Attachment and depressive symptoms in middle childhood

Abstract

The current study investigated whether biological sensitivity to emotional information moderates the link between attachment anxiety/avoidance and depressive symptoms. Sixty children (9–12 years old) completed questionnaires on attachment and depressive symptoms. Skin conductance level (SCL) was measured across three conditions: an emotionally neutral baseline condition, a negative mood induction condition, and a positive mood induction condition. SCL variability (SCLV) was calculated as the intraindividual variation across these conditions expressing the extent to which children are biologically sensitive to positive and negative emotional information. Results showed that SCLV moderated the association between depressive symptoms and attachment anxiety. Attachment anxiety was only linked with depressive symptoms when children showed more SCLV, suggesting that attachment anxiety is only a risk factor for children who are biologically sensitive to respond to emotional information. SCLV did not moderate the association between depressive symptoms and attachment avoidance. Instead, a significant correlation was found between attachment avoidance and SCLV, which replicated previous research and might be caused by more avoidantly attached children’s unsuccessful attempts to suppress emotional reactions.

Keywords

Attachment depressive symptoms middle childhood moderation skin conductance

The quality of parent–child interactions is important to understand middle childhood depressive symptoms (e.g., McLeod, Weisz, & Wood, 2007). Given the long-term detrimental effects of middle childhood depressive symptoms (e.g., Gillham, Reivich, Jaycox, & Seligman, 1995; Harrington, Fudge, Rutter, Pickles, & Hill, 1991; Nolen-Hoeksema, Girgus, & Seligman, 1992), understanding the underlying mechanisms is crucial for the development of effective prevention and intervention. Next to evidence that the observable quality of parenting behaviors and parental care is important (e.g., Mezulis, Hyde, & Abramson, 2006), other evidence increasingly suggests that an essential explanatory mechanism refers to how children perceive these interactions. More specifically, if less supportive parenting decreases children’s confidence in maternal support, fostering insecure attachment, they become more at risk to develop depressive symptoms (e.g., Bosmans, Braet, Beyers, Van Vlierberghe, & Van Leeuwen, 2011; Brumariu & Kerns, 2010; Groh, Roisman, van IJzendoorn, Bakermans-Kranenburg, & Fearon, 2012). However, not all insecurely attached children display depressive symptoms (Brumariu & Kerns, 2010). This finding raises the question which factors determine why some insecurely attached children show more depressive symptoms than other insecurely attached children.

Recent research suggests that one such factor might relate to individual differences in the extent to which children biologically respond to positive and negative information in the environment (Belsky & Pluess, 2009). One possible indicator of these individual differences is the level of children’s skin conductance (SC; Boyce & Ellis, 2005; Gilissen, Bakermans-Kranenburg, van IJzendoorn, & van der Veer, 2008). SC refers to electrodermal reactivity caused by the sweat glands, which are innervated by the sympathetic component of the autonomic nervous system (Fowles, Kochanska, & Murray, 2000). Variation in SC level (SCL) in response to different stimuli is a direct indicator of the extent to which individuals biologically respond to relevant stimuli. When children show more SCL variation, they experience more distress in response to negative stimuli, but more positive affect in response to positive stimuli (Erath, El-Sheikh, Hinnant, & Cummings, 2011). Instead, when they show little SCL variation, they are minimally affected by emotional stimuli. In line with the finding that experienced distress is an important risk factor for developing depressive symptoms (Vasey, Harbaugh, Mikolich, Firestone, & Bijttebier, 2013), enhanced SCL variability (SCLV) determines whether the environment has an impact on the development of depressive symptoms (El Sheikh, Keller, & Erath, 2007). In the latter study, high SCL-reactive children developed more symptoms in response to parent conflict compared to low-reactive children. Instead, if conflict was low, they developed the least symptoms.

Building on these findings, the current study aimed to test the hypothesis that the association between insecure attachment and depressive symptoms is moderated by SCLV. Because depression in insecurely attached children results from inadequate regulation of distress (Brenning, Soenens, Braet, & Bosmans, 2012; Brumariu & Kerns, 2010), it seems plausible that this effect is most pronounced in children who are biologically more sensitive to experience distress in response to negative stimuli. More specifically, distressed securely attached children are able to seek parental support (Ainsworth, Blehar, Waters, & Wall, 1978), which reduces distress and associated depressive symptoms (Brumariu & Kerns, 2010; Mezulis et al., 2006). Instead, infant observation research and middle childhood self-report research both demonstrated that insecurely attached children employ less successful, alternative, affect-regulation strategies (Brenning et al., 2012; Cassidy, 1994). Two strategies can be distinguished, depending on whether insecure attachment mainly leads to anxiety about parental abandonment and rejection (anxious attachment) or to a preference to independently solve distress to avoid having to rely on parental support (avoidant attachment) (Brenning et al., 2012; Cassidy, 1994). On the one hand, children expressing to be more anxiously attached are vulnerable for depressive symptoms because they tend to hyperactivate distressing emotions during negative experiences. On the other hand, children expressing to be more avoidantly attached are also vulnerable for depressive symptoms, because they try to deactivate emotions to reduce the impact of distress during negative experiences. As a result, they are less likely to solve the source of distress. The prolonged exposure to distress can lead to depressive symptoms (Brenning et al., 2012). As inadequate regulation of distress should be most problematic in children who are biologically sensitive to experience distress (e.g., see John & Gross, 2004), the current study investigated whether SCLV moderates the link between attachment anxiety and avoidance and depressive symptoms.

In the current study, SCLV was measured across three conditions: while children were not exposed to emotional information, while they were exposed to negative information, and while they were exposed to positive information. Calculating intraindividual variation across these conditions expresses the extent to which children are biologically responsive to positive and negative emotional information most adequately. Although both parents are attachment figures (Bowlby, 1969), the current study focused on attachment toward mother because attachment to mother is a stronger predictor for depressive symptoms in middle childhood (Bosmans et al., 2011). Using these variables, we predict that SCLV will moderate the link between insecure attachment and depressive symptoms such that the link will only be significant if children are high on SCLV.

Method

Participants

A sample of 60 children (30 boys and 30 girls) aged 9–12 (M = 10.45, SD = 1.06) from the general population participated in the study. All children lived with their biological mother. Forty-five children (75.0%) lived with both parents (married or cohabiting partners), and 15 children (25.0%) resided with their mother after divorce. Seven children lived with a single mother, and five lived in a reconstituted family. Regarding mothers’ and fathers’ level of education, 18.4% and 31.6% had a high school degree, 53.3% and 30.0% had a post-high school technical training or a bachelor degree, and 28.3% and 38.3% had a university master degree, respectively. All parents were Caucasian.

Procedure

Children together with their mother were invited to participate in a study regarding the role of maternal proximity in the alleviation of negative emotions, by means of a flyer distributed in the fourth, fifth, and sixth grade of elementary schools in Flanders. All invited children and their mothers gave their written accord for taking part in the study.

During a lab visit, data were collected as part of a larger study on attachment and emotional responses.¹ Children completed the tests, while mother was present in another room. Before the procedure started, the device (Dreamer-technology, Medatec^®) for SC measurement was put on the child and two SC sensors were applied to the middle and index finger. Since the current study was part of a larger study, the measurement device included several sensors and electrodes designed to assess other physiological measures that were not further used in the current study. Children were first given time to adjust to the measurement equipment while filling out questionnaires. Then, to acquire a baseline measure of children’s SCL, children were asked to sit calm and relax while staying alone in the room for 5 min. SCL was measured continuously throughout stimulus presentation.

During this period, children were presented a video fragment of remote beaches with a background of serene instrumental music (baseline). This baseline approach was chosen to avoid those children to feel bored without inducing positive or negative emotions. Next, children participated in an experiment on attachment and emotion responses. The experimental procedure first required the induction of negative emotions. At the end of the experiment, a mood repair paradigm was used. In order to induce mildly negative emotions, children were shown a negative video fragment (1 min 20 s) of a 7 p.m. news broadcast about the floods in Pakistan in 2010 reporting on children’s distress due to the disaster. To achieve mood repair, children watched a positive video fragment (1 min 40 s) of a 7 p.m. news broadcast on the birthday of a baby elephant in a zoo. To check whether the videos indeed induced negative and positive emotions, visual analog scale (VAS) was used. Children were asked to put a cross on three 100 mm lines to indicate how anxious, sad, and happy they felt, before the negative video, after the negative video, and after the positive video. The VAS scores after the negative video and the VAS scores after the positive video were compared with the VAS scores before the negative video. Results confirmed an increase in anxiety and sadness and a decrease in happiness after the negative video while showed a decrease in anxiety and sadness and an increase in happiness after the positive video (repeated-measures analyses, F-values range from 3.638 to 55.80; ps from .06 to <.0001). SC was measured during baseline, during the negative mood inducing video, and during the positive mood repairing video. This procedure was approved by the ethical board of local university.

Measures

Physiological measures

Children’s SC was constantly registered using Dream^® technology (Medatec Dream conductance amplifying and measuring module, Medical Data Technology, Brussels, Belgium), at a sampling rate of 200 Hz, with an exosomatic alternating current method with constant current (1, 5, or 25 µA, periodically adapted depending on the individual’s measured base impedance to optimize resolution), using two reusable 8 mm snap style silver–silver chloride electrodes (type SE-35, J&J Engineering) embedded into a hook and loop fastener band (Fowles et al., 1981). We applied SC electrodes to the volar (palmar) side of the medial phalanges of the middle and index finger of the nondominant hand using a salt- and chloride-free, hypoallergenic electrolyte paste (Spectra^® 360 electrolyte gel, Parker Laboratories Inc, Fairfield, New Jersey, USA.).

To calculate SCLV, first, average SC was calculated for three 1-min epochs during the procedure: during the baseline relaxation phase, during the negative mood induction video fragment, and during the positive mood induction fragment. To standardize the duration of SC measurement, this calculation was made for the last minute of each phase (baseline, positive, and negative phase). Next, SCLV was calculated as the within-subject standard deviations of the average SC across the three phases. This index was further used as our SCLV measure, with higher scores indicating a stronger tendency to vary in SC in response to emotional information (Doberenz, Roth, Wollburg, Maslowski, & Kim, 2011).

Attachment anxiety and attachment avoidance

Children filled out a child version of the Experiences in Close Relationships Scale–Revised (Brenning, Soenens, Braet, & Bosmans, 2011). The questionnaire consists of 36 items to evaluate two dimensions of insecure attachment, specifically Attachment Anxiety (e.g., “I worry about being abandoned”) and Attachment Avoidance (e.g., “I prefer not to show how I feel deep down”). All items focused on the attachment relationship with mother. The items were rated on a 7-point Likert-type scale ranging from “not at all” to “very much”. Higher scores indicate more Attachment Anxiety and Attachment Avoidance. Previous research shows high Cronbach α’s for both scales (e.g., Brenning et al., 2011). In our study, Cronbach’s α was .75 for anxious attachment and .79 for avoidant attachment.

Depressive symptoms

A short form of the Center for Epidemiologic Studies Depression scale (CES-D; Andresen, Malmgren, Carter, & Patrick, 1994; Radloff, 1977) was used to assess depressive symptoms. The instrument consists of nine negatively worded items and three positively worded items (reversely coded). The items ask how often symptoms, such as loneliness (e.g., People were unfriendly), feelings of fearfulness (e.g., I had the feeling people hate me), and restless sleep (e.g., I slept badly), were experienced during the past week. Responses were rated on a 4-point scale (0 = rarely; 3 = most of the time). The summative score of the 12 items, with a range of 0–36, was used, with higher scores suggesting more depressive symptoms. The reliability and validity of the CES-D have been established in both children and adolescents, showing good internal consistency and construct validity and distinguishing children with clinically significant depressive symptoms from healthy children (Poulin, Hand, & Boudreau, 2005). Moreover, a recent study confirmed that the CES-D is linked with middle childhood attachment anxiety and attachment avoidance and that the CES-D taps in the component of childhood and adolescent depression that is related to children’s reduced inclination to seek maternal support (Dujardin et al., in press). Using the cutoff scores provided by Poulin, Hand, and Boudreau (2005), 10% of the children had mild-to-severe depressive symptoms, which is in line with the prevalence of depression in middle childhood (Costello, Mustillo, Erkanli, Keeler, & Angold, 2003). Poulin et al. (2005) found a Cronbach α of .85 in a sample of 12,990 youngsters. In the current sample, Cronbach’s α was .69.

Plan of the analyses

First, effects of the video clips on SCL were investigated with Pairwise t-tests. Second, correlational analyses investigated interrelationships between the study variables and age effects on these variables. One-way analyses of variance (ANOVAs) were used to test the effects of gender and maternal civil status (whether or not she lived together with the biological father). Finally, the interaction between the two attachment dimensions and SCLV was tested using hierarchical multiple regression analysis (HMRA) approach with centered predictors (Aiken & West, 1991).

Results

Data were analyzed using the statistical package IBM SPSS statistics 22 for Windows (Armonk, New York, USA: IBM Corp). Table 1 shows the mean and standard deviation for SCL measured at three different time points: during baseline, during the negative mood induction video, and during the positive mood induction video. Pairwise t-tests demonstrated significant differences both between baseline and negative video, t(55) = −3.44, p < .001, and between baseline and positive video, t(55) = −3.20, p < .01. Table 2 displays the correlations and the descriptive information of the studied variables. The means and standard deviations are comparable to what is found in other studies in this age-group (e.g., Brenning, Soenens, Braet, & Bosmans, 2014; Dujardin et al., in press). Both anxious and avoidant attachment correlated with depressive symptoms. SCLV correlated significantly with avoidant attachment (even after controlling for anxious attachment, r = .39, p < .01) and depressive symptoms. No effects of age (see Table 1), gender (one-way ANOVAs revealed Fs ranging from .393 to 2.00; ps ranging from .533 to .164), or maternal civil status (Fs ranging from .327 to .756; ps ranging from .570 to .388) on the variables under study were found. Therefore, these variables were not included in further analyses.

Table 1.

Descriptive statistics of SCL at different time-point measurements.

	Minimum	Maximum	Mean	Standard deviation
SCL_baseline	.88	36.77	14.39	7.75
SCL_{negative video}	.90	37.35	15.02	7.79
SCL_{positive video}	.90	34.24	14.95	7.79

Note. SCL = skin conductance level. SCL is expressed in microsiemens (µS).

Table 2.

Correlations and descriptive statistics of the study variables.

	1.	2.	3.	4.	5.
1. Depressive symptoms	–
2. Attachment anxiety	.29*	–
3. Attachment avoidance	.26*	.27*	–
4. SCLV	.27*	.12	.41**	–
5. Age	−.22	−.13	−.03	.01	–
Mean	5.93	1.88	2.26	.18	10.45
Standard deviation	4.29	.65	.77	.22	1.06

Note. SCLV = skin conductance level variability.

*p < .05; **p < .01.

Next, HMRAs were carried out to investigate whether the relationship between insecure attachment (attachment anxiety and attachment avoidance) and depressive symptoms is moderated by SCLV. First, the interaction between attachment anxiety and SCLV was studied. As hypothesized, the interaction between attachment anxiety and SCLV significantly predicted depressive symptoms (see Table 3) with a medium effect size. The initial (correlational, see Table 2) effect of attachment anxiety was no longer significant after the interaction effect was taken into account (see Table 3). The interaction effect remained significant even after controlling for the three SCL scores (the SCLs during baseline, positive video, and negative video separately), suggesting that the effect was uniquely related to the variability of SCL across the three time points. Figure 1 was drawn according to the recommendations of Aiken and West (1991) to help interpreting the meaning of the interaction effect. The Johnson–Neyman procedure was used to calculate regions of significance to evaluate at which value of SCLV the slope of the association between attachment anxiety and depressive symptoms becomes significant (Hayes & Matthes, 2009). Results showed that the slope is significant when children have an SCLV score that is .05 SDs above the mean or more. Sixty-six percent of the children had a score below that cutoff. This suggests that attachment anxiety is only linked with depressive symptoms in 34% of the children.

Table 3.

Interactions between attachment anxiety or attachment avoidance and SCLV on depressive symptoms.

	R ²Δ	FΔ	df	β	f ²		R ²Δ	FΔ	df	β	f ²
Step 1	.14	4.50*	2, 56			Step 1	.10	3.03^†	2, 56
Attachment anxiety				.26*	.08	Attachment avoidance				.18	.03
SCLV				.23^†	.06	SCLV				.19	.03
Step 2	.14	10.67**	1, 55			Step 2	.00	0.02	1, 55
Attachment anxiety				.21	.06	Attachment avoidance				.18	.03
SCLV				.31**	.13	SCLV				.22	.02
Interaction				.39**	.19	Interaction				−.03	.00

Note. SCLV = skin conductance level variability; df = degree of freedom.

^† p < .10; *p < .05; **p < .01.

Figure 1.

Interaction between attachment anxiety and SCLV. SCLV: skin conductance level variability.

Second, HMRA was used to investigate the interaction between attachment avoidance and SCLV in predicting depressive symptoms. This analysis revealed that the main effects of attachment avoidance and SCLV on depressive symptoms became nonsignificant when put together in the model. Also, the interaction between both predictors was not significant (see Table 3).

Discussion

The current study aimed to investigate whether SCLV moderates the association between insecure attachment and depressive symptoms in middle childhood. We hypothesized that this link would only be significant in children with more SCLV. Our data confirmed this hypothesis for children who scored high on attachment anxiety, but not for children scoring high on attachment avoidance.

Preliminary analyses showed that in the current sample both attachment anxiety and avoidance were related to depressive symptoms. Although previous studies in middle childhood suggest that attachment anxiety is more consistently linked to depressive symptoms, several studies do provide some evidence for a less strong, but still significant link between attachment avoidance and emotional problems (Brenning et al., 2012; Brumariu & Kerns, 2010). One possible explanation for this pattern of effects could be found in recent research on attachment and children’s capacity to effortfully control their emotional responses (Heylen et al., in press). Results in the latter study suggested that attachment avoidance and the related deactivating emotion regulation strategies successfully protect children against the development of emotional problems until the point when their effortful control capacity is depleted. Only when that happens, these children become more vulnerable to develop depressive problems. Hence, effortful control could be a moderator that explains why in some studies links between attachment avoidance and depressive symptoms are found and in others not. More research is needed to test this hypothesis.

Also, SCLV is related to attachment avoidance, but not to attachment anxiety. At first sight, this pattern of associations could be considered surprising, because cortisol research has found opposite results. Those studies led to the suggestion that cortisol responses to emotional stimuli are less related to attachment avoidance because of associated deactivating emotion regulation strategies, whereas cortisol responses are more related to attachment anxiety, because of associated hyperactivation strategies (e.g., Luijk et al., 2010). This interpretation is supported by other studies showing that more suppression leads to lower cortisol levels (e.g., de Veld, Riksen-Walraven, & de Weerth, 2012). However, in the current study, we measured SC. This is the most autonomic psychophysiological variable (Boucsein, 2012) and, therefore, other links with attachment avoidance-related suppression are expected. More specifically, research suggests that suppression of positive and negative emotions (typical for avoidant attachment) leads to increases in SC (e.g., Gross & Levenson, 1997). In line with this expectation, research on attachment and SC responses to stress demonstrated unique links with attachment avoidance and not with attachment anxiety (Diamond, Hicks, & Otter-Henderson, 2006; Dozier & Kobak, 1992). Dozier and Kobak (1992) found these effects during confrontational attachment-related interview questions. They suggested that this effect reflected the distress avoidantly attached individuals experience when they are forced to expose themselves to memories they would prefer to suppress. The current results suggest that their results might have reflected a more general effect. More specifically, avoidantly attached individuals’ SCLs seem more likely to change in response to any emotionally relevant stimulus (positive, negative, and not necessarily attachment related).

The significant interaction between attachment anxiety and SCLV on depressive symptoms suggests that children who report being anxiously attached only report depressive symptoms if they are biologically more sensitive to experience emotions elicited by positive and negative events. The fact that the results were not affected by controlling for SCL at each time point, further supports the claim that the interaction effect is driven by variability in SCL response to different stimuli. Although the cross-sectional data do not allow inferring causal pathways, the current finding is in line with the expectation that biological sensitivity determines which anxiously attached children are vulnerable to develop depressive symptoms. This finding might help explain why studies on the link between attachment and depressive symptoms on average only find medium effect sizes (this study, Brumariu & Kerns, 2010; Burnette, David, Green, Worthington, & Bradfield, 2009; Simpson & Rholes, 2002). The fact that this main effect is not larger seems to suggest that a substantial number of more anxiously attached children do not develop depressive symptoms. The current results suggest that this might be because the latter children’s lower sensitivity to distress makes it less likely that their seemingly less adequate insecure attachment-related coping strategy puts them at risk to develop depressive symptoms. Such an interpretation would be in line with the idea that insecure attachment strategies are adaptive reactions to unresponsive environments and therefore are not a synonym to psychopathology (Hinde, 1988). However, because our study did not focus on other forms of psychopathology, we cannot rule out that attachment anxiety in children low on SCLV leads to other types of psychopathology. Contrary to the findings for attachment anxiety, no significant interaction between attachment avoidance and SCLV on depressive symptoms was found. This might be the result of the significant correlation between attachment avoidance and SCLV. If children who report being avoidantly attached in general show more SCLV, it is harder to find moderating effects of SCLV on the link between avoidant attachment and depressive symptoms.

These findings need to be interpreted taking into account several limitations, which should be addressed in future research. Next to the small sample size and the cross-sectional nature of the study, the first notable limitation of our study was the fixed order of the videos presented to the children. Indeed, if our research had not been part of a larger study, it would have been a better solution to randomize that fixed order. However, despite this shortcoming, the fixed order cannot fully explain the interaction effect we found. Moreover, we have to acknowledge that most studies use two time points instead of three, but recent studies have demonstrated that the more time points were used for SC, the more accurate the results were (Doberenz et al., 2011).

Second, another important limitation is the use of the self-report questionnaires for attachment, which only measure the easily and directly accessible perceptions of the respondent and possibly ignore information that is difficult to access strategically (De Haas, Bakermans-Kranenburg, & van Ijzendoorn, 1994). However, for middle childhood, a growing body of evidence suggests significant overlap between scores obtained by interviews and questionnaires (e.g., Kerns, Brumariu, & Seibert, 2011; Psouni & Apetroaia, 2013) and relationship (mother)-specific self-reported anxious and avoidant attachment links with children’s seeking of maternal proximity (Dujardin et al., in press). The latter studies support the use of relationship-specific self-reported attachment questionnaires at this age as an appropriate approach to measure attachment-related individual differences.

Third, we only examined the effect on a negative outcome (depressive symptoms). Consequently, we could only test whether SCLV is a risk factor in the development of children. However, recent research suggests that SCLV could be a differential susceptibility factor (Gilissen et al., 2008). According to the differential susceptibility model, higher variability levels should not only lead to more distress under adverse environmental conditions but also allow children to blossom in favorable environmental conditions (Belsky, 2005, Bakermans-Kranenburg, van IJzendoorn, Pijlman, Mesman, & Juffer, 2008). It would be worthwhile to investigate whether children who are low on attachment anxiety and who have high SCLV would also have significantly better scores on positive outcomes such as quality of life or positive emotions.

Finally, it would be worthwhile to design a study in which other measures of biological sensitivity can be used to investigate the current study’s hypotheses. Unfortunately, only SC offers a pure measure of autonomic sympathetic activity (Boucsein, 2012) and the interpretation of other indices, like heart rate variability, is affected by the nature of the tasks during which they are measured (Zisner & Beauchaine, in press). This made it impossible to use such indices in the current study. Nevertheless, it would be interesting to investigate whether effects replicate with other psychophysiological measures.

In conclusion, the current study sheds some light on the moderating effect of children’s biological responsiveness to emotional stimuli on the association between children’s insecure attachment and depressive symptoms in middle childhood. We found evidence for such an effect on the association between anxious attachment and depressive symptoms, suggesting that anxious attachment is only linked with depressive symptoms when children show more SCLV. From a clinical standpoint, this information may be used to develop more specific risk profiles for childhood depressive symptoms, thus developing specific interventions.

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by grants G.0934.12 and G.0774.15 of the Research Foundation Flanders (FWO) and grants OT/12/043 and CREA/12/004 from the Research Fund KULeuven, Belgium.

Note

References

Aiken

L. S.

West

S. G.

(1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage.

Ainsworth

M. D. S.

Blehar

M. C.

Waters

Wall

(1978). Patterns of attachment. Hillsdale, NJ: Erlbaum.

Andresen

E. M.

Malmgren

J. A.

Carter

W. B.

Patrick

D. L.

(1994). Screening for depression in well older adults: Evaluation of a short form of the CES-D. American Journal of Preventive Medicine, 10, 77–84.

Bakermans-Kranenburg

M. J.

Van Ijzendoorn

M. H.

Pijlman

F. T.

Mesman

Juffer

(2008). Experimental evidence for differential susceptibility: Dopamine D4 receptor polymorphism (DRD4 VNTR) moderates intervention effects on toddlers’ externalizing behavior in a randomized controlled trial. Developmental Psychology, 44, 293–300. doi:10.1037/0012-1649.44.1.293

Belsky

(2005). Differential susceptibility to rearing influences: An evolutionary hypothesis and some evidence. In Ellis

Bjorklund

(Eds.), Origins of the social mind: Evolutionary psychology and child development (pp. 139–163). New York, NY: Guildford.

Belsky

Pluess

(2009). Beyond diathesis-stress: Differential susceptibility to environmental influences. Psychological Bulletin, 135, 885–908. doi:10.1037/a0017376

Bosmans

Braet

Beyers

Van Vlierberghe

Van Leeuwen

(2011). Power assertive discipline and internalizing problems in adolescents: The role of attachment. Parenting: Science and Practice, 11, 34–55. doi:10.1080/15295192.2011.539507

Boucsein

(2012). Electrodermal activity. New York, NY: Plenum.

Bowlby

(1969). Attachment and loss: Vol. 1. Attachment. New York, NY: Basic Books.

10.

Boyce

W. T.

Ellis

B. J.

(2005). Biological sensitivity to context: I. An evolutionary–developmental theory of the origins and functions of stress reactivity. Development and Psychopathology, 17, 271–301. doi:10.1017/S0954579405050145

11.

Brenning

Soenens

Braet

Bosmans

(2011). An adaptation of the Experiences in Close Relationships Scale-Revised for use with children and adolescents. Journal of Social and Personal Relationships, 28, 1048–1072. doi:10.1177/0265407511402418

12.

Brenning

Soenens

Braet

Bosmans

(2012). Attachment and depressive symptoms in middle childhood and early adolescence: Testing the validity of the emotion regulation model of attachment. Personal Relationships, 19, 445–464. doi:10.1111/j.1475-6811.2011.01372.x

13.

Brenning

Soenens

Braet

Bosmans

(2014). Hechtingsangst en -vermijding bij kinderen en adolescenten. Psychopraktijk, 6, 28–31. doi:10.1007/s13170-014-0009-3

14.

Brumariu

L. E.

Kerns

K. A.

(2010). Parent-child attachment and internalizing symptomatology in childhood and adolescence: A review of empirical findings and future directions. Development and Psychopathology, 22, 177–203. doi:10.1017/s0954579409990344

15.

Burnette

J. L.

Davis

Green

J. D.

Worthington

E. L.

Jr. Bradfield

(2009). Insecure attachment and depressive symptoms: The mediating role of rumination, empathy and forgiveness. Personality and Individual Differences, 46, 276–280. doi:10.1016/j.paid.2008.10.016

16.

Cassidy

(1994). Emotion regulation: Influences of attachment relationships. Monographs of the Society for Research in Child Development, 59, 228–283. doi:10.1111/j.1540-5834.1994.tb01287.x

17.

Costello

Mustillo

Erkanli

Keeler

Angold

(2003). Prevalence and development of psychiatric disorders in childhood and adolescence. Archives of General Psychiatry, 60, 837–844. doi:10.1001/archpsyc.60.8.837

18.

De Haas

Bakermans-Kranenburg

M. J.

van IJzendoorn

M. H.

(1994). The Adult Attachment Interview and questionnaires for attachment style, temperament, and memories of parental behavior. Journal of Genetic Psychology, 155, 471–487. doi:10.1080/00221325.1994.9914795

19.

de Veld

D. M. J.

Riksen-Walraven

J. M.

de Weerth

(2012). The relation between emotion regulation strategies and physiological stress responses in middle childhood. Psychoneuroendocrinology, 37, 1309–1319.

20.

Diamond

L. M.

Hicks

A. M.

Otter-Henderson

K. A.

(2006). Physiological evidence for repressive coping among avoidantly attached adults. Journal of Social and Personal Relationships, 23, 205–229. doi:10.1177/0265407506062470

21.

Doberenz

Roth

W. T.

Wollburg

Maslowski

N. I.

Kim

(2011). Methodological considerations in ambulatory skin conductance monitoring. International Journal of Psychophysiology, 80, 87–95. doi:10.1016/j.ijpsycho.2011.02.002

22.

Dozier

Kobak

R. R.

(1992). Psychophysiology in attachment interviews: Converging evidence for deactivating strategies. Child Development, 63, 1473–1480. doi:10.2307/1131569

23.

Dujardin

Santens

Braet

De Raedt

Vos

Maes

Bosmans

(in press). Middle childhood support-seeking behavior during stress: Links with self-reported attachment and future depressive symptoms. Child Development.

24.

El-Sheikh

Keller

P. S.

Erath

S. A.

(2007). Marital conflict and risk for child maladjustment over time: Skin conductance level reactivity as a vulnerability factor. Journal of Abnormal Child Psychology, 35, 715–727.

25.

Erath

S. A.

El-Sheikh

Hinnant

J. B.

Cummings

E. M.

(2011). Skin conductance level reactivity moderates the association between harsh parenting and growth in child externalizing behavior. Developmental Psychology, 47, 693–706. doi:10.1037/a0021909

26.

Fowles

D. C.

Christie

M. J.

Edelberg

Grings

W. W.

Lykken

D. T.

Venables

P. H.

(1981). Publication recommendations for electrodermal measurements. Psychophysiology, 18, 232–239. doi:10.1111/j.1469-8986.1981.tb03024.x

27.

Fowles

D. C.

Kochanska

Murray

(2000). Electrodermal activity and temperament in preschool children. Psychophysiology, 37, 777–787. doi:10.1111/1469-8986.3760777

28.

Gilissen

Bakermans-Kranenburg

M. J.

van IJzendoorn

M. H.

van der Veer

(2008). Parent-child relationship, temperament, and physiological reactions to fear-inducing film clips: Further evidence for differential susceptibility. Journal of Experimental Child Psychology, 99, 182–195. doi:10.1016/j.jecp.2007.06.004

29.

Gillham

J. E.

Reivich

K. J.

Jaycox

L. H.

Seligman

M. E. P.

(1995). Prevention of depressive symptoms in school children: Two-year follow up. Psychological Science, 6, 343–351. doi:10.1111/j.1467-9280.1995.tb00524.x

30.

Groh

A. M.

Roisman

G. I.

van IJzendoorn

M. H.

Bakermans-Kranenburg

M. J.

Fearon

R. M. P.

(2012). The significance of insecure and disorganized attachment for children’s internalizing symptoms: A meta-analytic study. Child Development, 83, 591–610. doi:10.1111/j.1467-8624.2011.01711.x

31.

Gross

J. J.

Levenson

R. W.

(1997). Hiding feelings: The acute effects of inhibiting positive and negative emotions. Journal of Abnormal Psychology, 106, 95–103.

32.

Harrington

Fudge

Rutter

Pickles

Hill

(1991). Adult outcomes of childhood and adolescent depression: II. Links with antisocial disorders. Journal of the American Academy of Child & Adolescent Psychiatry, 30, 434–439. doi:10.1097/00004583-199105000-00013

33.

Hayes

A. F.

Matthes

(2009). Computational procedures for probing interactions in OLS and logistic regression: SPSS and SAS implementations. Behavior Research Methods, 41, 924–936. doi:10.3758/brm.41.3.924

34.

Heylen

Bosmans

Vasey

Dujardin

Vandevivere

Braet

De Raedt

(in press). Attachment and effortful control: Relationships with maladjustment in middle childhood. The Journal of Early Adolescence.

35.

Hinde

R. A.

(1988). Continuities and discontinuities: Conceptual issues and methodological considerations. In Rutter

(Ed.), Studies of psychosocial risk (pp. 367–383). Cambridge, England: Cambridge University Press.

36.

John

O. P.

Gross

J. J.

(2004). Healthy and unhealthy emotion regulation: Personality processes, individual differences, and lifespan development. Journal of Personality, 72, 1301–1334. doi:10.1111/j.1467-6494.2004.00298.x

37.

Kerns

K. A.

Brumariu

L. E.

Seibert

A. C.

(2011). Multi-method assessment of mother-child attachment: Links to parenting and child depressive symptoms in middle childhood. Attachment and Human Development, 13, 315–333. doi:10.1080/14616734.2011.584398

38.

Luijk

M. P.

Saridjan

Tharner

van Ijzendoorn

M. H.

Bakermans-Kranenburg

M. J.

Jaddoe

V. W.

… Tiemeier

(2010). Attachment, depression, and cortisol: Deviant patterns in insecure-resistant and disorganized infants. Developmental Psychobiology, 52, 441–452. doi:10.1002/dev.20446

39.

McLeod

B. D.

Weisz

B. B.

Wood

J. J.

(2007). Examining the association between parenting and childhood depression: A meta-analysis. Clinical Psychology Review, 27, 986–1003. doi:10.1016/j.cpr.2007.03.001

40.

Mezulis

A. H.

Hyde

J. S.

Abramson

L. Y.

(2006). The developmental origins of cognitive vulnerability to depression: Temperament, parenting, and negative life events. Developmental Psychology, 42, 1012–1025. doi:10.1037/0012-1649.42.6.1012.

41.

Nolen-Hoeksema

Girgus

J. S.

Seligman

M. E. P.

(1992). Predictors and consequences of childhood depressive symptoms: A 5-year longitudinal study. Journal of Abnormal Psychology, 101, 405–422. doi:10.1037/0021-843X.101.3.405

42.

Poulin

Hand

Boudreau

(2005). Validity of a 12-item version of the CES-D used in the National Longitudinal Study of Children and Youth. Chronic Diseases in Canada, 26, 65–72.

43.

Psouni

Apetroaia

(2013). Measuring scripted attachment-related knowledge in middle childhood: the Secure Base Script Test. Attachment & Human Development, 16, 22–41. doi:10.1080/14616734.2013.804329

44.

Radloff

L. S.

(1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385–401. doi:10.1177/014662167700100306

45.

Simpson

J. A.

Rholes

W. S.

(2002). Fearful-avoidance, disorganization, and multiple working models: Some directions for future theory and research. Attachment and Human Development, 4, 223–229. doi:10.1080/14616730210154207

46.

Vasey

M. W.

Harbaugh

C. N.

Mikolich

Firestone

Bijttebier

(2013). Positive affectivity and attentional control moderate the link between negative affectivity and depressed mood. Personality and Individual Differences, 54, 802–807. doi:10.1016/j.paid.2012.12.01

47.

Zisner

A. R.

Beauchaine

T. P.

(in press). Psychophysiological methods and developmental psychopathology. In Cicchetti

(Ed.), Developmental psychopathology (3rd. ed.). Hoboken, NJ: Wiley.