Maternal reflective functioning as a multidimensional construct: Differential associations with children’s temperament and externalizing behavior

Highlights

• Maternal reflective functioning should be regarded as a multidimensional concept.

• Dimensions of reflective functioning are differentially related to child behavior.

• Multidimensionality of reflective functioning can be observed pre- and postnatally.

Abstract

Maternal reflective functioning (RF) has been associated with children’s behavioral development. This study examined maternal prenatal and postnatal RF, as measured by the Pregnancy Interview and Parent Development Interview, as multidimensional constructs. It was also examined whether the RF-dimensions were associated with children’s temperament and externalizing behavior, as assessed by several questionnaires. The sample consisted of 123 first-time mothers (M age = 22.85 years, SD = 2.21) and their children (M age = 19.97 months, SD = 0.85, 56% male). Two related but distinct dimensions were found for prenatal RF, termed self-focused and child-focused mentalization. Three dimensions were observed for postnatal RF, termed self-focused, child-focused, and relation-focused mentalization. Results showed that prenatal RF negatively related to reported child physical aggression. Postnatal self-focused RF was positively linked to externalizing behavior and negative emotionality in offspring, while relation-focused RF scores were negatively associated with child physical aggression. Findings show that it is important to also look at the specific RF-dimensions when examining the effects of maternal RF on children’s behavioral development, as differential associations with behavioral outcomes exist. Discussion further focuses on the importance of these findings in prevention and clinical practice, and suggestions are being made to further improve the measurement of maternal RF-dimensions.

Introduction

Reflective functioning (RF), an operationalization of mentalization, has been defined as the ability to understand and interpret one’s own and others’ behavior in the light of mental states such as feelings, thoughts, fantasies, beliefs and desires (Fonagy, Gergely, Jurist, & Target, 2002). Ascribing (personal) meaning to underlying emotions and helping to clarify the cause and effect relations between those underlying mental states and behavior, are essential for adequate interpersonal functioning.

Mentalizing or RF in the context of parenting has been defined as a parent’s ability to understand their own mental states, to keep their child’s mental states in mind, and to understand how these mental states impact behavior (Ordway, Webb, Sadler, & Slade, 2015). More specifically, maternal RF is seen as the mother’s ability to think reflectively about herself as a parent, her child, and her relationship with the child (Slade, 2005). Maternal mentalizing starts to develop during pregnancy, as a woman’s representations of herself as a mother and of the baby become increasingly specific (Slade, Cohen, Sadler, & Miller, 2009). Prenatal maternal RF refers to the mother’s ability to think of the fetus, from at least the last trimester onwards, as a separate individual, with developing personal characteristics, needs, and temperament (Pajulo et al., 2015).

During pregnancy, a highly reflective mother prepares concretely and psychologically for the birth of the baby by making room for the infant both in mind and in practice (Pajulo et al., 2015). Postnatal maternal RF provides the mother an outline of how to respond when confronted with child-rearing issues (Ordway, Sadler, Dixon, & Slade, 2014). Specifically in times of elevated emotions, a reflective mother is likely to respond to her child’s signals with acceptance and in an appropriate manner. The reflective mother’s empathic responses serve a crucial function in organizing and regulating the child’s emotional states (Fonagy et al., 2002) and enable the child to develop the capacity to self-regulate. Furthermore, the mother is most likely to respond sensitively when she can understand the meaning and intention of children’s cues and see her children as separate from herself (Fonagy et al., 2002; Smaling et al., 2016). When the mother acts on incorrect assumptions about her child’s mental states or does not recognize her own and her child’s individual and separate emotional states, there is a risk for miscommunication (Ordway et al., 2015), which, in turn, could lead to poor emotion regulation, elevated stress responses, and behavioral problems (Bowlby, 1988; Ha, Sharp, & Goodyer, 2011; Sharp & Fonagy, 2008; Smaling et al.,in press).

Poor maternal RF or mentalizing has already been associated with behavioral problems in children (Benbassat and Priel, 2012, Ha et al., 2011; Meins, Centifanti, Fernyhough, & Fishburn, 2013), including attention problems, social withdrawal, anxiety, and dysfunctional mother-child interactions (Esbjørn et al., 2013, Fonagy et al., 2002; Smaling et al., 2016). The majority of studies linking maternal RF to child socio-behavioral development have focused on postnatal RF. Few studies have examined the role of prenatal maternal RF in the development of children’s behavioral development, although some evidence exists showing that this is predictive of children’s externalizing behavioral problems (specifically: physical aggression) as well (Smaling et al., in press).

Maternal RF has mostly been assessed postnatally, but it may be argued that for RF to develop optimally, this development should have started prenatally. Support for this notion comes from studies examining stability and change in maternal representations, part of maternal RF, which indicate moderate stability between prenatal and postnatal maternal representations (Benoit, Parker, & Zeanah, 1997; Theran, Levendosky, Bogat, & Huth-Bocks, 2005). Besides the moderate stability of the maternal representations, there are also studies suggesting more changeability of maternal representations (Aber, Belsky, Slade, & Crnic, 1999; Vizziello, Antonioli, Cocci, & Invernizzi, 1993). During the perinatal period mothers will slowly adapt to their new role. Likewise, during this period mothers might increasingly reflect upon their own childhood experiences. The birth of a healthy baby may also alleviate maternal anxieties activated during pregnancy, nurturing a more coherent and enriched representation of the woman as a mother and of her newborn baby. Furthermore, changes in maternal representations and mentalizing may be expected as these will be influenced by interactions with the actual baby. The importance of examining stability and change during the perinatal period is further underlined by the fact that prenatal maternal mentalizing predicts postnatal maternal mentalization, but not perfectly (Arnott and Meins, 2007, Arnott and Meins, 2008, Steele and Steele, 2008). In case pre- and postnatal RF indeed differ, it may be hypothesized that they will differentially predict child socio-behavioral outcomes as well.

When looking more closely at the definitions used to describe the mentalizing ability or maternal RF, the following characteristics of RF can be identified: (a) awareness, recognition, and acknowledgement of mental states in oneself and others, (b) an understanding of how mental states influence interpersonal interaction and behavior, (c) an understanding of mental state-dynamics in relationships, and (d) mental representations about relationships containing cognitive and emotional components (Fonagy and Bateman, 2008, Fonagy et al., 2002, Slade, 2005). Mentalizing with respect to infants and toddlers also often involves adopting a developmental perspective about the child’s growing capabilities and trying to make sense of the child’s internal world through observation of behavioral and affective cues (Slade, 2005, Slade, 2007). Whereas most studies to date have focused on maternal RF as a unitary construct, RF may have to be regarded as a multidimensional construct. Components may include (some of) the previously mentioned RF-characteristics. Whereas a distinction between these components appears to reflect a hierarchical structure of RF (with c being a more advanced form of RF than b, which in turn is more advanced than a), another plausible form of multidimensionality is one that distinguishes an intrapersonal dimension and interpersonal dimension (Benbassat & Priel, 2015). Awareness, recognition, and acknowledgement of mental states of oneself and of others may be related but distinguishable qualities, which, in turn, could differentially influence the understanding of mental-state dynamics of behavior and within relations.

Some empirical evidence for a two-dimensional structure of postnatal RF already exists. Suchman et al. (2010) identified a two-dimensional structure for postnatal RF in a sample of 47 substance-abusing mothers. A similar two-factor structure was identified by Borelli, St John, Cho, and Suchman (2016) in a high-risk community sample. The observed dimensions are in line with those theoretically suggested by Benbassat and Priel (2015). One dimension represented the maternal capacity to mentalize about her own emotions and behaviors (self-focused RF), while the second dimension represented the mother’s capacity to mentalize about her child’s mental states and behaviors, and about her interactions with the child (child-focused RF). Further indirect evidence for multidimensionality of concepts such as RF and mentalizing stems from studies showing separate though proximal neural networks for self-understanding and the understanding of others (Lieberman, 2007). These two forms of understanding may be related to different forms of interpersonal problems and psychosocial disorders (Luyten & Fonagy, 2012). For example, an impaired ability to self-mentalize appears to be an important characteristic of different forms of psychopathology (Fonagy et al., 2002), and may be present even when more general RF is intact (Bateman & Fonagy, 2006; Rudden, Milrod, Target, Ackerman, & Graf, 2006; Sharp et al., 2011).

No studies to date have investigated whether the two postnatally identified RF-dimensions can also be observed for prenatal reflective functioning and no studies to date have examined whether these RF-components (prenatally and postnatally) are differentially related to children’s temperament and externalizing behavior. A more detailed understanding of separate components of RF, and their implications for early behavioral development has the potential to enhance the efficacy of prevention and intervention programs aimed at reducing children’s behavioral problems (Smaling et al., 2016; Smaling et al., 2015).

To summarize, the first goal of this study was to examine maternal reflective functioning (RF) prenatally and postnatally as multidimensional constructs. We expected to find two distinct, but related dimensions for both prenatal and postnatal maternal mentalizing. More specifically, we expected to find a self-focused and a child-focused component of RF prenatally and postnatally.

The second goal was to investigate associations between the different dimensions of prenatal and postnatal RF with children’s temperament (i.e., negative emotionality and effortful control) and externalizing behavior (i.e., physical aggression and externalizing problems) at the age of 20 months. We expected that increased levels of RF, especially postnatal child-focused levels, would be associated with more optimal child behavior (less physical aggression, externalizing problems, and negative emotionality, and more effortful control).