1
Effects of Parental Anxiety and Intrusiveness on 7.5-year-old
Children’s Anxiety and Avoidance in a Social Referencing
Paradigm
Name: Maartje M. van Pagee
Studentnumber: 11059605
Master Orthopedagogiek
University: University of Amsterdam
Supervisor: Dr. Milica Nikolic
Date: 14-08-2018
2 Index
Abstract ... 3 Introduction ... 4
Social referencing paradigm 4
Parental intrusiveness 6
Method ... 9
Design and Population 9
Materials and Procedure 9
Measures 10 Data Analysis 11 Results ... 12 Preliminary Analyses 12 Main Analyses 14 Discussion ... 16 Maternal and paternal differences in anxiety and intrusiveness 17 Maternal and paternal expressed anxiety predicting child avoidance 18 Maternal and paternal intrusiveness predicting child avoidance 18
Limitations and future directions 19
Conclusion 20
3 Abstract
Introduction Social referencing, the process by which children take cues from parental behavior to form their responses to certain events or environment (SR), plays an important part in the intergenerational transmission of anxiety from parents to children. Even a short reaction from anxious parents in an SR-concept can increase children’s anxious and avoidance behaviors. This
cross-sectional study (part of a longitudinal investigation into the transmission of anxiety) investigates the predictive value of parental expressed anxiety and parental intrusiveness on 7.5-year-old children’s anxiety and avoidance behavior. Method In this study, 81 children (7.5-year-old) participated with their parents. Parental expressed anxiety, parental intrusiveness and children’s anxiety and avoidance behavior were measured in SR-concepts, based on Murray’s (2008) protocol and adapted for 7.5-year-olds. Each participant interacted with a stranger and with a robot toy. Meanwhile the behaviors of both the parents and children were observed. Results The levels between paternal and maternal expressed anxiety significantly differed in the robot concept (p <.01), but not in the stranger concept. The levels between maternal and paternal intrusiveness did not significantly differ in both the stranger and the robot concept. Maternal and paternal expressed anxiety had no predictive value for child avoidance. Maternal intrusiveness significantly predicted more child avoidance (p < .01) in the robot concept, while paternal intrusiveness predicted child avoidance in the stranger concept (p < .05). Discussion Fathers expressed equal levels of anxiety as mothers in a social context, but in a non-social context mothers expressed more anxiety than fathers. Maternal and paternal expressed anxiety did not matter for how much the child avoids both social and non-social contexts. Fathers and mothers expressed equal levels of intrusiveness in both social and non-social contexts. Paternal intrusiveness mattered more for how much the child avoids a social context, as maternal intrusiveness mattered more for how much the child avoids a non-social context.
Keywords: Social Referencing, Maternal and Paternal Expressed Anxiety, Maternal and
4
Effects of Parental Anxiety and Intrusiveness on 7.5-year-old Children’s Anxiety and Avoidance in a Social Referencing Paradigm
Anxiety is transmissible through generations: from parents to their children (Beidel & Turner, 1997). Through the years many studies have investigated the intergenerational transmission of anxiety and these studies showed that parental anxiety behaviors play an important role in the development of child anxiety and avoidance behavior (Aktar, Majdandžić, de Vente, & Bögels, 2013, 2014; Beidel & Turner, 1997; Chorpita & Barlow, 1998). Children with anxious parents are at a higher risk for developing anxious and avoidance behavior than children with non-anxious parents (Beidel & Turner, 1997; Chorpita & Barlow, 1998). But how do parents actually transmit their anxiety and influence the anxiety development of their children?
According to Murray, Creswell, and Cooper (2009), this transmission of anxiety can be explained in three different ways. The first model explains the intergenerational transmission of anxiety through genetic factors; children of anxious parents inherit a genetic predisposition for anxiety development (Eley, 2001). The second model explains that parents contribute to child anxiety
development through environmental factors (e.g., insensitive parenting, modeling anxious behavior or creating a social environment with lack of social support) (Murray et al., 2009). Furthermore, a combination between the genetic and environmental factors is possible (e.g., a genetic predisposition combined with an insensitive parenting style), which puts the child more at risk for developing anxiety and avoidance behavior (Fisak & Grills-Taquechel, 2007).
Through the years, many studies focused on the intergenerational transmission of anxiety in general through genetic or environmental factors in infancy (Belsky & Pluess, 2009; Eley, 2001; Fisak & Grills-Taquechel, 2007). Less is known about the intergenerational transmission of anxiety later in childhood and especially about the role and modeling of parental behaviors (e.g., anxious or intrusive behavior). Therefore, this study will investigate the role of parental behaviors beyond infancy, in 7.5-year-old children’s anxiety development, using a social referencing paradigm.
Social referencing paradigm
Social referencing (SR) is a mechanism that influences children’s behavior in anxious situations (Aktar et al., 2013, 2014; Pierce & Bandura, 1977). Parental anxiety can be transmitted to their children through modeling or observational learning of parental behavior in anxious situations (Aktar et al., 2013, 2014; Murray et al., 2008; Rapee, 2001). The way a parent behaves in a fearful situation influences the child’s perspective and behavior in that situation, because children learn from observing the behavior of their parents (Social learning theory, Pierce & Bandura, 1977). Social referencing (SR) of anxiety plays an important role in contributing to the intergenerational
transmission of anxiety (Aktar et al., 2013, 2014; Murray, de Rosnay, Pearson, Bergeron, Schofield, Royal-Lawson & Cooper, 2008). When a parent and child find themselves in an unfamiliar situation,
5
with a stranger or a novel toy (Murray et al., 2008), the child will refer to the parent (e.g., looking at the parents behavior) to check if the situation is ‘safe’ or if the situation is ‘dangerous’. In other words, if a parent communicates to the child that a particular situation is not safe (e.g., behaving anxious and avoiding), the child will behave accordingly and is likely to show anxious and avoidance behavior (Aktar et al., 2013; de Rosnay, Cooper, Tsigaras & Murray, 2006).
Studies that have investigated the link between parental anxiety and child fear and avoidance behavior via SR found that even a short reaction from a socially anxious parent in a social referencing concept (e.g., social or non-social context), increased the anxious and avoidance behavior of the child (de Rosnay et al., 2006; Gerull & Rapee, 2002). Parents with social anxiety disorder communicated more anxious and avoidance behavior in non-social and social contexts than parents without social anxiety disorder, therefore children of socially anxious parents were likely to show more typical anxious and avoidance behavior in SR situations than children of parents without social anxiety disorder (de Rosnay et al., 2006; Gerull & Rapee, 2002).
With regard to parental behavior, differences have been highlighted between fathers and mothers parenting behavior (Clarke-Steward, 1987; Lamb, 1977; Möller, Majdandžić, de Vente & Bögels, 2013). According to Lamb (1977) the parenting role of mothers in general in the child’s first year can be characterized by a caretaking perspective while the parenting role of fathers is in general more characterized by a playful perspective. Even more, the playful interactions of fathers and mothers are different (Lamb, 1977). Fathers in general will often play in a more physical way with their children (e.g., rough and tumble play), while mothers will play more in an intellectual way, like pretend-play (Clarke-Steward, 1987). Later in childhood, fathers are more likely to support their child to take initiative and to explore their child’s limits, while mothers are more likely to stimulate their child of taking care of others and generally show more protective behavior (Möller et al., 2013; Paquette, Carbonneau, Dubeau, Bigras & Tremblay, 2003, 2004).
Reading the above, we might wonder if these differences in parenting behaviors in fathers and mothers also have consequences in relation to the intergenerational transmission of anxiety. Murray et al. (2008) investigated the transmission of anxiety within a SR-paradigm for mothers with and without social anxiety disorders. In this study a stranger interacted with the mother in the SR concepts, which was observed by the child. Murray et al. found that mothers with a social anxiety disorder showed more socially anxious behavior towards the stranger than mothers without a social anxiety disorder. They also found that the children with socially anxious mothers, showed more anxiety and avoidance behavior, than the children with non-anxious mothers, proving that maternal anxiety behavior can be a predictor of child anxiety and avoidance behavior in a social referencing paradigm (de Rosnay et al., 2006; Murray et al., 2008).
6
Meanwhile the influence of paternal anxiety behavior on child anxious and avoidance behavior has been rarely studied, but is not less important. As mentioned before, fathers play an essential role in their children’s anxiety development. (Bögels & Phares, 2008; Majdandžić, Möller, de Vente, Bögels & van der Boom, 2014). The theory of Bögels & Phares (2008) suggests that the
father’s interaction with his child has a challenging perspective, while the mother-child interaction has a caretaking perspective. This is based on the evolutionary experience of fathers in exploring the external world (Bögels & Phares, 2008). The challenging behavior of the father influences the child to open up and explore the world. Challenging behavior can function as a buffer for developing child anxiety and avoidance behavior (Majdandžić, de Vente & Bögels, 2015). Fathers are more likely than mothers to show less anxious behavior while stimulating and challenging their children to explore. Möller et al. (2013) found that fathers expressed less anxious behavior and more challenging behavior while engaging in a non-social context than mothers, who expressed more anxious and protective behavior in the non-social context. According to Bögels and Phares (2008) and Majdandžić et al. (2015) anxious fathers have a lack of this challenging behavior, which increases the risk for developing anxious behavior in children. So, paternal anxious behavior seems as important as maternal anxious behavior for the transmission of anxiety in a social referencing paradigm. This is supported by recent studies of Aktar et al. (2014) who found that in social referencing concepts with toddlers the anxiety behaviors of fathers are as important as the anxiety behaviors of mothers.
To summarize, parental anxiety behavior of both fathers and mothers plays an important part in the development of child anxious behavior, through modeling their fears in a social context, like avoiding the conversation with strangers or expressing their fears (Murray et al., 2008; Pierce & Bandura, 1977). Anxious parents can also be unaffectionate towards their child or show a lack of stimulating the child to explore the world (Murray et al., 2008, 2009), which both are risk factors for increasing the child’s anxiety and avoidance behavior.
Parental Intrusiveness
Another way in which parental behavior influences their child’s anxiety and avoidance behavior is through a controlling and intrusive parenting style. Intrusive parents excessively try to regulate their children’s emotions and behaviors and instruct them on how to feel or think (McLeod, Wood, Sigman, Hwang & Chu, 2003). According to van der Bruggen, Stams and Bögels (2008), parental anxious psychopathology is a risk factor for developing a controlling parenting style. Parents with an anxiety disorder often show more controlling and intrusive behavior towards their children (McLeod, Wood, & Weisz, 2007). These parents not always consider their children as autonomous, they have trouble stimulating and encouraging their children to explore (Rapee, 2001) and they often make intrusive comments to their children (van der Bruggen et al., 2008). This intrusive parenting
7
behavior disrupts the children’s autonomous development and increases the anxious and avoidance behavior of the children (Hudson and Rapee, 2001).
In a meta-analytic study, McLeod et al. (2007) and van der Bruggen et al. (2008) found a low effect size association between parental anxiety and parental intrusiveness. Results from these studies suggest that parents with an anxiety disorder might develop an intrusive and controlling parenting style (McLeod et al., 2007; van der Bruggen et al., 2008). Aktar et al. (2014) studied this particular association in a social referencing paradigm with toddlers. They investigated the way parents expressed their anxiety and showed intrusive behavior in social and non-social contexts (e.g., a stranger and a robot concept). Aktar et al. found that socially anxious parents showed more intrusive behavior during the social referencing concepts than non-socially anxious parents, but found no evidence that parental intrusiveness predicted more anxiety and avoidance behavior in toddlers.
However, other studies state that intrusive parenting has a predictive value for increasing the child anxious and avoidance behavior (Hudson and Rapee, 2001; van der Bruggen et al., 2008). Intrusive parenting behavior has negative consequences for the social-emotional development of the child (McLeod et al., 2007; Szabo, Dekovic, van Aken, Verhoeven, van Aken & Junger, 2008), therefore it can be assumed that children with intrusive parents will show more anxious behavior and use more avoidance strategies in unfamiliar situations (McLeod et al., 2007; van der Bruggen et al., 2008), like the social referencing concepts, than children with non-intrusive parents. In the meta-analytic study of McLeod et al. (2007) and van der Bruggen et al. (2008) mentioned before, a medium effect size association was found between parental intrusiveness and child anxiety behavior.
The current study is a follow-up on the studies of Aktar et al. (2013, 2014), where the
predictive value of parental anxiety disorder and various parental behaviors on the behavior of infants and toddlers in a social referencing paradigm was investigated.
In the study with infants, Aktar et al. (2013) found no significant evidence for the predictive value of parental anxiety disorder trough social referencing on infant’s anxiety and avoidance behavior (Aktar et al., 2013). In the follow- up study with toddlers, they found that toddlers with anxious parents showed more anxiety and avoidance behavior than toddlers with non-anxious parents (Aktar et al., 2014). In both studies they found that fathers and mothers with a social anxiety disorder did not significantly differ in their behaviors towards their child. Furthermore, a positive association between parental anxiety and parental intrusiveness was found in the toddler studies (Aktar et al., 2014), which means that anxious parents showed more intrusive behavior during the social referencing concepts than non-anxious parents.
As mentioned before, less is known about the role of parental behaviors for children in a later lifetime, for example during middle childhood. This follow-up study will focus on the behaviors of
8
7.5-year-old children and their parents in a social referencing paradigm. The predictive value of parental expressed anxiety and parental intrusiveness on the anxiety and avoidance behavior of 7.5-year-olds and the differences between paternal and maternal expressed anxiety behaviors will be investigated.
First, the difference between maternal and paternal expressed anxiety levels and between the levels of paternal and maternal intrusive behavior in the SR-concepts will be investigated. Based on previous research of Möller et al. (2013), who found that fathers expressed less anxious behavior and more challenging behavior while engaging with a robot toy than mothers, the expectation is that fathers will express less anxiety behavior in the robot concept than mothers. Furthermore, mothers express more protecting and controlling behavior in general and stimulate their child to take care of others, while fathers have a more playful role (Paquette et al., 2003, 2004). Therefore, the expectation is that mothers will express more intrusive behavior than fathers.
Second, the predictive value of maternal and paternal expressed anxiety for anxious and avoidance behavior in 7.5-year-old children will be investigated. Children of anxious fathers and mothers are more at risk to manifest anxious and avoidance behavior than children of non-anxious fathers and mothers (Beidel & Turner, 1997; Chorpita & Barlow, 1998) and maternal and paternal anxiety behavior predicted more anxious and avoidance behavior in infants and toddlers (Aktar et al., 2013, 2014). The expectation is that 7.5-year-old children of mothers and fathers who express more anxiety will show more anxious and avoidance behavior during the social referencing concepts than children of non-anxious mothers and fathers. Based on findings of Aktar et al. (2013, 2014), who found that fathers and mothers anxiety is equally important for the child’s anxiety development, the expectation is that both mothers and fathers anxiety behavior will significantly predict child anxiety and avoidance behavior.
The third hypothesis investigates the predictive value of maternal and paternal intrusiveness on anxious and avoidance behavior in 7.5-year-old children. Based on previous research that proves positive associations between maternal and paternal intrusive behavior and the children’s anxious and avoidance behavior (Rapee, 2001; van der Bruggen et al., 2008), the expectation is that paternal and maternal intrusiveness is a predictor for child anxious and avoidance behavior at the age of 7.5-years. Furthermore, based on findings of Aktar et al. (2013, 2014), who found that fathers and mothers parenting behavior is equally important for the child’s anxiety development, the expectation is that both mothers and fathers intrusiveness will significantly predict child anxiety and avoidance behavior.
9 Method Design and Population
This study is part of a larger longitudinal investigation on the social development in children from birth to middle childhood at the University of Amsterdam (UvA). This longitudinal study followed parents, with and without anxiety disorders, and their children (from infancy to 7.5 years) and used social referencing concepts (e.g. social context and a non-social context). This research was approved by the ethics committee of the University of Amsterdam and took place at the UvA Family Lab. The participants were recruited via flyers in baby-shops, given out by midwives and via
advertisements on parenting websites and in parenting magazines. Parents who wanted to take part in the longitudinal study had to give a written informed consent before they could participate.
In this study the data collected from 7.5-year-old children (n = 81) and their mothers (n = 78) and fathers (n = 75) was used to investigate the influence of parental behavior on child behavior in a social referencing paradigm. The average age of the parents was 40 years and 9 months (M = 40.85,
SD = 4.66). Most of the participants in this study had a relatively higher educational level and higher
socioeconomic status. For the description of the sample, see Table 1.
Table 1
Participant Characteristics
Notes. ª M = mean and SD = standard deviation. ᵇ Assessed with an 8-point scale ranging from 1
(primary education) to 8 (university). ͨ Assessed with a 7-point scale ranging from 1 (<500 euros/month) to 7 (>5000 euros/month).
Materials and Procedure
Each 7.5-year-old child visited the lab twice, once with the mother and once with the father. About 50% of the children visited with their mother first and the other 50% first visited with their father first. To prevent a learning bias, this order was included as a control variable in the analyses. At
Parents (n = 152)
Children (n = 81)
Traits Father Mother Boys Girls
Sex n (%) 75 (49%) 78 (51%) 35 (43.2%) 46 (56.8%)
Age M (SD)ª 42.38 (4.842) 39.38 (3.99) 7.49 (0.06) 7.51 (0.17)
Dutch origin % 96.3 % 90.1 %
Educational levelᵇ M (SD) 6.54 (1.57) 7.14 (1.07) Monthly income ͨ M (SD) 4.62 (1.45) 4.2 (1.49)
10
the UvA Family Lab, the children and the parents participated in SR-concepts with a social context and a non-social context.
The SR-concept with a social context contained a stranger-situation; a stranger entered the room and started to ask the parent questions about his or her personality. Meanwhile the child was positioned on a chair in a 45 degrees angle from the parent and had to watch the conversation between the parent and the stranger. After two minutes the stranger asked the parent to switch chairs with the child and then asked the child questions about school and hobbies. The stranger left after a 2 minute conversation with the child and remained neutral throughout the whole concept. For both visits (mother and father visit) a different stranger was used.
The SR-concept with a non-social context contained a robot-situation; a remote controlled robot was used, build by the researchers from the UvA Family Lab. To prevent a learning bias by the child, the looks of the robot could be changed, by putting a different cover over the robot. The two robots were counterbalanced between the mother and father visits. The robot concept was divided into three phases in which the robot moved and made sounds (e.g., buzzing and annoying sounds). In the first phase, the parents were instructed to stay neutral and not to encourage the child to approach the robot. In phase two, the parent was allowed to talk about the robot and in phase three, the parent was instructed to actively encourage the child to play with the robot. For both SR-concepts, the participants were videotaped by researchers from the UVA Family Lab for further analysis and coding.
Measures
Two research observer-teams with each two researchers and under supervision of principal investigator followed an eight-week training program in coding the child’s or the parent’s behavior. Next the researchers analyzed and coded the child’s and the parent’s behavior in the SR-concepts, using Murray et al.’s (2008) coding protocol. Originally Murray’s protocol was used for coding parental behavior and the behavior of 4-year-old children (2008), but the two research-teams adapted the protocol for coding the behavior of 7.5-year-old children and their parents. Two coding schemes were developed; one for the stranger concept and a parallel coding scheme for the robot concept. To prevent any carry over effects, each observer coded either the father-condition or the
mother-condition. The SR-concepts were divided into four different episodes, each with time-intervals like 30 seconds or 1 minute. To measure the parents and the children’s first reaction to a strange person or a strange object, shorter episodes from 5 to 10 seconds were included (e.g., entry of stranger). In each episode the specific behavior of the participants was coded on a 5 point Likert-scale. The final score for each concept was the mean score of the separate scores given in each episode.
The two research-teams were blind to the diagnostic status of parental anxiety and double-coded about 20% of the SR-data for inter-observer reliability. The inter-observer reliability for the parent and child variables was good (intraclass correlations: parental expressed anxiety in stranger
11
concept .89 and in robot concept .90, parental intrusiveness in stranger concept .87 and in robot concept .93, child avoidance .97 for the stranger concept and .93 for the robot concept and child fearfulness in stranger concept .63 and in robot concept .92). The lower reliability in the stranger concept for child fearfulness can be explained by the little variance between the scores of the two observers. Although the reliability of child fearfulness in the stranger concept was lower, it was still acceptable. The double coded data was entered in the data collection by averaging the scores of the two observers. The variance was normal for most of the variables.
Parental expressed anxiety behavior. For measuring parental expressed anxiety behavior the observational symptoms of anxiety during the stranger and the robot concepts, divided into three categories: facial anxiety (e.g., biting or licking the lips), bodily anxiety (e.g., motionless or stiff body posture) and verbal anxiety (e.g., stiff responses), were scored on a 5 points Likertscale (no anxiety, mild anxiety, moderate anxiety, intense anxiety and very intense anxiety). Parents who showed more observational symptoms of anxiety received higher scores than parents who showed a few symptoms of anxiety.
Parental intrusive behavior. The intrusive behavior parents showed during the stranger and robot concepts, like interfering or taking control of the situation (e.g., talking to the child or gesturing for the child’s attention), were scored on a 5 points-Likertscale (no intrusiveness, mild intrusiveness, moderate intrusiveness, intense intrusiveness, and very intense intrusiveness). Very intrusive parents received higher scores on parental intrusiveness than parents who were not intrusive at all.
Child fearful behavior. The fearful behavior children showed was divided into three
categories: facial fear (e.g., wide open eyes, showing teeth), bodily fear (e.g., tension in the muscles or decreased bodily activity) and vocal fear (e.g., nervous laughter or fearful comments). Child
fearfulness was scored on 5 points-Likertscale (no fearfulness, mild fearfulness, moderate fearfulness, intense fearfulness, and very intense fearfulness) and children that showed more fearful behavior did receive higher scores than children that showed no fear.
Child avoidance behavior. The avoidance expressions (e.g., moving away from the stranger or the robot, hiding behind the parent) of the children were rated for duration and intensity. The avoidance expressions of the child during the SR-concepts were scored on a 5 points-Likertscale (no avoidance, mild avoidance, moderate avoidance, intense avoidance and very intense avoidance). Children showing more expressions of avoidance did receive higher scores than children that showed less avoidance expressions.
Data Analysis
For analyzing the parental and child behavior, the observations from both the SR concepts were used from both the father and mother visit, giving rise to four observation concepts for each child. Expressed maternal and paternal anxiety, maternal and paternal intrusiveness as well as child
12
fearfulness and avoidance in both robot and stranger concepts were entered as continuous variables and standardized. Unfortunately, child fearfulness was not normally distributed, especially in the stranger concept. The distribution was centered around the score of 3 (moderate fearful behavior), due to the very specific scoring mentioned earlier at reliabilities. This variable was not considered for further analyzing.
The difference in maternal and paternal levels of expressed anxiety and intrusiveness was investigated using an independent sample t-test with parental gender as a grouping variable and parental expressed anxiety and parental intrusiveness as testing variables. The second and third hypotheses were tested with multiple regression models. For each analysis a significance of p < .05 was used.
The relative predictive value of maternal and paternal expressed anxiety on child avoidance was investigated (separately for the stranger and robot concepts) using a multiple regression analysis with maternal and paternal expressed anxiety as predicting variables and child avoidance behavior as outcome variable.
The relative predictive value of maternal and paternal intrusive behavior on child avoidance was investigated using a multiple regression analysis with maternal and paternal intrusive behavior as predicting variables and child avoidance behavior as outcome variable (separately for the stranger and robot concepts).
Results Preliminary Analyses
Before analyzing the data, the variables were inspected for normal distribution, missing values and outliers as shown in Table 2. A visual inspection of the variables, including histograms, Q-Q Plots and a check for outliers, showed that parental expressed anxiety and child avoidance behavior were approximately normally distributed. The distribution of parental intrusiveness for the stranger was positively skewed and kurtotic, but after squared root transformation the skewness and kurtosis were improved. As stated above, child fearfulness was not normally distributed and therefore not considered for further analyzing. Because all cases were considered important, outliers were not deleted from the sample. No cases have been removed from the sample, due to missing data. Participants that only participated in one SR-concept (e.g., the stranger concept) were not removed because their data from the other SR-concept still was useful for analyzing. The assumptions of normality, equal variances, linearity and heteroscedasticity for the analysis were checked and found to be satisfactory.
13 Table 2
Descriptive Statistics of Parental Expressed Anxiety, Parental Intrusiveness, Child Fearfulness and Child Avoidance
Variable N M (SD) Min Max Skewness
(Std. Error) Kurtosis (Std. Error) Parental anxiety Stranger concept 152 3.34 (.24) 2.67 4 -0.18 (.22) 0.18 (.43) Robot concept 152 3.31 (.31) 2.5 3.94 -0.06 (.21) -0.36 (.43) Parental intrusiveness Stranger concept 152 1.08 (.11) 1 1.44 1.5 (.21) 1.73 (.42) Stranger concept-squaredª 152 1.04 (.05) 1 1.2 1.42 (.21) 1.43 (.42) Robot concept 152 1.29 (.23) 1 1.88 0.72 (.21) -0.05 (.43) Robot concept-squaredª 152 1.13 (.01) 1 1.37 0.55 (.21) -0.36 (.42) Child fearful behavior
Stranger concept 152 2.94 (.12) 2.33 3.33 -2.57 (.22) 10.18 (.43) Robot concept 152 3 (.24) 2.25 3.88 -0.40 (.22) 2.95 (.43) Child avoidance behavior
Stranger concept 152 2.93 (.31) 2.11 3.67 -0.03 (.22) -0.39 (.43) Robot concept 152 3.08 (.34) 2.1 3.8 -0.44 (.22) 0.32 (.43)
Notes. ªSquared root transformation.
Correlations between the parent variables were calculated separately per concept and parent as shown in Table 3. Only the significant correlations will be mentioned below.
Positive significant correlations were found between paternal anxiety in the stranger concept and paternal anxiety and intrusiveness in the robot concept. Paternal intrusiveness in the stranger concept correlated positively with paternal intrusiveness in the robot concept.
Maternal expressed anxiety in the stranger concept correlated positively with maternal anxiety in the robot concept and maternal intrusiveness in the robot concept correlated positively with child avoidance in the robot concept. Also, child avoidance in the stranger concept correlated positively with child avoidance in the robot concept.
Furthermore, positive correlations were found between maternal intrusiveness in the stranger concept and paternal intrusiveness in the robot concept and between paternal and maternal
14 Table 3
Correlations Between Maternal and Paternal anxiety, Paternal and Maternal Intrusiveness and Child Avoidance in Stranger and Robot Concepts
Notes. P_anx = Paternal anxiety, M_anx = Maternal anxiety, P_intr = Paternal intrusiveness, M_intr =
Maternal intrusiveness and C_Avoid = Child avoidance. ªPearson’s r. * p < .05. ** p < .01.
Main analyses
First, independent samples t-tests for two groups have been performed, to measure the differences between maternal and paternal expressed anxiety and to measure differences between maternal and paternal intrusiveness (see Table 4). The difference between maternal expressed anxiety behavior and paternal expressed anxiety behavior for the stranger concept was not significant, but the difference in the robot concept between maternal expressed anxiety and paternal expressed anxiety was significant. Further, Cohen’s effect size value (d = 0.50) suggested a moderate practical significance. This means that mothers and fathers equally expressed their anxious behavior in the stranger concept, but they expressed their anxiety differently in the robot concept. Mothers expressed significantly more anxiety than fathers in the robot concept. The differences between maternal and paternal intrusiveness for both the stranger and robot concept were not significant (see Table 4), which means that mothers and fathers equally expressed their intrusive behavior in both the stranger and robot concept.
Stranger concept Robot concept
P_anx M_anx P_intr M_intr C_avoid P_anx M_anx P_intr M_intr C_avoid Stranger concept P_anx _ M_anx -.14ª _ P_intr .01 -.06 _ M_intr .12 .21 -.08 _ C_avoid .02 .12 .24 -.07 _ Robot concept P_anx .46** -.01 .12 .11 .14 _ M_anx -.01 .38** .11 -.10 -.03 .03 _ P_intr .25* .15 .27* .27* -.13 .03 -.19 _ M_intr .12 -.16 .08 -.01 .03 -.08 -.17 .29* _ C_avoid .14 .10 .02 .08 .46** .06 .00 .19 .40** _
15 Table 4
Independent Sample T-test Between Maternal and Paternal Anxiety and Between Paternal and Maternal Intrusiveness in Stranger and Robot Concepts
Variables n M SD t df p
Expressed anxiety
Stranger concept Mother 77 3.38 0.26
-1.29 149 .200
Father 74 3.33 0.23
Robot concept Mother 72 3.37 0.30
-2.90 137 .004*
Father 67 3.22 0.29
Intrusiveness
Stranger concept Mother 77 1.10 0.16
-0.25 149 .806
Father 74 1.09 0.16
Robot concept Mother 72 1.30 0.27
0.69 137 .490
Father 67 1.34 0.27
Second, to measure the influences of maternal and paternal expressed anxiety on child
avoidance behavior multiple regression analyses were performed separately for the stranger and for the robot concept (as shown in Table 5), with paternal and maternal expressed anxiety as independent predictors and child avoidance as a dependent variable. The overall model was not significant. Also, neither mother’s nor father’s expressed anxiety significantly predicted child avoidance. This means that maternal and paternal expressed anxiety has no significant predictive value for 7.5-year-old children’s avoidance behavior in both the stranger and the robot concept.
Third, to measure the influences of maternal and paternal intrusive behavior on child
avoidance behavior multiple regression analyses were performed separately for the stranger and for the robot concept (as shown in Table 5), with paternal and maternal intrusiveness as independent
predictors and child avoidance as a dependent variable. In the stranger concept, the overall model for the predictive value of maternal and paternal intrusiveness on child avoidance behavior was not significant. Although the model was not significant, the unique effect of paternal intrusiveness on child avoidance behavior, after accounting for maternal intrusive behavior was significant. This result indicates that the more intrusive fathers behaved in the stranger concept, the higher the avoidance behavior of the child.
In the robot concept, the overall model for maternal and paternal intrusiveness on child avoidance behavior was significant (see Table 5), with maternal and paternal intrusiveness explaining for 18% of child avoidance. The unique effect of maternal intrusiveness on child avoidance behavior,
16
after accounting for paternal intrusiveness was statistically significant. The unique effect of paternal intrusiveness on child avoidance behavior, after accounting for maternal intrusive behavior was not statistically significant. These results indicate that the more intrusive mothers behaved in the robot concept, the higher the avoidance behavior of the child. However this relation was significant only for mothers, not for fathers. This means that maternal but not paternal intrusiveness has a significant predictive value for 7.5-year-old children’s avoidance behavior in the robot concept.
Table 5
Regression Analysis: Predictors of Child Avoidance
Notes. ªSSR = stranger concept. ᵇDSR = Robot concept.
Discussion
The central goal of this study was to investigate the predictive value of maternal and paternal
expressed anxiety and maternal and paternal intrusiveness on the anxiety and avoidance behaviors of 7.5-year-old children in a social referencing paradigm. The results and limitations of this study will be discussed below.
Independent variable Dependent variable F (df) R² t β p
Hypothesis 2 Parental anxiety SSRª Child Avoidance SSR 0.50 (2,68) -.02 .741 Mother 0.99 .15 .325 Father 0.28 .04 .779 Parental anxiety DSRᵇ Child Avoidance DSR 0.12 (2,56) .004 .890 Mother 0.02 .002 .989 Father .048 .06 .631 Hypothesis 3 Parental intrusiveness SSR Child avoidance SSR 2.21 (2,64) .07 .118 Mother -0.35 -.11 .726 Father 2.04 .76 .046* Parental intrusiveness DSR Child avoidance DSR 5.61 (2,56) .17 .006* Mother 2.96 .36 .004* Father 0.61 .09 .542
17
Maternal and paternal differences in anxiety and intrusiveness
First, the difference between the levels of maternal and paternal expressed anxiety in the social referencing concepts was investigated. The results in the robot concept supported the hypothesis that the levels of anxiety between fathers and mothers are significantly different. Mothers showed more anxiety than fathers in the robot concept. This finding is consistent with previous research of Möller et al. (2013), who found that fathers expressed less anxious behavior while engaging with a robot toy and supported their child to take initiative and to explore the robot, while mothers expressed more anxious behavior in the presence of the robot.
The results in the stranger concept did not support a difference in the levels of maternal and paternal expressed anxiety. Mothers and fathers showed equal levels of expressed anxiety behavior during the interaction with the stranger. A possible explanation for the different results between the stranger and the robot concept is that in the stranger concept parents were not specifically asked to actively challenge their child to engage with the stranger. According to Majdandžić et al. (2015), the challenging behavior of the father can be a buffer for anxiety and avoidance behavior. Fathers are more likely than mothers to show less anxious behavior while stimulating and challenging their children to explore. It can be assumed that fathers expressed less anxiety behavior in the robot concept than mothers because they stimulated and challenged their child’s interaction with the robot more than mothers. In the stranger concept parents were not asked to encourage the child, which can explain the equal levels of expressed anxiety between mothers and fathers.
No significant differences have been found between the levels of maternal and paternal intrusiveness in both the stranger and the robot task, which means that both fathers and mothers equally expressed their intrusive behavior (e.g., protecting behavior, negative verbal or physical interactions). These findings are inconsistent with previous research of Paquette et al. (2003, 2004), who stated that mothers express more protecting and controlling behavior in general.
A possible explanation for these results is the way parental intrusiveness was coded. In this study the quality of the intrusive behavior was thought more important than the quantity. Parents showing physical intrusiveness received higher scores for parental intrusive behavior than parents showing verbal intrusiveness. Because fathers’ interactions with the child are considered more physical (Clarke-Steward, 1987), fathers are more likely to engage in physical intrusive behavior than mothers. Thus, if mothers engaged in multiple verbal intrusive interactions and fathers in less physical intrusive interactions, they would both receive high scores for intrusive behavior. This makes it possible that fathers and mothers still received equal scores for intrusive behavior, because the quality and not the quantity of parental intrusiveness was coded.
18
Maternal and paternal expressed anxiety predicting child avoidance
Second, the predictive relative value of maternal and paternal expressed anxiety on the avoidance behavior in 7.5-year-old children was investigated. For both the stranger concept and the robot concept, levels of expressed maternal and paternal anxiety did not significantly predict levels of avoidance behavior in 7.5-year-old children. These results are inconsistent with earlier studies of Beidel & Turner (1997), who found that children from anxious parents are more likely to develop anxious and avoidance behavior than children of non-anxious parents. These findings are also
inconsistent with previous social referencing studies of Murray et al. (2008), who found that infants of anxious mothers, showed more anxious and avoidance behavior, than infants with mothers that
showed less anxious behavior and of Aktar et al. (2014), who found that toddlers with socially anxious parents showed more anxious and avoidance behavior than toddlers with non-anxious parents.
A possible explanation for the lack of support for the predictive value of parental anxiety behavior on child avoidance behavior is that this study only used the expressed parental anxiety behavior in the SR-concepts to measure the parents’ anxious behavior. It is likely to find stronger associations between parental anxiety and child avoidance behavior when the parental anxiety status, measured with the ADIS-IV-L (Brown, Barlow & Di Nardo, 1994) is used instead of the parental expressed anxiety. The clinical interview measures parental anxiety symptoms in multiple situations and is therefore less exposed to human errors and a more valid instrument to measure parental anxiety. Further research should consider using parental anxiety state in combination with parental expressed anxiety as predictor for child avoidance.
Another explanation to be considered is the developmental stage of the children that
participated in this study (7.5-year-olds). These 7.5-year-olds are currently becoming more capable of developing their own opinions, are more likely to think for themselves (Došen, 2007) and are less likely to be influenced by their parents’ anxious behavior towards a robot toy than infants and toddlers are. Piira, Champion, Bustos, Donnelly and Lui (2007) support the idea that children of a younger age are easier influenced by their parents behavior, while older children are more likely to think for themselves and to consider if a specific context is to be afraid or nervous for (Došen, 2007).
Furthermore, the end of infancy is considered to be the most sensitive period for the transmission of anxiety through social referencing (Aktar et al., 2013, 2014; Piira et al., 2007). This could imply that at 7.5-years the social referencing paradigm is not the only mechanism that plays an important role in the transmission and modeling of anxiety.
Maternal and paternal intrusiveness predicting child avoidance
Third, the predictive value of maternal and paternal intrusiveness for avoidance behavior in 7.5-year-old children was investigated. In the stranger concept, levels of maternal and paternal intrusiveness did not predict child avoidance behavior overall. However, the unique contribution of
19
paternal intrusiveness predicting child avoidance in the stranger concept was significant. A possible explanation for this result is that, according to Bögels, Stevens and Majdandžić (2011), the reactions and behavior of fathers in a social context matter more for the child than the reactions and behavior of mothers. This is based on the findings of Bögels et al. that fathers have the capacity to decrease child anxiety and avoidance and increase child self-confidence by stimulating the child to explore and push boundaries in the social world. Therefore, when fathers do exactly the opposite and make intrusive comments or show protective behavior in social contexts – what mothers are more likely to do according to Paquette (2004) – this paternal intrusiveness has a greater impact on the child’s anxiety and avoidance behavior than maternal intrusive behavior (Bögels et al., 2011).
In the robot concept higher levels of maternal intrusiveness significantly predicted more avoidance behavior in 7.5-year-old children. When the mother showed more intrusive behavior in the robot concept, the child showed significantly more avoidance behavior. This result suggests that intrusive parenting as in physical intrusions or verbal comments from the mother (e.g., do not go or touch the robot), often lead to avoidance behavior of the child (e.g., not touching or playing with the robot). This finding is consistent with previous researches of Hudson and Rapee (2001), where they found that maternal intrusiveness increases the child anxiety behavior, of McLeod et al. (2007) and van der Bruggen et al. (2008), where evidence was found that intrusive parenting behavior had a negative influence on the child’s avoidance behavior. A possible explanation for the fact that maternal intrusiveness was a significant predictor (and paternal intrusive behavior was not) for child avoidance in the robot concept is the difference in parenting behaviors between fathers and mothers. As
mentioned before, the role of maternal parenting behavior is more caring and protective of nature (Paquette, 2004), than paternal parenting behavior (e.g., challenging and stimulating behavior). Maternal caring and protective behavior may, thus, be of more importance than paternal protective behaviors and in turn it may easier increase the anxiety and avoidance behavior in the child (de Wilde & Rapee, 2008).
Limitations and future directions
This study has several limitations, the first one being that most of the participants were from a relatively higher social-economic status and had a more well-educated background than the general population. Also, most children that participated grew up with their biological parents. This has implications for the generalizability of the results toward people with a lower social economic status or children brought up with step-parents or foster families. Well-educated participants, and with a higher social-economic status, are likely to have learned more about effective parenting than the general population. For example, studies showed that parents with a higher social-economic status used less intrusive and controlling parenting styles than parents with a low social-economic status
20
Furthermore, the participants in this study were videotaped while participating in social referencing concepts. This could have an implication on parental anxious and intrusive behavior. Parents that in general show more anxious or intrusive behavior might have behaved differently knowing that they were being observed and videotaped. Third, in this study differences in child characteristics (e.g., gender of the child) and their influence on anxious and avoidance behavior were not analyzed. The effect of parental behaviors on the development of child anxiety and avoidance behavior might be different for boys and girls. According to Leve and Fagot (1997) the gender of the parent positively influences the reactions to opposite-sex children. Additional research should consider analyzing parenting behaviors in relation to the parental and the child’s characteristics.
As mentioned before, this study took part in a larger longitudinal study of the intergenerational transmission of anxiety in a social referencing paradigm. The current study is cross-sectional and used only the data when the children were 7.5-years-old. Therefore, the effects of maternal and paternal anxiety and intrusiveness in children’s infancy and early childhood on child anxiety and avoidance have not been taken into account. According to Murray et al. (2008) the end of infancy is a sensitive period for the learning of anxiety. It can be assumed that parental expressed anxiety or intrusiveness at infancy has a more predictive value for child anxiety and avoidance behavior at 7.5-years. For further research, it is recommended to study the effects of parental expressed anxiety at infancy and
toddlerhood on child anxiety and avoidance at a later age to receive a more complete picture of the intergenerational transmission of anxiety in a social referencing paradigm and the relations between parenting behaviors and children’s anxiety behaviors.
Conclusion
The results in this study did not support the general idea that maternal and paternal anxiety can be transmitted toward 7.5-year-old children through social referencing. While maternal and paternal expressed anxiety did not matter for how much the child avoids both social and non-social contexts, paternal intrusiveness in the stranger concept and maternal intrusiveness in the robot concept predicted more avoidance behavior of the child. Further research should consider measuring the effects of parental expressed anxiety, parental lifetime-disorder and parental intrusiveness at infancy and toddlerhood on child avoidance at 7.5 years, as the end of infancy is a sensitive period for learning anxiety in a social referencing paradigm. Moreover, it would be interesting to measure opposite-sex relations, because the effects of maternal and paternal expressed anxiety and intrusiveness on child anxiety and avoidance behavior might be different for boys and girls.
21 References
Aktar, E., Majdandžić, M., de Vente, W., & Bögels, S.M. (2013). The interplay between expressed parental anxiety and infant behavioral inhibition predicts infant avoidance in a social referencing paradigm. The Journal of Child Psychology and Psychiatry, 54(2), 144–156. doi:10.1111/j.1469-7610.2012.02601.x
Aktar, E., Majdandžić, M., Vente, W., & Bögels, S. M. (2014). Parental social anxiety disorder prospectively predicts toddlers' fear/avoidance in a social referencing paradigm. Journal of
Child Psychology and Psychiatry, 55(1), 77-87. doi:10.1111/jcpp.12121
Beidel, D. C., & Turner, S. M. (1997). At risk for anxiety: I. psychopathology in the offspring of anxious parents. Journal of the American Academy of Child and Adolescent Psychiatry, 36(7), 918–924. doi:10.1097/00004583-199707000-00013
Belsky, J., & Pluess, M. (2009). Beyond diathesis–stress: differential susceptibility to environmental influence. Psychological Bulletin, 135(6), 885–908. doi:10.1037/a0017376
Bögels, S. M., & Perotti, E. C. (2011). Does father know best? A formal model of the paternal influence on childhood social anxiety. Journal of Child and Family Studies, 20(2), 171–181. doi:10.1007/s10826-010-9441-0
Bögels, S.M., & Phares, V. (2008). Fathers’ role in the etiology, prevention and treatment of child anxiety: A review and a new model. Clinical Psychology Review, 28(4), 539–55.
doi:10.1016/j.cpr.2007.07.011
Bögels, S. M., Stevens, J., & Majdandžić, M. (2011). Parenting and social anxiety: fathers’ versus mothers’ influence on their children’s anxiety in ambiguous social situations. Journal of Child
Psychology and Psychiatry, 52(5), 599-606. doi: 10.1111/j.1469-7610.2010.02345.x
Brown, T. A., Barlow, D. H., & DiNardo, P. A. (1994). Anxiety Disorders Interview Schedule for
DSM-IV (ADIS-IV): Client Interview Schedule. Graywind Publications Incorporated.
Chorpita, B. F. (2001). Control and the development of negative emotion. In M. W. Vasey & M. R. Dadds (Eds.), psychopathology of anxiety (pp. 112-142). Oxford, UK: Oxford University Press. doi:10.1093/med:psych/9780195123630.003.0006
Chorpita, B. F., Barlow, D. H. (1998). The development of anxiety: the role of control in the early environment. Psychological Bulletin 124(1), 3–21. doi:10.1037/0033-2909.124.1.3
Clarke-Stewart, K.A. (1978). And daddy makes three: the father’s impact on mother and young child.
Child Development, 49(2), 446 –478. doi:10.2307/1128712
de Rosnay, M., Cooper, P.J., Tsigaras, N., & Murray, L. (2006). Transmission of social anxiety from mother to infant: An experimental study using a social referencing paradigm. Behaviour
22
de Wilde A., & Rapee, R. M. (2008). Do controlling maternal behaviors increase state anxiety in children’s responses to a social threat? A pilot study. Journal of Behavior Therapy and
Experimental Psychiatry, 39(4), 526–537. doi:10.1016/j.jbtep.2007.10.011
Došen, A. (2007). Psychische stoornissen, gedragsproblemen en verstandelijke handicap. Een integratieve benadering bij kinderen en volwassenen. Tijdschrift voor Psychiatrie, 49, 456-457.
Dotterer, A. M., Iruka, I. U., & Pungello, E. (2012). Parenting, race, and socioeconomic status: Links to school readiness. Family Relations, 61(4), 657-670. doi:10.1111/j.1741-3729.2012.00716-x Eley, T.C. (2001). Contributions of behavioral genetics research: Quantifying genetic, shared
environmental and non-shared environmental influences. In M. W. Vasey & M. R. Dadds (Eds.), The developmental psychopathology of anxiety, (pp. 45–59). New York, US: Oxford University Press. doi:10.1093/med:psych/9780195123630.003.0003
Fisak, B. Jr, & Grills-Taquechel, E. A. (2007). Parental modelling, reinforcement, and information transfer: Risk factors in the development of child anxiety? Clinical Child and Family
Psychology, 10(3), 213–231. doi:10.1007/s10567-007-0020-x
Gerull, F. C., Rapee, R. M. (2002). Mother knows best: effects of maternal modelling on the
acquisition of fear and avoidance behavior in toddlers. Behavior Research and Therapy, 40(3), 279–287. doi:10.1016/s0005-7967(01)00013-4
Hudson, J. L., Rapee, R. M. (2001). Parent–child interactions and anxiety disorders: an observational study. Behavior Research and Therapy, 39(12), 1411–1427.
doi:10.1016/s0005-7967(00)00107-8
Lamb, M. E. (1977). Father–infant and mother–infant interaction in the first year of life. Child
Development, 48(1), 167–181. doi:10.2307/1128896
Leve, L. D., & Fagot, B. I. (1997). Gender-role socialization and discipline processes in one- and two-parent families. Sex Roles, 36(1-2), 1–21. doi:10.1007/bf02766236
Majdandžić, M., de Vente, W., & Bögels, S. M. (2015). Challenging parenting behavior from infancy to toddlerhood: Etiology, measurement and differences between fathers and mothers. Infancy,
21(4), 423-452. doi:10.111/infa.12125
Majdandžić, M., Möller, E. L., de Vente, W., Bögels, S. M., & Boom, van den, D. C. (2014). Fathers’ challenging parenting behavior prevents social anxiety development in their 4-year-old children: A longitudinal observational study. Journal of Abnormal Child Psychology, 42 (2), 301-310. doi:10.007/s10802-013-9774-4
McLeod, B. D., Wood, J. J., & Weisz, J. R. (2007). Examining the association between parenting and childhood anxiety: A meta-analysis. Clinical Psychology Review, 27(2), 155–172.
23
Möller, E. L., Majdandžić, M., de Vente, W., & Bögels, S. M. (2013). The evolutionary basis of sex differences in parenting and its relationship with child anxiety in Western societies. Journal of
Experimental Psychopathology, 4(2), 88. doi:10.5127/jep.026912
Murray, L., Creswell, C., & Cooper, P.J. (2009). The development of anxiety disorders in childhood: An integrative review. Psychological Medicine, 39(9), 1413–1423.
doi:10.1017s0033291709005157
Murray, L., de Rosnay, M., Pearson, J., Bergeron, C., Schofield, E., Royal-Lawson, M., & Cooper, P. J. (2008). Intergenerational transmission of social anxiety: The role of social referencing processes in infancy. Child Development, 79(4), 1049–1064.
doi:10.1111/j.1467-8624.2008.01175.x
Piira, T., Champion, G. D., Bustos, T., Donnelly, N., & Lui, K. (2007). Factors associated with infant pain response following an immunization injection. Early Human Development, 83(5), 319-326. doi:10.1016/j.earlhumdev.2006.06.007
Paquette, D., Carbonneau, R., Dubeau, D., Bigras, M., & Tremblay, R. E. (2003). Prevalence of father-child rough-and-tumble play and physical aggression in preschool children. European
Journal of Psychology of Education, 18(3), 171—189. doi:10.1007/bf03173483
Paquette, D. (2004). Theorizing the father-child relationship: Mechanisms and developmental outcomes. Human Development, 47(4), 193-217. doi:10.1159/000078723
Pierce, W. D., Bandura, A. (1977). Social learning theory. Canadian Journal of Sociology, 2(3), 321. doi:10.2307/3340496
Rapee, R. M. (2001). The development of generalized anxiety disorder. In M. W. Vasey and M. R. Dadds (Eds.), The Developmental Psychopathology of Anxiety, 481–504. Oxford University Press: New York, US. doi:10.1093/med:psych/9780195123630.003.0021
Szabo, N., Dekovic, M., van Aken, C., Verhoeven, M., van Aken, M. A. G., & Junger, M. (2008). The relations among child negative interactive behavior, child temperament, and maternal
behavior. Early Childhood Research Quarterly, 23, 366-377. doi:10.1016/j.ecresq.2008.01.004
van der Bruggen, C. O., Stams, G. J., & Bögels, S. M. (2008). Research review: the relation between child and parent anxiety and parental control: a meta-analytic review. Journal of Child
