Rumination,Excessive Reassurance Seeking,and Stress Generation Among Early Adolescent Girls

Abstract

Nolen-Hoeksema proposed that rumination increases stressful events and circumstances; however, few studies have examined this question. Thus, we explored whether (a) rumination predicted increases in the generation of chronic and acute stress, (b) excessive reassurance seeking (ERS) mediated links between rumination and stress generation, (c) rumination increased exposure to acute independent (uncontrollable) stress, and (d) rumination predicted chronic stress generation in certain domains, but not others. These questions were examined in a 1-year study of 126 early adolescent girls (M _age = 12.39 years) using contextual objective stress interviews. Findings indicated that rumination predicted increases in acute dependent interpersonal stress and chronic interpersonal stress, and ERS mediated these associations. Moreover, rumination was not associated with acute independent stress. Finally, the effect of rumination on chronic stress generation was most salient in adolescents’ romantic lives and in parent-adolescent relationships. These findings suggest that ruminators create stressful interpersonal environments.

Keywords

rumination stress generation chronic stress excessive reassurance seeking adolescence

There is increasing interest in understanding factors that confer risk for multiple forms of psychopathology. One such factor, rumination—the tendency to passively and repeatedly focus on one’s distress and the associated causes and consequences (Nolen-Hoeksema, 1991)—has been associated with the development of depression, anxiety, and eating disorders, as well as substance use, aggression, and self-injurious behavior (e.g., for a review, see Nolen-Hoeksema, Wisco, & Lyubomirsky, 2008). In the response styles theory, Nolen-Hoeksema (1991) proposed that one way rumination leads to the development of psychopathology is by inhibiting behaviors that may alleviate, or change the source of, one’s distress, thereby increasing stressful events and circumstances (Nolen-Hoeksema, 1991; Nolen-Hoeksema et al., 2008). However, few studies have explicitly tested whether ruminators actively create stress. The stress generation model (Hammen, 1991) provides a useful framework for doing so. The model posits that vulnerable individuals behave in ways that contribute to the occurrence of stress, particularly interpersonal stress, including both acute dependent interpersonal stress (i.e., interpersonal stressful life events caused in part by the person’s actions or behaviors; for example, conflicts) and chronic interpersonal stress (i.e., ongoing stressful interpersonal circumstances; Hammen, 2006; Liu & Alloy, 2010).

Because disorders associated with rumination (e.g., depression) often emerge in mid-adolescence (Rohde, Beevers, Stice, & O’Neil, 2009), it is particularly important to investigate whether rumination confers risk for stress generation in early adolescence, prior to the onset of psychopathology. Furthermore, as compared with early adolescent boys, early adolescent girls generate higher levels of interpersonal stress (Rudolph & Flynn, 2007; Rudolph & Hammen, 1999) and report higher levels of rumination (Hampel & Petermann, 2005). Indeed, girls’ higher levels of rumination partially contribute to the gender difference observed in depression (Hilt, McLaughlin, & Nolen-Hoeksema, 2010). Thus, in the present study, we examined whether rumination places early adolescent girls at risk for generating interpersonal stress.

Rumination as a Predictor of Stress Generation

Surprisingly, prior work examining whether rumination confers risk for stress generation is limited and mixed. Some evidence suggests that rumination leads to stress generation. For example, greater rumination predicts increases in acute dependent interpersonal stress (Flynn, Kecmanovic, & Alloy, 2010), peer victimization (McLaughlin & Nolen-Hoeksema, 2012; Shapero, Hamilton, Liu, Abramson, & Alloy, 2013), and family emotional abuse (Shapero, Hamilton, et al., 2013). Similarly, in one study of early adolescents, greater use of maladaptive coping strategies, including involuntary engagement (i.e., rumination, emotional, and psychological arousal), predicted increases in acute dependent interpersonal stress (Flynn & Rudolph, 2011). In contrast, other work has failed to find evidence that rumination leads to increases in acute dependent interpersonal stress (Hamilton et al., 2013; Shapero, Hankin, & Barrocas, 2013) and relational peer victimization (Hamilton et al., 2013).

In addition to mixed evidence, there are also gaps in the literature. First, previous studies examining rumination and stress generation have not specifically focused on ruminative brooding (i.e., passively focusing on one’s distress)—the component of rumination that has been consistently predictive of depression (Burwell & Shirk, 2007; Treynor, Gonzalez, & Nolen-Hoeksema, 2003) and does not overlap with depressive symptoms (Treynor et al., 2003). Second, prior work has exclusively examined acute stress generation; thus, although we have evidence that ruminators experience chronic problems in their relationships (e.g., Nolen-Hoeksema & Davis, 1999; Nolen-Hoeksema et al., 2008; Shapero, Hamilton, et al., 2013), we do not know whether ruminators generate chronic interpersonal stress. Third, most existing studies have relied on life-event checklists (for an exception, see Flynn & Rudolph, 2011). However, as compared with checklists, contextual life stress interviews offer enhanced validity (e.g., Hammen, 2006; Liu & Alloy, 2010) and are less biased by cognitive vulnerabilities (Simons, Angell, Monroe, & Thase, 1993). Moreover, checklists do not permit fine-grained contextual distinctions regarding event independence—the extent to which the event occurs as a result of the individual’s behavior (e.g., Hammen, 2006; Liu & Alloy, 2010). Given that the stress generation model posits that vulnerable individuals will experience increased dependent (i.e., events caused in part by the person’s behavior), but not independent (i.e., uncontrollable events; Hammen, 1991), events, failing to precisely make this distinction may weaken the effect of rumination on stress generation (Liu & Alloy, 2010).

Fourth, few studies have examined whether rumination specifically enhances risk for generating acute stress, or whether rumination is simply associated with increased exposure to acute stress (e.g., Rudolph & Hammen, 1999). Although initial evidence suggests that rumination does not increase exposure to acute independent stress (Flynn et al., 2010; Hamilton et al., 2013), this question has not been evaluated with contextual stress interviews. Fifth, it is unclear whether rumination confers risk for the generation of both interpersonal and non-interpersonal stress, or whether ruminators are particularly likely to generate interpersonal stress. The one prior study that has addressed this question using contextual life stress interviews demonstrated that ineffective stress responses (including rumination) predicted increases in both acute dependent interpersonal and non-interpersonal stress (e.g., academic events; Flynn & Rudolph, 2011). However, given the focus was on examining the effect of ineffective coping strategies, it is impossible to ascertain the extent to which rumination was driving the observed stress generation effect. Thus, more work is needed to clarify whether rumination leads to the generation of both interpersonal and non-interpersonal acute stress, along with both interpersonal and non-interpersonal chronic stress, the latter of which has not been investigated in prior work.

Finally, little is known about how rumination contributes to stress generation. The response style theory posits that ruminators exhibit behaviors that have maladaptive consequences for their relationships and erode social support (Nolen-Hoeksema, 1991; Nolen-Hoeksema et al., 2008). Supporting this, following a traumatic loss, those who engaged in high levels of rumination were more likely to seek support, but reported receiving less support, as compared with those who engaged in low levels of rumination (Nolen-Hoeksema & Davis, 1999). Similarly, McLaughlin and Nolen-Hoeksema (2012) demonstrated that rumination led to increases in relational peer victimization via increasing peer communication, which the authors posited might be because ruminators were engaging in excessive support seeking. In line with this, young adults who engaged in higher levels of rumination reported needing more, but also being less satisfied with their levels of, social support, which led to subsequent increases in acute dependent interpersonal stress (Flynn et al., 2010).

Together, these findings suggest that rumination may lead to stress generation by increasing excessive reassurance seeking (ERS)—repeatedly asking for reassurance that one is loveable and worthy in order to alleviate feelings of distress (Joiner, Alfano, & Metalsky, 1992). Indeed, it has been suggested that rumination may fuel ERS via enhancing negative mood (Joiner, 2000) and feelings of uncertainty (Nolen-Hoeksema, 2000), and prior work indicates that rumination and ERS are positively associated (Weinstock & Whisman, 2007). In addition, ERS has obtained support as a predictor of the generation of acute dependent interpersonal stress (e.g., Eberhart & Hammen, 2009). Thus, both theory and research suggest that rumination may breed ERS, leading to interpersonal stress generation.

The Present Study

Drawing upon stress generation theory (Hammen, 1991), interpersonal theories of depression (e.g., Joiner et al., 1992), and response styles theory (Nolen-Hoeksema, 1991), we investigated whether rumination (a) directly predicted interpersonal stress generation and (b) indirectly predicted interpersonal stress generation via ERS. Expanding upon prior work, we simultaneously examined acute and chronic interpersonal stress generation to elucidate whether rumination predicted each type of stress generation (i.e., acute dependent interpersonal stress, chronic interpersonal stress), as well as whether ERS mediated the link between rumination and each type of stress generation. In addition, we accounted for current and lifetime history of depressive symptoms along with past year acute and chronic stress to evaluate the effects of rumination over and above the effects of initial stress and depression (e.g., Hammen, 2006). These questions were examined in a 1-year longitudinal study of early adolescent girls.

Based on prior work (e.g., Eberhart & Hammen, 2009; Flynn & Rudolph, 2011), we expected that greater rumination would predict increases in acute dependent interpersonal stress and chronic interpersonal stress, and that ERS would mediate these associations. In addition, we examined whether rumination places adolescents at risk for generating stress, specifically interpersonal stress, or whether rumination also increases exposure to independent forms of stress (e.g., Rudolph & Hammen, 1999). Based on prior work (Flynn & Rudolph, 2011) and the stress generation model (Hammen, 1991), we predicted that rumination would lead to increases in interpersonal and non-interpersonal stress generation (i.e., chronic interpersonal and non-interpersonal stress, acute dependent interpersonal and non-interpersonal stress), but not to increases in independent stress. Finally, we explored whether rumination was a particularly potent predictor of chronic stress generation in specific domains, including four interpersonal (parent-child relationship, close friendships, romantic relationships/dating, and peer social circle) and two non-interpersonal (academics, academic behavior). Given limited prior work, we did not have predictions about the pattern of chronic stress generation effects.

Method

Participants and Procedure

One hundred twenty-six early adolescent girls and their primary female caregivers participated in the present study. Adolescent girls who were currently in or entering (i.e., summer before) sixth or seventh grade or entering eighth grade and a primary caregiver (herein called “mother”) willing to participate were eligible.¹ Participants were recruited from two predominately rural counties in New England. Given the rural location and the small population of the communities from which participants were drawn, participants were recruited through multiple methods, including advertisements or flyers (10.3%), referrals and word-of-mouth (13.1%), and local schools (76.6%).

At Time 1 (T1), participation included (a) a laboratory visit during which mothers and daughters each completed separate diagnostic and objective stress interviews and a packet of questionnaires, including a measure of pubertal status; and (b) online questionnaires completed separately at home, including measures of rumination and ERS. Of the 126 dyads who completed the laboratory visit, 116 completed the online questionnaires. Approximately one year later, 86% (n = 108) participated in a follow-up (T2) that included the same objective and diagnostic stress interviews. There were no significant differences between those who did and did not participate in T2 on the T1 variables, except that those who did not participate had higher levels of chronic stress at T1 (p < .05). All 126 participants were retained in analyses (see below).

The 126 adolescents ( $\bar{X}$ _age = 12.39 years, SD = 0.76 years) included in the present study were in or entering sixth (48.4%) or seventh (33.3%) grade, or entering eighth (15.9%) grade (2.4% missing). Most families were middle to upper class, <US$10,000 (4.0%); US$10,000-US$20,000 (4.8%); US$21,000-US$40,000 (10.3%); US$41,000-US$60,000 (19.0%); US$61,000-US$100,000 (26.2%); >US$100,000 (35.7%). Consistent with the ethnic composition of the counties from which participants were drawn (94% and 97% Caucasian), most participants were Caucasian (mothers: 94.5%; daughters: 82.6%). Most mothers ( $\bar{X}$ _age = 44.32 years, SD = 6.75 years) were married (76.2%), were the biological parent of the adolescent (94.4%), and had bachelor’s or advanced degrees (65.1%).

Measures

Family income

Family income was assessed via mother-report and examined as a potential covariate.

Perceived pubertal timing

At T1, adolescents completed the Pubertal Development Scale (PDS; Petersen, Crockett, Richards, & Boxer, 1988) to assess pubertal development. The five items, which assess growth spurt in height, skin and body hair changes, breast development, and age at menarche, are rated on a 4-point scale, from no development (1) to development seems completed (4), except for menarche, which is rated dichotomously. The PDS has shown good psychometric properties (Petersen et al., 1988). The items are summed and divided by the number of questions (five) to obtain a Pubertal Development Scale score, which reflects pubertal status ( $\bar{X}$ = 2.64, SD = 0.63; α = .70). To convert pubertal development scores to pubertal timing scores, age was partialled out by conducting a regression analysis predicting PDS scores from age in months and saving the standardized residuals (e.g., Rudolph, 2008). Higher residual scores indicated earlier timing. Based on prior work indicating that pubertal timing, but not age or pubertal status, enhances stress generation effects among early adolescents (Rudolph, 2008), pubertal timing was examined as a potential covariate.

Depressive symptoms

Adolescents were interviewed with the Schedule for Affective Disorders and Schizophrenia for school-aged children-present and lifetime version (K-SADS-PL; Kaufman et al., 1997). The interview assessed current (past month) and lifetime history of depressive symptoms. For past depressive symptoms, the worst period of depressive symptoms was coded. Consistent with prior work (e.g., Stroud & Davila, 2008), symptoms were rated: 0 = no symptoms; 1 = mild symptoms (1-2 symptoms); 2 = moderate, sub-threshold symptoms of depression (3-4 symptoms); 3 = meets Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; American Psychiatric Association, 1994) criteria for depression, using specific behavioral anchors for number of symptoms and degree of impairment. Interrater reliability (assessed via electronic recordings for 27% of interviews) was good (T1—current: intraclass correlation [ICC] = 1.00, past: ICC = .97; T2—current: ICC = 1.00, past: ICC = .98). At T1, seven girls (5.6%) had current symptoms (rated 1-2) and one (0.8%) had current diagnosable depression. Prior to T1, 25 (19.8%) had symptoms (rated 1 or 2) and six (4.8%) had a depressive episode. Given theory (Hammen, 1991, 2006) and evidence (e.g., Hammen, 1991; Rudolph et al., 2000) suggesting that both current and prior depression predict stress generation, current and lifetime history of depressive symptoms were each examined as potential covariates.

Rumination

At T1, five items from the 22-item Ruminative Responses Scale (Nolen-Hoeksema & Morrow, 1991) were used to assess ruminative brooding or the tendency to passively focus on one’s situation (herein called rumination). The five items used loaded onto the brooding factor in Treynor et al. (2003; for example, “Think ‘Why do I always react this way?’”). The directions were modified for adolescents to report on their responses to feeling “stressed or upset” rather than “depressed” (Burwell & Shirk, 2007). For each item, participants indicated how often they “think or do each one” when they “feel stressed or upset” using a 4-point scale from 1 (almost never) to 4 (almost always). The mean was used in analyses (α = .80).

ERS

At T1, ERS was measured with the Depressive Interpersonal Relationship Inventory (Joiner et al., 1992). The four items assess the frequency of seeking reassurance from close others (e.g., “Do you find yourself often asking the people you feel close to how they truly feel about you?”) using a five-item scale from 1 (not at all) to 5 (extremely often). The mean was used in analyses (α = .78).

Life stress

A modified version of the UCLA Life Stress Interview (adapted from Hammen, 1991; Hammen et al., 1987; Rudolph & Flynn, 2007; Rudolph & Hammen, 1999; Rudolph et al., 2000) was used to assess adolescents’ levels of chronic (i.e., ongoing objective stress over the past year) and acute (i.e., events with a brief onset and relatively short duration) stress. The T1 interview assessed the year prior to T1; the T2 interview assessed the time period between T1 and T2. To assess adolescents’ level of life stress, mothers and daughters completed separate interviews with the same interviewer; interviewers were blind to all other study data.

Chronic life stress

Interviewers used behavioral probes to elicit descriptions of adolescents’ ongoing objective stress over the past year in several domains (academics, that is, academic performance; academic behavior, that is, ongoing behavior problems at school; parent-child relationship, close friendships, peer social life, romantic relationships/dating, parents’ marital [or cohabiting] romantic relationship, if applicable). Using behavioral descriptions, interviewers rated adolescents’ level of chronic stress over the past year in each domain on scale from 1 (excellent/optimal circumstances) to 5 (very bad circumstances) in half-point increments; rating points were anchored by specific behavioral indicators to provide an objective rating that was independent of participants’ subjective experiences. Interrater reliability was good (ICCs: T1— $\bar{X}$ = .80 [range = .72-.89], daughter; $\bar{X}$ = .83 [.70-.91], mother; T2— $\bar{X}$ = .93 [.90-.97], daughter; $\bar{X}$ = .96 [.93-.98], mother).

Two chronic stress composites were created for each time point: (a) chronic interpersonal stress: the mean of the parent-child relationship, close friendships, peer social life, romantic relationships/dating, and parents’ marital relationship (if applicable); and (b) chronic non-interpersonal stress: the mean of the academic and academic behavioral domains. Chronic stress composites were not stratified by independence, as all domains of chronic stress assessed were considered at least partially dependent upon participants’ behavior. At T1 and T2, the composites based on mother and daughter report were highly correlated for interpersonal, T1: r(124) = .74; T2: r(124) = .78; ps < .001, and non-interpersonal, T1: r(124) = .78; T2: r(124) = .84; ps < .001, chronic stress; thus, the mother and daughter ratings were combined (Rudolph et al., 2000), by taking the mean of each domain. The interpersonal domains were significantly intercorrelated (T1— $\bar{X}$ : r = .38, range: rs = .17-.72; ps < .05; T2— $\bar{X}$ : r = .42, range: rs = .19-.68; ps < .05), with the exception of the correlations between romantic relationships/dating and parents’ marital (or cohabiting) romantic relationship, which reached marginal significance, T1: r(110) = .17, p = .08; T2: r(110) = .19, p = .06. In addition, the non-interpersonal domains (academics, academic behavior) were significantly intercorrelated, T1: r(124) = .32; T2: r(124) = .50; ps < .001. The T1 interpersonal and T1 non-interpersonal chronic stress composites were combined by taking the mean of all of the chronic stress domains; this composite was examined as a potential covariate.

In addition to the chronic stress composites, we also used the mean of the mother and daughter ratings for each T2 chronic stress domain (academics, academic behavior, parent-child relationship, close friendships, peer social life, romantic relationships/dating) in examining the effect of rumination on specific chronic stress domains. The T2 domain ratings based on mother and daughter report were significantly correlated ( $\bar{X}$ : r = .73, range: rs = .64-.89; ps < .001).

Acute life stress

For each event reported, participants provided information about the context surrounding the event (e.g., circumstances and resources to cope with it, predictability, and prior experience with similar events), the duration, and the consequences to obtain the degree of impact for a typical individual given the context. Integrating information from mothers and daughters (e.g., Rudolph et al., 2000), interviewers prepared narrative accounts of each event (excluding participants’ subjective reactions). Narratives were presented to an independent rating team who were blind to participants’ diagnostic information and subjective reactions.

The team rated the objective impact, independence, and interpersonal nature of each event. Objective impact was rated on a scale of 1 (no negative impact) to 5 (extremely severe negative impact). Independence, or the degree to which the event resulted from the participant’s behavior, was rated on a scale of 1 (fully independent of the person’s behavior) to 5 (fully dependent; occurred strictly as a result of the person’s own actions). Events rated as 3 or higher were coded as dependent and those rated 2.5 or lower were considered independent (e.g., Rudolph & Flynn, 2007). The rating team also rated the event as interpersonal or non-interpersonal. Interpersonal events referred to those events in which the primary context involved relations with other people (e.g., conflicts) or events that happened to others that affected relations with the participant (e.g., friend moving away); non-interpersonal events included all events not rated as interpersonal (e.g., failed test). A second team, blind to the original ratings, rerated a set of events (n = 132) on objective impact (ICC = .92), independence (ICC = .99), and interpersonal status (ICC = .98).

Consistent with prior work (e.g., Rudolph & Hammen, 1999), for each time point, four acute stress composites were created by summing the objective ratings for each type of event for each interview period: (a) acute dependent interpersonal stress (i.e., dependent interpersonal events; for example, conflicts, relationship break-up), (b) acute dependent non-interpersonal stress (i.e., dependent non-interpersonal events; for example, academic failure, extracurricular disappointment), (c) acute independent interpersonal stress (e.g., parental job loss), and (d) acute independent non-interpersonal stress (e.g., moved, changed schools). To reduce the number of variables used in analyses, the four T1 acute stress variables were summed and examined as a potential covariate.

Data Analytic Plan

Path analysis was conducted using Mplus 7 (Muthen & Muthen, 1998-2012). Little’s MCAR test indicated that data were missing completely at random, χ²(33) = 40.935, p = .161; thus, missing data were estimated with full information maximum likelihood (FIML), consistent with current recommendations for handling missing data (Enders, 2013).

First, in preliminary analyses, we examined the correlations between potential covariates (pubertal timing, family income, initial acute stress, initial chronic stress, current depressive symptoms, past depressive symptoms) and the T2 stress outcome variables (T2 acute dependent interpersonal stress, T2 chronic interpersonal stress). For each model, potential covariates that were marginally significantly (p < .10) associated with at least one of the T2 stress outcome variables were covaried.

Second, we examined whether rumination directly, and indirectly via ERS, predicted increases in acute dependent interpersonal stress and chronic interpersonal stress. Thus, we specified a model that included direct paths from rumination to each T2 chronic interpersonal stress and T2 acute dependent interpersonal stress, as well as indirect paths from rumination to each T2 chronic interpersonal stress and T2 acute dependent interpersonal stress through ERS (Model 1; Figure 1). Covariances were included between each of the covariates and between the disturbances of T2 chronic interpersonal stress and T2 acute dependent interpersonal stress. Third, we examined the specificity of the observed effects in separate analyses for each type of stress (Models 2-5). Models 2 to 5 were identical to the primary analysis, except only one T2 stress variable was included as an outcome variable (acute independent interpersonal stress, acute independent non-interpersonal stress, acute dependent non-interpersonal stress, or chronic non-interpersonal stress).

Figure 1.

Conceptual model.

Finally, in Model 6, we explored whether rumination predicted chronic stress in certain domains, but not others.² We repeated the primary analyses (Figure 1), but used the chronic stress domains (parent-child relationship, close friendships, peer social life, romantic relationships/dating, academics, academic behavior)³ instead of the T2 chronic stress composites. Thus, this model included T2 acute dependent interpersonal stress and all of the T2 chronic stress domains as outcomes.

All models were fully saturated; thus, fit indices were not applicable (as saturated models show perfect fit). Consistent with recommendations (MacKinnon, 2008; MacKinnon, Lockwood, & Williams, 2004), significance of direct and indirect effects was evaluated using bias-corrected bootstrapping (n = 5,000). Significant effects are those not including zero in the 95% asymmetric confidence intervals (CI).

Results

Preliminary Analyses

See Table 1 for descriptive statistics and intercorrelations. Rumination and ERS were significantly positively correlated (p < .01) and were each significantly positively correlated with T2 acute dependent interpersonal stress and T2 chronic interpersonal stress. Rumination, but not ERS, was significantly associated with T2 chronic non-interpersonal stress (p < .01). However, rumination and ERS were unrelated to other forms of T2 acute stress (ps > .05).

Table 1.

Intercorrelations Among and Descriptive Statistics for the Variables Included in the Models (N = 126).

	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. T1 acute stress	—
2. T1 chronic stress	.32**	—
3. T1 current depressive symptoms	.01	.15	—
4. T1 past depressive symptoms	.06	.15^†	.25**	—
5. T1 pubertal timing (standardized)	.24**	.16^†	.07	.10	—
6. T1 family income	−.34**	−.36**	−.10	−.20*	−.21*	—
7. T1 rumination	.12	.33**	.04	.10	.14	−.16^†	—
8. T1 ERS	.05	.27**	.04	.02	.18^†	−.15	.50**	—
9. T2 acute dependent interpersonal stress	.29**	.30**	.36**	.24*	.16	−.15	.36**	.40**	—
10. T2 acute dependent non-interpersonal stress	.23*	.09	.23*	−.02	.08	.00	.15	.12	.38**	—
11. T2 acute independent interpersonal stress	.40**	.37**	.28**	.18^†	.13	−.31**	.13	.12	.25**	.13	—
12. T2 acute independent non-interpersonal stress	.21*	.13	.15	.03	−.04*	−.05	−.01	.06	.19*	.28**	.27**	—
13. T2 chronic interpersonal stress	.26**	.63**	.26**	.23**	.18^†	−.20*	.39**	.36**	.63**	.30**	.38**	.20*	—
14. T2 chronic non-interpersonal stress	.36**	.50**	.33**	.16^†	−.18^†	−.45**	.26**	.11	.48**	.26**	.35*	.10	.54**	—
$\bar{X}$ (SD)	11.85 (7.63)	1.97 (0.44)	0.10 (0.44)	0.43 (0.84)	0.00 (1.00)	4.66 (1.00)	2.02 (0.67)	1.70 (0.72)	3.89 (3.64)	2.44 (2.42)	9.17 (5.54)	1.67 (1.71)	2.19 (0.50)	1.79 (0.60)
n	126	126	126	126	126	126	111	110	108	108	108	108	108	108

Note. T1 and T2 stress variables assessed with the Life Stress Interview (adapted from Hammen, 1991; Hammen et al., 1987; Rudolph & Flynn, 2007; Rudolph & Hammen, 1999; Rudolph et al., 2000) with separate interviews with mother and daughter. T1 current and past depressive symptoms assessed with the K-SADS-PL (Kaufman et al., 1997). Pubertal timing = standardized residuals converted from pubertal status scores (Petersen et al., 1988). Rumination assessed with five items from Ruminative Responses Scale (Nolen-Hoeksema & Morrow, 1991). Excessive reassurance seeking (ERS) was assessed with the Depressive Interpersonal Relationship Inventory (Joiner et al., 1992). Ns vary due to missing questionnaire data at T1 and attrition at T2. T1 = Time 1; T2 = Time 2; K-SADS-PL = Schedule for Affective Disorders and Schizophrenia for school-aged children-present and lifetime version.

†

p < .10. *p < .05. **p < .01.

All of the potential covariates (T1 current depressive symptoms, T1 past depressive symptoms, T1 acute stress, T1 chronic stress, family income, perceived pubertal timing) were significantly (ps < .05) or marginally significantly (ps < .10) associated with either the T2 chronic stress variable or the T2 acute stress variable included in Models 1 to 5. In addition, all of the potential covariates were significantly (ps < .05) associated with at least one of the T2 chronic stress domains. Thus, all potential covariates were retained in analyses.

Does Rumination Predict Increases in Acute and Chronic Interpersonal Stress Generation?

As hypothesized, rumination predicted increases in T2 acute dependent interpersonal stress, standardized estimate (β) = .275; p = .007, and T2 chronic interpersonal stress (β = .224, p = .010; Table 2).

Table 2.

Models 1 To 6: Direct and Indirect (Via ERS) Effects of Rumination on Various Forms of Stress (N = 126).

Outcome	β (95% CI)	SE	z	p
Model 1
T2 acute dependent interpersonal stress
Direct effect	.275 [.076, .475]	.102	2.701	.007
Indirect effect	.167 [.038, .295]	.066	2.542	.011
T2 chronic interpersonal stress
Direct effect	.224 [.054, .395]	.087	2.587	.010
Indirect effect	.097[.009, .185]	.045	2.164	.030
Model 2
T2 acute independent interpersonal stress
Direct effect	.025 [−.172, .222]	.100	0.253	.801
Indirect effect	.039 [−.048, .125]	.044	0.876	.381
Model 3
T2 acute independent non-interpersonal stress
Direct effect	−.039 [−.226, .147]	.095	−0.415	.679
Indirect effect	.056 [−.072, .184]	.065	0.851	.395
Model 4
T2 acute dependent non-interpersonal stress
Direct effect	.159 [−.100, 1.384]	.099	1.606	.108
Indirect effect	.041 [−.266, .503]	.053	0.783	.433
Model 5
T2 chronic non-interpersonal stress
Direct effect	.123 [−.080, .326]	.103	1.189	.234
Indirect effect	−.011[−.266, .503]	.057	−0.188	.851
Model 6
T2 acute dependent interpersonal stress
Direct effect	.274 [.076, .473]	.101	2.714	.007
Indirect effect	.152 [.026, .279]	.065	2.355	.019
Romantic relationships/dating chronic stress
Direct effect	.227 [.029, .425]	.101	2.252	.024
Indirect effect	.090 [−.005, .186]	.049	1.849	.064
Close friendships chronic stress
Direct effect	.036 [−.160, .231]	.100	0.356	.722
Indirect effect	.080 [−.031, .191]	.057	1.410	.159
Peer social life chronic stress
Direct effect	.155 [−.015, .326]	.087	1.791	.073
Indirect effect	.089 [−.016, .193]	.053	1.666	.096
Parent-child relationship chronic stress
Direct effect	.336 [.155, .516]	.092	3.647	<.001
Indirect effect	.019 [−.091, .128]	.056	0.334	.738
Academic chronic stress
Direct effect	.076 [−.105, .257]	.092	0.822	.411
Indirect effect	−.032 [−.132, .068]	.051	−0.628	.530
Academic behavior chronic stress
Direct effect	.108 [−.132, .349]	.123	0.883	.377
Indirect effect	.016 [−.099, .131]	.059	0.268	.788

Note. Pubertal timing, family income, current depressive symptoms, past depressive symptoms, and initial acute and chronic stress were covaried. ERS = excessive reassurance seeking; β = standardized coefficient; CI = confidence interval; T2 = Time 2.

Does ERS Mediate the Prospective Associations Between Rumination and T2 Acute and Chronic Interpersonal Stress Generation?

Consistent with predictions, greater rumination was indirectly associated with increases in T2 acute dependent interpersonal stress (β = .167, p = .011) and T2 chronic interpersonal stress (β = .097, p = .030) through greater ERS (see Figure 2 and Table 2). The direct effects of rumination on T2 acute dependent interpersonal stress (β = .108; SE = .112; z = .964; p = .335; 95% CI = [−.112, .329]) and T2 chronic interpersonal stress (β = .127; SE = .090; z = 1.421; p = .155; 95% CI = [−.048, .303]) were no longer significant, further supporting the mediating role of ERS.

Figure 2.

Mediation of rumination on acute and chronic interpersonal stress generation by ERS (N = 126).

Does Rumination Directly and Indirectly (Via ERS) Predict Increases in Independent and Non-Interpersonal Forms of Stress?

The direct effects for T2 non-interpersonal chronic stress (β = .123, p = .234), acute dependent non-interpersonal stress (β = .159, p = .108), acute independent non-interpersonal stress (β = −.039, p = .679), and acute independent interpersonal stress (β = .025, p = .801) were small and not significant. In addition, the indirect effects for these forms of stress were not significant (ps > .381; Table 2).

Does Rumination Directly and Indirectly Predict Chronic Interpersonal and Non-Interpersonal Stress in Specific Domains of Chronic Stress?

Rumination directly predicted increases in romantic relationships/dating (β = .227; p = .024) and parent-child relationship (β = .336, p < .001) chronic stress, and the direct effect of rumination on peer social life chronic stress approached significance (β = .155, p = .073). In addition, the indirect effects of rumination on romantic relationships/dating (β = .090, p = .064) and peer social life (β = .089, p = .096) chronic stress via ERS each approached significance. However, rumination did not predict increases in chronic stress in the close friendships (β = .036, p = .722), academic (β = .076, p = .411), or academic behavior (β = .108, p = .377) domains. Finally, the indirect effects of rumination on the other domains of chronic stress generation were not significant (ps > .15; see Table 2).

Discussion

Guided by response styles theory (Nolen-Hoeksema, 1991), interpersonal models of depression (e.g., Joiner et al., 1992), and the stress generation model (Hammen, 1991), this study explored links between rumination, ERS, and stress generation. Four main findings emerged. First, greater rumination directly predicted increases in acute and chronic interpersonal stress generation. Second, rumination indirectly predicted both forms of interpersonal stress generation through its association with ERS. Third, rumination was a particularly potent predictor of stress generation in adolescents’ romantic/dating and parent-child relationships. Fourth, rumination was not associated with increases in independent and non-interpersonal forms of stress.

Rumination Predicts Increases in Acute and Chronic Stress Generation

The findings of the present study suggest that early adolescent girls who engage in greater levels of rumination generate increases in both acute and chronic interpersonal stress over time. This finding is particularly noteworthy because it emerged when we accounted for initial stress, as well as current and lifetime depressive symptoms, and simultaneously examined acute and chronic forms of interpersonal stress generation. Thus, rumination predicted increases in each form of interpersonal stress generation, over and above the effects of initial stress and depressive symptoms. In addition, acute and chronic interpersonal stress were measured via objective stress interviews with adolescents and their mothers, reducing concerns that rumination contributes to increased perceptions of acute (e.g., perceiving rejection and conflict) and chronic (e.g., perceiving low support and closeness) stress, but not actual increases in these forms of stress. Together with the one prior study that used contextual stress interviews (Flynn & Rudolph, 2011), this suggests that ruminating about negative emotions, problems and stress may lead early adolescents to behave in ways that generate interpersonal stress.

When we explored the links between rumination and each domain of chronic stress, rumination predicted chronic interpersonal stress generation in adolescents’ parent-child relationships, romantic lives, and their peer social circle (though the latter effect was small and reached only marginal significance). Such findings are consistent with prior work demonstrating that rumination uniquely predicts increases in family emotional abuse and peer victimization (Shapero, Hamilton, et al., 2013), and further suggest that rumination predicts other aspects of chronic stress in these domains (e.g., parent-child relationship qualities, such as closeness and conflict resolution), as well as chronic stress in adolescents’ romantic lives. Surprisingly, however, rumination did not predict increases in close friendship chronic stress. Although counterintuitive, this finding aligns with prior work indicating that rumination was positively associated with self-reported friendship quality and closeness (e.g., A. J. Rose, 2002), as well as research examining co-rumination. Co-rumination is a dyadic process that is conceptually related to rumination and involves excessive discussion of problems and negative emotions within friendships (A. J. Rose, 2002). Co-rumination has also been positively associated with the quality and closeness of adolescents’ friendships (e.g., A. J. Rose, 2002), despite conferring risk for acute stress generation and internalizing symptoms (Hankin, Stone, & Wright, 2010). Thus, although speculative, it may be that ruminators are engaging in high levels of co-rumination in their close friendships, thereby enhancing the positive qualities (e.g., closeness, support, trust, self-disclosure) tapped by our measure of close friendship chronic stress. Indeed, rumination and co-rumination are positively correlated (A. J. Rose, 2002). Thus, future work examining links between rumination, co-rumination, and chronic stress generation may shed light on why rumination did not confer risk for the generation of chronic stress in girls’ close friendships.

Rumination and Stress Generation: The Role of ERS

The present findings suggest that ERS plays a role in the link between rumination and interpersonal stress generation. This finding is consistent with research demonstrating that the interpersonal consequences of rumination may render individuals at risk for generating stress. For example, prior work demonstrated that rumination leads to stress generation by decreasing social support satisfaction (Flynn et al., 2010), as well as by increasing the (perceived) degree and quality of peer communication (McLaughlin & Nolen-Hoeksema, 2012), which the authors posited may be because ruminators engage in clingy, dependent, and aggressive behavior. Expanding upon prior work focusing on acute interpersonal stress generation and peer victimization, the present findings suggest that rumination also contributes to chronic interpersonal stress generation via ERS. Interestingly, however, ERS only mediated the link between rumination and chronic stress in certain domains: romantic relationship/dating and peer social life (though both effects were only marginally significant). Although little is known about whether ERS may be particularly toxic to certain types of relationships, one study indicated that ERS predicted increases in acute stress (measured with a checklist) within undergraduates’ romantic relationships, but not within their friendships, roommate relationships, or family relationships (Shahar, Joiner, Zuroff, & Blatt, 2004). Similarly, a meta-analysis showed that the link between ERS and interpersonal rejection was stronger for romantic relationships than non-romantic relationships (Starr & Davila, 2008).

These findings, however, should be viewed tentatively given that rumination and ERS were measured concurrently. Future research that examines whether rumination predicts increases in ERS, which in turn predicts increases in subsequent interpersonal stress generation, is needed. Moreover, it may be particularly informative to examine these associations using daily diary methods (Eberhart & Hammen, 2009). Such research would permit the evaluation of a transactional model wherein rumination, ERS, and acute interpersonal dependent stress reciprocally influence one another over time. Furthermore, this approach may be useful in delineating factors that modify associations between rumination, ERS, and stress generation. For example, it may be that frequent use of problem solving or distraction buffer the effect of rumination on stress generation, as brief distraction and mindfulness interventions help youth disengage from a ruminative state (Hilt & Pollak, 2012); thus, these strategies may interrupt links between rumination, ERS, and interpersonal stress generation.

Specificity of Predictive Effects

Consistent with the stress generation model (Hammen, 1991), rumination did not predict acute independent (uncontrollable) stress, suggesting that adolescent girls who engage in high levels of rumination are not simply exposed to higher levels of stressful events, but instead contribute to their occurrence. The specificity of this effect is particularly important during adolescence, as many of the interpersonal events that adolescents experience, such as divorce or parental job loss, occur to their parents, independent of adolescents’ behavior (Harkness et al., 2010). This suggests that adolescents who engage in high levels of rumination actively contribute to increasing acute interpersonal stress in their lives, rather than simply inheriting a stressful environment created by their parents (e.g., Rice, Lewis, Harold, & Thapar, 2013). Moreover, rumination was not associated with non-interpersonal forms of acute and chronic stress. This contradicts the one prior study that used objective life stress interviews to address this question: Flynn and Rudolph (2011) showed that ineffective stress responses (including rumination) predicted increases in acute dependent non-interpersonal stress (e.g., academic events). This mixed and limited evidence highlights the need for future research examining whether rumination confers risk for non-interpersonal stress generation.

Role of Developmental Context and Gender

These findings may have been influenced by the early adolescent developmental stage of participants and may be specific to early adolescent girls. In line with this, Liu and Alloy (2010) posited that stress generation effects might intensify between early adolescence and adulthood, as individuals increasingly play a direct role in shaping their social experiences as cognitive vulnerabilities solidify. Given that the use of rumination increases from childhood through adolescence (Hampel & Petermann, 2005), rumination might play a lesser role in stress generation in children. Furthermore, Liu and Alloy (2010) posited that stress generation patterns may be most evident in family contexts in childhood and in peer contexts during adolescence, reflecting the increasing influence of peer relationships coupled with the decreasing focus of parental relationships characteristic of adolescence (e.g., Steinberg, 1987). In the present study, the effect of rumination was most evident in peer and family contexts, perhaps reflecting this transition. Moreover, the present findings may be specific to early adolescent girls, who exhibit higher levels of rumination (Hampel & Petermann, 2005) and generate higher levels of interpersonal stress (Rudolph & Flynn, 2007; Rudolph & Hammen, 1999), as compared with their male counterparts. Thus, future research exploring whether these findings generalize to other developmental periods and to early adolescent boys is needed.

Strengths and Limitations

The results of this study are strengthened by our use of contextual objective stress interviews with adolescents and their mothers, as well as diagnostic interviews to account for current and lifetime history of depressive symptoms. In addition, building upon prior work, we considered both acute and chronic forms of stress generation, allowing us to examine whether rumination uniquely predicted—directly and indirectly via ERS—each form of stress generation. Moreover, by examining the unique effect of rumination on different domains of chronic interpersonal stress, we demonstrated that rumination was particularly toxic for adolescents’ romantic lives and the parent-child relationship.

There are also limitations. First, the sample was self-selected and composed of early adolescent girls who were mostly Caucasian. Thus, it will be important to replicate these findings in a representative sample of early adolescents from diverse racial and ethnic backgrounds. Second, our two-wave longitudinal design allowed us to examine whether rumination predicted prospective increases in stress generation, but rumination and ERS were measured concurrently. Thus, future work is needed to fully test ERS as a mediator. Furthermore, given the design was correlational, we cannot rule out the possibility that a third variable (e.g., neuroticism, genetic factors) contributed to the observed associations. Third, power may have been limited due to the small magnitude of the stress generation effects and the modest sample size. Furthermore, given that the direct and indirect effects for Model 1 were significant, but some of the domain findings (particularly the indirect effects) reached only marginal significance, it is likely that power may have been further limited in these analyses. This may have been due to lower variation within each chronic stress domain, as compared with that for the chronic interpersonal and non-interpersonal stress composites, as well as low levels of chronic stress. Thus, replication in a larger high-risk sample of individuals experiencing varied levels of chronic stress is needed. Fourth, we could not examine whether rumination directly and indirectly (via ERS) leads to the generation of acute stressful events in certain domains (e.g., romantic or social events) or to the generation of specific types of interpersonal events. For example, consistent with the link between ERS and rejection (Starr & Davila, 2008), it may be that high ruminators generate particularly high levels of rejection-related events in their relationships by engaging in high levels of ERS. Fifth, we did not assess chronic independent stress (e.g., financial difficulties, neighborhood conditions), and thus, we could not examine if the effect of rumination on chronic stress was specific to chronic stress generation, or if rumination also increased exposure to chronic independent stress.

Finally, we relied on self-report measures to assess rumination and ERS, and thus, their association may be inflated by shared method variance. In addition, although prior work indicates that self-reported and observer-rated ERS are moderately correlated (Joiner & Metalsky, 2001), future laboratory studies that include observational assessments may be useful. For example, adolescents’ ERS behavior with a close other (i.e., parent, peer, romantic partner) could be measured prior to and following a rumination induction. This type of investigation may also elucidate how close others evoke, respond to, and maintain adolescents’ ERS. Such studies may also reveal how, and under what circumstances, rumination contributes to ERS. For example, it may be that feelings of uncertainty (Nolen-Hoeksema, 2000) or negative mood (Joiner, 2000) mediate and/or enhance the rumination-ERS link. Shedding light on whether these emotions fuel or increase risk for ERS will identify additional targets for prevention.

Conclusion

The findings of the present study suggest that early adolescent girls who engage in high levels of rumination create stressful environments. Such findings support the use of prevention approaches targeting rumination—such as school-based mindfulness programs (Mendelson et al., 2010)—during early adolescence, prior to the onset of most forms of psychopathology that are associated with rumination (Rohde et al., 2009). Along with targeting rumination, the present findings underscore the importance of targeting adolescents’ maladaptive interpersonal behaviors through the development of adaptive interpersonal skills (K. Rose, Hawes, & Hunt, 2014). Indeed, targeting rumination and maladaptive interpersonal behaviors may not only prevent early adolescent girls from responding to stress and problems in their lives in a ruminative manner but also, more importantly, prevent them from generating the stress and problems in the first place.

Footnotes

Acknowledgements

The authors thank the families who generously gave their time to participate in this project as well as the staff of the Williams College Youth Emotion Center.

Authors’ Note

A portion of this research was presented at the 2014 annual convention of the Association for Cognitive and Behavioral Therapies, Philadelphia, PA.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

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 institutional funds from Williams College. S.W. was supported by the National Institute of Drug Abuse of the National Institutes of Health under Award Number K01DA037280.

Notes

Author Biographies

Catherine B. Stroud is an assistant professor of psychology at Williams College. Her research examines stress, depression, and interpersonal relationships.

Effua E. Sosoo is a graduate student in the clinical psychology PhD program at the University of North Carolina at Chapel Hill. She was formerly a student at Williams College. Her research examines how racism and discrimination impacts the mental and physical well-being of Black college students.

Sylia Wilson is a research assistant professor in the Department of Psychology at the University of Minnesota. Her research focuses on risk factors for and consequences of psychopathology in childhood, adolescence, and adulthood.

References

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

Burwell

R. A.

Shirk

S. R.

(2007). Subtypes of rumination in adolescence: Associations between brooding, reflection, depressive symptoms, and coping. Journal of Clinical Child and Adolescent Psychology, 36, 56-65.

Eberhart

N. K.

Hammen

(2009). Interpersonal predictors of stress generation. Personality and Social Psychology Bulletin, 35, 544-556.

Enders

C. K.

(2013). Dealing with missing data in developmental research. Child Development Perspectives, 7, 27-31. doi:10.1111/cdep.12008

Flynn

Kecmanovic

Alloy

L. B.

(2010). An examination of integrated cognitive-interpersonal vulnerability to depression: The role of rumination, perceived social support, and interpersonal stress generation. Cognitive Therapy and Research, 34, 456-466.

Flynn

Rudolph

K. D.

(2011). Stress generation and adolescent depression: Contribution of interpersonal stress responses. Journal of Abnormal Child Psychology, 39, 1187-1198.

Hamilton

J. L.

Stange

J. P.

Shapero

B. G.

Connolly

S. L.

Abramson

L. Y.

Alloy

L. B.

(2013). Cognitive vulnerabilities as predictors of stress generation in early adolescence: Pathway to depressive symptoms. Journal of Abnormal Child Psychology, 41, 1027-1039.

Hammen

(1991). Generation of stress in the course of unipolar depression. Journal of Abnormal Psychology, 100, 555-561.

Hammen

(2006). Stress generation in depression: Reflections on origins, research, and future directions. Journal of Clinical Psychology, 62, 1065-1082.

10.

Hammen

Adrian

Gordon

Burge

Jaenicke

Hiroto

(1987). Children of depressed mothers: Maternal strain and symptom predictors of dysfunction. Journal of Abnormal Psychology, 96, 190-198.

11.

Hampel

Petermann

(2005). Age and gender effects on coping in children and adolescents. Journal of Youth and Adolescence, 34, 73-83.

12.

Hankin

B. L.

Stone

Wright

P. A.

(2010). Corumination, interpersonal stress generation, and internalizing symptoms: Accumulating effects and transactional influences in a multiwave study of adolescents. Development and Psychopathology, 22, 217-235.

13.

Harkness

K. L.

Alavi

Monroe

S. M.

Slavich

G. M.

Gotlib

I. H.

Bagby

R. M.

(2010). Gender differences in life events prior to onset of major depressive disorder: The moderating effect of age. Journal of Abnormal Psychology, 119, 791-803.

14.

Hilt

L. M.

McLaughlin

K. A.

Nolen-Hoeksema

(2010). Examination of the response styles theory in a community sample of young adolescents. Journal of Abnormal Child Psychology, 38, 545-556.

15.

Hilt

L. M.

Pollak

S. D.

(2012). Getting out of rumination: Comparison of three brief interventions in a sample of youth. Journal of Abnormal Child Psychology, 40, 1157-1165.

16.

Joiner

T. E.

(2000). Depression’s vicious scree: Self-propagating and erosive processes in depression chronicity. Clinical Psychology: Science and Practice, 7, 203-218.

17.

Joiner

T. E.

Alfano

M. S.

Metalsky

G. I.

(1992). When depression breeds contempt: Reassurance seeking, self-esteem, and rejection of depressed college students by their roommates. Journal of Abnormal Psychology, 101, 165-173.

18.

Joiner

T. E.

Metalsky

G. I.

(2001). Excessive reassurance seeking: Delineating a risk factor involved in the development of depressive symptoms. Psychological Science, 12, 371-378.

19.

Kaufman

Birmaher

Brent

Rau

Flynn

Moreci

. . . Ryan

(1997). Schedule for Affective Disorders and Schizophrenia for School-Age Children- Present and Lifetime version (K-SADS-PL): Initial reliability and validity data. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 980-987.

20.

Liu

R. T.

Alloy

L. B.

(2010). Stress generation in depression: A systematic review of the empirical literature and recommendations for future study. Clinical Psychology Review, 30, 582-593.

21.

MacKinnon

D. P.

(2008). Introduction to statistical mediation analysis. New York, NY: Lawrence Erlbaum.

22.

MacKinnon

D. P.

Lockwood

C. M.

Williams

(2004). Confidence limits for the indirect effect: Distribution of the product and resampling methods. Multivariate Behavioral Research, 39, 99-128.

23.

McLaughlin

K. A.

Nolen-Hoeksema

(2012). Interpersonal stress generation as a mechanism linking rumination to internalizing symptoms in early adolescents. Journal of Clinical Child & Adolescent Psychology, 41, 584-597.

24.

Mendelson

Greenberg

M. T.

Dariotis

J. K.

Gould

L. F.

Rhoades

B. L.

Leaf

P. J.

(2010). Feasibility and preliminary outcomes of a school-based mindfulness intervention for urban youth. Journal of Abnormal Child Psychology, 38, 985-994.

25.

Muthen

L. K.

Muthen

B. O.

(1998-2012). Mplus user’s guide (7th ed.). Los Angeles, CA: Author.

26.

Nolen-Hoeksema

(1991). Responses to depression and their effects on the duration of depressive episodes. Journal of Abnormal Psychology, 100, 569-582.

27.

Nolen-Hoeksema

(2000). Further evidence for the role of psychosocial factors in depression chronicity. Clinical Psychology: Science and Practice, 7, 224-227.

28.

Nolen-Hoeksema

Davis

C. G.

(1999). “Thanks for sharing that”: Ruminators and their social support networks. Journal of Personality and Social Psychology, 77, 801-814.

29.

Nolen-Hoeksema

Morrow

(1991). A prospective study of depression and posttraumatic stress symptoms after a natural disaster: The 1989 Loma Prieta earthquake. Journal of Personality and Social Psychology, 61, 115-121.

30.

Nolen-Hoeksema

Wisco

B. E.

Lyubomirsky

(2008). Rethinking rumination. Perspectives on Psychological Science, 3, 400-424.

31.

Petersen

A. C.

Crockett

Richards

Boxer

(1988). A self-report measure of pubertal status: Reliability, validity, and initial norms. Journal of Youth and Adolescence, 17, 117-133. doi:10.1007/bf01537962

32.

Rice

Lewis

Harold

G. T.

Thapar

(2013). Examining the role of passive gene-environment correlation in childhood depression using a novel genetically sensitive design. Development and Psychopathology, 25, 37-50.

33.

Rohde

Beevers

C. G.

Stice

O’Neil

(2009). Major and minor depression in female adolescents: Onset, course, symptom presentation, and demographic associations. Journal of Clinical Psychology, 65, 1339-1349.

34.

Rose

A. J.

(2002). Co-rumination in the friendships of girls and boys. Child Development, 73, 1830-1843. doi:10.1111/1467-8624.00509

35.

Rose

Hawes

D. J.

Hunt

C. J.

(2014). Randomized controlled trial of a friendship skills intervention on adolescent depressive symptoms. Journal of Consulting and Clinical Psychology, 82, 510-520.

36.

Rudolph

K. D.

(2008). Developmental influences on interpersonal stress generation in depressed youth. Journal of Abnormal Psychology, 117, 673-679.

37.

Rudolph

K. D.

Flynn

(2007). Childhood adversity and youth depression: Influence of gender and pubertal status. Development and Psychopathology, 19, 497-521.

38.

Rudolph

K. D.

Hammen

(1999). Age and gender as determinants of stress exposure, generation, and reactions in youngsters: A transactional perspective. Child Development, 70, 660-677.

39.

Rudolph

K. D.

Hammen

Burge

Lindberg

Herzberg

D. S.

Daley

S. E.

(2000). Toward an interpersonal life-stress model of depression: The developmental context of stress generation. Development and Psychopathology, 12, 215-234.

40.

Shahar

Joiner

T. E.

Zuroff

D. C.

Blatt

S. J.

(2004). Personality, interpersonal behavior, and depression: Co-existence of stress-specific moderating and mediating effects. Personality and Individual Differences, 36, 1583-1596.

41.

Shapero

B. G.

Hamilton

J. L.

Liu

R. T.

Abramson

L. Y.

Alloy

L. B.

(2013). Internalizing symptoms and rumination: The prospective prediction of familial and peer emotional victimization experiences during adolescence. Journal of Adolescence, 36, 1067-1076.

42.

Shapero

B. G.

Hankin

B. L.

Barrocas

A. I.

(2013). Stress generation and exposure in a multi-wave study of adolescents: Transactional processes and sex differences. Journal of Social and Clinical Psychology, 32, 989-1012.

43.

Simons

A. D.

Angell

K. L.

Monroe

S. M.

Thase

M. E.

(1993). Cognition and life stress in depression: Cognitive factors and the definition, rating, and generation of negative life events. Journal of Abnormal Psychology, 102, 584-591.

44.

Starr

L. R.

Davila

(2008). Excessive reassurance seeking, depression, and interpersonal rejection: A meta-analytic review. Journal of Abnormal Psychology, 117, 762-775.

45.

Steinberg

(1987). Impact of puberty on family relations: Effects of pubertal status and pubertal timing. Developmental Psychology, 23, 451-460.

46.

Stroud

C. B.

Davila

(2008). Pubertal timing and depressive symptoms in early adolescents: The roles of romantic competence and romantic experiences. Journal of Youth and Adolescence, 37, 953-966. doi:10.1007/s10964-008-9292-9

47.

Treynor

Gonzalez

Nolen-Hoeksema

(2003). Rumination reconsidered: A psychometric analysis. Cognitive Therapy and Research, 27, 247-259.

48.

Weinstock

L. M.

Whisman

M. A.

(2007). Rumination and excessive reassurance-seeking in depression: A cognitive-interpersonal integration. Cognitive Therapy and Research, 31(3), 333-342.