The Inconsolable Doll Task: Prenatal coparenting behavioral dynamics under stress predicting child cognitive development at 18 months

Highlights

• This study introduces a novel procedure - the Inconsolable Doll Task- and is the first to observe prenatal coparenting behaviors during a stressful interaction context.

• Co-parenting couples’ dynamic of negative escalation predicted the child’s cognitive development at 18-months.

• These predictions remained significant when accounting for prenatal and postnatal assessments of playful coparental behavioral patterns and self-reported coparental perceptions, as well as when controlling for parental education.

• The IDT can be useful in prevention and early intervention of expectant parents.

Abstract

Studies have demonstrated that coparenting can be assessed prenatally through playful observational conditions, including simulated baby enactments. Regrettably, there is a lack of empirical research examining how prenatal coparenting under the emotional stress elicited by the distress of a simulated infant predicts children’s cognitive development. The current longitudinal study introduces a novel procedure—the Inconsolable Doll Task—to assess prenatal coparenting behavioral dynamics under the stress of a non-responsive doll simulator, and examines the extent to which prenatal interaction patterns predict the child’s cognitive development at 18 months. The sample consists of 105 community-based, co-living, expectant fathers and mothers. Data were collected prenatally, at three, six, and 18 months in home and lab visits. Results indicate that the prenatal coparenting dynamic of negative escalation explains a unique variance in children’s cognitive development at 18 months. This effect is evident even when accounting for both prenatal and postnatal assessments of low-stress coparenting behavioral patterns or self-reported coparenting perceptions, and when controlling for parental education. These findings are discussed in terms of their methodical, empirical, and clinical implications.

Introduction

An examination of the child development field reveals exciting shifts moving from monadic to dyadic, and most recently, triadic conceptual and empirical approaches to child development. This shift corresponds with two central theories. The first is the ecological systems theory, according to which, development is a function of forces emanating from multiple settings and from the relations among these settings (Bronfenbrenner & Morris, 2006). The second is the transactional theory of development (Sameroff, 1975; Sameroff & Fiese, 2000), according to which there is a bidirectional influence in transactions between children and their environments that shape their development. The integration of these two theories implies that it is useful to study child development within multi-person family group dynamics, while taking into account the mutual influence of the different family members—including the child—on the child’s developmental trajectory (Belsky, 1984; Belsky, Putnam, & Crnic, 1996; Bronfenbrenner, 1986; Cowan & Cowan, 2000; McHale, Kuersten-Hogan, & Rao, 2004). One facet of the family system is the coparenting unit.

Coparenting refers to the degree of collaboration, affirmation, and support between adults raising children for whom they share responsibility (Feinberg, 2003; McHale, 2007). Appreciating the importance of coparenting patterns, studies have been able to successfully activate the coparenting system even prior to the baby’s birth through self-reports and interviews (e.g., McHale & Rotman, 2007). Researchers were even able to activate the family dynamics in a playful situation using a doll simulating the unborn baby (e.g., Favez, Frascarolo, Lavanchy Scaiola, & Corboz-Warnery, 2013). Nonetheless, infants often present parents with stressful situations, such as when the baby cries inconsolably.

Attempting to capture a fuller understanding of coparenting and its impact on child development, this study sought to assess observed prenatal coparenting under conditions that might evoke stress in parents, namely, a doll simulating an inconsolable baby that expectant parents were asked to care for and soothe. The research was designed to investigate whether these prenatal coparenting patterns under stress could account for individual differences in the child’s development at 18 months, even when considering assessments of prenatal and postnatal coparenting family dynamics under low emotional arousal or coparenting perceptions.

Throughout history and around the world, one of the essential tasks identified with adulthood is parenting young children, in most cases, within a coparenting system. According to Minuchin’s (1974) theory of family structure, the family system is headed by the coparenting relationship, or the “executive subsystem,” which is related to but distinct from the preexisting relationship between the partners (Altenburger, Schoppe-Sullivan, Lang, Bower, & Kamp Dush, 2014; Schoppe-Sullivan, Mangelsdorf, Frosch, & McHale, 2004). Coparenting is concerned principally with the degree of collaboration, affirmation, and support between adults raising children for whom they share responsibility (Feinberg, 2003; McHale, 2007). Given the focus on the shared aim of rearing the child (Burney & Leerkes, 2010), coparenting includes only the aspects of the marital relationship that are relevant for parenting (Abidin, 1992; Feinberg, 2003; Katz & Gottman, 1996; McHale & Rasmussen, 1998). At its core, coparenting is a triadic structure involving the coordination of two adults responsible for the caring, education, and nurturing of their children (McHale, Kazali et al., 2004; McHale, Kuersten-Hogan et al., 2004), as well as the extent to which the parents support or undermine each other’s parenting efforts (Belsky, Woodworth, & Crnic, 1996).

Irrespective of its operationalization (self-reports, observations, or interviews), coparenting has been shown to impact children's development, psychopathology, and adjustment at different ages (Teubert & Pinquart, 2010). For example, self-reported coparenting communication, conflict, and shared decision-making when the child was 24-months-old was directly linked to children’s math, literacy, and social skills at 48 months (Cabrera, Scott, Fagan, Steward-Streng, & Chien, 2012). Similarly, supportive coparenting predicted better social skills and attention, less passivity and dependence, and higher grades in the classroom in the following year (Dopkins Stright & Neitzel, 2003).

In terms of socio-emotional difficulties, studies relying both on self-reports and observational assessments of coparenting quality have shown that children growing up in families experiencing disruptions in coparenting exhibit higher levels of externalizing behavior problems in the second year of life (Belsky, Putnam et al., 1996; Belsky, Woodworth et al., 1996; Jacobvitz, Hazen, Curran, & Hitchens, 2004; Jones, Shaffer, Forehand, Brody, & Armistead, 2003; Schoppe, Mangelsdorf, & Frosch, 2001; Schoppe-Sullivan, Weldon, Claire Cook, Davis, & Buckley, 2009), less prosocial behavior (Scrimgeour, Blandon, Stifter, & Buss, 2013), and inhibition difficulties as early as 18 months (Belsky, Putnam et al., 1996; Belsky, Woodworth et al., 1996; Lindsey & Caldera, 2005).

There is also evidence that coparenting is related to parent-child and family relationships (McHale & Rasmussen, 1998). In a longitudinal study, the perceived absence of coparenting cooperation predicted an increase in negative child-to-mother behavior and negative child-to-father behavior between 18 and 24 months (Floyd, Gilliom, & Costigan, 1998). Brown, Schoppe-Sullivan, Mangelsdorf, and Neff (2010) found that children from families exhibiting higher levels of observed supportive coparenting in early infancy were more likely to be securely attached to their fathers (but not mothers) at 13 months of age. Taken together, these findings clearly demonstrate that acrimonious parental interactions and disagreements about childrearing can be detrimental to young children’s development (Feinberg, Kan, & Hetherington, 2007).

In contrast to the body of work demonstrating associations between coparenting and children’s socio-emotional development, little is known about the impact of coparenting on children’s cognitive functioning. Favez et al. (2012) documented the influence of observed coparenting and family alliance in a playful interaction in the first two years of children’s cognitive development at the age of five years. Relatedly, Jacobvitz et al. (2004) found that observed hostile family interactions in a nonclinical sample of toddlers forecasted the development of ADHD in middle childhood. However, to the best of our knowledge, there is little research addressing the possible impact of the quality of coparenting on children’s cognitive development. One of the goals of this study is to address this question as early as possible, namely, in the prenatal phase.

Appreciating the significance of coparenting dynamics for child development, researchers have aimed to identify coparenting patterns as early as possible, and examine whether coparenting could be assessed reliably even prenatally. Doing so would inform early intervention and perhaps even prevention programs geared to enhance child development (Altenburger et al., 2014; Feinberg, 2003). To that end, and using a variety of assessments, researchers attempted to document associations between prenatal and postnatal coparenting interactions.

One way to explain the continuity between prenatal and postnatal coparenting dynamics is via the parents’ mental representations of coparenting that form during pregnancy and perhaps even earlier. These mental representations involve various aspects of family processes, including expectations of the degree of harmony vs. conflict, support, and endorsement vs. undermining that will be experienced in the coparenting relationship. Consequently, parental representations of coparenting involve cognitive facets of the coparenting relationship such as the caregivers’ perceptions of the overall quality of their coparenting relationship, and appraisals and anticipations of their own and their partners’ specific coparenting behaviors. These prenatal representations of the future coparenting relationship are hypothesized to be associated with their postpartum coparenting relationship (Kuersten-Hogan, 2017).

Researchers have often used questionnaires to capture the cognitive facts of the coparenting relationship. McHale and Rotman (2007) used Cowan and Cowan’s “Ideas About Parenting” questionnaire (IAP) in pregnancy, which contains 46 different child-rearing items, to assess discrepancies in the scores between fathers’ and mothers’ parenting ideas. McHale and Rotman (2007) found that larger discrepancies between partners’ parenting beliefs during the third trimester of pregnancy predicted less overall coparenting solidarity (high in cohesion and low in conflict) at both 3 and 12 months on the LTP (LTP; Fivaz-Depeursinge & Corboz-Warnery, 1999) and Still-Face Paradigm (SFP; Tronick & Gianino’s, 1986).

One technique for obtaining self-reports about perceptions about the quality of coparenting is the Coparenting Relationship Scale (CRS; Feinberg, Brown, & Kan, 2012), whose postpartum version we adapted to create prenatal coparenting self-report measures. The CRS provides a rich and multidimensional assessment of the perceptions and attitudes towards coparenting including endorsement, undermining, agreement, support, division of labor, and closeness.

Attempting to obtain a fuller understating of expectant parents’ coparenting views and attitudes, McHale and colleagues (McHale, Kazali et al., 2004, McHale, Kuersten-Hogan et al. (2004) and McHale and Rotman, 2007) developed the prenatal Coparenting Interview, in which participants reflected on their own parents’ coparenting relationship, and on their expectations for the coparenting relationship they would soon begin forming with their partner. The researchers reported that when negative outlooks and expectancies colored the future parents' representations, coparenting cohesion at 3 months and coparenting solidarity at 12 months were lower than in other families. In addition, they found that between-parent differences in parenting beliefs evident during pregnancy predicted the extent of postnatal solidarity (McHale, Kazali et al., 2004; McHale, Kuersten-Hogan et al., 2004; McHale & Rotman, 2007).

Researchers have also used the Who Does What discussion (WDW), based on Cowan and Cowan’s (1988) self-reported questionnaire “Who Does What” (Elliston, McHale, Talbot, Parmley, & Kuersten-Hogan, 2008; Schoppe-Sullivan et al., 2004) to examine the level of agreement between partners’ coparenting representations more dynamically. Initially, each parent individually completes the 24-item WDW list of childcare roles (e.g., diapering, feeding), and rates how much of each item is their or their partner’s responsibility. After completing the questionnaire independently, partners are asked to share their responses and reach consensus on each item. Research shows that expectant parents’ perceptions and beliefs about expected and ideal childcare responsibilities predicted their postpartum representations and functioning of coparenting at 3 and 12 months (Kuersten-Hogan, 2017).

To assess coparenting prenatally on a behavioral level, researchers developed the prenatal LTP (PLTP; Carneiro, Corboz-Warnery, & Fivaz-Depeursinge, 2006), a semi-standardized paradigm used extensively to assess prenatal family alliance and coparenting in a playful, low-stress context, in which parents are asked to play out an encounter with their child, simulated by a fabric doll (with a baby’s body, but without distinct facial features) (Carneiro et al., 2006; Corboz-Warnery & Fivaz-Depeursinge, 2001; Fivaz-Depeursinge, Frascarolo-Moutinot, & Corboz-Warnery, 2010). Beyond the role-playing abilities, the task calls on the parents’ mutual support of and cooperation with each other as parents (Carneiro et al., 2006).

Using the PLTP, Favez et al. (2013) demonstrated that the interactive qualities necessary for mother–father–infant coordination could even be detected during pregnancy. Specifically, they found that a couple’s cooperation in the prenatal playful PLTP was correlated with almost all dimensions of postnatal play at 3 and 18 months. Furthermore, warmth and intuitive parenting were also correlated with several postnatal dimensions, as well as with the child’s emotional and behavioral wellbeing (Carneiro et al., 2006; Favez et al., 2012, 2013).

The variety of methods designed to assess coparenting suggests its multifaceted nature, composed of cognitions, expectations, and behaviors – while considering the different aspects of the relationship, as well as its manifestation in various contexts. One question that arises pertains to the extent to which the different aspects of prenatal coparenting correspond with each other. Regrettably, data addressing this issue are only now accumulating and are currently inconclusive. Recently, Kuersten-Hogan (2017) was the first to explore cognitive as well as behavioral facets of the prenatal coparenting relationship simultaneously and found no correlation between prenatal coparenting representations and behaviors. One possibility for this lack of association is that parental coparenting behaviors were measured within a low emotional arousal context. Perhaps representations of prenatal coparenting correspond more with behavioral coparenting during high emotional arousal states. It is possible that reports of coparenting may be more consistent with observed behavior during stressful rather than relaxed coparenting interactions.

The findings pertaining to the PLTP clearly demonstrate the importance of positive, playful, and supportive coparenting for subsequent family dynamics and a child’s emotional development. Unfortunately, the desire to answer a young child’s immediate and demanding needs, which are expressed nonverbally and are often hard for parents to interpret, can be difficult and stressful.

From an ecological perspective, one of the sources that influences parental functioning, including coparenting, is the amount of stress or support the parent experiences (Belsky, 1984). Indeed, the impaired functioning of various dimensions of coparenting is associated with higher levels of parenting stress (Margolin, Gordis, & John, 2001; Shai, Dollberg, & Szepsenwol, 2017). Conversely, well-functioning coparenting systems can reduce parental stress. In fact, coparenting has been conceptualized as a significant component of parenthood on which partners can rely for support when stressed by the many frustrations of parenting (Weissman & Cohen, 1985). The risk of engaging in emotionally unsupportive coparenting is most likely to surface when families face greater stress in the form of daily aggravations (Belsky, Crnic, & Gable, 1995). This is especially true when considering the coparenting of younger children, who, compared with older children, may require more cooperation and teamwork to meet the moment-to-moment demands of parenting young children (Margolin et al., 2001) and keeping parental stress at bay (Durtschi, Soloski, & Kimmes, 2017).

Fascinating work has been conducted investigating the impact of coparenting under stressful conditions on child development. Brown et al. (2010), for example, gave parents of 3.5 month-old infants a “onesie” and asked them to change the infant into this outfit together. According to the researchers, this task was designed to assess coparenting behavior during a joint child care task, a situation that is arguably more stressful than triadic free play (see also Schoppe-Sullivan, Mangelsdorf, Brown, & Sokolowski, 2007).

Moreover, and in line with the transactional model (Sameroff, 1975), the infant itself is sometimes the source of stress and strain on the coparenting unit. Cook, Schoppe-Sullivan, Buckley, and Davis (2009) found that children’s negative affect was a significant predictor of undermining coparenting, such that more temperamentally difficult children had parents who undermined each other's parenting more frequently and intensely. These findings also underscore the importance of including the child, or the represented child, in the coparenting assessment, because the child’s presence and his or her level of arousal activate and challenge the functioning of the coparenting.

We believe that the well-documented influence of coparenting on child development stems, at least to some extent, from what the child is exposed to, on the behavioral level, in the family dynamics. Furthermore, we contend that the significance of the coparenting alliance in terms of its impact on the child is at its peak when the coparenting alliance is challenged in times of tension, difficulty, or disagreement between the parents who also need to consider their child’s wellbeing in their actions. Indeed, there is work suggesting that a significant portion of marital conflict takes place in the presence of children (Papp, Cummings, & Goeke-Morey, 2002), and that marital and parental conflict can “spill over” to the parent-child subsystem, thereby influencing the child’s development (Katz & Gottman, 1996; Kitzmann, 2000).

Taken together, we argue that in an ecologically valid measurement of prenatal coparenting dynamics that taps into how coparenting patterns actually influence the child’s developmental trajectory should simultaneously meet two criteria: a) including the represented infant in the assessment, and b) activating and challenging the coparenting alliance.

To the best of our knowledge, researchers have investigated prenatal coparenting dynamics meeting one of these criteria, but not both integrated within one study. Some studies used a task designed to activate and challenge the coparenting system, but did so in the absence of a representation of the infant (e.g., Cowan & Cowan, 1990; Kuersten-Hogan, 2017; McHale, Kazali et al., 2004; McHale, Kuersten-Hogan et al., 2004; McHale & Rotman, 2007; Schoppe-Sullivan et al., 2004), while other studies included the represented infant during a playful coparenting interaction, but not in one designed to challenge the coparenting alliance (e.g., Carneiro et al., 2006; Corboz-Warnery & Fivaz-Depeursinge, 2001; Favez, Frascarolo, & Fivaz-Depeursinge, 2006; Favez et al., 2012, 2013; Fivaz-Depeursinge et al., 2010).

To fill this gap in the literature, we designed a study to simulate a prenatal coparenting high-arousal and stressful behavioral task using the RealCare Baby^® II-Plus infant simulator. Our goal was to create a realistic, high-arousal context in as ecologically valid a way possible to illuminate how the presence of an inconsolable infant affects coparenting dynamics and the child’s development. More specifically, the aim of the study was twofold. First, we aimed to establish convergent validity for the newly developed Inconsolable Doll Task (IDT). With this goal in mind, we examined concurrent associations of coparenting dynamics with behavioral family alliance dynamics (LTP; Corboz-Warnery & Fivaz-Depeursinge, 2001), behavioral spousal support (Collins & Feeney, 2000), and self-reported coparenting (CRS; Feinberg et al., 2012). Second, we sought to expand our understating of how coparenting impacts children’s cognitive development and establish the procedure’s predictive validity. To that end, the current work was designed to examine the extent to which prenatal coparenting dynamics predict the child’s cognitive development at 18 months (Mullen Scales of Development; Mullen, 1995). To address the possibility that infants’ cognitive development results from either heredity or general environmental influences (e.g., Rowe, Jacobson, & Van den Oord, 1999; Van Bakel & Riksen-Walraven, 2002) rather than coparenting per se, we controlled for parents’ education when predicting children’s cognitive functioning from prenatal coparenting dynamics.

Therefore, we constructed three hypotheses:

H1

Prenatal coparenting behavior will be positively associated with self-reported coparenting quality. Specifically, supportive coparenting detected during the Inconsolable Doll Task will be positively correlated with self-reported supportive co-parenting assessed by self-reported coparenting perceptions.

H2

Prenatal coparenting dynamics under high-arousal conditions will be associated with prenatal and postnatal coparenting quality under low-arousal conditions. Specifically, supportive coparenting identified during the Inconsolable Doll Task will be positively associated with coordinated family alliance behavioral dynamics assessed both prenatally and postnatally.

H3

Supportive coparenting behavior detected during the Inconsolable Doll Task will predict various aspects of the child’s cognitive development at 18 months, above and beyond the influence of the parents’ education levels.