Rumination Versus Distraction: Dyadic Implementation Eliminates the Response Manipulation Emotion Regulation Effect

The Response Manipulation Task (RMT) is a popular laboratory protocol for inducing rumination and distraction. Across published studies of dysphoric participants who undergo negative mood inductions when no other people are present, only once has the RMT induction failed in its purpose. The present experiment tested the robustness of the RMT under dyadic conditions (N = 135 pairs of same sex friends). When administered in the presence of another person, the RMT showed no differential effects on subsequent negative mood or state rumination. The negative mood induction successfully induced negative mood; the effect of the manipulation did not depend on depressive symptoms; and the state rumination measure was reliable and valid. In light of this pattern of effects, nonsignificant findings on manipulation checks and substantive hypothesis tests are attributed to failure of the RMT to produce rumination and distraction under these specific study conditions. The Discussion explores constraints on the generalizability of the RMT effect due to the presence of others, including the influence of dyadic emotion regulation, interpersonal distress avoidance, and secure attachment relationships.

Introduction

Stress generation theory holds that depression typical behavior produces stress, particularly interpersonal stress [7; 11]. The cognitive and affective mechanisms underlying this process are not well understood [11]. Rumination has been proposed as
a candidate mechanism, possibly as a precursor to stress generating behaviors, such as excessive reassurance seeking and negative feedback seeking [8; 24]. Rumination has been identified as a mechanism behind stress generation in adolescents, with high levels of rumination associated with subsequent peer victimization [17]. We aimed to compare the interpersonal consequences of rumination to distraction, an emotion regulation strategy that reduces negative affect more effectively than rumination [23]. We used an established experimental manipulation, the response manipulation task [RMT; 13; 16], to induce rumination and distraction in friend dyads. We hypothesized that rumination would lead to interpersonal difficulties, as operationalized by lower friend-reported rapport, willingness to affiliate, and friend worth. Because the RMT has not previously been used with dyadic samples, we also explore the effectiveness of the RMT for inducing differential negative mood in a large dyadic sample.

The Response Manipulation Task

The RMT was designed to elicit different emotion regulation strategies by directing participants’ thought processes toward either rumination or distraction. The RMT has been used frequently, particularly in tests of the response styles theory. Response styles theory suggests that rumination is a harmful cognitive affective process that promotes the development and maintenance not only of depressive symptoms but also of a host of other forms of psychopathology [23; 26].

Effects of the RMT

Initial studies of response manipulation established that participants in distraction conditions experience declines in depressed mood induced sadness, whereas participants in rumination conditions maintain a relatively stable level of sadness [19]. The robust effect of the RMT on sad or depressed mood supported later use of differential mood change (distraction vs. rumination) as a manipulation check. Indeed, in subsequent studies of sufficiently dysphoric samples, the RMT nearly always produced similar results [13; 14; 22; 25; 31].

Under what conditions has the RMT failed?

In only one previous study has the RMT failed to produce statistically significant differential mood change from pre to post task [10]. In this study there were 29 and 31 participants in the rumination and distraction conditions, respectively, and despite the small sample size the RMT effect was very nearly significant (p < .06), was in the expected direction, and a similar sized effect would have been statistically significant with even
a slightly larger sample. Other studies have shown the RMT effect to hold only for participants experiencing high levels of dysphoria or depressed mood. Specifically, past studies have split their samples into “dysphoric” and “nondysphoric” or “depressed” and “nondepressed” groups on the basis of either structured diagnostic interviews or, more often, self report depressive symptom measures. Across seven reports encompassing 16 separate samples, the RMT has successfully produced differential mood effects in every dysphoric sample and failed to produce differential mood effects in every nondysphoric sample [12–14; 16; 25; 30; 31]. In one additional study, the RMT produced similar effects across dysphoric and nondysphoric samples [20]. In no published study to date has the manipulation produced nonsignificant effects in a dysphoric sample.

The Present Study

The present study used the RMT twice as part of a within subjects experiment addressing the interpersonal consequences of emotion regulation. This study is similar to previous studies using the RMT, with the key difference that participants in this study were dyads – specifically, they were same sex pairs of friends. In addition to the standard manipulation check for differential effects on negative mood, this study also assessed state rumination following the RMT.

Methods

Participants

135 participants were recruited either from psychology courses at the University of Notre Dame or from the South Bend, Indiana community using flyers and other community announcements (e.g., email newsletters). Each participant brought a same sex friend to the experiment (N = 270). However, only one participant in each pair completed the RMT. Two participants were excluded for being under 18 years of age, and one additional participant was excluded for blatant rushing through questionnaires, resulting in a final sample of 267 individuals, 133 of whom completed the RMT. The majority of the sample was female (77%). Participants ranged in age from 18 to 61 years (M = 20.39, SD = 5.45). The majority were white (87%), with 13% Asian, 5% African American, 2% Native Hawaiian or Pacific Islander, and 1% American Indian. A minority of the sample identified as Hispanic or Latino (12%).

Materials and measures

Negative mood induction. It is necessary for participants to be experiencing at least moderate levels of negative emotion so that a difference in the effectiveness of rumination versus distraction for reducing negative emotion can be observed [26]. Video clips with
a negative focus have been used previously to induce negative mood states prior to rumination [10; 32]. For this study, two video clips were used to induce negative feelings (e.g., sadness, anger, fear). The first was a 5-minute animated scene depicting one couple’s tragic lives (Pixar/Disney’s “Up”), and the second was a 9-minute compilation of news coverage of the September 11 attacks on the World Trade Center. To our knowledge, these videos have not been previously used in research, and both were chosen because they were shown to induce negative emotions in a pilot test.

RMT [15; 19]. The RMT comprises separate rumination and distraction tasks that each consist of 45 items on which participants are directed to focus their attention for exactly 8 minutes. Participants were handed a paper packet with the following instructions to read: “For the next few minutes, try your best to focus your attention on each of the ideas on the following pages. Read each item slowly and silently to yourself. As you read the items, use your imagination and concentration to focus your mind on each of the ideas. Spend a few moments visualizing and concentrating on each item. Please continue until the experimenter returns.” The experimenters then asked participants if they had any questions, then stated, “It is important that you do not mention your task or any of its contents to your friend.” The rumination condition, includes a series of items related to thoughts, feelings, and self concept (e.g., “Think about whether you feel stressed right now”). In the distraction condition, participants are asked to attend to a series of externally focused items not related to their current mood (e.g., “Think about the shape of the torch on the Statue of Liberty”).

Perseverative Thinking Questionnaire – State Version [PTQ-S; 30]. The PTQ-S is a 15 item self report scale measuring state rumination. Participants are asked to describe their thinking process during a specified period of time (e.g., in the past 5 minutes) by rating the frequency of occurrences of thoughts such as “I kept thinking about the same issue all the time” on a 0 to 4 scale. The PTQ-S has been shown to reflect changes in state rumination in response to a laboratory manipulation [33]. Therefore, the scale is expected to reflect within subjects fluctuations in state rumination following the RMT. In the present sample, PTQ-S Cronbach’s alphas were excellent, being .91 and .93 at the first and second time points, respectively. Data were approximately normal at the first (M = 2.58, SD = .72, n = 133) and second (M = 2.65, SD = .78, n = 132) time points.

Mood rating scales. Participants were asked to rate their mood using seven 5 point scales (0 = “not at all”, 4 = “very”). Ratings assigned to “sad” and “depressed” were averaged to form a negative mood variable. The other five scales (e.g., “impatient”, “energized”) were filler items designed to distract from the true purpose of the scales.
A similar measure has been used to assess mood states in a previous study on rumination, in which negative mood was shown to decrease as time elapsed following a negative mood induction and to be higher following a rumination than a distraction condition [32]. Cronbach’s alphas for the two item negative mood scale were fair, ranging from .72 to .77 across three administrations. Negative mood was positively skewed at the first (M = 1.41, SD = .67, n = 132), second (M = 1.87, SD = .84, n = 133), and third (M = 2.10, SD = .88, n = 131) administrations.

Center for Epidemiologic Studies Depression Scale (CES-D; 23]. The CES‑D assesses current symptoms of depression by self report. Responses range on a 0 to 3 scale, with higher scores indicating more severe depression. Internal consistency was good (Cronbach’s alpha = .87). The scale also possesses good convergent validity, being able to discriminate between psychiatric inpatients and the general population, showing decreases as a result of treatment, and correlating highly with other measures of depression [27]. In the present sample, 19.8% of participants met the cutoff for “possible depression” (scoring 16 or greater), consistent with a previous estimate of 21% in the general population [27].

Rapport scales [1]. Participants rated their rapport with their friend during the discussions using three 0 to 10 items. Scores on this scale were shown to be lower in dyads where one partner suppressed emotions than in control dyads [1]. In the present sample, Cronbach’s alphas were .76 and .78 at the first and second time points, respectively, demonstrating fair internal consistency. Rapport was negatively skewed at the first
(M = 6.90, SD = 1.57, n = 133) and second (M = 7.18, SD = 1.73, n = 132) time points.

Willingness to affiliate scales. Five items were written to assess participants’ willingness to maintain a relationship with their friend in the future. Participants rated items on 4 point scales ranging from “strongly disagree” to “strongly agree”, with higher total scores representing higher willingness to affiliate with friends. Internal consistency was fair (Cronbach’s alpha = .70), and scores were negatively skewed (M = 3.28, SD = .45,
n = 133).

Evaluation of friend on revision of Rosenberg Self Esteem Questionnaire (R-SEQ [28; 29]. The R-SEQ measures global self worth and has been adapted to measure perceptions of roommates’ worth [29]. Participants rate 10 items on a 4-point scale, with higher scores representing more negative impressions of roommates’ worth. The scale correlates strongly with measures of rejection, and scores correlate with depression among individuals high in reassurance seeking [9]. Items were reworded to refer to friends instead of roommates for the present study. Internal consistency was fair (Cronbach’s alpha = .75), and scores were negatively skewed (M = 3.76, SD = .25, n = 132).

Procedure

As a cover story, participants were told the study focused on “reactions to adverse events in friend pairs”. After providing informed consent, participants independently completed a battery of questionnaires in separate, private rooms, including demographic information, the mood rating scales, and the CES-D. One participant in each pair was assigned randomly to the target group, which received the RMT, and the other to the partner group (hereafter referred to as the target and partner, respectively). The order of rumination and distraction tasks and the order of negative mood induction videos were randomized, and there was no significant association between the outcomes of these two randomizations, χ²(1, 133) = 2.72, p = .12. Both participants watched the first negative mood induction video together, as they were later asked to discuss their reactions to it. Next, they were separated into different rooms. The target participant received either the rumination or distraction RMT, while the partner was asked to wait until the experimenter returned. After the 8 minutes allocated to complete the RMT had passed, participants completed the mood rating scales. As part of the larger study, participants spent the next 10 minutes discussing with each other their thoughts and feelings in response to the video they had watched earlier. Following the discussion, participants were led into separate rooms again to complete the PTQ‑S, rapport scales, willingness to affiliate scales, and
R-SEQ. Participants each received $10-20, except for psychology students who instead elected to receive credits that would increase their course grade.

Participants next reconvened to watch the second video together and then were separated once more. Target participants completed whichever RMT they had not completed following the first video, and partners waited again. Following this task, both participants rated their moods. Next, both participants again took part in a 10-minute discussion about the second video. After the second discussion, participants were separated to complete the PTQ-S and rapport scales, then debriefed and awarded credit or payment. See fig. 1 for a diagram detailing the flow of the experiment.

Results

All data analyses were conducted with SPSS version 24. Missing data were handled using listwise deletion. To evaluate whether randomization resulted in approximately equal groups, we compared participants who completed the rumination task first with the participants who completed the distraction task first on all measures assessed in the baseline questionnaire. No differences were detected between two groups (Table 1).

Figure 1. Experiment flow diagram

Notes. Parentheses indicate a counterbalanced randomization. CES-D – Center for Epidemiological Studies Depression Scale. PTQ-S – Perseverative Thinking Questionnaire – State Version. R‑SEQ – Evaluation of friend on revision of Rosenberg Self-Esteem Questionnaire.

Table 1

Comparison of means of target participant scores on demographic
and baseline measures by randomly assigned order of conditions

Notes. Standard deviations are shown in parentheses. Degrees of freedom for t tests were 131. CES-D – Center for Epidemiological Studies Depression Scale, total scale score corrected for missing items. ^a – z scores are reported for differences in proportions. No group differences were significant (all p’s > .05).

Manipulation checks

Because previous research has found that distraction is more effective than rumination at reducing negative emotional experience following a negative mood induction 12], targets’ negative mood should be lower immediately following the distraction task than immediately following the rumination task. A 2 (condition: rumination vs. distraction) × 2 (task order: rumination distraction vs. distraction rumination) repeated measures ANOVA was conducted to test for the expected main effect of condition on negative mood. Contrary to expectations, no significant main effect of condition on negative mood following rumination or distraction was detected, F(1, 129) = 1.58, p = .21, η_p² = .01.

Second, target participants should report higher state rumination during discussions that follow rumination than during discussions that follow distraction. Another 2×2 repeated measures ANOVA was conducted with PTQ-S state rumination scores as the dependent variable. No significant main effect of condition on PTQ-S state rumination during discussions was detected, F(1, 130) = .37, p = .55, η_p² = .00. Although this result may suggest that the manipulation failed to affect actual rumination during discussions, another possibility is that the PTQ-S lacks validity to detect induced differences in state rumination. However, target participants’ PTQ-S scores correlated positively with CES-D depression scores and negative mood before corresponding discussions, providing evidence for the scale’s construct validity in relation to these two different but theoretically related constructs (Table 2).

Table 2

Bivariate correlations between target PTQ-S state rumination scores
and related measures

Notes. Ns range from 130 to 133. The PTQ-S was administered following each discussion and refers to participants’ state rumination during each discussion. “Rumination discussion” and “distraction discussion” refer to the response manipulation task condition preceding the discussions. CES-D – Center for Epidemiological Studies Depression Scale. PTQ-S – Perseverative Thinking Questionnaire – State Version. * – p < .05, ** – p < .01, *** – p < .001.

To investigate whether the negative mood induction had increased targets’ negative mood from baseline to the period following the RMT, we conducted a 2 (video: September 11^th vs. tragic couple) × 2 (time: baseline vs. after RMT) repeated measures ANOVA.
A significant interaction was detected between video and time, F(1, 130) = 9.19, p = .003, η_p² = .07, suggesting changes in negative mood depended on which video participants watched. However, simple effects analyses revealed that negative mood increased significantly from baseline to the period following RMT for both videos (Figure 2); this increase was larger following the September 11^th video, F(1, 130) = 380.28, p < .001, d = .80, than following the tragic couple video, F(1, 130) = 380.28, p < .001, d = .39. Because negative mood was not measured immediately after the negative mood induction, these effect sizes do not necessarily represent the actual mood changes produced by the induction.

Figure 2. Negative mood induction effects by video

Notes. Video: black line – September 11^th, gray line – Tragic Couple. “After RMT” data are from the first half of the experiment only. Error bars represent standard error of the mean. RMT – response manipulation task.

Moderating effect of depressive symptoms

To examine the possibility that the manipulation only altered mood in depressed participants, CES-D depression scores were tested as a moderator in a one way ANCOVA testing the effects of condition (rumination vs. distraction) on negative mood following the RMT. There was no interaction between CES-D scores and condition, F(1, 129) = .89, p = .35, η_p² = .00, suggesting depressive symptoms were irrelevant to the success of the manipulation. Furthermore, CES-D scores did not depend on the order in which participants watched videos, t(123.04) = -.57, p = .57, d = -.10, ruling out this potential confound. In another analysis, only those participants who were above the recommended CES-D cutoff score of 16 or greater for possible depression [27] were included, resulting in a sample of 27 participants. Even after allowing for a liberal type I error rate (α = .10) to accommodate the small sample size, there was no significant within subjects difference in negative mood following the rumination versus the distraction task among possibly depressed participants, t(27) = 1.26, p = .22, d_z = .24.

Test of study hypotheses

The study’s primary hypotheses were that experimentally induced rumination would be followed by lower partner rated scores on measures of rapport, willingness to affiliate, and friend worth than would experimentally induced distraction. A two tailed repeated measures t test revealed no significant effects of condition on rapport, t(132) = -.66, p = .51, d_z = -.06. Two independent samples two tailed t tests revealed no significant effects of condition on willingness to affiliate, t(132) = ‑1.30, p = .20, d = -.22, or R-SEQ friend worth, t(122.52) = 1.63, p = .11, d = .28. In all three cases, effect sizes were small by Cohen’s [3] conventional standards. Adding video order to these three models did not affect the finding that condition was a non-significant predictor of each outcome.

Discussion

In this within subjects dyadic experiment, target participants completed the well established RMT on two occasions. Both an established manipulation check based on change in mood ratings and a unique manipulation check based on state rumination ratings failed to show any effect of the RMT. Owing to the failed manipulation checks, the main study hypotheses could not be tested adequately. Planned hypothesis tests returned non significant results. Because this manipulation has only very infrequently failed to produce expected effects, follow up analyses were conducted to shed light on the cause of this surprising finding.

Previous studies have found that the RMT only induces differential moods in dysphoric or depressed participants [13; 25; 31]. Although the mood induction in the current study appears to have induced dysphoria, one possibility was that not enough participants were actively depressed for the RMT to create a difference in negative mood between the rumination and distraction conditions. However, follow up analyses showed that depressive symptoms did not moderate the effect of the manipulation on negative mood and that the manipulation failed to produce differential moods even in participants who met an established CES‑D depression cutoff. In no previous study has the RMT failed to produce differences in negative mood in a depressed sample.

Before concluding manipulation failure, we must rule out the possibility that our manipulation check measures were themselves not valid. The negative mood manipulation check we used has been used previously to capture differences in negative mood following the RMT [32]. This manipulation check also closely resembles other commonly used negative mood checks administered following the RMT [13; 19; 31]. The second manipulation check found that PTQ-S state rumination did not differ between rumination and distraction tasks. Although the PTQ‑S has not been used previously as a manipulation check, the RMT was created precisely to induce state rumination [19]. Therefore, if the PTQ-S measures state rumination reliably and validly, it is a suitable manipulation check. Previous research has shown that the PTQ-S captures variability in state rumination following a different rumination versus control manipulation [33]. In our sample, the PTQ-S showed excellent reliability and convergent validity with depressive symptom and negative mood measures. Thus, two independent manipulation checks, both supported by validity evidence, failed to show expected effects of the RMT.

Having ruled out that our sample was not dysphoric enough for the RMT to show differential mood effects, and that the manipulation checks were not sound psychometrically, the most likely remaining explanation for our pattern of results is that the RMT failed to induce rumination and distraction in participants. Because this manipulation has rarely failed in published research, we next document the circumstances unique to this study that we propose led to the manipulation failure.

Circumstances under which the RMT may fail

Our procedures matched previous procedures closely, using the same wording, presented in paper packets of materials, and allotting exactly 8 minutes for participants to complete the task. The most obvious procedural difference between the present study and past studies is that pairs of friends participated simultaneously in the present study. Although friends were separated into different rooms before the target participant completed the RMT, watching the negative mood induction videos with a friend may have altered participants’ emotional responses. In an attempt to promote negative emotions in response to the videos and limit potential interpersonal emotion regulation, participants were instructed not to communicate during, or share any reactions immediately following, the videos. However, watching a video with a friend, even in silence, may be enough to change the way participants process emotions in response to the video. At least one study has shown that participants avoid strong emotions when they know they will interact with another participant [5]. This finding suggests participants in friend pairs may activate alternative emotion regulation strategies, such as avoidance, that prevent rumination during the RMT. Future work replicating and extending these findings with other dyadic relationships would not only provide guidance about using the RMT to manipulate emotion regulation, it would also inform interpersonal models of emotion regulation and depression with implications for how people regulate emotions in interpersonal situations [8; 11; 24].

Extensive research also shows that stress responses generally are buffered even by another person’s physical presence [2; 4]. From an attachment perspective, participants with secure attachment relationships with their study partner would be expected to experience lessened rumination and negative affect following a negative mood induction, whereas participants with anxious attachment relationships with their partner may experience heightened rumination and negative affect [18]. Unfortunately, data on interpersonal attachment were not collected. Nevertheless, it is possible that, although participants still reported increased negative mood following negative mood induction, exposure to the physical presence of their friend overrode or changed their emotion regulatory response so that the RMT had little effect on participants’ emotion regulation strategies, particularly rumination.

The results of the present study suggest the RMT may not produce expected effects in dyads. Because this is the first study to use the RMT in a dyadic sample, it is currently unknown whether the RMT would have similarly small effects in other types of dyads
(e.g., spouse pairs, stranger dyads). The question of whether the RMT failed due to the dyadic sample alone, due to other unmeasured relational factors (e.g., secure attachment), or due to another factor such as watching the negative mood induction video together is left for future research. It is possible that the literature on the RMT suffers from a “file drawer” problem, in which non significant results are not published [6; 21]. Although such an effect is difficult to demonstrate, the present manuscript aims to weaken any file drawer effects in the RMT literature.

Limitations

The present study was limited in a few ways. First, the depressed subsample was small, which lowered power for detecting RMT effects in depressed versus nondepressed participants. Second, we used novel mood induction videos, which limits our ability to compare our results directly to those of previous RMT studies using different mood inductions. Third, some of our measures (e.g., negative mood) were only fairly internally consistent (.70 ≤ α < .80). Fourth, our sample was majority white, young, and female, which may limit the generalizability of our findings to diverse populations.

1. Butler E.A., Egloff B., Wilhelm F.H. et al. The social consequences of expressive suppression. Emotion, 2003. Vol. 1, no. 3, pp. 48–67. DOI: 10.1037/1528-3542.3.1.48
2. Coan J.A., Schaefer H.S., Davidson R.J. Lending a hand: Social regulation of the neural response to threat. Psychological Science, 2006. Vol. 12, no. 17, pp. 1032–1039. DOI: 10.1111/j.1467-9280.2006.01832.x
3. Cohen J. A power primer. Psychological Bulletin, 1992. Vol. 1, no. 112, pp. 155–159. DOI: 10.1037/0033-2909.112.1.155
4. DeVries A.C., Glasper E.R., Detillion C.E. Social modulation of stress responses. Physiology and Behavior, 2003. Vol. 3, no. 79, pp. 399–407. DOI: 10.1016/S0031-9384(03)00152-5
5. Erber R., Wegner D.M., Therriault N. On being cool and collected. Mood regulation in anticipation of social interaction. Journal of Personality and Social Psychology, 1996. Vol. 4, no. 70, pp. 757–766. DOI: 10.1037//0022-3514.70.4.757.
6. Fanelli D. “Positive” results increase down the hierarchy of the sciences. PLoS ONE, 2010. Vol. 4, no. 5, 10 p. DOI: 10.1371/journal.pone.0010068
7. Hammen C.L. Generation of stress in the course of unipolar depression. Journal of Abnormal Psychology, 1991. Vol. 4, no. 100, pp. 555–561. DOI:10.1037/0021-843X. 100.4.555
8. Joiner T.E. Depression’s vicious scree: Self-propagating and erosive processes in depression chronicity. Clinical Psychology: Science and Practice, 2000. Vol. 2, no. 7,
   pp. 203–218. DOI: 10.1093/clipsy/7.2.203
9. Joiner T.E., Alfano M.S., Metalsky G.I. When depression breeds contempt: Reassurance seeking, self-esteem, and rejection of depressed college students by their roommates. Journal of Abnormal Psychology, 1992. Vol. 1, no. 101, pp. 165–173. DOI: 10.1037/0021-843X.101.1.165
10. Joormann J., Siemer M. Memory accessibility, mood regulation, and dysphoria: Difficulties in repairing sad mood with happy memories? Journal of Abnormal Psychology, 2004. Vol. 2, no. 113, pp. 179–188. DOI: 10.1037/0021-843X.113.2.179
11. Liu R.T. Stress generation: Future directions and clinical implications. Clinical Psychology Review, 2013. Vol. 3, no. 33, pp. 406–416. DOI: 10.1016/j.cpr.2013.01.005
12. Lyubomirsky S., Tucker K.L., Caldwell N.D. et al. Why ruminators are poor problem solvers: Clues from the phenomenology of dysphoric rumination. Journal of Personality and Social Psychology, 1999. Vol. 5, no. 77, pp. 1041–1060. DOI: 10.1037//0022-3514. 77.5.1041
13. Lyubomirsky S., Caldwell N.D., Nolen-Hoeksema S. Effects of ruminative and distracting responses to depressed mood on retrieval of autobiographical memories. Journal of Personality and Social Psychology, 1998. Vol. 1, no. 75, pp. 166–177. DOI: 10.1037//0022-3514.75.1.166
14. Lyubomirsky S., Kasri F., Zehm K. Dysphoric rumination impairs concentration on academic tasks. Cognitive Therapy and Research, 2003. Vol. 3, no. 27, pp. 309–330. DOI: 10.1023/A:1023918517378
15. Lyubomirsky S., Nolen-Hoeksema S. Self-perpetuating properties of dysphoric rumination. Journal of Personality and Social Psychology, 1993. Vol. 2, no. 65, pp. 339–349. DOI: 10.1037/0022-3514.65.2.339
16. Lyubomirsky S., Nolen-Hoeksema S. Effects of self-focused rumination on negative thinking and interpersonal problem solving. Journal of Personality and Social Psychology, 1995. Vol. 1, no. 69, pp. 176–190. DOI: 10.1037/0022-3514.69.1.176
17. McLaughlin K.A., Nolen-Hoeksema S. Interpersonal stress generation as
    a mechanism linking rumination to internalizing symptoms in early adolescents. Journal of Clinical Child & Adolescent Psychology, 2012. Vol. 5, no. 41, pp. 584–597. DOI:10.1080/15374416.2012.704840
18. Mikulincer M., Shaver P.R., Pereg D. Attachment theory and affect regulation: The dynamics, development, and cognitive consequences of attachment-related strategies. Motivation and Emotion, 2003. Vol. 2, no. 27, pp. 77–102. DOI: 10.1023/A:1024515519160
19. Morrow J., Nolen-Hoeksema S. Effects of responses to depression on the remediation of depressive affect. Journal of Personality and Social Psychology, 1990. Vol. 3, no. 58, pp. 519–527. DOI: 10.1037/0022-3514.58.3.519
20. Moulds M.L., Kandris E., Williams A.D. The impact of rumination on memory for self-referent material. Memory, 2007. Vol. 8, no. 15, pp. 814–821. DOI: 10.1080/ 09658210701725831
21. Munafò M.R., Nosek B.A., Bishop D.V.M. et al. A manifesto for reproducible science. Nature Human Behaviour, 2017. Vol. 1, no. 1, pp. 1–9. DOI:10.1038/s41562-016-0021
22. Naumann E., Tuschen-Caffier B., Voderholzer U. et al. Rumination but not distraction increases eating-related symptoms in anorexia and bulimia nervosa. Journal of Abnormal Psychology, 2015. Vol. 2, no. 124, pp. 412–420. DOI: 10.1037/abn0000046
23. Nolen-Hoeksema S. Responses to depression and their effects on the duration of depressive episodes. Journal of Abnormal Psychology, 1991. Vol. 4, no. 100, pp. 569–582. DOI: 10.1037/0021-843X.100.4.569
24. Nolen-Hoeksema S. Further evidence for the role of psychosocial factors in depression chronicity. Clinical Psychology: Science and Practice, 2000. Vol. 2, no. 7,
    pp. 224–227. DOI: 10.1093/clipsy.7.2.224
25. Nolen-Hoeksema S., Morrow J. Effects of rumination and distraction on naturally occurring depressed mood. Cognition and Emotion, 1993. Vol. 6, no. 7, pp. 561–570. DOI: 10.1080/02699939308409206
26. Nolen-Hoeksema S., Wisco B.E., Lyubomirsky S. Rethinking rumination. Perspectives on Psychological Science, 2008. Vol. 5, no. 3, pp. 400–424. DOI: 10.1111/j.1745-6924.2008.00088.x
27. Radloff L.S. The CES-D Scale: A self-report depression scale for use in the general population. Applied Psychological Measurement, 1977. Vol. 1, no. 3, pp. 385–400. DOI: 10.1177/014662167700100306
28. Rosenberg M. Society and the adolescent self-image. Princeton, NJ: Princeton University Press, 1965. 338 p.
29. Swann W.B., Wenzlaff R.M., Krull D.S. et al. Allure of negative feedback: Self-verification strivings among depressed persons. Journal of Abnormal Psychology, 1992.
    Vol. 2, no. 101, pp. 293–306. DOI: 10.1037/0021-843X.101.2.293
30. Vickers K.S., Vogeltanz-Holm N.D. The effects of rumination and distraction tasks on psychophysiological responses and mood in dysphoric and nondysphoric individuals. Cognitive Therapy and Research, 2003. Vol. 3, no. 27, pp. 331–348. DOI: 10.1023/ A:1023970501448
31. Whitmer A.J., Gotlib I.H. Switching and backward inhibition in major depressive disorder: The role of rumination. Journal of Abnormal Psychology, 2012. Vol. 3, no. 121,
    pp. 570–578. DOI: 10.1037/a0027474
32. Yoon K.L., Joormann J. Is timing everything? Sequential effects of rumination and distraction on interpersonal problem solving. Cognitive Therapy and Research, 2012. Vol. 3, no. 36, pp. 165–172. DOI: 10.1007/s10608-010-9330-2
33. Zetsche U., Ehring T., Ehlers A. The effects of rumination on mood and intrusive memories after exposure to traumatic material: An experimental study. Journal of Behavior Therapy and Experimental Psychiatry, 2009. Vol. 4, no. 40, pp. 499–514. DOI: 10.1016/ j.jbtep.2009.07.001.