic respiratory sinus arrhythmia (RSA) concordance on the associations between parenting behaviors (intrusive and nurturing) and children’s externalizing and internalizing behavior problems reported by the teachers
Note. CI = Confidence interval; RSA = respiratory sinus arrhythmia; Child sex: (male = 0, female = 1); Child race: (Non-White = 1, White = 2).
Externalizing and internalizing behavior problems were examined as covarying outcomes.
* p<0.05.
Table 3. The multilevel model of the moderating effects of parent–child dyadic respiratory sinus arrhythmia (RSA) concordance on the associations between parenting behaviors (intrusive and nurturing) and children’s externalizing and internalizing behavior problems reported by the parents
Note. CI = Confidence interval; RSA = respiratory sinus arrhythmia; Child sex: (male = 0, female = 1); Child race: (Non-White = 1, White = 2).
Externalizing and internalizing behavior problems were examined as covarying outcomes.
* p<0.05; **p<0.01.
Teacher-reported outcomes
There were no statistically significant main effect associations between either parenting behaviors or dyadic RSA concordance and children’s internalizing and externalizing behavior problems, as reported by teachers (Table 2). However, tests of the moderation effects revealed a statistically significant interaction between parental nurturance (but not intrusiveness; b = −0.02, 95% CI [−0.13, 0.09]) and dyadic RSA concordance in the association with children’s internalizing behavior problems (b = 0.20, 95% CI [0.00, 0.39]). We found no moderation effects between parental intrusiveness (b = −0.05, 95% CI [−0.23, 0.13]) or nurturance (b = 0.15, 95% CI [−0.23, 0.54]) and dyadic RSA concordance in their association with children’s externalizing behavior problems.
Tests of the simple slope and the RoS provided additional insight into the moderating effects of dyadic RSA concordance on the association between parental nurturance and children’s internalizing problems. Figure 2 displays the effects of Parental nurturance on children’s internalizing problems for children with negative (− 1SD) and positive (+ 1SD) dyadic RSA concordance. Simple slopes analyses revealed that the association between parental nurturance and internalizing problems was significant for children who exhibited positive (+ 1SD simple slope = 0.25, t = 2.51, p = 0.012) but not negative RSA concordance (− 1SD simple slope = −0.15, t = 0.91, p = 0.36) with their caregivers. The positive RSA concordance in this context indicates that the mother’s and child’s RSA changed in the same direction across the tasks at the same time; both the child’s and mother’s RSA increased or decreased simultaneously. The RoS analyses for the moderator (i.e., dyadic RSA concordance) further indicated that the significant association between parental nurturance and internalizing problems was significant for children exhibiting RSA concordance at the mean and above (M = 0 and SD = 1). However, the lower bound of the RoS was outside our range of observed data and was not interpreted. The RoS analyses for the independent variable (i.e., parental nurturance) further indicated the lower and upper bounds of dyadic RSA concordance at −0.59 and 8.71, respectively. The upper bound of the RoS was outside our range of observed data and was not interpreted. Together, these results suggest that in the context of low parental nurturance, teachers reported fewer internalizing problems among children who showed a positive (but not negative) RSA concordance with their caregiver across the laboratory social challenges. In other words, the concordant or more similar pattern of epoch-to-epoch RSA change between children and caregivers (i.e., parent and child RSA changed in the same direction, either both decreased or increased) appeared as a resilience factor against internalizing psychopathology in the context of low parental nurturance.
Figure 2. Association between nurturing parenting and children’s internalizing problems, reported by teachers, moderated by dyadic RSA concordance. Slopes are plotted at negative (− 1SD) and positive (+ 1SD) dyadic RSA concordance. Negative concordance refers to lower concordance of parent–child dyadic RSA, or rather increased discordance (i.e., change in RSA for one member of the dyad is associated with a change in the opposite direction for the other member of the dyad). In contrast, positive concordance reflects a high concordance of parent–child dyadic RSA (i.e., change in RSA for one member of the dyad is associated with a change in the same direction for the other person in the dyad). Gray shaded areas refer to regions where the two lines are significantly different.
Parent-reported outcomes
Results indicated a significant positive association between intrusive (but not nurturing) parenting and parents’ reports of their children’s externalizing problems (b = 0.13, 95% CI [0.00, 0.26]). There were no statistically significant main effect associations between either parenting behaviors or RSA concordance and children’s internalizing behavior problems (Table 3). However, tests of the moderation effects revealed a statistically significant interaction between intrusive (but not nurturing; b = −0.10, 95% CI [−0.39, 0.19]) parenting and dyadic RSA concordance in the association with children’s externalizing behavior problems (b = −0.20, 95% CI [−0.35, −0.04]). We found no moderation effects between intrusive (b = −0.08, 95% CI [−0.20, 0.05]), nurturing (b = −0.04, 95% CI [−0.20, 0.13]) parenting, and dyadic RSA concordance in their association with children’s internalizing behavior problems.
Tests of the simple slope and the RoS provided additional insight into the moderating effects of dyadic RSA concordance on the association between parental intrusiveness and children’s externalizing problems. Figure 3 displays the effects of parental intrusiveness on children’s externalizing problems for children with negative (− 1SD) and positive (+ 1SD) dyadic RSA concordance. Simple slopes analyses revealed that the positive associations between parental intrusiveness and externalizing problems were significant only for children who exhibited negative (−1SD simple slope = 0.33, t = 3.45, p <0.001), but not positive, RSA concordance (+ 1SD simple slope = −0.07, t = 0.72, p = 0.47) with their parents. The negative concordance in this context represents that mothers and children demonstrated dissimilar patterns of epoch-to-epoch RSA change across tasks; as the child’s RSA increased, the mother’s RSA decreased simultaneously, and vice versa. The RoS analyses for the moderator (i.e., dyadic RSA concordance) further indicated that the positive association between parental intrusiveness and children’s externalizing problems was significant for children just above the average on concordance and below. The upper bound of the RoS was outside our range of observed data and was not interpreted. These results suggest that higher parental intrusiveness was associated with more externalizing problems, but only for children who showed negative RSA concordance with their caregiver. In other words, the negative parent–child dyadic RSA concordance, or the parent–child’s dissimilar patterns of epoch-to-epoch RSA change across tasks exacerbated the risk of externalizing problems in the context of high parental intrusiveness. Conversely, low intrusiveness was associated with fewer externalizing problems for children who showed negative (but not positive) dyadic RSA concordance. The RoS analyses for the independent variable (i.e., intrusive parenting) further revealed lower and upper bounds of −0.38 and 1.052, respectively, suggesting that intrusive parenting approximately 0.5 standard deviations below and 1 standard deviations above the mean was positively associated with externalizing problems for children with low to average levels of dyadic RSA concordance.
Figure 3. Association between intrusive parenting and children’s externalizing problems, reported by parents, moderated by dyadic RSA concordance. Slopes are plotted at negative (− 1SD) and positive (+ 1SD) dyadic RSA concordance. Negative concordance refers to lower concordance of parent–child dyadic RSA, or rather increased discordance (i.e., change in RSA for one member of the dyad is associated with a change in the opposite direction for the other member of the dyad). In contrast, positive concordance reflects a high concordance of parent–child dyadic RSA (i.e., change in RSA for one member of the dyad is associated with a change in the same direction for the other person in the dyad). Gray shaded areas refer to regions where the two lines are significantly different.
Discussion
This study characterized dyadic RSA concordance between children with elevated BI and their caregivers across tasks uniquely designed to mimic exposure to novel stimuli and social interactions while parents observed their children navigating these stressful experiences. Furthermore, it tested whether dyadic RSA concordance, which was operationalized as the within-dyad similarities or dissimilarities in the pattern of epoch-to-epoch RSA change across socially challenging laboratory stressors, moderates the links between intrusive and nurturing parenting behaviors and adjustment problems (Fox et al., Reference Fox, Zeytinoglu, Valadez, Buzzell, Morales and Henderson2023; Shakiba et al., Reference Shakiba, Lynch, Propper, Mills-Koonce and Wagner2023; Ugarte et al., Reference Ugarte, Miller, Weissman and Hastings2023; Viana et al., Reference Viana, Trent, Raines, Woodward, Storch and Zvolensky2021). Consistent with past research (Eisenberg et al., Reference Eisenberg, Taylor, Widaman and Spinrad2015; Murray et al., Reference Murray, Dwyer, Rubin, Knighton-Wisor and Booth-LaForce2014; Yan et al., Reference Yan, Ansari and Wang2019), the main effect results indicated that parents who practiced more intrusive parenting behaviors reported higher externalizing behaviors in their children. Findings further suggested that the associations between parental intrusiveness and nurturance and risk for externalizing and internalizing behavior problems were moderated by dyadic RSA concordance (i.e., if the parent–child dyad showed a similar or dissimilar pattern of RSA changes across the social stressor tasks).
Results from the teacher-report models indicated that positive dyadic RSA concordance served as a resilience factor, buffering the negative effects of low nurturing parenting on children’s risk of exhibiting internalizing problems, as reported by teachers. These findings align with those of Lan et al. (Reference Lan, Zhang, Lunkenheimer, Chang, Li and Wang2024), which suggest that high parent–infant dyadic RSA synchrony functions as a resilience factor for intergenerational transmission of internalizing problems. Mother–child physiological concordance is shaped and develops through the repeated interaction that a mother and child have with one another over time, in which they dynamically regulate their physiological responses to match their partner’s physiological state (Feldman, Reference Feldman2012). A concordant pattern of physiological regulation in a parent–child dyad may contribute to and provide the biological basis for children’s resilience by promoting child’s effective emotion and self-regulation capacities, survival, and socioemotional adjustment (Feldman, Reference Feldman2012; Morris et al., Reference Morris, Cui, Criss and Simmons2018). However, the current study and existing research suggest that the adaptiveness of dyadic RSA concordance depends, in part, on the nature and context of parent–child interaction and the characteristics of both members of a dyad (DePasquale, Reference DePasquale2020; Fuchs et al., Reference Fuchs, Lunkenheimer and Lobo2021; Lan et al., Reference Lan, Zhang, Lunkenheimer, Chang, Li and Wang2024; Lunkenheimer et al., Reference Lunkenheimer, Busuito, Brown and Skowron2018).
However, our results appeared to be inconsistent with the findings of Oshri et al. (Reference Oshri, Liu, Suveg, Caughy and Huffman2023) in which they reported negative parenting to be associated with decreased internalizing symptoms among youth exhibiting negative RSA synchrony with their parents. By contrast, high positive dyadic RSA synchrony exacerbated the positive associations between negative parenting and youth behavior problems. Differences in developmental stages, task contexts, and the nature of the two samples (i.e., children with elevated BI vs. a community sample) may have contributed to these opposing findings. It is possible that, in the context of high adversity, the relative influence of dyadic RSA concordance may be moderated by children’s age. For example, the current study and the work by Lan and colleagues (Reference Lan, Zhang, Lunkenheimer, Chang, Li and Wang2023) suggest that dyadic RSA concordance operates as a resilience factor for young children, but dyadic physiological concordance may exacerbate the risk for adjustment problems in older children and adolescents. The context in which dyadic concordance is assessed is also important to consider. For example, Oshri and colleagues (Reference Oshri, Liu, Suveg, Caughy and Huffman2023) assessed dyadic RSA concordance during a conflict discussion task, which may have induced significantly higher levels of physiological stress in both members of a dyad, compared to the current study’s tasks that did not involve interactions between a parent–child dyad, or the Lan et al., study that assessed dyadic coregulation during the low-stress free play task. While the current study cannot directly address these possibilities, it does add to the growing literature that identifies dyadic physiological concordance as a moderator of the influences of early experience on children’s adjustment problems.
Results from the parent-report models indicated that parental intrusiveness was associated with higher externalizing problems, but only for children exhibiting a negative pattern of dyadic RSA concordance with their caregiver. In other words, negative dyadic RSA concordance served as an additive risk factor that exacerbated the negative effects of intrusive parenting on children’s increased risk of externalizing problems. Notably, parents reported fewer externalizing problems in the context of low levels of intrusiveness for children exhibiting negative dyadic RSA concordance. We consider these findings in the context of coercion theory (Patterson, Reference Patterson1982), which posits a process of mutual reinforcement, during which caregivers unintentionally reinforce children’s disruptive and oppositional behaviors (e.g., child disobeys or resists parental requests and directions), which in turn, elicits more coercive and intrusive behaviors from the caregiver towards the child (e.g., the parent gets angrier and harsher). Coercive family processes are one important mechanism influencing the development of conduct problems and oppositionality in typically developing young children (Patterson, Reference Patterson, Dishion, Snyder and Patterson2016 Reference Patterson1982; Shaw & Bell, Reference Shaw and Bell1993; Smith et al., Reference Smith, Dishion, Shaw, Wilson, Winter and Patterson2014). Compared to the typically developing children, parents of children with elevated BI are more likely to view their children as vulnerable and to practice more coercive and overcontrolling types of parenting behaviors as a result, particularly when they are exposed to unfamiliar, novel, and socially threatening situations (Rubin & Mills, Reference Rubin, Mills, Sigel, Goodnow and McGillicuddy-deLisi1992; Rubin et al., Reference Rubin, Coplan and Bowker2009). The within-dyad negative pattern of RSA concordance observed in some of the parent–child dyads in the current study may indicate a mismatch between parents and children in their perception of environmental threats or stress. These findings further indicate that such discordance intensifies the adverse effects of parental intrusive and controlling behaviors on children’s socioemotional adjustment. Another important point to consider when interpreting these findings is that the simple slope for the influence of parental intrusiveness on children’s externalizing behaviors cannot be distinguished from zero for children who showed a positive pattern of dyadic RSA concordance. In other words, for these children, dyadic RSA concordance attenuates the associations between parental intrusiveness and externalizing outcomes. In this sense, both models suggest that positive dyadic concordance may be a resiliency factor.
Theoretical and empirical work (Porges, Reference Porges1992, Reference Porges2007; Thayer & Lane, Reference Thayer and Lane2009) suggests that RSA represents a key neurophysiological substrate of individual differences in self-regulation, social communication, and attention. As indexed by RSA functioning, individual differences in the PNS functioning are thought to provide insight into how individuals perceive their environment and leverage regulatory resources to adaptively respond to navigate changing demands. The results of the current study suggest that the relative impact of dyadic RSA concordance varies as a function of the parenting behavior and informants of children’s adjustment problems (i.e., parent vs. teacher). In addition, our findings suggest that the dyadic pattern of regulation protects against children’s oppositionality and externalizing psychopathology in the context of parenting that is low on control and intrusion. Conversely, the combination of negative dyadic physiological concordance and intrusive, controlling parenting resulted in higher oppositional and externalizing behavior problems. Within the dyads exhibiting negative patterns of RSA concordance, there might also be meaningful links between parents’ regulatory difficulties and coercive and intrusive parental disciplinary practices, which in turn, predict and increase the risk for externalizing behaviors in children (Lunkenheimer et al., Reference Lunkenheimer, Sturge-Apple and Kelm2023; Sturge-Apple et al., Reference Sturge-Apple, Skibo, Rogosch, Ignjatovic and Heinzelman2011). However, this was not directly tested in the current study and should be a focus for future investigations.
When interpreting and contrasting parent- versus teacher-reported assessments of children’s adjustment problems, it is crucial to consider that children with elevated BI tend to exhibit higher levels of anxiety in the presence of both familiar and unfamiliar peers in contrast to when they are in the company of their parents and siblings. As in the present sample, teachers reported a greater prevalence of internalizing problems in children during daycare or preschool hours, particularly among those who showed a positive dyadic RSA concordance with their parents and experienced low levels of nurturing parenting. Moreover, parents with a more intrusive parenting style reported more externalizing behaviors in their children. Taken together, the present study’s findings provide new and compelling insights into dyadic RSA concordance as one underlying mechanism of the heterogeneity in the associations between parenting behaviors and the risk of childhood adjustment problems in a sample of children with elevated BI.
Several strengths and limitations associated with the present study are worth noting as they could provide important directions for future investigations. First, the contrastive interaction effects observed in this study may be due, in part, to the limited statistical power of the current sample to reliably and accurately detect moderated associations in the multilevel models (Vize et al., Reference Vize, Baranger, Finsaas, Goldstein, Olino and Lynam2023). Such contrastive patterns are likely to emerge in underpowered analyses, where individuals high on the moderator exhibit an association in one direction. In contrast, those low on the moderator show a linear association in the opposite direction (Roisman et al., Reference Roisman, Newman, Fraley, Haltigan, Groh and Haydon2012). Therefore, we urge readers to interpret the interaction findings with caution and recommend that future researchers attempt to replicate the key associations in a larger sample of behaviorally inhibited children.
Second, although we employed multiple informants to provide a comprehensive overview of children’s adjustment problems in various contexts, future studies should incorporate measures that explicitly ask about behaviors observed across different contexts. We assume that teachers and parents are reporting on behaviors that they are more likely to observe in the classroom and home contexts, respectively, but the measures used in the current study do not explicitly make these distinctions.
Third, the present study is the first to examine the dyadic physiological concordance between children with elevated BI and their caregivers across a series of tasks purposefully designed to mimic exposure to novel stimuli and social interactions in these high-risk children. Future longitudinal studies are needed to investigate whether BI and non-BI samples may show different patterns of RSA concordance (i.e., positive vs. negative concordance) across both stressful and nonstressful conditions. These studies should also explore whether patterns of dyadic RSA concordance may change throughout child development within both BI and non-BI community samples. Furthermore, future research should investigate how the parent-BI child dyadic concordance pattern may vary depending on parenting behaviors, caregiver anxiety and depression levels, and family demographic characteristics.
Fourth, the use of a sample of highly inhibited young children in the current study may limit the extent to which these findings are generalizable to broader community-based and clinical populations, or other age groups. Future studies should advance this line of inquiry by examining whether synchrony across multiple biological regulatory systems (e.g., the sympathetic and PNSs, the neuroendocrine system) may also serve as an index of biological sensitivity to caregiving experiences and children’s and adolescents’ subsequent adjustment outcomes. It is important to note that the inhibitory influence of the PNS on cardiac functioning does not occur in isolation. Rather, it is strongly shaped and influenced by environmental factors, challenge contexts, underlying cognitive and neurodevelopmental mechanisms, and its dynamic interaction with other regulatory systems, particularly the sympathetic nervous system (Berntson & Cacioppo, Reference Berntson and Cacioppo2004; McEwen, Reference McEwen2007; Porges, Reference Porges2007). Therefore, we recommend that future research examine whether dynamic autonomic coordination may also serve as an index of a child’s level of sensitivity to the caregiving environment and their subsequent risk or resilience for developing adjustment problems.
In addition, we recommend that, in the future, researchers should examine dyadic physiological concordance across other caregiving contexts (e.g., parental psychopathology, child maltreatment, and neglect). Another limitation of the study is that dyadic RSA concordance was assessed using tasks that did not involve direct parent–child interaction. Future research should explore these questions using tasks that involve more direct interaction between parents and behaviorally inhibited children.
In summary, we provide new insight into dyadic RSA concordance as one underlying mechanism of the heterogeneity in the associations between intrusive and nurturing parenting behaviors and childhood risks of developing adjustment problems in a sample of children with elevated BI. Taken together, our findings highlight the importance of considering both neurobiological and environmental factors when examining the developmental origins and causes of mental health problems in all children, including children with BI.
Acknowledgments
This project was funded by NIH R01MH103253 awarded to A.C-T. And K.H.R. Manuscript preparation was partially supported by NIH R03MH123762 awarded to N.J.W.
Competing interests
No conflicts declared.
Open access
