Most parenting research focuses on parenting behavior—typically either reported by the parent or observed by researchers. Although this research is clearly important, it is also important to understand the thought processes and beliefs that underlie and influence those parenting behaviors. As Goodnow (1988) points out, “To focus only on parents’ overt behaviors is to treat parents as unthinking creatures, ignoring the fact that they interpret events, with these interpretations probably influencing their actions and feelings” (p. 287).

Researchers have long been interested in parents’ beliefs and ideas about parenting (e.g., Goodnow 1988; Miller 1988; Murphey 1992; Sigel et al. 1992). More recently, researchers have also become increasingly interested in parents’ reflection on their child’s thoughts and feelings through measures like reflective functioning (e.g., Rutherford et al. 2013; Slade et al. 2005). Parents differ in how often (and how accurately) they think about what is going on in their child’s mind, and they differ in how they think about their parental role. Further, research indicates that parents’ thoughts and feelings about their child’s mind (e.g., Kelly et al. 2005) and their thoughts about parenting (e.g., Kochanska et al. 1989; Sigel et al. 1992) are both associated with parenting behavior. Researchers, however, have not yet examined how these two types of thinking are related to each other and how each is uniquely related to parenting behavior and child emotional development.

There are many ways to examine parents’ thoughts and feelings about their child’s mind and about parenting. The current investigation focuses on two key ways: parental reflective functioning and parents’ beliefs about parenting. Reflective functioning (RF) assesses an individual’s thinking about one’s own and others’ thoughts and feelings and an understanding that behaviors are meaningfully connected to underlying mental states (Fonagy et al. 1991). Parental RF focuses on the parents’ thoughts about their own and their child’s mental states in the context of their relationship (Fonagy et al. 1991; Slade 2005). Parental RF is often assessed through the Parent Development Interview (Aber et al. 1985) which is modeled on the Adult Attachment Interview (George et al. 1984) and includes questions about the parent–child relationship, times that the parent and child were “clicking,” and about separations from the child. Parents high on RF are able to think about both their own and their child’s emotions and understand the connections between emotions, thoughts, and behaviors (Fonagy et al. 1998). RF allows parents to anticipate how their children might respond emotionally to situations and to make sense of their children’s behaviors (Gergely and Unoka 2008). Thus, it makes sense that parents with higher levels of reflective functioning might have more child-focused beliefs about parenting.

Parents’ beliefs about parenting are often divided into being either progressive or traditional (e.g., Lansford and Bornstein 2011; Schaefer and Edgerton 1985). Parents with more progressive (modern) parenting beliefs tend to focus more on children’s minds (e.g., ideas, feelings); whereas parents with traditional (authoritarian) beliefs tend to focus more on child behavior (e.g., conformity, misbehavior; Schaefer 1991). Progressive parents are more likely to appreciate child self-direction and curiosity. Traditional parents, on the other hand, are more concerned with children following adult instructions (Schaefer and Edgerton 1985).

Researchers have proposed that considering the child’s mind is often the basis upon which sensitive caregiving is built (e.g., Fonagy et al. 1991; Meins 1997, 1999). Using reflective functioning, parents are able to be more appropriately responsive, empathic, and nurturing with their child. Reflective functioning is more relationship-focused than behavior-focused which allows parents to focus on “connection over correction” while parenting (Baylin 2015, p. 176). This empathy and connectedness in the parent–child relationship is a key aspect of nurturing parenting.

Multiple studies have examined the association between parental RF and caregiving quality (see Camoirano 2017 for a review) including specifically examining maternal sensitivity (Rosenblum et al. 2008) and atypical maternal behaviors (Kelly et al. 2005). Researchers have not yet, however, examined associations between parental reflective functioning and nurturing parenting. Additionally, much of the research examining links between RF and parenting behavior studied mothers and infants (Kelly et al. 2005; Rosenblum et al. 2008), toddlers (Buttitta et al. 2019; Jessee et al. 2016; Suchman et al. 2010) or school-aged children (e.g., Rostad and Whitaker 2016)—few studies have examined these associations in preschool-aged children. In one exception, researchers found that preschool teachers with higher levels of RF reported engaging in more behaviors that promote social emotional development (e.g., explaining to one child how another child was feeling; Stacks et al. 2013). Another study examined this association between RF and parenting of preschool children in fathers and found that paternal RF was not associated with fathers’ self-reported parenting practices (Stover and Kiselica 2014). The researchers proposed that this null finding could be because of the specific parenting measure that they used (in which fathers reported on the quality of their attachment relationship with the child) and/or because men, on average, tend to be lower on RF (Bouchard et al. 2008; Esbjørn et al. 2013; Jessee et al. 2016) which may be unrelated to their parenting quality (Stover and Kiselica 2014).

This dearth of investigations into the association between parental RF and parenting during the preschool years is somewhat surprising given that preschool is a particularly important developmental period for children’s developing understanding of their own and others’ minds (e.g., Bartsch and Wellman 1995; Carpendale and Lewis 2015; Wellman et al. 2001). Children are coming to understand more about emotions—including identifying emotion expressions (e.g., Denham and Couchoud 1990) and their causes (Fabes et al. 1988). Thus, parents’ ability to reflect on their child’s mind seems especially important to investigate during this period.

Parents’ thoughts and beliefs about parenting shape how they view their child and how they parent (Bornstein and Lansford 2009; Kochanska et al. 1989; Sigel et al. 1992). Parents with more progressive (less traditional) parenting beliefs tend to be warmer (Luster et al. 1989), more positive and supportive (Jessee et al. 2016), less intrusive (Ispa et al. 2004), more sensitive (Shears and Robinson 2005), and have less dysfunctional parent–child interactions (Deater-Deckard and Scarr 1996). These parents are more likely to consider their child’s perspective and are more interested in their child’s opinions; thus, they are less likely to focus only on obedience and “good” behavior and more likely to focus on the parent–child relationship. Given these findings, it seems logical that these more progressive parents would also nurture a closer, more intimate relationship with their child that focused on emotional connections.

Broadly, mentalization is the imagining of what others are thinking and feeling (Fonagy and Allison 2012), and this includes both adult reflective functioning and children’s developing understanding of their own and others’ emotions. Thus, it makes sense that parents who are more likely to consider others’ thoughts and feelings (and how those thoughts and feelings are related to behaviors) would have children who were also better able to understand others’ thoughts and feelings. Studies examining mind-mindedness—a construct related to reflective functioning—have found that parents’ mind-mindedness is indeed associated with children’s theory of mind (e.g., Lundy 2013; Meins and Fernyhough 1999).

Several studies have examined the link between parental RF and children’s mentalizing in older children and teens. In a study on preadolescents, researchers found that when mothers had higher levels of reflective functioning, preteens had more advanced mentalization abilities (Rosso et al. 2015). Relatedly, in a sample of children (around 10 years old) who had experienced childhood sexual abuse, maternal reflective functioning was associated with children’s reflective functioning (Ensink et al. 2015). Finally, both maternal and paternal RF were associated with teens’ RF (Benbassat and Priel 2012). However, the one study that reported directly on the association between reflective functioning (about one’s early experiences rather than about one’s current relationship with one’s child) and young children’s theory of mind found that mothers’ RF was associated with children’s reasoning about others’ beliefs and desires (Fonagy et al. 1998; also described in Steele and Steele 2008).

Given that parents with progressive beliefs tend to encourage children to think for themselves and express their ideas, it makes sense that their children would have a more advanced theory of mind than children of more traditional parents. Although researchers have not explicitly examined this association, they have found that mothers with more progressive parenting beliefs used more mental-state talk when interacting with their 24-month-olds (Jessee et al. 2016). Relatedly, toddler teachers who had more progressive beliefs about child care used more emotion labeling when interacting with children (King 2016). This use of mental state language could lead children to develop more advanced emotion understanding. It could also be that progressive parents—who encourage their children to express their autonomy and differing opinions—allow for their child to consider multiple perspectives—both their own and their parents’ (Vinden 2001). These skills may allow children to more accurately understand what is going on in their own and others’ minds. Parents who have higher expectations about conformity, on the other hand, may discourage the child from thinking through differing perspectives and instead focus on the parents’ perspective (Vinden 2001).

Two studies have examined traditional or authoritarian parenting beliefs—the inverse of progressive beliefs—and children’s theory of mind. Mothers with more traditional parenting beliefs had preschool children who were less accurate in attributing intent to others (i.e., they were more likely to attribute hostile intent to other children; Runions and Keating 2007), and White mothers with more authoritarian parenting attitudes had preschool-aged children with less developed understanding of their own and others’ minds (Vinden 2001). Thus, research indicates that progressive (non-authoritarian) parenting beliefs are associated with children’s developing theory of mind.

In sum, the current investigation examined the associations between reflective functioning and progressive parenting beliefs and between each of these two variables and both parental behavior (i.e., nurturing) and child emotion understanding. Based on previous research, the hypotheses are that reflective functioning and parenting beliefs will be associated and that each will be associated with both maternal nurturing and child emotion understanding.

Method

Participants

Participants were 52 preschool-aged children (27 boys, 25 girls) and their mothers. Participants were recruited from an advertisement in a free local parenting magazine. Children were between 36 and 71 months old (M = 51.34; SD = 9.87 months; see Table 1). Most children (85%) had at least one sibling. Average mother age was 35.44 years (SD = 5.02; range = 26.72–50.97 years), and most mothers (71%) were White. Most (65%) of mothers worked (part-time or full-time) outside the home, and most (81%) had completed college or a higher level of education. With respect to family income, 17 (33%) reported less than $50K, 14 (27%) reported between $50K and $80K, 10 (19%) reported between $80K and $100K, and 11 (21%) reported over $100K.

Table 1 Descriptives of study variables
Full size table

Procedure

Mothers and their children visited the lab together. Before the visit, mothers were sent a set of questionnaires to be completed and brought in to the lab. Questionnaires include demographics, a measure of parenting beliefs, and a measure of nurturing parenting. Mothers consented both verbally and in writing to both their own and their child’s participation in the study. Mothers were interviewed using the Parent Development Interview-Revised (Slade et al. 2004), which was audio recorded and later coded for reflective functioning. During the mother’s interview, children completed a series of tasks (in a separate room) with a research assistant including an affective perspective taking task which assessed emotion understanding.

Measures

Parent development interview-revised: reflective functioning

The Parent Development Interview-Revised (PDI-R; Slade et al. 2004) is a 20-question interview that asks mothers about their relationship with their child (for mothers with more than one child, they were asked to focus on the child in the study). The PDI-R is similar to the Adult Attachment Interview (George et al. 1984) in that it asks individuals to use adjectives to describe their relationship and offer stories or examples to support those adjectives, but for the PDI-R the focus is on the relationship with the child. Parents are also asked to describe instances in which they and their child did or did not “click” and how having a child has changed them.

Interviews were transcribed and globally coded for reflective functioning (RF) by the author. RF focuses on an individual’s appropriate understanding and explanations of others’ emotions and behaviors (Slade 2005). RF coding includes four categories: awareness of the nature of mental states (e.g., demonstrating understanding that mental states can be disguised), an effort to understand the mental states underlying one’s own and others’ behavior (e.g., describing the emotions that led to one’s child throwing a tantrum), a recognition of the developmental aspects of mental states (e.g., describing changes in mental states that occur with typical development), and mental states with respect to the interviewer (e.g., acknowledging that it must be hard for the interviewer to listen to an emotionally challenging story). Reflective functioning is coded on a scale from −1 (bizarre) to 9 (high). Descriptives for study variables are presented in Table 1. Twelve transcripts (23%) were coded by a reliability coder, and reliability was good (ɣ = 0.733; ICC = 0.766). Both coders are certified reliable RF coders and were blind to participants’ demographic information and responses and scores on parenting beliefs, nurturing parenting, and child emotion understanding.

Parent modernity scale: progressive parenting

Mothers completed a modified version of the Parent Modernity Scale (PMS; Schaefer and Edgerton 1985), which assesses parenting beliefs and yields two scales: progressive and traditional parenting beliefs. The original PMS has 30 items including “Children will not do the right thing unless they must,” “Children learn best by doing things themselves rather than listening to others,” and “Children should not question the authority of their parents.” The original PMS includes seven questions related to school and teachers (e.g., “Teachers should discipline all the children the same”). These questions were removed because they were not relevant for the preschool-aged children in the current study. This left 23 questions in the version of the PMS used in the current investigation. Mothers indicated their level of agreement with items on a 5-point scale, from 1 (strongly disagree) to 5 (strongly agree). Item ratings were summed and yielded two subscales: progressive beliefs (seven items; M = 29.94, SD = 3.36; α = 0.73; e.g., “Children should be allowed to disagree with their parents if they feel their own ideas are better”) and traditional beliefs (16 items; M = 39.94, SD = 9.92; α = 0.83; e.g., “Children must be carefully trained early in life or their natural impulses will make them unmanageable”). Mothers who endorsed progressive parenting beliefs were less likely to endorse traditional parenting beliefs, r(52) = −0.446, p < 0.001. Thus, the subscales were combined to create a global composite of progressive parenting beliefs (α = 0.85). Because the two subscales differed on the number of items rated, each subscale was standardized first, and then the standardized traditional beliefs score was subtracted from the standardized progressive beliefs score. Thus, mothers high on the progressive parenting beliefs composite held beliefs that children should express themselves freely and be able to question parents, whereas mothers low on this scale valued child conformity and held beliefs that children should not be allowed to disagree with parents. This measure has demonstrated predictive validity and high test–retest reliability (Schaefer and Edgerton 1985) as well as significant stability between 1 month and first grade, r = 0.77 (NICHD Early Child Care Research Network and Arsenio 2004).

Parenting dimensions inventory: nurturing parenting

The Parenting Dimensions Inventory (PDI; Power 2002; Slater and Power 1987) is a self-report measure assessing parenting behavior. The current investigation focused on the 6-item nurturing subscale of the PDI which includes questions like “I encourage my child to talk about his or her troubles” and “My child and I have intimate moments together.” Mothers answered questions on a scale from 1 (not at all like me) to 6 (exactly like me). Reliability was acceptable (α = 0.653).

Affective perspective taking task: child emotion understanding

Children completed an affective perspective taking task which assessed their ability to predict their own and others’ emotional reactions (Harwood and Farrar 2006). Children were shown a laminated sheet of paper with drawings of two faces: one smiling and one frowning. In order to establish a baseline of understanding, children were asked to identify which face was happy and which was sad. Once this baseline was established, children were asked who their best friend was, and the experimenter explained that the child was going to hear short stories about themselves and their best friend and that they would be asked how they and their friend would each feel in that situation. Children could either verbally respond or point to the happy and sad faces. Answers were scored as either correct (1 point) or incorrect (0 points). If children gave an answer other than “happy” or “sad” (or other than pointing to one of the two [smiling or frowning] faces), the researcher asked the child again whether the child in the story would feel happy or sad. If a child did not verbally give one of the two answer choices or point to one of the two faces, their answer was scored as incorrect.

In some stories both the child and the friend experience the same emotion (e.g., “You and [friend] are coloring pictures together and your teacher comes over to tell you what a good job you are both doing), and in others the two children experience different emotions (e.g., “You and [friend] are playing ‘Candyland’ together. You win the game and [friend] loses the game;” see Harwood and Farrar 2006 for a fuller description of the task including a script). Children generally performed well on the task (M = 18.511 [out of 24 possible]; SD = 4.472).

Results

Preliminary Analyses

Preliminary analyses were conducted to examine associations between demographic variables (i.e., child gender and age, mother age and education, family income) and variables of interest (i.e., reflective functioning, progressive beliefs, nurturing parenting, emotion understanding).

Progressive parenting was marginally associated with both maternal age, r(52) = 0.258, p = 0.064, and maternal education, r(52) = 0.264, p = 0.058. Older and better educated mothers had more progressive parenting beliefs. Although these associations did not reach significance, in order to be conservative, both maternal age and education were included as covariates for all analyses involving progressive parenting beliefs.

As expected, older children performed significantly better on the emotion understanding task, r(50) = 0.585, p < 0.001. Thus, child age was included as a control in all analyses involving emotion understanding. Children of older mothers also performed better on the emotion understanding task, r(50) = 0.287, p = 0.043.

None of the demographic variables were significantly associated with either reflective functioning or nurturing parenting. None of the variables of interest differed based on child gender, all p’s > 0.215.

Associations between RF, Parenting Beliefs, Nurturing Parenting, and Child Emotion Understanding

Bivariate associations between variables of interest are presented in Table 2, below the diagonal. Associations between variables of interest—controlling for relevant covariates—are presented in Table 2, above the diagonal. Controlling for maternal age and education, mothers who had higher levels of reflective functioning held more progressive parenting beliefs, r(48) = 0.282, p = 0.048. These mothers with higher RF also reported higher levels of nurturing parenting, r(52) = 0.285, p = 0.040, and had children who performed better on the emotion understanding task (controlling for child age; r[47] = 0.316, p = 0.027). Mothers with more progressive parenting beliefs reported more nurturing parenting, r(48) = 0.397, p = 0.004.

Table 2 Correlations among study variables
Full size table

Regression Analysis: Nurturing Parenting

Given that reflective functioning and progressive parenting beliefs were both associated with nurturing parenting, regression analyses predicting nurturing parenting were conducted in order to examine the unique contributions of each variable to nurturing parenting. Maternal age and education were entered in the first step as covariates. The predictor variables, maternal reflective functioning and mothers’ progressive beliefs, were entered in the second and final step. Results are presented in Table 3. The overall model was significant, F(4,47) = 3.032, p = 0.027. Progressive parenting beliefs (p = 0.014)—but not reflective functioning (p = 0.259)—significantly predicted nurturing parenting behavior.

Table 3 Predictors of nurturing parenting
Full size table

Discussion

The current investigation examined the associations between parents’ reflective functioning, parenting beliefs, nurturing parenting, and child emotion understanding. Mothers with higher levels of reflective functioning had more progressive parenting beliefs—indicating that these mothers who make more of an effort to consider their child’s inner world (and think about their relationship with their child in more mentalistic terms) also held more child-centered parenting beliefs. These mothers with higher levels of reflective functioning also had preschool-aged children who were better able to accurately think about their own and others’ emotions. Finally, mothers with progressive beliefs also reported that they were more nurturing in their parenting. Thus, the ways that mothers think about their child’s mind and about their role as parents are associated with both how they interact with their child and how their children come to understand emotions.

Parents higher on RF reflect more on their child’s inner world. This reflection may allow parents to focus more on the child’s thoughts and feelings than on their behaviors—an important facet of progressive parenting (Schaefer 1991). These parents’ high RF may also cause them to encourage children to think for themselves and express their opinions (Schaefer 1991). However, the directionality of the association between progressive parenting beliefs and reflective functioning is not clear. It could be that having progressive parenting beliefs encourages or allows parents to think about their child in mentalistic terms. It may also be that this association is bidirectional and that these constructs influence one another across development. Or, certainly, there could be some third factor influencing both; however, both maternal age and education level were included as covariates in these analyses.

Additionally, mothers with more progressive beliefs reported that they were more nurturing in their parenting. This is consistent with previous research indicating that parents with more progressive parenting beliefs are warmer (Luster et al. 1989) and more supportive (Jessee et al. 2016) in their parenting. These nurturing parents are more likely to foster intimacy in their relationship with their children, encourage their children to talk about their feelings, and respect and be interested in their child’s opinions—all parenting behaviors that are consistent with progressive parenting beliefs.

Although bivariate correlations demonstrated that mothers higher on reflective functioning reported higher levels of nurturing parenting, this association between RF and nurturing became nonsignificant when progressive parenting was included in the model. Previous research has found that RF is associated with maternal sensitivity (e.g., Rosenblum et al. 2008), but this was the first study to examine associations between RF and nurturing parenting. It seems that the ways that RF is associated with nurturing parenting overlap with the ways that progressive beliefs are associated with nurturing. For example, both parents high in RF and those with progressive parenting beliefs tend to consider their child’s perspective and take interest in their child’s thoughts and feelings. It is not clear why progressive parenting beliefs would be more strongly associated with nurturing parenting than reflective functioning. Perhaps more traditional (less progressive) parents’ focus on child conformity precludes an intimate, nurturing relationship in a way that parents’ low RF (not much thought about the child’s inner world) does not. Perhaps these parents low in RF foster warmth and nurturance in their relationships with their children even if they do not actively consider their child’s thoughts and feelings. The fact that both progressive parenting beliefs and nurturing parenting were measured using self-report may also play a role. It could be that parents who report having more progressive beliefs may also report that they engage in more nurturing parenting. Future research should examine this association using observations of nurturing parenting.

Finally, mothers higher on reflective functioning—mothers who think about their child and their relationship with their child in mentalistic ways—have children who have more advanced emotion understanding. These reflective mothers are likely interacting with their children in ways that address and respond to the children’s inner worlds which, in turn, allows children to better understand their own and other’s emotions. This is consistent with work with older children and teens (e.g., Ensink et al. 2015; Rosso et al. 2015) and with the one study examining links between RF and younger children’s understanding of others’ beliefs and desires (Fonagy et al. 1998). However, this is the first study to explicitly examine the association between parental RF and preschool children’s emotion understanding which is notable considering the importance of the preschool period for the development of children’s theory of mind. During the preschool period, children are advancing in their understanding of the causes of emotions (e.g., Fabes et al. 1988) and identifying emotional expression in themselves and others (e.g., Denham and Couchoud 1990). Other research has established that parents play an important role in children’s emotion understanding (e.g., Dunn et al. 1991), and the current investigation adds to our understanding by demonstrating that parents’ reflective functioning is an important factor in children’s developing emotion understanding.

Limitations

There are several limitations to this study. The sample size was relatively small, homogenous, and low-risk. Future research should examine the associations between RF and parenting beliefs and parenting behavior and child emotion understanding in a larger, more diverse sample. These associations may play out differently in more diverse and/or at-risk samples. Further, this study was cross-sectional—it would be interesting to see how RF and parenting beliefs are associated with child emotion understanding over time. Future research should also examine possible mediators (e.g., emotion socialization, mental-state talk) of the associations between RF and child emotion understanding. Further, recent research (e.g., Suchman et al. 2010) has examined different types of reflective functioning—self-focused and child-focused. Researchers should examine whether and how these different types of RF might be differentially associated with both parenting behavior and child socioemotional development. Finally, as discussed above both parental beliefs and nurturing parenting were self-reported which could explain some of the association between these variables. Future research should examine associations between reflective functioning, parenting beliefs, and parenting behavior using observations of parenting behavior.

In sum, these findings indicate that mothers who were higher on reflective functioning reported that they had more progressive parenting beliefs, and these more reflective mothers also had children who were better able to understand their own and others’ minds. Further, mothers who reported more progressive parenting beliefs also reported that they were more nurturing in their parenting. These findings are particularly important given that researchers are increasingly interested in helping parents consider their child’s thoughts and feelings (e.g., Slep and O’Leary 1998; Suchman et al. 2008) in order to improve parenting quality and child outcomes. This research indicates that both parenting beliefs and reflective functioning may be important targets for clinicians working to improve outcomes for families.