1
Parental anxiety and infants’
attention bias towards
fearful faces
Master thesis Clinical Psychology - University of Leiden J.C.E. Jeurissen
Studentennummer: s2321548 Supervisor: C. Nimphy Leiden, July 2020
2 Abstract
Although there are indications that parental anxiety is related to the development of anxiety among children, direct evidence for the specific mechanisms that underlie the parent-child transmission of anxiety is limited. The interaction between anxious parents and their infant is a potentially important mechanism for the transmission of fear from the parent to the infant. This study aims to investigate the relationship between parental anxiety (mothers and fathers) and infants’ attention bias towards fearful faces (vs. happy faces). Both mothers (N = 155) and fathers (N = 74) were included in the study and parental anxiety was measured with the DASS questionnaire. Attention bias towards fearful faces (vs. happy faces) was measured with an eye-tracker in infants between 5 and 19 months old (N = 220; 112 girls, 108 boys). Results show that infants looked longer at fearful faces than at happy faces. However, no relation between parental anxiety and infants’ attention bias towards fearful faces was found. Moreover, the gender of the parent did not influence the relationship between parental anxiety and an infant’s attention bias. The findings do not support the notion that attention bias in infants is a mechanism that could explain the transmission of anxiety from parents to their children. Our study contributes to the literature about the transmission of anxiety from parents to their children and the role of infants’ attention bias in this transmission.
3 The relationship between parental anxiety and infants’ attention bias towards
fearful faces
Anxiety disorder is one of the most common diseases among adults and children (Kessler et al., 2005). The Netherlands Mental Health Survey and Incidence study (NEMESIS-2) investigated that in 2011, almost 1.1 million adults had been diagnosed with an anxiety disorder (Graaf, Have & Dorsselaar, 2011). Among children, the prevalence of anxiety disorders varies from 4 to 20%. Anxiety in children can lead to a reduction in social and intellectual functioning (Pine, 1997). In addition, Bittner and colleagues (2007) state that there is an increased risk that anxious children develop a psychiatric disorder as an adolescent or adult, for example an anxiety disorder, a depression or an attention-deficit/hyperactivity disorder (ADHD).
Evidence supports that anxiety runs in the family (Telman, Steensel, Maric & Bögels, 2018). Children who have parents with an anxiety disorder are at greater risk for developing an anxiety disorder themselves (Warner, Mufson & Weissman, 1995). A number of processes are probably involved in the intergenerational transmission of anxiety in childhood. Firstly, genetic mechanisms contribute in the transmission process. Adoption studies and twin studies have shown that heritability seems to play a role in the development of anxiety disorders (Eley, 2001). Also, environmental mechanisms - and parenting practices in particular - play a role in the transmission of anxiety. Research suggests that several parenting practices, such as parenting styles and the modelling of fearful behaviour, may be risk factors for the
development of child anxiety (Ginsburg & Schlossberg, 2002). However, although some studies find indications that these parenting practices are related to the development of anxiety among children, direct evidence for this relationship is limited and more research needs to be conducted (Mcleod, Wood & Weisz, 2007).
One factor of the normal infant developmental process that possibly plays a role in the development of anxiety in children is social referencing (SR). SR is ‘a process of emotional communication in which one’s perception of others person’s interpretations of events is used to form one’s own understanding of that event’ (Feinman, 1982). To understand and form an interpretation of the situation, the infant often uses the communication (through emotions) of their parents. That is, infants learn through parental modelling and observational learning from their parents (Fisak & Grills-Taquechel, 2007). This starts with infants at the end of their first year. The behaviour of parents is often imitated by the infant, which may indicate that parents’ responses are being used by the infants as a guide for developing their own
4
response to the situation. For example, the study of Feinman and Lewis (1983) states that infants (1 year old) act less friendly to strangers when their mothers express negative faces and voices towards them. This indicates that an infant may use emotional cues from their caregiver as a source of information when they face unfamiliar people or situations (Feinman, 1982).
A study of Gerull and Rapee (2002) suggests that parental modelling and
observational learning also play a role in the transmission of fear. They found that when the mother of an infant shows negative facial expressions towards a fear-relevant toy (spider or snake), the infant demonstrates significantly more avoidance and fear to the toy. Their results indicate that infants may retain fear-related information by observing their parents. As another example, Hornik, Risenboover & Gunnar (1987) found that when mothers react with a fearful face to a stimulus, the infant is more likely to respond fearfully to the stimulus as well.
To summarize, in the process of SR, infants use their parents’ responses to create their own emotional responses and behaviour. This process (of parental modelling) may cause infants to take over the fearful responses that their parents show.
If infants use their parents’ responses to form their own emotional responses and behaviour, this may also have consequences for the shaping of their attention. When
processing emotions, young infants generally show an attention bias towards negative stimuli (Pérez-Edgar et al., 2017). Cognitive anxiety models describe this attention bias as:
‘individuals selectively direct cognitive resources towards threat stimuli, especially during the very early stages of stimulus processing’ (Mogg & Bradley, 1998). Nelson and Dolgin (1985) found for example that infants (7-months old) looked longer at fearful faces than at happy faces. Various studies support these findings and find an attention bias to threat-related stimuli compared to neutral stimuli as well (Peltola et al., 2009). If this attention bias does not diminish over time, it could possibly lead to maladaptive developmental pathways later in life (Morales, Fu, & Pérez-Edgar, 2016).
One possible intergenerational transmission mechanism of anxiety is shared attention bias between parents and their infants. Anxious parents are more likely to show negative facial expressions to their infant (such as fear). Moreover, De Haan and Nelson (1999) demonstrate that infants show an increased attention to their mother’s faces versus the faces of strangers. That implies that when infants have anxious parents, they are repeatedly exposed to fearful expressions. As a consequence, they might develop an enhanced sensitivity to that expression as well (de Haan et al., 2004). Moreover, they may learn that these expressions are important emotions.This may increase the risk of developing an increased negativity attention
5
bias towards these faces (Lipsitt, 1998). Clinical literature indicates that attention bias towards threats is an indicator for the development and maintenance of anxiety (Beck & Clark, 1988).
Thus far, it has become clear that infants show an attention bias towards threatening faces, and that this attention bias may be indicative for the development of anxiety in the infant. However, it is not clear yet whether the strength of this attention bias is dependent on the presence of anxiety symptoms in the parent. Insight in the relationship between an infants’ attention bias towards threatening faces and parental anxiety may lead to a better
understanding of why some infants develop anxiety later in life and others do not. Previous studies suggest that maternal anxiety may affect attention biases towards threats in infants. A few studies found that attention to fearful faces is positively related to maternal anxiety symptoms (Aktar et al., 2016; Taylor, Slade & Herbert, 2014). By using eye-tracking, Morales and colleagues (2017) found that maternal anxiety leads to a stronger attention bias among infants towards threatening faces (compared to neutral faces). Similar results were also found in children. Daughters of anxious mothers showed more attention bias towards threatening faces (vs neutral faces) compared to children of non-anxious mothers (Montagner et al., 2015).
Previous studies mainly focused on comparing fearful versus neutral faces. However, neutral facial expressions can be interpreted as ambiguous expressions (Thomas et al., 2001) and children often reported neutral as a negative emotion. This could result in a
misinterpretation of the face (Melfsen & Florin, 2002). In addition, in studies who use dynamic facial expressions as stimuli, neutral faces typically have fewer movements than happy or threatening faces. Happy and fearful faces start at neutral and slowly become more happy or fearful. For example, happy faces change from neutral to happy with the opening of the mouth, widening of the eyes and a larger smile. It is likely that if a face changes, infants are more drawn to look at it compared to a neutral face which does not change during the presentation. This study compares fearful faces with happy faces, so that a misinterpretation of the neutral face (and the fact that a neutral face is more static) cannot influence the results.
Moreover, most of the research has focused on maternal anxiety. Therefore, not much is known about the effect of fathers’ anxiety on the attention bias of infants. This is
noteworthy, because according to Möller, Majdanzic and Bögels (2013), fathers and mothers play a different role in the development of anxiety in children. They found that fathers show more overinvolvement than mothers. This overinvolvement, in turn, is associated with more anxiety in their infants (Möller, Majdanzic & Bögels, 2015). Furthermore, fathers spend less time with their infants than mothers, making them more insecure in their parental-infant
6
interaction (Geary, 2010). This insecurity may result in more fearful behaviour and more fearful expressions by the father towards their infant, compared to the mother. Therefore, the effect of the anxiety of a father on an infant’s attention bias may be stronger than the effect of a mother’s anxiety.
Currently, relatively little is known about whether maternal anxiety relates differently to infants’ attention bias than father’s anxiety. Aktar and colleagues (2018) found that more negative emotions (anxiety) in fathers (and not in mothers) were related to more interest in fearful faces in highly temperamentally negative infants. Based on these findings, it can be expected that attention bias in infants’ towards fearful faces increases more when they have an anxious father, than when they have an anxious mother. This study includes both parents, so that a more complete understanding of the link between an infant’s attention bias and parental anxiety can be developed.
Present study
From the previous paragraphs, it follows that the interaction between anxious parents and their infant is a potentially important mechanism for the transmission of fear from the parent to the infant. There is evidence that maternal anxiety leads to an increased attention bias towards threats among infants (Morales et al., 2017). This study aims to find out whether these findings can be replicated and extends it by including fathers in the study. Furthermore, this study investigates whether there is a difference between mothers and fathers in how parental anxiety relates to an infant’s attention bias towards fearful faces. The possible relationship is investigated through an explorative study that uses eye-tracking. In this study the following research question will be investigated: ‘What is the relationship between infants’ attention bias towards fearful faces and mothers’ and fathers’ anxiety?’. To answer this question, the following hypotheses are tested:
(1) Infants show an attention bias towards images that show fearful faces compared to images with happy faces.
(2) There is a positive relationship between parental anxiety and the attention bias of the infant towards images that show fearful faces (versus happy faces).
(3) The relationship between parental anxiety and the attention bias of the infant towards images that show fearful faces (versus happy faces) is stronger when the gender of the parent is male.
7 Method
Participants
This study included 220 infants between 5 and 19 months old (106 boys and 114 girls) and 229 caregivers (74 fathers and 155 mothers). Only primary caregivers who speak English or Dutch fluently were included. Participants were recruited through invitation letters that were sent by the municipality of Amsterdam. The project has been approved by the Ethics
Committee Psychology of Leiden University.
The original sample of the study consisted of 251 families. For 14 infants and their parents, the fixation data was not available due to child fussiness, tracking problems, software failure or experimenter errors. For these reasons, these 14 infants and their parents dropped out of the study. In addition, another 8 parents and 17 infants were excluded because of missing data (these participants had not completed the eye-tracking experiment and/or the questionnaires) or dropout.
Materials
Parental Anxiety of the caregivers was assessed with the Depression, Anxiety and Stress
Scales (DASS-42). The DASS-42 is a self-reported questionnaire (42 items), used to assess the negative emotional state of depression, anxiety and stress. Each of the scales contains 14 items (depression, anxiety and stress). In this study, only the scores on the anxiety scale were used. Each item is scaled on a Likert-scale that runs from zero (‘not at all’) to three
(‘severity’), so that a total score between 0 and 42 results. The total scores from 0-7 are interpreted as normal anxiety symptoms, 8-9 as mild anxiety, 10-14 as moderate anxiety, 15-19 as severe anxiety and 20 or more as extreme anxiety (Lovibond, 15-1995). So, a higher score on the scale indicates a higher anxiety severity. The total score on the anxiety scale was used to measure parental anxiety in this analysis, and it was used as a continuous variable.
The reliability and validity of the Netherlands DASS varies from sufficient to good. The test-retest for anxiety scale is 0.89 (Beurs, Dyck, Marquenie, Lange & Blonk, 2001). In the current study, the subscale anxiety was found reliable with a Cronbach’s alpha of 0.73.
Eye tracking
Attention bias of infants towards emotional faces was assessed with eye-tracking during the
lab visit. The attention bias was measured by comparing the dwell time of the infant towards fearful faces to the dwell time towards happy faces in milliseconds (ms). The total dwell time of the infant towards fearful faces was subtracted from the total dwell time towards happy
8
faces. The score on the attention bias scale is a continuous variable and the score shows the difference between the total dwell times.
The experimenters showed the infant dynamic videos of 2 male and 2 females faces: happy, fearful, angry, sad and neutral. The current study only focused on fearful and happy faces. The faces that were used in the experiment were extracted from the NimStim database (Tottenham et al., 2009). There were 40 trials. Each trial started with 500 ms attention-getter followed by 1000 ms of blank screen. Subsequently, a 1500 ms presentation of blurred face was shown, and then the video ended with dynamic stimuli. In this part, a neutral expression was randomly followed by the four emotional stimuli.
Time taking care of the infant was assessed with the following question: ‘On average,
how many days a week do you take care of your child alone (without your partner)?’The score on this scale is a continuous variable and could vary between zero and seven days. The score shows how many days the parent (mother or father) takes care of the infant alone. Procedure
Before the experiment, the caregivers filled in the DASS questionnaire and gave their consent to participate in the experiment. Then, the parents and the infant visited the lab. The
experimenter gave the caregivers an information letter about the experiment in advance. Before the measurement of the attention bias, the infant was placed in a car seat. This car seat was located approximately 60 cm away from the computer screen. The parent of the infant sat next to the infant (to the right of the infant). The experimenter calibrated the eye-tracker (Tobii T120), which measured the dwell times and the infant’s attention. After the attention bias task was completed, the parent received a debriefing and a payment.
Statistical analysis
All statistical analyses were conducted using IBM SPSS statistics 24.0. A significance level of 0.05 was used. Prior to the main analyses, the assumptions of linear regression (normality, equal distribution of variances and linearity) were checked for the main variables: 'Attention bias infants' and 'Parental anxiety symptoms'. Furthermore, outliers were identified and the influence of the outliers was checked.
The first hypothesis has been tested with a paired sample test. The paired sample t-test checked whether there was a significant difference between the dwell time of infants towards happy and fearful faces. If there is no significant difference between the two faces, then there is no difference in dwell time, the null hypothesis would be supported. If the results showed a significant difference between the dwell time of the infant towards happy and
9
fearful faces, then there is an attention bias in infants towards fearful faces. That would imply that H0 can be rejected and H1 is supported.
To answer the second hypothesis, a hierarchical linear regression analysis was used. This regression tested whether a relationship between parental anxiety (X) and attention bias among infants towards fearful faces (Y) exists. If no significant correlation was found between X and Y, the H0 hypothesis could not be rejected. If the results show a significant correlation between parental anxiety and infants’ attention bias towards fearful faces,
however, H0 could be rejected and H1 would be supported. This would indicate that there is a relationship between parental anxiety and attention bias among infants towards threat-related faces. Also, the variable ‘time taking care of the child’ was added as control variable. It was expected that the control variable would partly explain the relationship between parental anxiety and infants' attention bias towards fearful faces.
To check the third hypothesis, a PROCESS analysis has been run. This analysis was used to test whether the relationship between infants’ attention bias towards threat-related faces (Y) and parental anxiety (X) is influenced by the gender of the parent. A significant interaction-effect would indicate that the relation between parental anxiety and infants’ attention bias towards threat-related faces depends on the gender of the parent. If an infant’s attention bias towards threat faces is significantly stronger when they have anxious fathers than when they have anxious mothers, this would support H1. If the interaction variable has no significant effect on the relationship between variables X and Y (no interaction-effect is found), H0 cannot be rejected. Also, if an infant’s attention bias towards threat faces is significantly stronger when they have anxious mothers (than anxious fathers), then also H1 cannot be supported.
10 Results
Preliminary analyses
The first and second table gives an overview of the characteristics for the parents and infants of the sample. Table 1 Infants’ characteristics N % M (SD) Gender Boys 108 49.19% Girls 112 50.91% Age 6 months old 73 33.19% 6.08 (0.52) 12 months old 78 35.45% 12.08 (0.60) 18 months old 69 31.36% 17.88 (0.64)
Note. M; mean, SD; standard deviation.
Table 2
Parents’ characteristics
Mother (N = 155) Father (N = 74)
Age parents in years, M (SD) 34.29 (4.11) 35.55 (4.85)
Hours of work per week (payed), M (SD)
24.3 (13.34) 33.1 (12.83)
Days in child care/week, M (SD) 2.78 (1.89) 1.49 (1.40)
Educational level(%)
Never worked 0 0
Predominantly manual labor without vocational education 0.6 1.6 Predominantly manual labor requiring vocational education 1.3 1.6 Predominantly principal work requiring vocational education 2.6 3.1 Predominantly main work at LBO or MBO level and not managerial 4.5 1.6
Self-employed with a maximum of 4 employees 11.0 14.1
Self-employed with more than 4 employees 2.6 4.7
Salaried at LBO or MBO level and managerial 3.9 1.6
Salaried at HBO level and not managerial 25.2 25.0
Salaried at HBO and managerial 9 7.8
Salaried employment requiring scientific education 39.4 39.1
Professional level(%)
Primary school/education 0 3.1
Lower vocational education (LBO) 0 1.6
General secondary education(MAVO) 1.3 0
Higher general secondary education(HAVO) 1.9 6.3
Pre-university education(VWO) 1.3 3.1
Secondary vocational education (MBO) 8.4 4.7
Higher professional education (HBO) 23.2 23.4
Scientific education(WO, university) 63.2 56.3
Others 0.6 1.6
11 <500 7.7 2.7 500-1000 9.0 4.1 1000-2000 17.4 6.8 2000-3000 23.2 20.3 3000-4000 20 18.9 4000-5000 7.1 14.9 >5000 7.1 14.9 I don’t know 0.6 0
I don’t want to say 7.7 4.1
Note. M, Mean; SD, standard deviation. N = 229.
The assumptions for linear regression were checked. The Kolmogorov Smirnov test and the histogram showed that the attention bias of the infants was normally distributed (p = 0.200). However, the Kolmogorov and the histogram showed a non-normal distribution for the variable parental anxiety (p = <0.001), with a skewed distribution to the right (skewness = 2.783, SE = 0.176). Due to the large sample size, the sample is robust against the violation of normality. The assumptions for linearity and homoscedasticity seemed to be violated,
therefore a log-transformation was applied. Finally, multicollinearity was checked using the VIF (variance inflation factor). The VIF value was 1.014, which means that there is no multicollinearity (between parental anxiety symptoms and time spent by the parent taking care of the child) and the assumption for no multicollinearity was met. For these reasons, the regression analyses could be carried out using the log-transformed parental anxiety.
Outliers
The main variables were checked for outliers. The scores were converted to z-scores (standardized) and scores that deviated three standard deviations or more from the mean |3| were considered outliers. For the variables infants’ attention bias and parental anxiety, outliers were found. For infants’ attention bias, one outlier was found, with a z-score of -3.145. That infant had only completed 12 trials (for the eye-tracking). Based on previous research, it was decided that participants who had completed < 30% of the trials were not valid enough for the study (Amso, Haas and Markant, 2014) and for that reason this infant was removed from the sample.
In addition, 6 outliers were found for parental anxiety symptoms (z = |3.8-5.8|). The respondents with outlier scores, contained scores between 9 and 13 points. According to Lovibond (1995), these scores would correspond to mild and moderate anxiety, respectively. It is likely that these scores are appropriate for parents belonging to a non-clinical community, which is the focus of this study (instead of clinical). For this reason, it was decided not to remove the outliers in the study.
12
Main analyses
To investigate whether infants have an attention bias towards images that show the facial expression of fear compared to images that show a happy facial expression, a paired sample t-test (one-sided) was conducted. The analysis shows that there is a significant difference between the dwell time from infants to fearful faces (M = 3273.24, SD = 1102.48) and happy faces (M = 3149.67, SD = 1124.34), t (410) = 3.067, p = 0.002). This means that the mean dwell time of an infant towards fearful faces differs significantly from the mean dwell time towards happy faces. Thus, there seems to be an attention bias towards fearful faces (versus happy faces) in infants.
Secondly, a hierarchical linear regression was performed to investigate if there is a positive relationship between parental anxiety and the attention bias of the infant towards fearful versus happy faces. Model 1 (Table 3) shows that there is no significant relationship between parental anxiety symptoms and infants' attention bias (β = -0.06, t = -0.812, p = 0.418). In the second model, the control variable ‘days taking care of the child’ was added, which causes an increase in the explained variance of the model. However, this increase is not significant (F (2, 187) = 1.264, p = 0.285), which means that the variable ‘days taking care of the child’ does not significantly improve the model and is not a predictor for infants’ attention bias.
Table 3
Hierarchical regression analysis for the relation between parental anxiety symptoms and infants’ attention bias.
Attention bias infants’
Β p R2 ΔR2
F
Model 1 0.056 0.003 0.581
Parental anxiety symptoms -0.056 0.447
Model 2 0.116 0.013 1.264
Parental anxiety symptoms -0.067 0.357
Days taking care of the infant -0.102 0.165
Note. β = Beta, standardized regression coefficient; P = Probability, significance 0.05; R2 = explained variance; ΔR2 = change in R2, F = f-distribution. N = 189.
In order to investigate whether the relationship between parental anxiety and the attention bias of the infant towards fearful faces is stronger when the gender of the parent is male, the PROCESS analysis of Hayes was carried out. Table 4 shows the results of the PROCESS analysis. The PROCESS analysis shows that the gender of the parent is not a
13
significant moderator for the relationship between parental anxiety and infants’ attention bias (β = -0.001, t = -0.007, p = 0.994). This means that the gender of the parent does not influence the relationship between parental anxiety symptoms and infants' attention bias and as such that the relationship is not stronger for fathers (gender male) or mothers (gender female). Table 4
PROCESS-analysis for moderation-effect gender of the parent.
Attention bias infants
Β p t R2 ΔR2
Model 1 0.092 0.009
Parental anxiety symptoms -0.064 0.387 -0.867
Gender 0.072 0.330 0.977
Gender*Parental anxiety symptoms -0.001 0.994 -0.007
Note. β = Beta, standardized regression coefficient; P = Probability, significance 0.05; t = test statistic; R2 =
explained variance; ΔR2 = change in R2,. N = 190.
Discussion
The current study aims to obtain more insight into the development of attention bias in infants and to further investigate the possible relationship between infants’ attention bias and parental anxiety. Results suggest that infants have an attention bias towards fearful faces (compared to happy faces). However, infants’ attention bias towards fearful faces was not significantly related to parental anxiety. Moreover, the gender of the parent did not influence the strength of the relationship between infants’ attention bias and parental anxiety.
In this study, we found evidence that infants have an attention bias towards fearful faces (compared to happy faces). This is in accordance with earlier findings, which stated that infants between 7 and 14 months have an attention bias towards fearful facial expressions (De haan et al., 2004; Peltola et al., 2009). In addition, Nelson and Dolgin (1985) found that infants from the age of 7 months looked longer towards fearful faces compared to happy faces. Our study replicates previous findings showing the attention bias of infants’ towards fearful faces (versus happy faces).
Although infants have an attention bias towards fearful faces, we did not find that the attention bias of the infants towards fearful faces is positively related to parental anxiety. In the study of Morales and colleagues (2017), they did find a positive relation between maternal
14
anxiety and infants’ attention bias towards threat. However, in their study, they used angry faces as threat-related stimuli (instead of fearful faces). Possibly, angry faces are perceived as more threatening. The reason is that angry faces constitute a more direct form of a threat (the expresser itself represents the threat), whereas a fearful face indicates that the expresser observes an external threat in the environment (Ewbank et al., 2009). Because angry faces are possibly perceived as greater threats (compared to fearful faces), angry faces might induce a greater attention bias. Consequently, Morales and colleagues (2017) may have found a positive relationship.
In contrast to Morales and colleagues (2017), Creswell and colleagues (2011) did find a negative relation between maternal social anxiety disorder (SAD) and infants’ attention to low- versus high-intensity negative facial expressions. They compared the difference in total looking time at fearful and angry facial expressions among infants of mothers with and without SAD. Infants with mothers who have SAD looked longer at low-intensity fearful faces; infants of non-anxious mothers showed the opposite result. Thus, infants of mothers with SAD have less attention for highly fearful faces. This reduced attention may indicate avoidance of the infant of fearful faces. A possible explanation for this avoidance is that repeated exposure to parental fearful/anxious expressions results in more avoidant responses away from the fearful face (Murray et al., 2009). A reason that Creswell and colleagues (2011) found an effect in their study, whereas this study does not find an effect, could be that they compared mothers who are diagnosed with a social anxiety disorder (according to the DSM-IV) to non-clinical mothers. Most likely, in the current study there were few or even no parents with a clinical anxiety disorder, given that none of the parents achieved an (extremely) severe score on the anxiety questionnaire. The difference between the severity of anxiety in the two studies could explain the difference. After all, parents with a clinical anxiety disorder most likely show more frequent anxiety expressions/behaviour towards their infants than parents with non-clinical anxiety, which may increase the effect of these anxiety symptoms on the infant. Future studies should investigate the differences in anxious expressions and
behaviour between clinical and sub/non-clinical anxious parents and the impact on their infant’s attention bias towards threat-related faces.
Similar to the findings in the current study, Aktar and colleagues (2016) did not find a relation in their eye-tracking study between paternal of maternal anxiety and increased attention of the infants towards fearful objects (faces and objects). In addition, other studies focused on infants with different ages than the subjects of the current study. For example, the study of Creswell and colleagues (2011) focused on infants who are 2.5 month old and
15
Morales and colleagues (2017) included infants with a wide range of ages from 4 to 24 months old. It is possible that the age of the infants influences the results, because emotion processing and attention bias develop strongly with age (Field & Lester, 2010). Therefore, the effect of an anxious parent on the attention bias of a 4 month old infant may differ from the effect on the attention bias of an infant that is 24 months old. Possibly, the differences in findings can also be explained by the different methodologies of the studies. For example, the other studies (Creswell et al., 2011; Morales et al., 2017) used angry faces as threat stimuli, while the current study and Aktar and colleagues (2016) did not use angry faces (but used fearful faces instead). In further research it might be interesting to investigate the effect of the different threat stimuli. When investigating the attention bias of infants towards threat stimuli, it might be useful to make a distinction between angry faces and fearful faces.
No differences were found between the mothers and fathers. Although Aktar and colleagues (2018) found that more paternal anxiety was related to more interest in fearful faces in infants (with negative temperament), we did not find the same results. Similar to Geary (2010) we did find though that fathers spend less time taking care of the infant. The results show that fathers take care of the infant (on their own) almost 1.5 days less than the mother. However, the gender of the parent nor the time that the parent spends taking care of the infant (on his or her own) appeared to be related to the attention bias in infants towards fearful faces. This contradicts our expectations. We expected that fathers would become more insecure in their parent-infant interaction because they spend less time with the infant, and that this insecurity would result in more fearful expressions and overinvolvement towards their infants. In turn, this would lead to an increased attention bias in the infant. A first explanation for the fact that our hypothesis was not confirmed could be that the fact that fathers spend less time taking care of their infant does not necessarily result in insecurity and overinvolvement, and does therefore also not lead to an increased attention bias. Another explanation could be that the fathers were under-represented in the study, so that the
differences between fathers and mothers could be measured less well. Despite the fact that the current study found no evidence that fathers play a greater role in the attention bias of infants’ than mothers, it might be useful to further investigate the role of the father. In the study of Möller, Majdanzic and Bögels (2015), it seems that overinvolvement of the father is an
important predictor for attention bias in infants towards fearful faces. It may be useful to carry out observations to measure the overinvolvement of fathers and it is important that a sufficient amount of fathers and mothers participates in the study.
16
taken into account. First, anxiety symptoms of the parents were only assessed via one questionnaire. Despite the fact that a self-reported questionnaire is a good instrument to measure anxiety, the answers can be influenced by the parents’ views. An alternative instrument to measure parental anxiety is a diagnostic interview by a trained psychologist, making the participant’s own perspective less influential. In follow-up research, it is
recommended to add another measuring instrument, for example observations. During these observations, naturalistic emotional expressions of the parents can be observed.
Second, in the current study, the fathers (32.3%) were underrepresented compared to the mothers (67.7%). As a result, it may have been more difficult to measure the influence of the gender of the parents, which may have distorted the results. Due to the under-representation of fathers, the external validity has been impaired. As a result, the findings are less
representative for fathers and the extent to which the result can be generalized to fathers in the population (in general) is reduced. Third, the majority of the participating parents in this study are highly educated. Moreover, 57% of the mothers and 69% of the fathers has an individual income above EUR 2000 per month. In the study of Melchior and colleagues (2012) they made a distinction between the total income of the parents, in high (> 1500 euro) and low income (< 1500 euro). Thus, in this study, the majority of our participants belong to the people with high incomes. However, maternal psychopathology and low socioeconomic circumstances of the family could affect the infant negatively. This could lead to more stress for the mother, which results in less attention and a more negative attitude towards the child. The child might therefore develop emotional difficulties early in life (Melchior et al., 2012). If more parents with a lower socio-economic status had been included in the current study, this might have yielded different results. It could be that anxious parents with a lower
socio-economic status are less capable of 'hiding' their anxiety for their infants (for example because of stress). This, in turn, might cause the infant to develop an attention bias toward threats. If more parents with a lower socio-economic status had been included in this study, we would potentially have found a relationship between parental anxiety and the attention bias. To verify this presumption, future studies should include more parents with a lower socio-economic status.
Despite the limitations, the current study has been able to collect a large sample. This results in great statistical power, whereby the probability that a significance test rightly rejects the null hypothesis and adopts the alternative hypothesis is higher. In addition, it is a benefit of the current study that both fathers and mothers are studied, so that any differences between their roles in the transmission of fear could be revealed. Finally, in the current study, dynamic
17
stimuli (emotional faces) were used and both female and male facial expressions were shown. The use of dynamic facial expressions has a stronger influence on the emotional processing of infants than static facial expressions (Grossmann, 2010).
In conclusion, we did not find a relationship between parental anxiety and attention bias among infants. This holds for both mothers and fathers. This finding does not support the notion that attention bias in infants is a mechanism that could explain the transmission of anxiety from parents to their children. It is possible that the severity of the parental anxiety influences the relationship and therefore it is recommended to include parents with a clinical and sub or non-clinical anxiety disorder in further research.In addition, little is known about the role of the father in the transmission of fear towards the infant. In order to better
understand how fathers with an anxiety disorder interact with their infants, it is valuable to make use of observations in further research. Further research on the relationship between parental anxiety and infants’ attention bias towards fearful faces can increase our
understanding in the transmission of anxiety from parent to child, which can help prevention and intervention efforts.
18 References
Aktar, E., Colonnesi, C., de Vente, W., Majdandžić, M., & Bögels, S.M. (2016). How do parents’ depression and anxiety, and infants’ negative temperament relate to parent–infant face-to-face interactions? Development and Psychopathology, 29(3), 697–710.
doi:10.1017/s0954579416000390
Aktar, E., Mandell, D.J., de Vente, W., Majdandžić, Oort, F.J., Renswoude, D.R., Raijmakers, M.E.J., & Bögels, S.M. (2018). Parental negative emotions are related to behavioral and pupillary correlates of infants’ attention to facial expressions of emotion. Infant behavior and development, 53, 101-111. doi: 10.1016/j.infbeh.2018.07.004
Amso, D., Haas, S., & Markant, J. (2014). An eye tracking investigation of developmental change in bottom-up attention orienting to faces in cluttered natural scenes. PLOS one, 9(1). Beck, A.T., & Clark, D.A. (1988). Anxiety and depression: An information processing perspective. Anxiety Research, 1(1), 23–36. doi: 10.1080/10615808808248218
Bittner, A., Egger, H.L., Erkanli, A., Jane Costello, E., Foley, D.L., & Angold, A. (2007). What do childhood anxiety disorders predict? Journal of Child Psychology and Psychiatry, 48(12), 1174–1183. doi:10.1111/j.1469-7610.2007.01812.x
Creswell, C., Cooper, P., Giannakakis, A., Schofield, E., Woolgar, M., & Murray, L. (2011). Emotion processing in infancy: Specificity in risk for social anxiety and associations with two-year outcomes. Journal of Experimental Psychopathology, 2(4), 490-508. doi: 10.5127/jep.018911
De Beurs, E., van Dyck, R., Marquenie, L.A., Lange, A., & Blonk, R.W.B. (2001). De DASS: een vragenlijst voor het meten van depressie, angst en stress. Gedragstherapie, 34, 35-53. de Graaf, R., ten Have, M.M., & van Dorsselaer, S. (2010). De psychische gezondheid van de Nederlandse bevolking. NEMESIS-2: Opzet en eerste resultaten. Utrecht: Trimbos-instituut. de Haan, M., Belsky, H., Reid, V., Volein, A., & Johnson, M.H. (2004). Maternal personality and infants’ neural and visual responsivity to facial expressions of emotion. Journal of child psychology and psychiatry, 45(7), 1209-1218. doi: 10.1111/j.1469-7610.2004.00320.x
de Haan, M., & Nelson, C.A. (1999). Brain activity differentiates face and object processing in 6-month-old infants. Developmental Psychology, 35(4), 1113–1121. doi: 10.1037/0012-1649.35.4.1113
Eley, T.C. (2001). Contributions of behavioral genetics research: Quantifying genetic, shared environmental and nonshared environmental influences. The developmental psychopathology of anxiety, 45-59. doi: 10.1093/med:psych/9780195123630.003.0003
Ewbank, M., Lawrence, A.D., Passamonti, L., Keane, J., Peers, P.V., Calder, A.J. (2009). Anxiety Predicts a Differential Neural Response to Attended and Unattended Facial Signals of Anger and Fear. NeuroImage, 44(3), 1144-1151. doi: 10.1016/j.neuroimage.2008.09.056 Feinman, S. (1982). Social referencing in infancy. Merrill-Palmer Quarterly, 28(4), 445–470.
19
Feinman, S., & Lewis, M. (1983). Social referencing at then months: a second-order effect on infants’ responses to strangers. Child development, 54(4), 878-87. doi:
10.1111/j.1467-8624.1983.tb00509.x
Field, A.P., & Lester, K.J. (2010). Is there room for ‘Development’ in developmental models of information processing biases to threat in children and adolescents? Clinical child family psychology review, 13, 315-332. doi: 10.1007/s10567-010-0078-8
Fisak, B., & Grills-Taquechel, A.E. (2007). Parental Modeling, 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
Geary, D.C. (2010). Male, Female: The evolution of human sex differences (2nd ed.). Washington: American Psychological Association.
Gerfull, F.C., & Rapee, R.M. (2002). Mother knows best: effects of maternal modelling on the acquisition of fear and avoidance behaviour in toddlers. Behaviour, Research and Therapy, 40(3), 279-287. doi: 10.1016/s0005-7967(01)00013-4
Ginsburg, G.S., & Schlossberg, M.C. (2002). Family-based treatment of childhood anxiety disorders. International Review of Psychiatry, 14(2), 143-154. doi:
10.1080/09540260220132662
Grossmann, T. (2010). The development of emotion perception in face and voice during infancy. Restorative Neurology and Neuroscience, 28, 219–236. doi:10.3233/RNN-2010-0499
Hornik, R., Risenhoover, N., & Gunnar, M. (1987). The effects of Maternal Positive Neutral, and Negative Affective Communications on Infant Responses to New Toys. Child
development, 58(4), 937-944
Kessler, R.C., Berglund, P., Demler, O., Jin, R., Merikangas, K.R., & Walters, E.E. (2005). Lifetime Prevalence and Age-of-Onset Distributions of DSM-IV Disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 62(6), 593.
doi:10.1001/archpsyc.62.6.593
Lipsitt, L.P. (1998). Learning and emotion in Infants. Pediatrics, 102, 1262-1267.
Lovibond, S.H., & Lovibond, P.F. (1995). Manual for the Depression Anxiety Stress Scales (2nd. Ed.). Sydney: Psychology Foundation
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. doi: 10.1016/j.cpr.2006.09.002
Melchior, M., Chastang, J-F., Lauzon, B., Galéra, C., Saurel-Cubizolles., & M-J., Larroque, B. (2012). Maternal depression, socioeconomic position, and temperament in early childhood: The EDEN mother-child cohort. Journal of Affective Disorders, 137, 165-169. doi:
10.1016/j.jad.2011.09.018
Melfsen, S., & Florin, I. (2002). Do Socially Anxious Children Show Deficits in Classifying Facial Expressions of Emotions? Journal of Nonverbal Behavior, 26, 109-126.
20
Mogg, K., & Bradley, B.P. (1998). A cognitive-motivational analysis of anxiety. Behaviour Research and Therapy, 36(9), 809–848. doi:10.1016/s0005-7967(98)00063-1
Möller, E.L., Majdandžić, M., & Bögels, S.M. (2015). Parental anxiety, parenting behavior, and infant anxiety: Differential associations for fathers and mothers. Journal of Child and Family Studies, 24(9), 2626–2637. doi: 10.1007/s10826-014-0065-7
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, 88–117
Montagner, R., Mogg, K., Bradley, B.P., Pine, D.S., Czykiel, M.S., Miguel, E.C., Rohde, L.A., Manfro, G.G., & Salum, G.A. (2015). Attentional bias to threat in children at-risk for emotional disorders: role of gender and type of maternal emotional disorder. European child & adolescent psychiatry, 25, 735-742. doi: 10.1007/s00787-015-0792-3
Morales, S., Brown, K.M., Taber-Thomas, B.C., LoBue, V., Buss, K.A. & Pérez-Edgar, K.E. (2017). Maternal anxiety predicts attentional bias towards threat in infancy. Emotion, 7(5), 874-883.
Morales, S., Fu, X., & Pérez-Edgar, K.E. (2016). A developmental neuroscience perspective on affect-biased attention. Developmental Cognitive Neuroscience, 21, 26-41. doi:
10.1016/j.dcn.2016.08.001
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:
10.1017/s00332917-9005157
Nelson, C.A., & Dolgin, K.C. (1985). The generalized discrimination of facial expressions by seven-month-old infants. Child development, 56(1), 58-61.
Peltola, M.J., Leppänen, J.M., Maki, S., & Hietanen, J.K.(2009). Emergence of enhanced attention to fearful faces between 5 and 7 months of age. Social Cognitive and Affective Neuroscience, 4, 134–142.
Pine, D.S. (1997). Childhood anxiety disorders. Current Opinion in Pediatrics, 9(4), 329– 338. doi:10.1097/00008480-199708000-00006
Pérez-Edgar, K., Morales, S., LoBue, V., Taber-Thomas, B.C., Allen, E.K., Brown K.M., & Buss, K.A. (2017). The impact of negative affect on attention patterns to threat across the first two years of life. Development psychology, 53(12), 2219-2232. doi: 10.1037/dev0000408. Stein, M.B., & Stein, D.J. (2008). Social anxiety disorder. The Lancet, 371(9618), 1115– 1125. doi:10.1016/s0140-6736(08)60488-2
Taylor, G., Slade, P., & Herbert, J.S. (2014). Infant face interest is associated with voice information and maternal psychological health. Infant Behavior and Development, 37(4), 597–605. doi:10.1016/j.infbeh.2014.08.002
Telman, L.G.E., Van Steensel, F.J.A., Maric, M., & Bögels, S.M. (2018). What are the odds of anxiety disorders running in families? A family study of anxiety disorders in mothers, fathers, and siblings of children with anxiety disorders. Europe Child Adolescents Psychiatry, 27, 615-624. doi: 10.1007/s00787-017-1076-x
21
Thomas, K.M., Drevets, W.C. Whalen, P.J., Eccards, C.H., Dahl, R.E., Ryan, N.D. (2001). Amygdala response to facial expressions in children and adults. Biological Psychiatry, 49(4), 309-316.
Warner, V., Mufson, L., & Weissman, M.M. (1995). Offspring at high and low risk for depression and anxiety: mechanisms of psychiatric disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 3(6), 786–797.
