Journal of Clinical Child & Adolescent Psychology

ISSN: 1537-4416 (Print) 1537-4424 (Online) Journal homepage: http://www.tandfonline.com/loi/hcap20

Longitudinal Associations Between Reactive and Regulatory Temperament Traits and Depressive Symptoms in Middle Childhood Yuliya Kotelnikova, Sarah V. M. Mackrell, Patricia L. Jordan & Elizabeth P. Hayden To cite this article: Yuliya Kotelnikova, Sarah V. M. Mackrell, Patricia L. Jordan & Elizabeth P. Hayden (2015) Longitudinal Associations Between Reactive and Regulatory Temperament Traits and Depressive Symptoms in Middle Childhood, Journal of Clinical Child & Adolescent Psychology, 44:5, 775-786, DOI: 10.1080/15374416.2014.893517 To link to this article: http://dx.doi.org/10.1080/15374416.2014.893517

Published online: 17 Apr 2014.

Submit your article to this journal

Article views: 126

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=hcap20 Download by: [University of Lethbridge]

Date: 05 November 2015, At: 13:51

Journal of Clinical Child & Adolescent Psychology, 44(5), 775–786, 2015 Copyright # Taylor & Francis Group, LLC ISSN: 1537-4416 print=1537-4424 online DOI: 10.1080/15374416.2014.893517

Longitudinal Associations Between Reactive and Regulatory Temperament Traits and Depressive Symptoms in Middle Childhood Yuliya Kotelnikova, Sarah V. M. Mackrell, Patricia L. Jordan, and Elizabeth P. Hayden Downloaded by [University of Lethbridge] at 13:51 05 November 2015

Department of Psychology, Western University

Although a large literature has examined the role of temperament in adult and adolescent depression, few studies have investigated interactions between reactive and regulatory temperament traits in shaping depressive symptoms in children over time. Child temperament measures (laboratory observations and maternal reports) and depressive symptoms were collected from 205 seven-year-olds (46% boys), who were followed up 1 (N ¼ 181) and 2 (N ¼ 171) years later. Child participants were Caucasian (87.80%), Asian (1.95%), or other ethnicity (7.80%); 2.45% of the sample was missing ethnicity data. Multilevel modeling was used to investigate within- and between-person variance in intercepts and slopes of child depressive symptoms. A steeper increase in depressive symptoms was found for children lower in laboratory-assessed effortful control (EC). Lower mother-reported surgency and higher mother-reported NE predicted increases in child depressive symptoms in the context of lower mother-reported EC. Our findings implicate EC as having main and moderating effects related to depressive symptoms in middle childhood. We emphasize the importance of developing prevention programs that enhance EC-like abilities.

Depression ranks among the most common and debilitating psychological disorders (Kessler et al., 2005; Kessler, McGonagle, Zhao, & Nelson, 1994), making the development and refinement of models of risk a crucial goal. Although many models of depression vulnerability have been developed, representing biological, cognitive, and interpersonal theories (e.g., Joiner & Timmons, 2009; Joormann, 2009; Thase, 2009), interest in the role of temperament is extensive and longstanding. Understanding relations between temperament and depression may provide a number of benefits, including the identification of early-emerging risk for the disorder (Klein, Durbin, & Shankman, 2009), thus potentially enhancing prevention and early intervention efforts. However, surprisingly few studies have examined longitudinal associations between temperament and depressive symptoms in childhood. We address this Correspondence should be addressed to Yuliya Kotelnikova, Department of Psychology, Western University, 361 Windermere Road, Westminster Hall, London, Ontario, N6A3K7, Canada. E-mail: [email protected]

gap, taking a multitrait, multimethod, longitudinal approach to testing associations between temperament and depressive symptoms in middle childhood, an important developmental period that directly precedes the well-established adolescent increase in depression (Lewinsohn, Hops, Roberts, Seeley, & Andrews, 1993). In particular, we expand extant research and theory by testing interactions between reactive and regulatory traits, following Carver and colleagues’ model (Carver, Johnson, & Joormann, 2009). In recent decades, the Big Three model of Clark and Watson (1991; Watson & Clark, 1993) has driven much of the research on adult temperament–psychopathology links. This model consists of three broad superfactors: neuroticism=negative emotionality (NE), extraversion= positive emotionality (PE), and disinhibition versus constraint. Of importance, the Big Three model maps well onto contemporary frameworks for child temperament. More specifically, Rothbart and colleagues conceptualize child temperament in terms of extraversion=surgency, negative affectivity, and effortful

Downloaded by [University of Lethbridge] at 13:51 05 November 2015

776

KOTELNIKOVA, MACKRELL, JORDAN, HAYDEN

control (EC; Rothbart, Ahadi, Hershey, & Fisher, 2001; Rothbart, Sheese, & Posner, 2007). Extraversion= surgency and negative affectivity, analogous to PE and NE from Clark and Watson’s model, are held to reflect reactive aspects of temperament. On the other hand, EC is thought to play a regulatory function, consisting of lower order traits that capture executive attention (e.g., voluntary attentional focusing and shifting) as well as the capacity to inhibit dominant responses or initiate behavior based on social norms (e.g., inhibitory and activation control; Kieras, Tobin, Graziano, & Rothbart, 2005; Rothbart & Bates, 2006); extant work supports the role of inhibitory control facets of EC as especially relevant to disorder risk (Eisenberg et al., 2005; Nigg, 2006). EC (and its inhibitory control facet in particular) is related to lower levels of disinhibition and higher constraint and is thought to play an important moderating role of reactive tendencies (De Pauw & Mervielde, 2010; Nigg, 2006; Rudolph, TroopGordon, & Llewellyn, 2013). There is evidence for the role of higher NE and lower PE in youth depression, primarily from self- and parentreport questionnaire assessments of temperament in community and psychiatric adolescent-aged samples (e.g., Joiner & Lonigan, 2000; Lonigan, Phillips, & Hooe, 2003; Phillips, Lonigan, Driscoll, & Hooe, 2002), although the research on PE is less consistent than that supporting links between NE and psychopathology (Chorpita, Plummer, & Moffitt, 2000; De Bolle & De Fruyt, 2010). These trait-depression associations may emerge through an array of mechanisms (see Klein et al., 2009, for a review of models of how temperament and depression are linked). With respect to EC, although much work has focused on linking EC-like traits in adults and children to externalizing psychopathology (e.g., Clark, 2005; Eisenberg et al., 2009; Kochanska & Knaack, 2003; Lengua, 2006), the relationship between EC and related traits and depression remains unresolved. More specifically, some researchers (e.g., Eisenberg et al., 2009; Lengua, 2006; Muris, 2006; Oldehinkel, Hartman, Ferdinand, Verhulst, & Ormel, 2007) suggest that lower EC is related to elevated internalizing symptoms, whereas others (e.g., Murray & Kochanska, 2002; Oosterlaan, Logan, & Sergeant, 1998; Rydell, Berlin, & Bohlin, 2003) find either no relationship or evidence that higher EC is a risk factor for developing internalizing symptoms. Despite the extensive research on associations between temperament and depression, unanswered questions remain. In particular, there is a surprising paucity of studies addressing the possibility that EC works in conjunction with other traits, such as PE and NE, in relating to emerging depression risk over time. Although such interactions are not typically addressed in accounts of the Big Three models, Carver and

colleagues (2009) recently proposed a conceptually related model describing how EC relates to both internalizing and externalizing symptoms by virtue of its regulatory function on reactive traits. Although these authors refer to reward and punishment sensitivity, rather than PE and NE, it has been noted that these constructs show substantial conceptual overlap (Klein et al., 2009), with PE related to reward sensitivity and NE to punishment sensitivity. Carver and colleagues (2009) and others (e.g., Beauchaine, 2001) have suggested that functioning of reactive systems (i.e., reward and punishment sensitivity; PE and NE) is particularly salient in the context of insufficient top-down regulation (or low EC; Rudolph et al., 2013). In such models, individuals who are temperamentally low in reward sensitivity or heightened punishment sensitivity are likely to develop internalizing psychopathology in the context of a limited capacity to regulate such tendencies due to lower EC. In other words, lower reward sensitivity (and related constructs, such as PE), and higher punishment sensitivity (and related constructs, such as NE), may not exert main effects on psychopathological outcomes if their impact is contingent on individual differences in the capacity to regulate these tendencies. Thus, children with temperamental risk in terms of reactive traits may not experience negative outcomes when they have sufficient regulatory capacities. Similar models have been postulated by others (e.g., Eisenberg et al., 2005; Nigg, 2006) in which regulatory capacities moderate associations between reactive tendencies and negative outcomes during development; the psychophysiological and neural correlates of these traits have also been described (e.g., Beauchaine, 2001; Nigg, 2006). Thus, although some of these models have not focused on NE and PE specifically, considered as a whole, this literature suggests the possibility that EC may interact with NE and PE in predicting the development of internalizing disorders in childhood. As noted, few empirical studies have tested interactions between temperament traits in predicting depression, although a small literature based on questionnaire measures exists. For example, in a cross-sectional sample of 9- to 13-year-olds, Muris, Meesters, and Blijlevens (2007) found that higher self-reported fear (an NE facet) and lower self-reported EC were associated with internalizing symptoms. Oldehinkel et al. (2007) found a similar pattern of results across time in a sample of 11-year-olds followed up at age 13. Rudolph et al. (2013) also provided evidence of lower levels of parent-reported inhibitory control (a facet of EC) predicting higher levels of self-reported depressive symptoms in the context of high self-reported avoidance motivation (a construct related to punishment sensitivity) in a sample of 8-year-old girls. However, Mezulis, Simonson, McCauley, and Stoep (2011) failed to replicate the

Downloaded by [University of Lethbridge] at 13:51 05 November 2015

TEMPERAMENT AND DEPRESSIVE SYMPTOMS

moderating effect of self-reported EC on self-reported NE or PE in predicting depressive symptoms concurrently in a large community sample of 12-year-olds. The lack of consistency in findings on links between child temperament and depressive symptoms may stem from the complexities regarding assessment that arise in research on children. More specifically, most work on child temperament has relied on parent reports, which have a number of advantages, including their affordability and their capacity to capitalize on parents’ extensive exposure to their children’s behavior across diverse settings (Rothbart & Bates, 2006). However, parent reports of child temperament likely capture both objective aspects of child behavior as well as parent factors (Durbin & Wilson, 2012; Hayden, Durbin, Klein, & Olino, 2010), and parent reports of child temperament have typically shown poor convergent validity with teacher reports and observational and laboratory measures (Durbin, Hayden, Klein, & Olino, 2007; Seifer, Sameroff, Barrett, & Krafchuk, 1994). Despite being expensive and time-consuming to administer, laboratory measures of child temperament may have advantages over parent reports, such as the use of standardized stimuli, specific coding procedures that minimize rater bias, and the capacity to elicit child behaviors that may be present at lower rates in naturalistic settings (Durbin et al., 2007). Only a few studies have investigated the relationship between temperament assessed in laboratory settings and markers of child depression risk (e.g., Durbin, Klein, Hayden, Buckley, & Moerk, 2005; Olino, Klein, Dyson, Rose, & Durbin, 2010) and fewer still have looked at the relationship between laboratoryassessed temperament and child depressive symptoms (Dougherty et al., 2011; Dougherty, Klein, Durbin, Hayden, & Olino, 2010). These studies generally support the notion that NE and PE may reflect an important vulnerability for depression, although only one examined child behavior relevant to EC (Dougherty et al., 2011), and this study did not test interactions between traits. Given the gaps in the current literature, the goal of the current investigation was to characterize the role of reactive and regulatory child temperament traits (i.e., NE, PE, and EC) in predicting initial levels and change in depressive symptoms in middle childhood, using a multi-informant, multimethod approach to the assessment of temperament. We expected that higher NE and lower PE would be associated with higher levels of depressive symptoms concurrently and steeper increases in depressive symptoms over time (Dougherty et al., 2010). Given the mixed evidence with regard to the role of EC in internalizing symptoms, we did not make firm predictions about main effects of this trait on depressive symptoms. However, we predicted that lower EC would moderate both PE and NE to predict

777

elevated depressive symptoms concurrently and increases across time (Carver et al., 2009), such that low PE and high NE would be more closely linked to depressive symptoms in the context of low EC. Given its importance for important child outcomes (Eisenberg et al., 2009; Lengua, 2006; Rothbart et al., 2007), we focused on inhibitory control aspects of EC in the present study.

METHOD Participants A community sample of 205 7-year-old children (46% boys, Mage ¼ 7.41 years, SD ¼ .30) and their parents (Mage mothers ¼ 37.48 years, SD ¼ 8.96; Mage fathers ¼ 40.43 years, SD ¼ 11.50) were recruited for a study of children’s emotional and cognitive development through a psychology department database of research volunteers, and advertisements placed in local newspapers and online. Children with major psychological and medical concerns, as determined by trained study personnel during recruitment, were ineligible. Children performed within the normal range (M ¼ 111.92, SD ¼ 12.15) on the Peabody Picture Vocabulary Test, Fourth Edition (PPVT-IV; Dunn & Dunn, 2007). Children were Caucasian (87.80%), Asian (1.95%), or other ethnicity (7.80%); 2.45% of the sample was missing ethnicity data. Approximately half (50.24%) of the families participating reported a family income ranging from $40,000 to $100,000; 26.83% of families reported a family income greater than $100,000, and 15.12% of families reported a family income of less than $40,000; 7.81% of the sample was missing family income data. These sample characteristics are comparable to data pertaining to race and income reported in the 2008 census for London, Ontario (Statistics Canada, 2008), the census closest to when baseline data were collected. Parents of the children in the study were provided with monetary compensation at all three time points of assessment; children were given smalls toys as gifts for participation. All parents provided signed consent for their own and their children’s participation at all time points of assessment; all children provided their verbal assent at the second follow-up (age 9). No participants dropped out at the point of providing consent=assent. This study was approved by the Ethics Review Board of the University of Western Ontario. Laboratory Temperament Assessment Child temperament was assessed using an hour-long video-recorded battery of laboratory tasks based on the Laboratory Temperament Assessment Battery (Goldsmith, Reilly, Lemery, Longley, & Prescott, 1995)

778

KOTELNIKOVA, MACKRELL, JORDAN, HAYDEN

adapted to be appropriate for older children. In support of their validity, temperament indices derived from this battery show significant homotypic continuity with the same traits assessed using the original Laboratory Temperament Assessment Battery (Durbin et al., 2007) and are meaningfully associated with children’s symptoms (Kotelnikova, Olino, Mackrell, Jordan, & Hayden, 2013).

Downloaded by [University of Lethbridge] at 13:51 05 November 2015

Exploring new objects (Fear, PE). The child was left alone to play freely in a room containing several ambiguous or mildly ‘‘scary’’ objects: a cloth tunnel and tent, a remote-controlled spider, a plastic skull covered with a red cloth, a Halloween mask, and a box containing a plastic beating heart and fake spiderwebs. Racing cars (Anger, Sadness, PE). The child was given photographs of a desirable toy (a remotecontrolled race car) and a relatively boring toy (a small plastic doll with unmoving parts) and was told to choose which she or he wanted to play with. Next, the child was told that the requested toy was lost and was given the non-preferred toy. Stranger approach (Fear). The child was left alone in the main experimental area to play with a toy golf set. Following a short delay, a friendly, unfamiliar male research assistant entered the room and spoke to and approached the child. Frustrating puzzle (Anger, Sadness). The child was left alone to complete a puzzle that the experimenter said was easy but contained pieces that did not fit together. After 3 min, the experimenter returned and explained that she had accidentally given the child the wrong pieces. Practical joke (PE). The experimenter showed the child how to use a remote-controlled whoopee cushion, and the child was invited to surprise his or her parent with the toy. Object fear (Fear). The child was left alone in the main experimental area and instructed to look inside a pet carrier, after being told that it contained ‘‘something scary.’’ Children’s fear was assessed both before and during exploration of the carrier, which actually contained a small, stuffed animal obscured by shredded newspaper. Toy parade (PE). The child was given a bell and told that each time they rang it, a research assistant would bring them a new toy, but that they would have to trade in the toy they had for the new toy. Toys were intended to be fun and included Mr. Potato Head, a Fun

Hop, a Gearation Toy, a floor piano and guitar, and LEGOs. Coding Procedures Undergraduate, postbaccalaureate, and graduate student raters blind to other study data coded all videos. As part of the training process, raters coded videos with a trained ‘‘master’’ coder. Trainees then coded sets of 10 to 15 videos independently until they were able to code five videos with a minimum ICC of .80. Ongoing checks were done to maintain interrater reliability (minimum ICC ¼ .80) for all episodes; half of all coders’ affect coding was also coded by the master coder (a senior lab member, who received training from, and was supervised by Dr. Hayden, the senior author of this article), and if the ICC for a given episode was below .80, the episode was discussed and final consensus ratings were made. Each instance of facial, bodily, and vocal positive affect; fear; sadness; and anger exhibited by children in each episode was rated on a 3-point scale as low, moderate, or high. Instances of moderate- and high-intensity behaviors were weighted to account for their greater intensity (e.g., total sample size of moderate intensity smiles  2; total sample size of high intensity vocal sadness  3). After weighting, the total number of low, moderate, and high intensity behaviors was summed separately within each channel (facial, bodily, vocal) across the seven episodes, standardized, and summed across the three channels to derive total scores for positive affect (referred to as PE henceforth; a ¼ .75), sadness, fear, and anger (asadness ¼ .67; afear ¼ .68; aanger ¼ .68). NE was the average of the standardized sadness, fear, and anger variables for each episode. In addition to affective codes, a single rating for each episode was made on a 3-point scale for two behaviors relevant to inhibitory control aspects of EC (Dougherty et al., 2011; Olino et al., 2010), impulsivity and compliance. Impulsivity was based on the child’s tendency to respond and=or act without reflection. Compliance was based on ‘‘rule-breaking’’ behavior and the persistence of the noncompliance. Impulsivity ratings from each task were reverse-scored and aggregated with compliance ratings from each task to obtain an index of inhibitory control aspects of EC (a ¼ .78). These internal consistencies are highly comparable to other laboratory measures of temperament (e.g., Dougherty et al., 2011; Durbin et al., 2005; Durbin et al., 2007). Temperament in Middle Childhood Questionnaire (TMCQ) The TMCQ (Simonds & Rothbart, 2004) is a 157-item parent-report measure of temperament for children ages 7 to 10. Using guidelines from a factor analysis of the 17

Downloaded by [University of Lethbridge] at 13:51 05 November 2015

TEMPERAMENT AND DEPRESSIVE SYMPTOMS

TMCQ scales (Simonds, 2006), we derived indices of higher order traits. The Negative affectivity (NA) factor included the following TMCQ scales: Anger=frustration (a ¼ .81), Discomfort (a ¼ .72), fear (a ¼ .72), Sadness (a ¼ .77), and Soothability, reverse scored (a ¼ .71). The EC factor consisted of Inhibitory control (a ¼ .74), Attentional focusing (a ¼ .90), Low intensity pleasure (a ¼ .61), Perceptual sensitivity (a ¼ .80), and Activation control (a ¼ .80). The TMCQ does not have a scale or factor that specifically assesses children’s expressions of positive affect. At the same time, the factor analysis yielded a factor that showed reasonable conceptual overlap with trait PE, labeled surgency by Simonds (2006), an aggregate of High intensity pleasure (a ¼ .83), Activity (a ¼ .88), Assertion=dominance (a ¼ .71), Impulsivity (a ¼ .88), and reverse-scored Shyness (a ¼ .85) scales. Internal consistencies were moderate to good for Surgency, NA, and EC (Cronbach’s as ¼ .88, .77, and .90, respectively). Given the significant construct overlap with our laboratory measures of NE and EC, the TMCQ NA and EC scales are referred to as TMCQ or mother-reported NE and EC. However, as the TMCQ surgency scale taps different behaviors than our laboratory measure of PE, we refer to it as surgency.1 Assessment of Child Depressive Symptoms Following the laboratory assessment, a home visit took place with each family an average of 40.02 days (SD ¼ 29.65) later. With the help of a research assistant, who read items aloud and recorded the child’s responses, children completed the Depression Self-Rating Scale (DSRS; Birleson, 1981), a 24-item self-report measure of depression in children and youth. We elected to focus on child self-reported depression as research on children’s emotional and cognitive development has established that even children as young as 6 are capable of accurately reporting on their moods, emotions, cognitions, and other symptoms relevant to depression (Durbin, 2010; Luby et al., 2009; McGrath & Repetti, 2002) and as researchers have noted the critical importance of self-reported measures of child depressive symptoms (Klein et al., 2009). The DSRS has been validated with children as young as 6 (e.g., Asarnow & Carlson, 1985). Baseline DSRS scores demonstrated good internal consistency (Cronbach’s a ¼ 0.73), and scores in the current sample (see Table 1) were com1

To address the issue of overlap between the TMCQ surgency and observed EC variables, we reran our models excluding TMCQ impulsivity from the surgency scale. Similarly, because low intensity pleasure and perceptual sensitivity are not traditionally identified as aspects of EC (e.g., Kochanska & Knaack, 2003) we also reran our models excluding these scales from the EC composite. The results obtained proved almost identical to those presented here and may be obtained by request from the first author.

779

parable to other nonclinical samples (e.g., Asarnow & Carlson, 1985; Hayden, Klein, Durbin, & Olino, 2006). At the initial assessment, 24% (n ¼ 50) of children showed clinically significant levels of self-reported depressive symptoms, that is, scores of 17 or higher, which have been previously found to show acceptable sensitivity, specificity, and accuracy with regard to depression diagnoses (Asarnow & Carlson, 1985). Of the initial sample of 205 children, 181 (88.3%) completed the DSRS again during home visits 1 year after the baseline assessment, when children were an average of 8.48 years old (SD ¼ .32). At this point, 19% (n ¼ 34) had DSRS scores of 17 or higher. Comparing participants who participated in the first follow-up to those who did not showed no significant differences in proportion of boys participating, depressive symptoms, or family income (all ps > .37; mean Cohen’s d for comparisons ¼ .12, range ¼ .03–.20). However, children who participated in the first follow-up had higher PPVT scores than those who did not, F(1, 202) ¼ 4.63, p ¼ .03; Cohen’s d ¼ .45). A third assessment occurred at a psychology research laboratory when children were an average of 9.63 years old (SD ¼ .38); 171 children (83% of the original sample) participated, with children again completing the DSRS. Those participants who completed the third assessment compared to those who completed the baseline assessment only did not differ in terms of number of boys participating, PPVT scores, depressive symptoms, or family income (all ps > .19; mean Cohen’s d ¼ .16, range ¼ .06–.26). At this last assessment, 26% (n ¼ 45) of children had DSRS scores of 17 or higher. The DSRS showed acceptable internal consistency at the first (Cronbach’s a ¼ .72) and second follow-up (Cronbach’s a ¼ .86); average scores were low and consistent with nonclinical samples (see Table 1). Data Analytic Plan First, we examined associations between all study variables (Table 1). Next, multilevel modeling (MLM) was conducted using Mplus 7 (Muthe´n & Muthe´n, 1998–2012) to examine the relationship between child temperament and depressive symptoms over time. MLM permits the examination of intraindividual (withinperson) change over time, as well as interindividual (between-person) variability in intraindividual change. Two parameters are estimated to represent aspects of change: the intercept (I), which represents the level of the outcome measure at which the time variable equals zero (in the present case, baseline assessment), and the slope (S), which represents the linear rate at which the outcome measure changes. The age variable was centered around children’s average age at baseline in order for the growth parameters to represent the average

780

KOTELNIKOVA, MACKRELL, JORDAN, HAYDEN TABLE 1 Bivariate Correlations Between Child Depressive Symptoms and Temperament Traits at Baseline

Downloaded by [University of Lethbridge] at 13:51 05 November 2015

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

TMCQSurg TMCQNE TMCQEC LPE LNE LEC DSRST1 DSRST2 DSRST3 Child Sex Income PPVT

1

2

3

4

5

6

7

8

9

10

11

12

M

SD

—

.04 —

.27 .39 —

.18 .02 .05 —

.01 .23 .13y .09 —

.32 .06 .25 .38 .23 —

.09 .09 .26 .02 .05 .04 —

.11 .05 .23 .04 .08 .14y .35

.10 .16 .28 .10 .06 .18 .33

.18 .08 .25 .02 .15 .27 .09 .12 .09 —

.06 .13y .11 .03 .01 .13y .02 .09 .13 .06 —

.08 .05 .05 .15 .04 .01 .04 .20 .03 .09 .18 —

129.21 68.08 166.81 .00 .00 .00 12.95 13.49 14.53 — 3.63 111.92

18.53 19.97 17.73 .82 .46 1.69 5.50 5.00 7.15 — 1.18 12.15

Note. Child sex was coded as 0 for boys and 1 for girls; family income at baseline was coded as follows: 1 ¼ $100,000; TMCQ ¼ Temperament in Middle Childhood Questionnaire; TMCQSurg, NE, EC ¼ mother-reported child surgency, negative emotionality, and effortful control; LPE, LNE, LEC ¼ laboratory-assessed child positive emotionality, negative emotionality, and effortful control; DSRS ¼ The Depression Self-Rating Scale; PPVT ¼ Peabody Picture Vocabulary Test. y p < .10  p < .05  p < .01.

child’s true initial status (intercept at age 7.41) and annual rate of true change (slope) across time. After establishing the significance of unexplained variance in intercepts (p < .001) and slopes (p < .001) following Singer and Willett (2003), the following two models were examined as predictors of child depressive symptoms: (a) laboratory-assessed PE, NE, EC, ECNE, and EC  PE interaction terms, and (b) maternally reported surgency, NE, EC, EC  NE, and EC-surgency interactions terms.2,3 To test interactions, each predictor variable was centered and interaction terms were formed as the product of the two centered predictors (Aiken & West, 1991). For the sake of parsimony and to conserve power, nonsignificant interaction terms were dropped from final models. As relevant, we report the proportional variance reduction (PVR) for significant predictors of intercepts and slopes, a local effect size estimate that can be used in MLM analyses (Peugh, 2010). Significant interaction terms were interpreted by plotting the equations and conducting post hoc analyses of simple slopes (i.e., by recentering the variables involved in creating the interaction terms at þ1 SD and 1 SD above the means for those traits; Aiken & West, 1991). The models were then rerun with these 2 The generally low convergence across the two methods of assessing child temperament made combining laboratory and motherreported traits into a single model untenable. This low convergence is typical of studies incorporating observational and parent-report methods (Durbin & Wilson, 2012; Hayden et al., 2010; Olino, Durbin, Klein, Hayden, & Dyson, 2013). 3 Although here we present only the results involving temperament traits as predictors of child self-reported depressive symptoms, we also tested the same models with child sex as a covariate, which did not change the results and was, hence, dropped for parsimony. Relevant data are available upon request from the authors.

recentered variables to obtain the mean intercepts and slopes for depressive symptoms across time at þ and 1 SD on both the relevant reactive trait and EC.

RESULTS Bivariate Associations Between Temperament and Child Depressive Symptoms Table 1 shows bivariate correlations between temperament measures and child depressive symptoms as well as demographic variables. As expected based on past research, between-method correlations between temperament traits showed significant, albeit modest, associations between mother-reported surgency and laboratory PE, mother-reported and laboratory NE, and motherreported and laboratory EC. Laboratory-Assessed Temperament and Children’s Depressive Symptoms A full model predicting child self-reported depression with laboratory temperament traits (NE, PE, and EC), the EC  NE interaction, and the EC  PE interaction as predictors (Akaike’s information criterion [AIC] ¼ 3446.59) showed that neither the EC  NE nor the EC  PE interaction terms was significant; thus, these interactions were removed from the final model and only main effects of the three traits were tested, which resulted in a lower AIC value, suggesting a better model fit (AIC ¼ 3440.55). None of the laboratory-assessed traits predicted children’s self-reported baseline levels of depressive symptoms. However, laboratory EC was a predictor of slope at the level of a strong trend

TEMPERAMENT AND DEPRESSIVE SYMPTOMS

781

(S ¼ .27, p ¼ .06; PVRslopes ¼ 2.87%). Figure 1 presents the relationship between laboratory EC and child selfreported depression over time. This figure shows that children lower on laboratory EC at baseline showed a steeper increase in self-reported symptoms of depression across time than children higher in laboratory EC. Simple slopes analysis (Aiken & West, 1991) indicated that children lower in laboratory EC at baseline showed a significant increase in their self-reported symptoms of depression over time, whereas children higher in EC did not.

Downloaded by [University of Lethbridge] at 13:51 05 November 2015

Mother-Reported Temperament and Children’s Depressive Symptoms Next, a model including mother-reported temperament traits (NE, surgency, and EC) and their interactions (EC  NE and EC–surgency) as predictors of child self-reported depression was tested (AIC ¼ 3434.35). TMCQ EC was the only significant (negative) predictor of child self-reported symptoms of depression at baseline (I ¼ .08, p < .001; PVRintercepts ¼ 22.91%). With respect to change over time, interactions between TMCQ EC and surgency (S ¼ .01, p < .001; PVRslopes ¼ 15.38%) and EC and NE (S ¼ .01, p ¼ .04; PVRslopes ¼ 2.16%) were significant (Figures 2 and 3). Simple slope analyses indicated that children higher on mother-reported NE and lower on mother-reported EC at baseline showed a significant increase in their self-reported depressive symptoms across time (M slope ¼ .95, p ¼ .03), whereas children higher on mother-reported EC and lower on mother-reported NE at baseline did not (see Figure 2). Children higher in both mother-reported NE and EC at baseline also showed a significant increase on their depression symptoms over

FIGURE 1 Plot of laboratory effortful control at 1 SD as it relates to changes in child self-reported depressive symptoms across time. Note. High LEC ¼ laboratory effortful control at þ1 SD; low LEC ¼ laboratory effortful control at 1 SD; y axis ¼ child self-reported depression measured by Depression Self-Rating Scale (DSRS) scores.  p < .01.

FIGURE 2 Plot of Temperament in Middle Childhood Questionnaire (TMCQ) negative emotionality at  1 SD interacting with TMCQ effortful control at  1 SD in predicting changes in child-selfreported depressive symptoms across time. Note. HighNE=highEC ¼ mother-reported negative emotionality and effortful control plotted at þ1 SD; lowNE=lowEC ¼ mother-reported negative emotionality and effortful control plotted at 1 SD; highNE=lowEC ¼ motherreported negative emotionality plotted at þ1 SD and mother-reported effortful control plotted at 1 SD; lowNE=highEC ¼ mother-reported negative emotionality plotted at 1 SD and mother-reported effortful control plotted at þ1 SD; y axis ¼ child self-reported depression measured by Depression Self-Rating Scale (DSRS) scores. y p < .10.  p < .05.

time (M slope ¼ 1.39, p ¼ .02), whereas children lower on these traits showed a marginally significant increase on these symptoms (M slope ¼ 1.18, p ¼ .07). Children lower in NE and higher in EC, the children at theoretically lowest temperamental risk for depressive symptoms, were the only group who did not show evidence for an increase in depressive symptoms over time. With respect to the EC–surgency interaction (Figure 3), simple

FIGURE 3 Plot of Temperament in Middle Childhood Questionnaire (TMCQ) surgency at  1 SD interacting with TCMQ effortful control at  1 SD in predicting changes in child-self reported depressive symptoms; highSur=highEC ¼ mother-reported surgency and effortful control plotted at þ1 SD; lowSur=lowEC ¼ mother-reported surgency and effortful control plotted at 1 SD; highSur=lowEC ¼ motherreported srugency plotted at þ1 SD and mother-reported effortful control plotted at 1 SD; lowSur=highEC ¼ mother-reported surgency plotted at 1 SD and mother-reported effortful control plotted at þ1 SD; y axis ¼ child self-reported depression measured by Depression Self-Rating Scale (DSRS) scores.  p < .01.

782

KOTELNIKOVA, MACKRELL, JORDAN, HAYDEN

slope analyses indicated that children lower on both mother-reported surgency and EC increased significantly on their self-reported symptoms of depression across time (M slope ¼ 2.1, p < .01). Similarly, children who scored higher on both of those traits also increased significantly on their self-reported depressive symptoms over time (M slope ¼ 1.52, p < .01). Children who were lower on mother-reported surgency and higher on EC maintained stable, lower levels over the course of the study. Similarly, children higher on mother-reported surgency and lower on mother-reported EC at baseline did not show a significant increase in their self-reported depressive symptoms.

Downloaded by [University of Lethbridge] at 13:51 05 November 2015

DISCUSSION Although a large empirical literature has developed over the past few decades relating temperament to depression, very few studies have focused on such links in middle childhood, a potentially critical period for understanding emerging depression risk. Further, the extant literature has not consistently examined potentially important interactions between temperament traits in predicting depression risk (Carver et al., 2009). We addressed this gap by examining main and interactive effects of reactive (i.e., NE and PE=surgency) and regulatory (i.e., EC) traits on child depressive symptoms. To our knowledge, this is the first report of findings linking observational and mother reported measures of temperament in middle childhood to child self-reports of depressive symptoms over time. The use of independent measures of study constructs and an extensive follow-up period constitutes a strict test of links between temperament and children’s emerging depressive symptoms. Both laboratory and maternally reported EC were related to children’s depressive symptoms, with laboratory EC predicting change in child self-reported symptoms across time; simple slope analyses indicated that children who were rated as lower on laboratory EC at baseline showed a significant increase in depressive symptoms across time. Maternal reports of EC indicated that children higher in TMCQ EC showed lower levels of self-reported depressive symptoms at baseline. These findings are consistent with the literature suggesting that lower levels of EC represent a risk marker for depression (Eisenberg et al., 2009; Lengua, 2006; Muris, 2006; Oldehinkel et al., 2007); our findings extend this work by examining EC–depressive symptom associations across a lengthy follow-up period and by using independent methods of assessing EC and depressive symptoms. Ample evidence exists for the capacity of specific programs designed to enhance EC-like abilities (Rueda, Posner, & Rothbart, 2005; van Veen & Carter, 2002).

Although these have been applied toward enhancing children’s ability to manage behaviors relevant to externalizing disorders, our findings suggest that such programs may reduce depression risk as well. Mother-reported EC also interacted with NE and surgency in predicting depressive symptoms. First, consistent with recent theory (Carver et al., 2009), children lower in both mother-reported surgency and EC had increased depressive symptoms over time. Carver et al. (2009) proposed that such a relationship might emerge when children with decreased reward sensitivity (i.e., lower surgency) are less able to upregulate low responsivity to rewards due to lower EC. This surgency–EC interaction was found only for maternally reported temperament, not for laboratory-assessed traits. In contrast to laboratory PE, which was defined solely in terms of positive emotion expression, TMCQ surgency is a broader construct, including facets such as high intensity pleasure and activity. These may map more closely onto reward responsivity as described in Carver and colleagues’ model, thus explaining why an interaction was found between EC and TMCQ surgency and not laboratory EC and PE. Despite research implicating PE in depression risk (Clark, 2005; Lonigan et al., 2003; Phillips et al., 2002), no main effects were found for either laboratoryassessed child PE or mother-reported child surgency predicting children’s depressive symptoms. Some findings suggest that longitudinal associations between PE and depressive symptoms are weaker than those for NE (e.g., Clark, 2005; De Bolle & De Fruyt, 2010; Kendler, Gatz, Gardner, & Pederse, 2006); our findings are consistent with previous studies failing to find a main effect of PE and related traits on depression (Anderson & Hope, 2008; Laurent & Ettelson, 2001). At the same time, previous studies with children (e.g., Chorpita et al., 2000) have found that low PE was related to symptoms of depression and social phobia. Because anxiety may precede depression in youth (Keenan, Feng, Hipwell, & Klostermann, 2009; Pine, Cohen, Gurley, Brook, & Ma, 1998), it is possible that we might have found stronger effects for lower PE had we been predicting socially anxious, rather than depressive, symptoms. Mother-reported EC and NE interacted to predict depressive symptoms; more specifically, children higher on NE and lower on EC at baseline showed a significant increase in depressive symptoms across time. These results are also consistent with Carver and colleagues’ (2009) suggestion that greater punishment sensitivity, an aspect of NE, may lead to developing internalizing symptoms when regulatory skills are inadequate. Furthermore, our findings are broadly consistent with a literature on anxious adults indicating that attention training, a construct related to EC, can reduce symptoms of anxiety (Schmidt, Richey, Buckner, & Timpano,

Downloaded by [University of Lethbridge] at 13:51 05 November 2015

TEMPERAMENT AND DEPRESSIVE SYMPTOMS

2009). However, we also had some findings that were less consistent with theory, including the fact that children high in maternally reported EC and low in other temperamental risks also showed increasing levels of depression. It is important to note that, as might be expected of children in this developmental period, mean depression scores for the sample were increasing at each time point. Thus, it is perhaps unsurprisingly to see this increase, even in temperamentally low-risk groups. In analyses not presented here, the two follow-ups were treated as baselines in MLMs in order to examine mean differences in depression scores (i.e., ‘‘intercepts’’) based on temperament; these showed that the main effect of EC on depressive symptoms persisted at ages 8 and 9, with lower EC predicting higher depressive symptoms at each time point. Similarly, although children generally increased in depressive symptoms across the follow-ups, those low in maternally reported EC tended to be those with the highest levels of depressive symptoms, other temperamental risks aside. Although additional follow-up into the age of risk is needed to better understand differences in group trajectories, these analyses further underscore the importance of EC in children’s depressive symptoms. Although we observed several interactions between reactive and regulatory mother-reported temperament traits in predicting child self-reported symptoms of depression, we did not observe these for observed temperament traits. Unfortunately, findings often do not replicate across different methods of temperament assessment, in particular when relating temperament to psychopathology (e.g., Dougherty et al., 2010; Durbin et al., 2007; Hayden, Klein, & Durbin, 2005). As previously discussed, differences in how traits were operationalized between our laboratory methods and maternal report measure may have contributed to the disparity in findings across methods. It is also possible that having a broader context for the assessment of PE and NE, as that obtained by maternal report, is important for detecting interactions with effortful control. However, it has also been demonstrated that mothers’ reports of child temperament are influenced by an array of maternal characteristics, including maternal psychopathology and personality (Durbin & Wilson, 2012). Which means of assessing child temperament shows the greatest predictive validity for important child outcomes requires additional longitudinal tests. It is also possible that, like the literature on children’s symptoms (De Los Reyes & Kazdin, 2005), the best source of information on children varies as a function of the target construct, with both methods having strengths and weaknesses. This study had a number of strengths, including its use of a multimethod, longitudinal approach. However, the present study also has limitations. First, we were

783

able to account for only the inhibitory and behavior activation aspects of EC in our observational measure; although research indicates that these are crucial facets of EC regarding psychopathology risk (Eisenberg et al., 2009; Nigg, 2006), our study cannot speak to other facets of this trait that may also be important for understanding children’s depression risk. Also, some of the reliability statistics for our observed measures of temperament were low, albeit acceptable. In the case of a construct such as temperament, which shapes behavior across contexts along with other state-related factors, lower internal consistency statistics may be more acceptable (Hogan & Roberts, 1996; Ones & Viswesvaran, 1996). Although anxiety is highly comorbid with depression during adolescence (Steer, Clark, Kumar, & Beck, 2008), we did not include child anxious symptoms in models; however, analyses not reported here indicated that adding anxious symptoms as a covariate did not change results. We did not investigate interactions between reactive temperament traits (i.e., NE and PE) or three-way interactions due to limited power; similarly, we were unable to thoroughly explore sex differences in the pattern of findings obtained by testing interactions between temperament and child sex in predicting depression. Although our sample was young, we did not conduct lifetime clinical interview assessments to rule out previous or current episodes. Finally, for reasons unknown, more child participants endorsed elevated rates of depressive symptoms than might have been expected in a community sample; the impact of this subgroup of children on the pattern of findings obtained is unclear. Although families and children were representative of the region in which the study was conducted, additional work that aims to replicate these findings in other community samples is needed. Findings indicate that EC shows both main and interactive links to depressive symptoms. Programs have been developed toward the goal of fostering children’s EC-related traits (Rueda et al., 2005; van Veen & Carter, 2002); such approaches may be useful components of preventative programs. Our findings may help identify children at particularly high risk for depression by virtue of other temperament traits, such as low surgency or high NE, who could be targeted for preventative efforts focused on improving EC skills.

FUNDING This project was supported by the Social Sciences and Humanities Research Council of Canada and Children’s Health Research Institute (London, Ontario, Canada).

784

KOTELNIKOVA, MACKRELL, JORDAN, HAYDEN

Downloaded by [University of Lethbridge] at 13:51 05 November 2015

REFERENCES Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Thousand Oaks, CA: Sage. Anderson, E., & Hope, D. A. (2008). A review of the tripartite model for understanding the link between anxiety and depression in youth. Clinical Psychology Review, 28, 275–287. doi:10.1016=j.cpr.2007. 05.004 Asarnow, J. R., & Carlson, G. A. (1985). Depression self-rating scale: Utility with child psychiatric inpatients. Journal of Consulting and Clinical Psychology, 53, 491–499. doi:10.1037=0022-006X.53.4.491 Beauchaine, T. P. (2001). Vagal tone, development, and gray’s motivational theory: Toward an integrated model of autonomic nervous system functioning in psychopathology. Development and Psychopathology, 13, 183–214. doi:10.1017=S0954579401002012 Birleson, P. (1981). The validity of depressive disorder in childhood and the development of a self-rating scale: A research report. Journal of Child Psychology and Psychiatry, 22, 73–88. doi:10.1111= j.1469-7610.1981.tb00533.x Carver, C. S., Johnson, S. L., & Joormann, J. (2009). Two-mode models of self-regulation as a tool for conceptualizing effects of the serotonin system in normal behavior and diverse disorders. Current Directions in Psychological Science, 18, 195–199. doi:10.1111= j.1467-8721.2009.01635.x Chorpita, B. F., Plummer, C. M., & Moffitt, C. E. (2000). Relations of tripartite dimensions of emotion to childhood anxiety and mood disorders. Journal of Abnormal Child Psychology, 28, 299–310. doi:10.1023=A:1005152505888 Clark, L. A. (2005). Temperament as a unifying basis for personality and psychopathology. Journal of Abnormal Psychology, 114, 505–521. doi:10.1037=0021-843X.114.4.505 Clark, L. A., & Watson, D. (1991). Tripartite model of anxiety and depression: Psychometric evidence and taxonomic implications. Journal of Abnormal Psychology, 100, 316–336. doi:10.1037=0021843X.100.3.316 De Bolle, M., & De Fruyt, F. (2010). The tripartite model in childhood and adolescence: Future directions for developmental research. Child Development Perspectives, 4, 174–180. doi:10.1111=j.17508606.2010.00136.x De Los Reyes, A., & Kazdin, A. E. (2005). Informant discrepancies in the assessment of childhood psychopathology: A critical review, theoretical framework, and recommendations for further study. Psychological Bulletin, 131, 483–509. doi:10.1037=0033-2909.131. 4.483 De Pauw, S. S. W., & Mervielde, I. (2010). Temperament, personality and developmental psychopathology: A review based on the conceptual dimensions underlying childhood traits. Child Psychiatry and Human Development, 41, 313–329. doi:10.1007=s10578-009-0171-8 Dougherty, L. R., Bufferd, S. J., Carlson, G. A., Dyson, M., Olino, T. M., Durbin, C. E., & Klein, D. N. (2011). Preschoolers observed temperament and psychiatric disorders assessed with a parent diagnostic interview. Journal of Clinical Child and Adolescent Psychology, 40, 295–306. doi:10.1080=15374416.2011.546046 Dougherty, L. R., Klein, D. N., Durbin, C. E., Hayden, E. P., & Olino, T. M. (2010). Temperamental positive and negative emotionality and children’s depressive symptoms: A longitudinal prospective study from age three to ten. Journal of Social and Clinical Psychology, 29, 462–488. Dunn, L. M., & Dunn, L. M. (2007). Peabody Picture Vocabulary Test (4th ed.). Circle Pines, MN: American Guidance Service. Durbin, C. E. (2010). Validity of young children’s self-reports of their emotion in response to structured laboratory tasks. Emotion, 10, 519–535. doi:10.1037=a0019008 Durbin, C. E., Hayden, E. P., Klein, D. N., & Olino, T. M. (2007). Stability of laboratory-assessed temperamental emotionality traits

from ages 3 to 7. Emotion, 7, 388–399. doi:10.1037=1528-3542. 7.2.388 Durbin, C. E., Klein, D. N., Hayden, E. P., Buckley, M. E., & Moerk, K. C. (2005). Temperamental emotionality in preschoolers and parental mood disorders. Journal of Abnormal Psychology, 114, 28–37. doi:10.1037=0021-843X.114.1.28 Durbin, C. E., & Wilson, S. (2012). Convergent validity of and bias in maternal reports of child emotion. Psychological Assessment, 24, 647–660. doi:10.1037=a0026607 Eisenberg, N., Valiente, C., Spinrad, T. L., Cumberland, A., Liew, J., Reiser, M., . . . Losoya, S. H. (2009). Longitudinal relations of children’s effortful control, impulsivity, and negative emotionality to their externalizing, internalizing, and co-occurring behavior problems. Developmental Psychology, 45, 988–1008. doi:10.1037= a0016213 Eisenberg, N., Zhou, Q., Spinrad, T. L., Valiente, C., Fabes, R. A., & Liew, J. (2005). Relations among positive parenting, children’s effortful control, and externalizing problems: A three-wave longitudinal study. Child Development, 76, 1055–1071. doi:10.1111= j.1467-8624.2005.00897.x Goldsmith, H. H., Reilly, J., Lemery, K. S., Longley, S., & Prescott, A. (1995). Laboratory temperament assessment battery: Preschool version. Unpublished manuscript. Hayden, E. P., Durbin, C. E., Klein, D. N., & Olino, T. M. (2010). Maternal personality influences the relationship between maternal reports and laboratory measures of child temperament. Journal of Personality Assessment, 92, 586–593. doi:10.1080=00223891.2010. 513308 Hayden, E. P., Klein, D. N., & Durbin, C. E. (2005). Parent reports and laboratory assessments of child temperament: A comparison of their associations with risk for depression and externalizing disorders. Journal of Psychopathology and Behavioral Assessment, 27, 89–100. doi:s10862-005-5383-z Hayden, E. P., Klein, D. N., Durbin, C. E., & Olino, T. M. (2006). Positive emotionality at age 3 predicts cognitive styles in 7-year-old children. Development and Psychopathology, 18, 409–423. doi:10.1017=S0954579406060226 Hogan, J., & Roberts, B. W. (1996). Issues and non-issues in the fidelity-bandwidth trade-off. Journal of Organizational Behavior, 17, 627–637. Joiner, T. E., & Lonigan, C. J. (2000). Tripartite model of depression and anxiety in youth psychiatric inpatients: Relations with diagnostic status and future symptoms. Journal of Clinical Child Psychology, 29, 372–382. doi:10.1207=S15374424JCCP2903_8 Joiner, T. E., & Timmons, K. A. (2009). Depression in its interpersonal context. In I. H. Gotlib & C. L. Hammen (Eds.), Handbook of depression (pp. 322–339). New York, NY: Guilford. Joormann, J. (2009). Cognitive aspects of depression. In I. H. Gotlib & C. L. Hammen (Eds.), Handbook of depression (pp. 298–321). New York, NY: Guilford. Keenan, K., Feng, X., Hipwell, A., & Klostermann, S. (2009). Depression begets depression: Comparing the predictive utility of depression and anxiety symptoms to later depression. Journal of Child Psychology and Psychiatry, 50, 1167–1175. doi:10.1111= j.1469-7610.2009.02080.x Kendler, K. S., Gatz, M., Gardner, C. O., & Pederse, N. L. (2006). Personality and major depression. Archives of General Psychiatry, 63, 1113–1120. doi:10.1001=archpsyc.63.10.1113 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, 593–602. doi:10.1001= archpsyc.62.6.593 Kessler, R. C., McGonagle, K. A., Zhao, S., & Nelson, C. B. (1994). Lifetime and 12-month prevalence of DSM–III–R psychiatric

Downloaded by [University of Lethbridge] at 13:51 05 November 2015

TEMPERAMENT AND DEPRESSIVE SYMPTOMS disorders in the United States: Results from the national comorbidity study. Archives of General Psychiatry, 51, 8–19. doi:10.1001= archpsyc.1994.03950010008002 Kieras, J. E., Tobin, R. M., Graziano, W. G., & Rothbart, M. K. (2005). You can’t always get what you want: Effortful control and children’s responses to undesirable gifts. Psychological Science, 16, 391–396. doi:10.1111=j.0956-7976.2005.01546.x Klein, D. N., Durbin, C. E., & Shankman, S. A. (2009). Personality and mood disorders. In I. H. Gotlib & C. L. Hammen (Eds.), Handbook of depression (pp. 93–112). New York, NY: Guilford. Kochanska, G., & Knaack, A. (2003). Effortful control as a personality characteristic of young children: Antecedents, correlates, and consequences. Journal of Personality, 71, 1087–1112. doi:10.1111= 1467-6494.7106008 Kotelnikova, Y., Olino, T. M., Mackrell, S. V. M., Jordan, P. L., & Hayden, E. P. (2013). Structure of observed temperament in middle childhood. Journal of Research in Personality, 47, 524–532. doi:10.1016=j.jrp.2013.04.013 Laurent, J., & Ettelson, R. (2001). An examination of the tripartite model of anxiety and depression and its application to youth. Clinical Child and Family Psychology Review, 4, 209–230. Lengua, L. J. (2006). Growth in temperament and parenting as predictors of adjustment during childrens transition to adolescence. Developmental Psychology, 42, 819–832. doi:10.1037=0012-1649. 42.5.819 Lewinsohn, P. M., Hops, H., Roberts, R. E., Seeley, J. R., & Andrews, J. A. (1993). Adolescent psychopathology: I. Prevalence and incidence of depression and other DSM–III–R disorders in high school students. Journal of Abnormal Psychology, 102, 133–144. Lonigan, C. J., Phillips, B. M., & Hooe, E. S. (2003). Relations of positive and negative affectivity to anxiety and depression in children: Evidence from a latent variable longitudinal study. Journal of Consulting and Clinical Psychology, 71, 465–481. doi:10.1037= 0022-006X.71.3.465 Luby, J., Belden, A., Sullivan, J., Hayen, R., McCadney, A., & Spitznagel, E. (2009). Shame and guilt in preschool depression: Evidence for elevations in self-conscious emotions in depression as early as age 3. Journal of Child Psychology and Psychiatry, 50, 1156–1166. doi:10.1111=j.1469-7610.2009.02077.x McGrath, E. P., & Repetti, R. L. (2002). A longitudinal study of children’s depressive symptoms, self-perceptions, and cognitive distortions about the self. Journal of Abnormal Psychology, 111, 77–87. doi:10.1037=0021-843X.111.1.77 Mezulis, A., Simonson, J., McCauley, E., & Stoep, A. V. (2011). The association between temperament and depressive symptoms in adolescence: Brooding and reflection as potential mediators. Cognition and Emotion, 25, 1460–1470. doi:10.1080=02699931.2010.543642 Muris, P. (2006). Unique and interactive effects of neuroticism and effortful control on psychopathological symptoms in nonclinical adolescents. Personality and Individual Differences, 40, 1409–1419. doi:10.1016=j.paid.2005.12.001 Muris, P., Meesters, C., & Blijlevens, P. (2007). Self-reported reactive and regulative temperament in early adolescence: Relations to internalizing and externalizing problem behavior and ‘‘big three’’ personality factors. Journal of Adolescence, 30, 1035–1049. doi:10.1016= j.adolescence.2007.03.003 Murray, K. T., & Kochanska, G. (2002). Effortful control: Factor structure and relation to externalizing and internalizing behaviors. Journal of Abnormal Child Psychology, 30, 503–514. doi:10.1023= A:1019821031523 Muthe´n, L. K., & Muthe´n, B. O. (1998). Mplus user’s guide (6th ed.). Los Angeles, CA: Muthe´n & Muthe´n. Nigg, J. T. (2006). Temperament and developmental psychopathology. Journal of Child Psychology and Psychiatry, 47, 395–422. doi:10.1111=j.1469-7610.2006.01612.x

785

Oldehinkel, A. J., Hartman, C. A., Ferdinand, R. F., Verhulst, F. C., & Ormel, J. (2007). Effortful control as modifier of the association between negative emotionality and adolescents mental health problems. Development and Psychopathology, 19, 523–539. doi:10.1017= S0954579407070253 Olino, T. M., Durbin, C. E., Klein, D. N., Hayden, E. P., & Dyson, M. W. (2013). Gender differences in young children’s temperament traits: Comparisons across observational and parent-report methods. Journal of Personality, 81, 119–129. doi:10.1111=jopy.12000 Olino, T. M., Klein, D. N., Dyson, M. W., Rose, S. A., & Durbin, C. E. (2010). Temperamental emotionality in preschool-aged children and depressive disorders in parents: Associations in a large community sample. Journal of Abnormal Psychology, 119, 468– 478. doi:10.1037=a0020112 Ones, D. S., & Viswesvaran, C. (1996). Bandwidth-fidelity dilemma in personality measurement for personnel selection. Journal of Organizational Behavior, 17, 609–626. Oosterlaan, J., Logan, G. D., & Sergeant, J. A. (1998). Response inhibition in AD=HD, CD, comorbid AD=HD þ CD, anxious, and control children: A meta-analysis of studies with the stop task. Journal of Child Psychology and Psychiatry, 39, 411–425. doi:10.1017=S0021963097002072 Peugh, J. L. (2010). A practical guide to multilevel modeling. Journal of School Psychology, 48, 85–112. doi:10.1016=j.jsp.2009. 09.002 Phillips, B. M., Lonigan, C. J., Driscoll, K., & Hooe, E. S. (2002). Positive and negative affectivity in children: A multitrait–multimethod investigation. Journal of Clinical Child and Adolescent Psychology, 31, 465–479. doi:10.1207=153744202320802142 Pine, D. S., Cohen, P., Gurley, D., Brook, J., & Ma, Y. (1998). The risk for early-adulthood anxiety and depressive disorders in adolescents with anxiety and depressive disorders. Archives of General Psychiatry, 55, 56–64. doi:10.1001=archpsyc.55.1.56 Rothbart, M. K., Ahadi, S. A., Hershey, K. L., & Fisher, P. (2001). Investigations of temperament at three to seven years: The children’s behavior questionnaire. Child Development, 72, 1394–1408. doi:10.1111=1467-8624.00355 Rothbart, M. K., & Bates, J. E. (2006). Temperment. In Handbook of child psychology: Vol. 3 Social, emotional, and personality development, (6th ed., pp. 99–166). Hoboken, NJ: Wiley. Rothbart, M. K., Sheese, B. E., & Posner, M. I. (2007). Executive attention and effortful control: Linking temperament, brain networks, and genes. Child Development Perspectives, 1, 2–7. doi:10.1111=j.1750-8606.2007.00002.x Rudolph, K. D., Troop-Gordon, W., & Llewellyn, N. (2013). Interactive contributions of self-regulation deficits and social motivation to psychopathology: Unraveling divergent pathways to aggressive behavior and depressive symptoms. Development and Psychopathology, 25, 407–418. doi:10.1017=S0954579412001149 Rueda, M. R., Posner, M. I., & Rothbart, M. K. (2005). The development of executive attention: Contributions to the emergence of self-regulation. Developmental Neuropsychology, 28, 573–594. doi:10.1207=s15326942dn2802_2 Rydell, A., Berlin, L., & Bohlin, G. (2003). Emotionality, emotion regulation, and adaptation among 5- to 8-year-old children. Emotion, 3, 30–47. doi:10.1037=1528-3542.3.1.30 Schmidt, N. B., Richey, J. A., Buckner, J. D., & Timpano, K. R. (2009). Attention training for generalized social anxiety disorder. Journal of Abnormal Psychology, 118, 5–14. Seifer, R., Sameroff, A. J., Barrett, L. C., & Krafchuk, E. (1994). Infant temperament measured by multiple observations and mother report. Child Development, 65, 1478–1490. doi:10.2307=1131512 Simonds, J. (2006). The role of reward sensitivity and response: Execution in childhood extraversion (Unpublished doctoral dissertation). Eugene, OR: University of Oregon.

786

KOTELNIKOVA, MACKRELL, JORDAN, HAYDEN

Downloaded by [University of Lethbridge] at 13:51 05 November 2015

Simonds, J., & Rothbart, M. K. (2004). The Temperament in Middle Childhood Questionnaire (TMCQ): A computerized self-report measure of temperament for ages 7–10. Poster session presented at the Occasional Temperament Conference, Athens, GA. Singer, J. D., & Willett, J. B. (2003). Applied longitudinal data analysis: Modeling change and event occurrence. New York, NY: Oxford University Press. Statistics Canada. (2008). Population groups, age groups, sex and selected demographic, cultural, labour force, educational and income characteristics, for the total population of Canada, provinces, territories, census metropolitan areas and census agglomerations, 2006 Census—20% sample data.

Steer, R. A., Clark, D. A., Kumar, G., & Beck, A. T. (2008). Common and specific dimensions of self-reported anxiety and depression in adolescent outpatients. Journal of Psychopathology and Behavioral Assessment, 30, 163–170. doi:10.1007=s10862-007-9060-2 Thase, M. E. (2009). Neurobiological aspects of depression. In I. H. Gotlib & C. L. Hammen (Eds.), Handbook of depression (pp. 187–217). New York, NY: Guilford. Van Veen, V., & Carter, C. S. (2002). The anterior cingulate as a conflict monitor: FMRI and ERP studies. Physiology & Behavior, 77, 477–482. doi:10.1016=S0031-9384(02)00930-7 Watson, D., & Clark, L. A. (1993). In D. M. Wegner & J. W. Pennebaker (Eds.), Behavioral disinhibition versus constraint: A dispositional perspective. Englewood Cliffs, NJ: Prentice-Hall.