1
Self-kindness : a buffer for depression?
The effect of self-kindness boosting on negative affect and self-criticism in youth
Annebelle Meijers
10166629
Master Thesis Clinical Developmental Psychology
December 2017
Supervisors: Dr. E. Salemink & A. Hagen, MSc
External Supervisor: M. Wrzesien, PhD
Second assessor: Dr. B. van Bockstaele
University of Amsterdam
Faculty of Social and Behavioral Sciences
Department of Psychology
2
ABSTRACT
The construct of self-compassion has been receiving a growing level of interest in the last years, as research suggests its possibility to be an adaptive way of both regulating negative emotions and relating to the self. Increasing self-kindness, an important component of self-compassion, could thus help prevent or treat psychopathology like depression. However, the literature on self-kindness in youth populations is limited and several issues demand further research. The aim of the present study was to explore (1) whether we could increase or boost self-kindness to help lower self-criticism and negative affect in youth (2) whether implicit boosting of self-kindness had a greater effect than explicit boosting of self-kindness as an emotion regulation strategy, and (3) whether the differential effects of the emotion regulation strategies would be moderated by the level of depressive symptoms. This was tested with a non-clinical sample of youth (N = 183) in a negative mood inducing experiment in which participants were randomly assigned to one of three conditions. In contrast to prior set expectations, results did not show an effect of both explicit and implicit self-kindness boosting on experienced negative affect or self-criticism, nor a moderating effect of depressive symptom levels. Limitations, implications and future directions are discussed.
Keywords: Self-kindness, Emotion Regulation, Self-criticism, Depression, Youth
3
PREFACE
This study is part of a bigger research by Maja Wrzesien (postdoctoral research fellow at the University of Amsterdam) and is funded by the Marie Skłodowska-Curie Individual Research Grant PEACEFUL MIND (MSCA-IF -656333). The methods of this study have been authorized by the guidelines of the Ethics Committee of Child Development and Education department of the University of Amsterdam (ERB: 2016-CDE-7242).
I want to thank Maja Wrsezien for guiding me throughout the data collection with a lot of warmth and knowledge. Also a big thank you to Elske Salemink and Annelieke Hagen, for offering your guidance throughout the writing process. It was a pleasure working with you!
4 Table of Contents
Abstract 2
Preface 3
1. Introduction 5-11
1.1 Depression and self-kindness 5-6
1.2 Emotion regulation 6-8
1.3 Self-criticism 8
1.4 Empirical status of self-kindness as an emotion regulation strategy 9-11
2. Methods 11-17
2.1 Participants 11
2.2 Design and Setting 12
2.3 Materials 12-15
2.4 Procedure 15
2.5 Data Analysis 16
2.6 Descriptive statistics and treatment of the data 16-17
2.7 Manipulation of self-kindness boosting and negative mood induction 17-18
3. Results 18-20
3.1 Self-kindness boosting effect on experienced negative affect 18 3.2 Self-kindness boosting effect on self-criticism state 19
3.3 Moderation of depression levels 19-20
4. Discussion & Conclusion 20-22
References 23-27
Appendix A – Tables of mean and standard deviation scores 28-29 Appendix B – Control questions debriefing 29
5
1. INTRODUCTION
1.1 Depression and self-kindness
Depression is one of the most occurring mental disorders in the world (World Health Organization, 2017) and can have an especially devastating effect when it occurs in youth. It not only impairs school and social functioning, but also generates serious family stress and provokes significant use of mental health services (Clarke, DeBar & Lewinsohn, 2003). Additionally, it is associated with a range of detrimental consequences into adulthood, including problems in physical and mental health (Weisz, McCarty & Valeri, 2006). Epidemiological studies show that almost a quarter of youths have suffered from at least one clinically significant depressive period by the time they are 18 years old (Patel et al., 2007). Even more disturbing is that the percentage of youths reporting a recent period of clinical depression, grew by 37% over the last ten years (Mojtabai, Olfson & Han, 2016). All these factors together make depression among youth a severe epidemiological problem and create a pressing need for the development of innovative therapeutic interventions and preventative programs.
In the last decade, there has been a growing interest to incorporate constructs like self-compassion and self-kindness into novel interventions or preventative programs. Examples are the Mindful Self-compassion (MSC) program of Neff and Germer (2013) and Compassion Focused Therapy (CFT, Gilbert, 2010). These therapies give more prominence to changing the person’s relationship with their problems, in order to alleviate psychological discomfort. A crucial aspect of this is the development of a compassionate or kind attitude ( MacBeth & Gumley, 2012). This cultivation of a different and more compassionate attitude towards one’s own difficulties and oneself in general, is thought to be the underlying mechanism for all mindfulness-based interventions (Neff, 2003; Kuyken et al., 2010). However, before delving into the topic of self-compassion, it is important to note that the exact definition of self-compassion and self-kindness and their relationship to each other is still a topic of debate between researchers (Strauss et al., 2016). The current study uses the terms of self-compassion and kindness in line with the definition of dr. Kristin Neff, a pioneer in the research field of self-compassion. Neff (2003, p.224) defines self-compassion as the ability to join one’s feelings of suffering with a sense of loving kindness, connection, and concern. In her model, Neff (2016) proposes that self-compassion is a composition of six dynamic and interacting components, of which three are considered protective factors and three that are vulnerabilities opposing these protective factors. The first – and for this study most important - component is self-kindness; the ability to treat oneself with care and understanding when confronted with personal failure, problems and stress. This is opposed to the component of self-judgment; when one is cold and critical towards oneself in these situations. The third component she describes as common humanity or the ability to recognize that all humans fail at times. This is opposed to the construct of isolation; feeling isolated by one’s failures – as if no one could possibly feel the same way or make the same mistakes. The fifth component is mindfulness; being aware of the present moment experience in a balanced way by putting one’s own situation into a larger perspective. This is opposite the construct of over-identifying with one’s emotions, in which these
6 emotions or situation gets exaggerated. Neff believes that self-compassion is best defined and
measured as ‘entailing less judgment, isolation, and over-identification, as well as more self-kindness, common humanity and mindfulness’ (2016, p.797). Even though most stated literature and previous research predominantly describe and research the construct of self-compassion in its whole, current study will solely focus on studying the concept of self-kindness. Reason for this is that the construct of self-kindness is more demarcated and thus easier to define and measure. Moreover, the recent meta-analysis of Muris & Petrocchi (2015) shows that self-kindness has the highest effect size as a positive indicator of self-compassion in relationship to psychopathology (r = -.34).
The biggest reason for the gained popularity of self-compassion and –kindness, is that these constructs are not only positively associated with psychological wellbeing by fostering emotional resilience and predicting higher levels of optimism and joy (Neff, 2011; Neff & McGehee, 2010). More importantly, they are also negatively associated with psychopathology like depression. Recent meta-analyses of research in this field show that individuals with higher levels of self-compassion display lower levels of depression and stress (MacBeth & Gumley, 2012, r = -0.54; Muris & Petrocchi, 2016, r = -0.53). Likewise, higher levels of self-kindness are negatively associated with depression (Muris & Petrocchi, 2016, r = -.34). To further investigate the negative association between self-kindness and depression, it is necessary to first get a good understanding about depression and which intrinsic factors are at play in predicting the development of this mental disorder. If these predicting factors are clear, it might be easier to understand exactly how self-kindness can possibly tackle or prevent
depression. Current study takes a closer look into poor emotion regulation skills and self-criticism, two constructs that are believed to be predictors of depression (e.g. Aldao & Nolen-Hoeksema, 2012; Diedrich et al., 2014; Ehret, Joorman & Berking, 2015).
1.2 Emotion regulation
The first predictor of depression that might be challenged through using self-kindness as an emotion regulation strategy, refers to deficits in emotion regulation skills. The regulation of emotions can be seen as organizing the intake of emotionally arousing information and helps to manage emotions after the experience of stressful events (Garnefski et al., 2007). This regulation happens at either a
conscious, deliberate level or on a unconscious, automatic level (Gyurak, Gross & Etkin, 2012). Adaptive emotion regulation strategies refer to a goal-directed and flexible application of these emotion regulation skills with regard to environmental demands (Diedrich et al., 2014; Gross, 2015). Experimental studies have shown that adaptive strategies like acceptance, problem solving and cognitive reappraisal lead to beneficial outcomes. These outcomes include effective interpersonal functioning (e.g. Richards & Gross, 2000), reductions in the experience of negative affect (e.g. Goldin, McRae, Rame, & Gross, 2007) and less chance on psychopathology (e.g. Aldao, Nolen-Hoeksema & Schweizer, 2010; Aldao & Nolen-Hoeksema, 2012). In contrast, deficits in emotion regulation like the use of maladaptive emotion regulation strategies refer to strategies that are ineffective at regulating emotions (Aldao et al., 2014). Examples of such strategies are suppression of thoughts or emotions,
7 avoidance of thoughts or emotions or an intense focus on the experienced negative emotions and its
causes and consequences, called rumination (Aldao & Nolen-Hoeksema, 2012). The use of
maladaptive emotion regulation skills have been hypothesized to be a transdiagnostic factor for most forms of childhood psychopathology (e.g. Gross, 1998, Zeman et al., 2006; Aldao & Hoeksema, 2012;
Diedrich et al., 2014) and are seen as a core component of depression and anxiety (e.g. Aldao & Nolen-Hoeksema, 2012). To illustrate, the study of Silk, Steinberg and Morris (2003) found that youth with diagnosed depression experience a lower self-efficacy in their emotion regulation abilities.
Moreover, these youths showed a variety of emotional competence deficits (e.g. poor understanding of emotions, impoverished awareness of emotions and dysregulated expression of emotions).
This leads to an important question; how does ineffective emotion regulation cause feelings of depression? Bradley’s (2003) model of dysregulation in emotion depicts that some people have a vulnerability to experience higher levels of arousal in general. This vulnerability is either learned (e.g. trauma or loss) or biological in nature (e.g. behavioral inhibition). When faced with stress, this vulnerability interferes with being able to regulate this reaction in adaptive ways. Ehret and colleagues (2014) carried out a randomized controlled trial with a clinical population of adults, that illustrates this point. In this study, they found that the absence of adaptive emotion regulation when one is feeling negative affect, leads to experiencing a loss of control over one’s feelings. This in turn gives rise to the feeling that these negative affective states will continue to impair one’s well-being. Theorists argue that not being able to effectively manage emotional responses to everyday events, will lead to more serious stages of distress which then can unfold into diagnosable depression (Aldao, Nolen-Hoeksema & Schweizer, 2010; Berking & Whitley, 2013). For a long time, the role of emotion regulation in psychological treatments of childhood disorders was largely ignored or not explicitly stated as being a central treatment goal (Zeman et al., 2006). Although recent years this has been shifting, the problem at this point is the unclarity about which adaptive emotion regulation strategies are most effective to fight psychopathology (Diedrich et al., 2014). Researchers found that maladaptive strategies have a stronger relationship with psychopathology than adaptive strategies do (e.g. Aldao, Nolen-Hoeksema & Schweizer, 2010; Aldao & Nolen-Hoeksema, 2012). This implies that the existence of a maladaptive emotion regulation strategy is more detrimental than the respective lacking of more adaptive emotion regulation strategies. This is a surprising and interesting find, as interventions for depression mostly focus on teaching more adaptive strategies to depressed individuals.
Recently, the use of self-compassion has been proposed as a potentially adaptive emotion regulation strategy when dealing with negative emotions and self-criticism (Gilbert & Procter, 2006; Petersen, 2014). Self-compassion can be regarded as an adaptive emotion regulation strategy, as it teaches an alternative way of relating to suffering that can be considered more psychologically adaptive and flexible (Jazaieri et al., 2014). Despite the interesting finding that maladaptive strategies seem to have a larger magnitude in their relationship with psychopathology than adaptive strategies, self-compassion still seems to combat maladaptive strategies in an innovative way. In general, people experience and tend to respond to pain and suffering in maladaptive ways (e.g. rumination or
8 suppression). Self-compassion encourages to experience suffering in the present moment with an attitude of curiosity and kindness towards oneself, without holding onto it throughout the day ( rumination) and without denying or pushing away (suppression) (Jazaieri et al., 2014). This specific approach towards suffering likely lowers general worry and may reinforce more positive coping strategies. This will help individuals to take a step back from their negative affect and thoughts, which in turn can prevent the escalation and maintenance of depressive thinking and behaviour (Feldman & Kuyken, 2011). In this sense, self-compassion could serve as a resilience factor, creating a buffer against the development and maintenance of depressive periods (Diedrich et al, 2014; Ehret, Joormann & Berking, 2015).
1.3 Self-criticism
The second predictor of depression that might be challenged by using self-kindness as an emotion regulation strategy, is self-criticism (e.g. Gilbert et al., 2006; Dunkley et al., 2009). Self-criticism can be defined as a response style to perceived failure that is characterized by negative judgment and self-evaluation (Ehret, Joorman & Berking, 2015). Having high levels of self-criticism is believed to be associated with extreme internal demands for achievement, perfection and recognition. Self-criticism involves rigid and constant self-analysis, extremely critical evaluations of one’s own behavior, an inability to derive satisfaction from successful performance and chronic concerns about others’ criticism and expectations (Dunkley, Zuroff & Blankenstein, 2003). As a consequence, self-criticism has extensive empirical support to be a transdiagnostic factor for multiple mental disorders (Cox et al., 2000; Zuroff & Mongrain, 1987), as well as to be a specific personality dimension or state involved in depression (Dunkley, Zuroff & Blankenstein, 2003). Research suggests that it is the strength of negative emotions towards oneself and an inability to adequately cope with these emotions, that put highly self-critical individuals at risk for the development and maintenance of depressive episodes (Adams et al., 2009; Ehret, Joorman & Berking, 2015). This way, self-criticism connects to the construct of maladaptive emotion regulation strategies. To illustrate, results of the clinical research of Adams et al. (2009, p.9) suggest that youth exhibiting higher levels of self-criticism are more likely to show an increase in depressive symptoms when exposed to a stressful situation in comparison to youth with low levels of self-criticism.
Given these points, one can conclude that increasing more effective emotion regulation skills in youth might interfere with the vicious cycles of both experienced negative affect and self-criticism. As self-compassion is seen as a more adaptive, alternative response to perceived failure, it has been stated to be an especially helpful emotion regulation strategy in self-critical individuals (Gilbert & Procter, 2006; Petersen, 2014). For example, instructing depression-prone students to use self-kindness in a laboratory setting has been found to help act as a buffer against depressogenic self-criticism following a stressful situation (Leary et al., 2007). As Bluth et al. (2016) state, self-compassion may be viewed as social support turned inwards, providing a way in which youth can directly support themselves emotionally.
9 1.4 Empirical status of self-kindness as an emotion regulation strategy
Even though this much is known about self-kindness today, several issues demand further research. First off, possible effects of self-compassion and -kindness might be moderated by the level of depressive symptoms, although not in the way one might think. Previous research suggests that while an increase of negative affect will generally interfere with the activation and usability of positive
cognitions (e.g., Joormann & Gotlib, 2007; Koster et al., 2010), self-compassion grows even stronger in this situation (Hein & Singer, 2008, Diedrich et al., 2016). Moreover, the positive effects of
self-compassion are less strongly affected by an increase in depressed mood than other adaptive emotion regulation strategies, like cognitive reappraisal (Berking & Whitley, 2014, p. 23). An explanation for this is that most adaptive strategies become more difficult to apply when suffering intensifies. Progressively higher levels of negative affect tend to lead to an increase in negative thoughts (Sheppes & Meiran, 2007) and a bigger discrepancy between the sensed current state and the demand for well-being (Diedrich et al., 2014). In contrast, self-compassion might even become easier to utilize as a person's suffering increases, as the intensity of the observed suffering is an important cause for compassion to be evoked (Diedrich et al., 2014; Hein & Singer, 2008). Moreover, Diedrich et al. (2016) show in their study with a clinical sample that the use of self-compassion even facilitates the usage of other emotion regulation strategies (e.g. cognitive reappraisal) in decreasing negative affect. In this sense, self-compassion and self-kindness may have an advantage in being used in comparison to other strategies to battle against depression in youth. As almost no research is done on this topic - let alone with youth, it is an important subject for further research.
Another topic of debate is on which level of conscious cognition these adaptive emotion
regulation strategies work best. Most research to date has focused on explicit (deliberate and effortful) forms of emotion regulation (Koole, Webb & Sheeran, 2015). However, there has been a growing research interest with regards to implicit (automatic) forms of emotion regulation, that operate without the need for conscious supervision or explicit intentions (e.g. Gyuarak, Gross & Etkin, 2011; Koole & Rothermund, 2011). Reason for this being that implicit regulation of emotions has an assumed advantage of requiring less effortful control; it consumes little or no attentional capacity or subjective effort. Moreover, automatic processes can be activated quickly and operate in an efficient manner (Mauss, Cook & Gross, 2006). It is important to see on which level of consciousness adaptive emotion regulation strategies are optimally acquired, when looking into their contribution to interventions or preventative programs. While multiple studies have found that priming of non-conscious or automatic emotion regulation strategies are as effective as applying conscious emotion regulation strategies in diminishing negative affect on a physiological level (Williams, Bargh, Nocera & Gray, 2009; Yuan et al. 2015), measuring the effect of implicit learning on subjective emotion experience leads to contradicting results. For instance, Yuan et al (2015) did a study with adults in which they activated cognitive
reappraisal both through priming (implicit) and through explicit instruction. Both implicit and explicit groups showed the same decrease in heart rate reactivity during a stressful task in comparison to people who did not receive an adaptive emotion regulation strategy to handle their emotional response
10 to the situation. However, the unconscious reappraisal group did not show reductions in subjective
negative emotion, whereas this was significantly decreased in the conscious reappraisal group. These results suggest that unconscious adaptive emotion regulation is only effective in decreasing
physiological consequences of frustrating emotion, but not in reducing subjective experience. In contrast, Mauss, Cook & Gross (2006) found that manipulating implicit emotion regulation through a word scramble task had not only a significant effect on physical arousal levels but also on subjective experience of negative emotions in adults. Participants that were primed with words representing emotion regulation or controlling strategies, reported less anger than the control group or participants that were primed with words representing the expression of emotions. Additionally, recent studies by Yang et al. (2014) and Wang & Li (2017) with very similar research design as Mauss et al. (2006) confirm these findings. They found that priming adult participants with emotion regulation goals through a word matching task lead to a decrease in experienced negative affect, obtained by
questionnaires in which participants rate to which extend they felt negative emotions after negative mood induction and comparing them with mood measured at baseline. These contradicting results show that more extensive research in this area is needed. This is critical, as many researchers argue that reducing subjective experience of negative affect is very important. Subjective experience of negative emotions play a major role in psychopathology like depression (e.g. Garnefski, Kraaij & Spinhoven, 2001; Liverant et al., 2008), and are even a big influence on the amount of emotion actually
experienced in the body (e.g. Bradley, 2003). We know from existing research that conscious
deployment of emotion regulation can reliably and positively shape the course of emotional responses, but it is less clear whether deploying and executing regulation without awareness can have similar benefits. An important issue to study is whether the effects of implicit emotion regulation can be measured on subjective experience without perturbing the process (Gyurak, Gross & Etkin, 2012).
Conclusively, recent research on self-compassion and self-kindness suggest their possibility to be an adaptive way of relating to the self when considering personal inadequacies or difficult life
circumstances, which could help prevent or treat psychopathology (Neff & McGehee, 2010). However, prior similar research has mostly focused on adults. This is an important gap to fill, as developing more adaptive emotion regulation strategies in childhood does not only have benefits in their current daily life (Zeman et al., 2006. Self-compassion and other adaptive emotion regulation strategies can form a protective buffer against possible psychopathology, so it even benefits youth’s psychological wellbeing later into adulthood (Garnefski et al, 2007).More research about which and how emotion regulation strategies work and if self-kindness boosting can help prevent or alleviate depressive symptoms, should help to further develop new or existing interventions and prevention programs against depression in youth.
Hence, the goal of the current study was to see whether boosting of kindness reduces self-reported negative affect and self-criticism state in youth. More specifically, this study focused on (1) whether we could increase or boost self-kindness to help lower self-criticism and negative affect in youth (2) whether implicit boosting of kindness had a greater effect than explicit boosting of
self-11 kindness as an emotion regulation strategy, and (3) whether the differential effects of the emotion
regulation strategies would be moderated by the level of depressive symptoms. Based on previous empirical literature, it was hypothesized that youth that were either explicitly instructed or primed to use self-kindness as an adaptive emotion regulation strategy would experience less negative affect and a lower state of self-criticism in comparison to youth that were not introduced to self-kindness at all after a negative mood induction. Furthermore, it is hypothesized that elevated levels of depressive symptoms would moderate the effect of self-kindness boosting (either through explicit instruction or priming) on experienced negative affect and self-criticism state in youth. It is expected that youth with elevated levels of depressive symptoms would have a smaller difference in experienced negative affect between baseline and after negative mood induction when they received explicit instructions to use self-kindness or were primed to do so, in comparison to youth with low levels of depressive symptoms.
2. METHOD
2.1 ParticipantsA total of 183 participants participated in this study; 93 boys (51%) and 90 girls (49%) between the age of 9 and 14 years old ( M = 10.8 years, SD = 1.38 years). For every participating child, one parent or primary caregiver was asked to fill in some questions about the child. Of all participating parents, 75% finished a HBO degree or higher. Participants consisted of visitors of the NEMO Science Museum, as this study is part of Science Live: an innovative research program of NEMO that enables scientists to carry out real, publishable, peer-reviewed research using NEMO visitors as volunteers. Participants were acquired through either letting them sign themselves up at the entrance of the Science Live Lab, or by approaching them when entering the NEMO Science Museum.
The following exclusion criteria were handled when recruiting participants : (a) a diagnosed mental disorder, (b) a primary language other than Dutch, (c) a physical limitation that obstructs the child in writing and / or using a computer. Furthermore, a child could only participate when it was
accompanied by a primary caregiver who was able to give consent and whose first language is Dutch. When testing the experiment before the data collection on three children between 7 and 10 years
old, it became clear that children below the age of 9 years 1) did not have a proper understanding of some of the questions regarding emotion regulation, and 2) would not be able to finish the experiment in time, as the time limitation established by NEMO was 20 minutes and children were doing it approximately 45 minutes. Hence this study was adjusted to examine children from the age of 9 and up. The age of 14 as a cut off age was decided based on the demographic statistics of NEMO Science Museum visitors.
12 2.2 Design and Setting
The present study employed one predictor variable: the boosting of self-kindness. Children were randomly assigned to one of three conditions: the implicit group - where self-kindness is boosted through a priming word task, the explicit group – in which self-kindness is boosted by explicit instruction, and the control group – in which self-kindness is not boosted. There were two dependent response variables: the level of self-reported negative affect and the level of self-reported self-criticism state. Both constructs were measured by questionnaires that youth filled in four times during the experiment. All three conditions endured a negative mood inducing task, to see whether the boosting of self-kindness would have an effect on provoked experience of negative affect. Levels of depressive symptoms were measured by a questionnaire parents filled in about their child during the experiment. In addition to the tasks described in this paper, other tasks were carried out in the overarching research project that current study did not make use of.
2.3 Materials
Implicit self-kindness boosting (priming)
To ensure an implicit way of self-kindness boosting, this study has used a word scramble task based on the studies of Williams et al. (2009)and Yuan et al. (2015) that allows to prime (i.e. implicitly activate) a specific concept that has an impact on the child’s behavior; in this case management of negative emotions during a stress related task. In this task, participants constructed a four word
sentence from five scrambled words. Depending on condition, participants would either engage in a (a) neutral priming task (control condition and explicit condition), (b) self-kindness priming task (implicit condition). In all three conditions, participants got 10 sentences to construct. In the neutral priming task, all the sentences contained neutral words (e.g. (a/how/he/box/sees). The correct sentence would be (‘he sees a box’). There is always only one way to construct the sentence in a grammatically correct manner. These words are depicted as neutral in a validated Dutch word list by Moors et al. (2013), meaning they are low in arousal and neutral in valence. In the self-kindness priming task, eight out of ten sentences consist of self-kindness related words (e.g. it/me/about/she/cares). The correct sentence would be ‘she cares about me.’) and two out of ten contain neutral words. The self-kindness related words are adapted from the Falconer (2015) and Neff(2003) self-kindness subscales. Both factors have good reliability, with a Cronbach’s α of .91 for self-compassionate attitude in Falconer’s scale (Costa et al., 2016), and a Cronbach’s α range from .74 to .88 for Neff’s self-kindness subscale (Kotsou & Leys, 2016).
Explicit self-kindness boosting
Prior to the negative mood induction phase, participants in all conditions were provided with the following instructions on the screen in front of them:
13 “Well done! Now you know the rules of the game. You can put the game back in box 1, close the box and put it away. Take box 2
in front of you but do not open it yet. For this game you will get only 2 minutes, the computer will let you know when the time is
up. Remember that your performance will be recorded with the camera and evaluated by an expert. When you’re ready, open the box,
and press NEXT.”
Only the participants in the explicit self-kindness emotion regulation condition were provided with additional instructions developed by us, in order to explicitly boost self-kindness:
“Now you are going to play a game that will be more challenging than the last one. Because of this you might experience some negative
emotions. For example you can get angry at yourself, you can get sad or irritated. Whatever happens please try to be kind and
friendly to yourself. Think about how a good friend would support and encourage you in this situation. Remember that the task is
difficult so you might not perform as you would like to. Try to remember that you are doing your best !
Negative Mood Induction
Negative mood was induced in this study by letting participants play an unsolvable version of a board game called ‘rush hour’ (see figure 1). Prior to the unsolvable
game, we included a training phase which involved a very easy version of the same game, so children could familiarize
themselves with it. The goal of the ‘rush hour’ game is to get the red car out of the parking lot as fast as possible. To do this, children had to move the other cars on the board, but only up-
and downwards or from side to side (depending on how the car Figure 1 Setup of the ‘Rush hour’ game was placed on the board). Children were not allowed to lift the cars. The first version of the game was a trial, intended for the child to learn the rules. Children got 30 seconds to play this simple version of the game. The program on the screen counted down and they heard a beep out of the headphones when their time was up. When the time was up, children were asked if they were able to solve the game. If not, the researcher approached the child to explain again and try it one more time to ensure the child knew the rules. After this, children played the second version of the game. This version was designed to be unsolvable, since the objective of the game was to induce a negative mood. Children got two minutes to solve this puzzle. Again the screen provided a countdown and the headphones gave a beep noise when the time was up, to further enhance negative mood induction.
In order to provide an additional social anxiety stressor to the negative mood induction
procedure, children were informed that their performance would be recorded by a camera in front of them and that they would be evaluated by an expert in another room.
Negative Affect
Participating children would fill in the Positive and Negative Affect Schedule (PANAS, Watson, D. & Clark, D. 1988) four times during the experiment, to check their level of self-reported negative mood at that current moment. Used as a psychometric scale, the PANAS can show relations between
14 positive and negative affect with personality states and traits. It comprises two mood scales, one that
measures positive affect (PA) and the other which measures negative affect (NA). Due to time limitation this study made use of the PANAS- Short Version, that consists of five descriptors for the PA scale and five descriptors for the NA scale to define their meanings. Although children filled in the whole PANAS, current study only uses the NA scale as we are interested in the decline of negative emotions. Participants were required to indicate to what extent they feel this way at the present moment, according to 5 negative affect items (miserable, angry, afraid, scared, sad) using 5-point scale that ranges from very slightly or not at all (1) to extremely (5). Higher scores for NA indicate higher negative affect.
Even though the last scoring of the PANAS by COTAN was in 1999, it is still widely used until this day as it has a strong reported validity with measures as general distress and dysfunction, depression, and state anxiety (Crocker, 1997; Harmon-Jones et al., 2009). The I-PANAS-SF NA subscale has an adequate reliability, with a Cronbach’s alpha of .76 . The NA subscale of the short version has a correlation with the full PANAS NA subscale of .59 (p < .01). This correlation is similar to the 2-month test-retest reliabilities Watson et al. (1988) report for the original PANAS, suggesting that the 10-item I-PANAS-SF compares well with the full 20-item original in terms of both correlating with the original full form and temporal stability, both important aspects of short form development (Thompson, 2007). ). As negative affect was measured four times during the experiment, Cronbach’s α for negative affect in current study varied between .59 and .82 between the four time points.
Self-criticism /self-kindness manipulation check
To measure self-kindness and self-criticism state, an adapted version of the Self Compassion Self Criticism Scale (SCCS, Falconer, C. King, J., & Brewin, C., 2015) was used. The SCCS is a newly developed measure to assess addressing both self-compassion and self-criticism in parallel at specific moments in time or under specific circumstances. Participants rate the extent to which they feel Harsh, Contemptuous, Critical, Soothing, Reassuring or Compassionate towards themselves at that current moment on a 7-point Likert Scale (1 = not at all; 7 = highly). Because this questionnaire is originally developed for adults, we adapted the language to be more simple and clear and made sure to put a small description next to every presented word. This way, participants would know exactly what the words meant. Before the experiment started, it was always stated clearly to each participant that when they did not understand certain words, they could ask one of the researchers for help.
The Self Criticism subscale was used to measure self-criticism state. The Self Compassion subscale was not initially designed to be a manipulation check prior to testing, but was decided afterwards to be used to check whether self-kindness boosting – both implicit and explicit- was successful. The manipulation check was done by comparing scores at baseline with scores after the priming phase and after negative mood induction. The ratings were summed to generate the adapted Self-Criticism Scale score (range 3–21).
(self-15 kindness), and are sensitive to change (Falconer, King & Brewin, 2015; Kamboj et al., 2015, Falconer
et al., 2016). As criticism was measured four times during the experiment, Cronbach’s α for criticism in current study varied between .72 and .89 . This also applies to the construct of self-kindness; it’s Cronbach’s α varied between .70 and .86 during the four phases of the experiment.
Depressive symptoms
The Children’s Depression Inventory 2- Parent Version (CDI-2 P, Multi-Health Systems, 2011) is a 17-item scale that assesses depression in youth aged 7–17 that is aimed to determine the occurrence and severity of depressive symptoms in children and adolescents. The list measures cognitive, affective and behavioral symptoms of depression. Parents rate to which extent the 17 statements fit to the observations they made of their child in the past two weeks (e.g. ‘My child looks sad’) on a 4-point scale that ranges from (1) not at all to (4) always. The crude total score is obtained by adding up the scores of the individual items. The score can be interpreted in four different ways: by means of a z-score, percentile score, score cut off or analysis at item level.
The CDI-2 P form shows high levels of internal consistency with the Cronbach’s α value of .91. The moderately high correlations among the subscales from .58 to .69 indicate a hierarchical association among total scales, and subscales (Bae, 2012). Current study found a Cronbach’s α of .78 for the CDI 2-P.
2.4 Procedure
When a child and accompanying parent would arrive at the Science Live lab, they were welcomed by one of the researchers. Once seated, the researcher explained shortly the goal of the study and what the child and parent could expect When the whole procedure was clear for both parent and child, an informed consent was signed for both. Parents would sign for themselves and their child when their child was under 12 years of age. Children of 12 years or older were asked to sign their own consent form too,. Next, the child would be accompanied by one researcher to a different desk with a desktop and the rush hour game hidden underneath a box. Another researcher would guide the parent to another room and install them behind a laptop to fill in some questions about their child. Parents were invited to leave their email addresses in order to be signed up for a newsletter to inform them about the progress and results of the research.
The experiment took approximately 25 minutes for each child, divided into four phases; a baseline phase, a priming phase, a negative mood induction phase and a recovery phase (see chart 1). All phases take between three and five minutes, depending on how fast the child goes through the program. Before the baseline phase starts, the researcher sits the child behind the desk with the monitor,
explains to the child how everything works and that if he/she wants to stop that’s possible at any given moment. Moreover, the child is asked to put their headphones on and keep them on during the whole experiment. All the instructions the child gets during the experiment as well as all the questionnaires, appear on the monitor in front of the child. After the child finished, it was taken by one of the
16 researchers to a desk for some final control questions (see Appendix B). The researcher asked the
questions verbally the child, and registered the answer in an excel sheet. After this, the child would be reunited with the participating parent or primary caregiver, for a joint debriefing.
Chart 1. Phases of the Experiment and their Tasks
2.5 Data Analysis
When looking into the data set and questions that were asked participants during debriefing (see appendix B), a total of 17 participants were excluded for multiple reasons. Seven participants were excluded because the program stopped before the end of the experiment. Four participants were excluded because of problems with the rush hour game (e.g. wrong placement of cars, cars fell off when grabbing the game). One participant was excluded because he wanted to stop during the experiment. Through control questions that were asked during debriefing, two participants were excluded because of ADHD (the parents of these participants did not read or interpret the exclusion criteria correctly). Additionally, two were excluded because they indicated that they did not read the explicit boosting text. This means that of the total of 183 initial participants that took part in the experiment, the data of 166 participants were used for analysis (Control = 59; Explicit = 53; Implicit = 54 ). All analyses have been executed with the help of IBM SPSS Statistics version 22.
A 3x4 repeated measures ANOVA was performed with group (implicit, explicit and control) as the between-subjects factor and time (baseline phase, priming phase, negative mood induction phase and recovery phase) as the within-subject factor. Even though the time point comparison of most interest is between before and after negative mood induction (time point 2 and 3), we looked at the entire four-point pattern as it could be interesting in explaining findings (e.g. maybe negative affect is decreased after implicit priming , but this result did not last until after negative mood induction). When the assumption of sphericity was violated during analyses, the degrees of freedom were corrected using Greenhouse–Geisser estimates of sphericity.
2.6 Descriptive statistics and treatment of the data
The three groups were adequately randomized on pre-experiment emotional state (negative affect and self-criticism state), age distribution and gender composite (all p > .05). Table 2 presents sample characteristics for all participants.
baseline phase
• SCCS + PANAS
priming phase
•word scramble task • SCCS + PANAS
negative mood induction
phase
• 'Rush hour' game • SCCS + PANAS
recovery phase
17 Table 2. Sample Characteristics and Comparison of Baseline State between Groups
Experimental groups
Control Explicit Implicit Total Test statistic p Demographics Sample size 59 (35.6%) 53(31.9%) 54 (32.5%) 166 (100%) Age -M (SD) 10.92 (1.55) 10.87 (1.23) 10.84 (1.34) 10.84 (1.38) F(2, 163) = .29 .75 Gender χ2 (2) = .35 .84 Male 29 (49.2%) 29 (54.7%) 28 (51.9%) 86 (51.8%) Female 30 (50.8%) 24 (45.3%) 26 (48.1%) 80 (48.2%) Baseline state- M (SD) Self-criticism 5.71 (3.27) 5.02(2.31) 5.07(2.91) 5.28 (2.88) F(2, 163) = 1.02 .36 Negative affect 6.08 (1.6) 5.83 (1.12) 6.15 (1.70) 6.02 (1.50) F(2, 163) = .68 .51
Note. M = Mean; SD = standard deviation
2.7 Manipulation of self-kindness boosting and negative mood induction
The first manipulation check was to examine whether the experiment successfully boosted
self-kindness in both an implicit way (through the word scramble task) and an explicit way (through explicit instruction). Therefore, a repeated measures ANOVA was conducted to compare scores on the SCSC Self-kindness Scale between the three experimental groups during the four time points of the
experiment. The results suggest that there was no difference in self-kindness between groups during the four different timepoints of the experiment ( F (6, 324) = .94, p = .47). For a detailed description of the results, see Appendix A.
The second manipulation check examines whether the ‘rush hour’ puzzle game successfully induced negative mood. For this, the scores of the PANAS NA Scale in the control group (where affect induction was free of regulatory influences) during the four time points of the experiment were compared to each other. It was expected to see an increase in negative mood after negative mood induction compared to after the priming phase and that this will likely go back to baseline during recovery phase. The results suggest that there is a main effect of time (See figure 1). participants in the control condition scored significantly higher on the NA scale of PANAS after negative mood
induction than after the word scramble task, and lower again in the recovery phase (F (2.16, 125.25) = 4.93, p < .01). This means there is a proper manipulation of negative mood; children experienced higher levels of negative feelings after playing the ‘rush hour’ game than they did in other phases of the experiment. The fact that participants scored significantly lower in the priming phase compared to the baseline phase, can be explained by possible tension at the start of the experiment. For the exact mean scores and their standard deviations on the four time points, there is an overview in table 2 in the appendix.
18 Figure 1. Experienced Negative Affect in the Control Group over four Time Points
Note: NMI = negative mood induction.
3. RESULTS
3.1 Self-kindness boosting effect on experienced negative affect
The first Repeated Measures ANOVA was conducted to see whether self-kindness boosting led to lower levels of experienced negative affect in the two experimental groups in comparison to the control group.Results revealed that there was no difference between the groups on self-reported negative affect at different time points (F (4.02, 327.94) = 1.14, p = .35). Analyses showed a main effect of time, but no interaction effect between condition and time (see figure 2). This means that the amount of change in experienced negative affect over time was not dependent of in which group a participant was placed; the change happened in all three experimental groups. For the exact mean scores and their standard deviations on the four time points, an overview can be found in table 2 of Appendix A.
Figure 2: Experienced Negative Affect through three Experimental Groups, over four Time Points.
Note: NMI = negative mood induction. 4.0 4.5 5.0 5.5 6.0 6.5 7.0
Baseline phase Priming phase NMI phase Recovery phase ne ga tive a ffe ct (P AN AS) control 4.0 4.5 5.0 5.5 6.0 6.5 7.0 Baseline
phase Primingphase NMI phase Recoveryphase
ne ga tive a ffe ct (P AN AS) control implicit explicit
19 3.2 Self-kindness boosting effect on self-criticism state
To see whether self-kindness boosting led to lower levels of experienced self-criticism state in the two experimental groups in comparison to the control group, a Repeated Measures ANOVA was
conducted, comparing the SCSC Self-Criticism Scale scores of the three groups to each other at four time points. Results revealed that there was no difference between the groups on self-reported negative affect at different time points (F (4.58, 372.94) = 1.11, p = .36). Analyses show that there was only a main effect of time, but no significant interaction effect between condition and time. These results can also be seen in figure 3. Even though figure 3 seems to show that self-criticism state in the implicit group was lower than the two other groups after negative mood induction, and that both self-kindness boosting groups seem to differ from the control group in the recovery phase, this was not significant. This means that the amount of change in experienced self-criticism state over time was not dependent of in which group a participant was placed. For the exact mean scores and their standard deviations on the four time points, an overview can be found in table 2 of Appendix A.
Figure 3. Experienced Self-criticism State through three Experimental Groups, over four Time Points
Note: NMI = negative mood induction. 3.3 Moderation of depression levels
Moreover, this study wanted to examine whether the level of depressive symptoms had an influence on the relationship between self-kindness boosting and experienced negative affect after negative mood induction. For this, a linear regression with dummy codes for the interaction between the CDI scores and experimental conditions was conducted. A two-tailed alpha level of .05 was used, as this is an exploratory moderation analysis. Levels of depressive symptoms were quantified by CDI sum scores. The results did not show any evidence of a moderating effect of depressive symptoms on the relationship between self-kindness boosting and experienced negative affect, with R2 = .02 and F (4, 156) = .94, p = .44. Against our predictions, the interaction between CDI scores and implicit self-kindness boosting did not significantly predict negative affect (β = .70, p = .22). The same goes for the interaction between CDI scores and explicit self-kindness boosting (β = .36, p = .23).
4.0 4.5 5.0 5.5 6.0 6.5 7.0 baseline
phase primingphase NMI phase recoveryphase
se lf-cir itic sm (S CC S) control implicit explicit
20 Lastly, this study examined the possible moderating effect of depressive symptoms on the
relationship between self-kindness boosting and experienced self-criticism state after negative mood induction. The analysis for this moderation was conducted following the same guidelines as the previous analysis. The results revealed that there was no possible moderation of depressive symptoms on the relationship between self-kindness boosting and experienced self-criticism state, with R2 = .02
and F (4, 156) = .92, p = .45 . Against our predictions, the interaction between CDI scores and implicit self-kindness boosting did not significantly predict self-criticism (β = .87, p = .13). The same goes for the interaction between CDI scores and explicit self-kindness boosting (β = .43, p = .15).For a detailed description about the moderation analysis results see table 3 in the appendix. For a detailed description about the interaction between all variables see table 4 in the appendix.
4. CONCLUSION AND DISCUSSION
The current study aimed to see whether self-kindness could be boosted to help lower self-criticism and negative affect in youth by potentially serving as an adaptive emotion regulation strategy. In contrast to our hypotheses, the boosting of self-kindness–whether it was implicit or explicit- did not have an effect on experienced negative affect or self-criticism in participating youth. Moreover, no real indication could be given if self-kindness boosting has a differential effect when it’s explicitly instructed in comparison to when it is implicit, nor in which direction this difference would manifest itself. Lastly, the results could not give an indication of moderating properties of depression levels on the
relationship between self-kindness boosting and negative affect or self-criticism. All in all, the obtained results could not indicate support for the proposition arising from existing literature set out in the introduction (Diedrich et al., 2016; Gilbert & Procter, 2006; Petersen, 2014).
Nevertheless, it is important to note that caution must be used in interpreting these results. One can find several plausible explanations as to why the predictions were not met. The most logical reason for the insignificant findings in this study is that – as tested by means of a manipulation check - it seems that the manipulation of self-kindness boosting in the experiment was not sufficient. With the boosting of self-kindness not being sufficiently implemented, no real detection of its possible effect can take place. Reason for this could be time limitation; the duration of self-kindness boosting in this experiment might have been too short to have any feasible effect on both explicit and implicit boosting of self-kindness. Firstly, for the implicit boosting of self-kindness the limited time resulted in a set of 10 sentences, of which eight contained priming words. This was initially not thought to be a problem, as word scramble tasks are often around 10 to 15 sentences while having significant priming effects. For instance, the study of Yuan et al (2015) - which was used as an example for the current study- used a word scramble task with 15 sentences of which 10 contained priming words. Additionally, Williams et al. (2009) used a word scramble task of 10 sentences of which 5 contained priming words. One has to note however, that Yuan et al. (2015) only found an effect of the priming task on physical outcome measures, not on subjective measures. The same goes for Williams et al. (2009); they found a priming
21 effect, but their outcome values only contained physical measures. On the other hand, Mauss, Cook & Gross (2006) did find that priming participants with words representing emotion regulation or
controlling strategies, reported less anger than the control group or participants that were primed with words representing the expression of emotions. However, their study design allowed for 42 sentences to be ‘unscrambled’ by participants. It is possible that using a priming task to implicitly boost self-kindness as an adaptive emotion regulation strategy, is only sufficient when participants are exposed to more self-kindness related priming words. A critical note for this argument is saying that the implicit self-kindness boosting was not sufficient, should be considered with caution. One goal of this study was to explore if unconscious boosting would have an effect on subjective experience. As results could not confirm this, it is also not sure to say whether the implicit self-kindness boosting succeeded when controlled for with a subjective measure. Additional physiological measurements would be necessary to be sure if the manipulation was or was not effective.As mentioned before, present study is part of a larger study in which also physical data such as heart rate and alpha amylase levels have been collected. As it was beyond the scope of this study, the physical data was not included. Knowing the results of the current study however, it would be interesting to involve that part of the data as well in a future study regarding this subject.
Secondly, due to time limitation the explicit boosting of self-kindness was put into practice through a short instruction in text. Participants of the explicit boosting group read an instruction on the screen in front of them on how to regulate possibly upcoming negative emotions by using self-kindness while performing the task. We aimed to control for participants understanding of the explicit instruction by asking afterwards about participants’ thoughts during the negative mood and if they used positive self-talk. However, it is possible that only providing an instruction in text was not enough for children to successfully implement self-kindness as an emotion regulation strategy. As Bluth & Eisenlohr-Moul (2017) state in a very recent study, the lack of a significant decrease in self-criticism and negative affect evidenced in this study, is possibly caused by the fact that youths need a more extended or larger “dose” of self-kindness. Consequently, they might be able to build their inner resources enough in order to establish a significant reduction in self-criticism or experienced negative affect. Furthermore, it may be that it takes longer for depressive symptoms like self-criticism or experienced negative affect to change. More precisely, the temporal order of the constructs may be such that perceived stress decreases first (Bluth & Eisenlohr-Moul, 2017). It is possible that over time with continued self-kindness practice, which saves more time for resource building, depressive symptoms subsequently decrease. One can argue that as for all human competencies, to be routinely used, self-kindness needs to not only be stimulated but also practiced and elaborated over time in order to predominate.
Some other study limitations should be acknowledged. First off, the sample population consisted for 75.5% of parents with a completed education level of HBO (university of applied sciences) or higher. Consequently, our results should be interpreted as preliminary and cannot easily be generalized for people that finished a lower level of education. Further, the relatively low baseline
22 scores on reported depressive symptoms also reduced the probability in finding (large) moderating
effects. As we did not work with a clinical sample, no hard statements can be made about the possible effect on youth with an actual diagnosis of depression which are in need for interventions. Another limitation is that due to time restraints this study did not include the physiological data collected in the overarching study, as this was beyond the scope of current study. However, as the results in the research of Williams et al. (2009) and Yuan et al. (2015) suggest, there is a possibility that the effect of implicit self-kindness boosting as an emotion regulation strategy is only detectable through
physiological responses like heartrate variability.
Nevertheless, the limitations mentioned above do help to shine light on the areas in which additional research is needed in order to establish the relationship between self-kindness and depressive symptoms. In future research, it is important to see whether a more extended or larger “dose” of self-kindness boosting would allow for a different outcome. For instance, one could allow for a word scramble task with more sentences containing priming words, or another priming task that has a longer priming time. Secondly, exposing youth to explicit self-kindness boosting for a longer period or through different means than instructions on a screen may give different results. For example, one could try to give instructions verbally or even give an extensive self-kindness training to youth with depressive symptoms. Another suggestion is to try audio instruction, which might be more appropriate for this age range. Lastly, future research could provide much more information about the performance of implicit self-kindness boosting if physiological data is included. Another option is to look into subjective measures that are specifically designed to pick up on emotions on an implicit level, like the Implicit Positive and Negative Affect Test (IPANAT, Quirin, Kazén & Kuhl, 2009). The present study did not include such measures, as we were interested to see if standard subjective measures could also provide information on implicit boosting based on previous literature. However, no evidence could be found for this expectation, suggesting that implicit emotion regulation might be easier to detect through implicit measures.
All in all, this study has shed light on ways that self-kindness boosting can possibly help lower depressive symptoms as experienced negative affect and self-criticism state. Eventhough current study did not see any significant effects of self-kindness boosting under the specific settings of the
experiment, it did shine light on which direction more extensive research is needed. By gaining insight into self-kindness boosting and how it could possibly serve as an adaptive emotion regulation strategy, one can develop more targeted and effective interventions and preventative programs to form a buffer against or tackle depressive symptoms in youth. This might eventually lead to reduction in the high prevalence and level of relapse of depression in youth. Continuing research into the concept of self-kindness –like current study- is therefore very important and could have significant theoretical and clinical implications.
23
REFERENCES
Adams, P., Abela, J. R., Auerbach, R., & Skitch, S. (2009). Self-criticism, dependency, and stress reactivity: An experience sampling approach to testing Blatt and Zuroff’s (1992) theory of personality predispositions to depression in high-risk youth. Personality and Social Psychology Bulletin, 35(11), 1440-1451. DOI: 10.1177/0146167209343811.
Aldao, A., Jazaieri, H., Goldin, P. R., & Gross, J. J. (2014). Adaptive and maladaptive emotion regulation strategies: Interactive effects during CBT for social anxiety disorder. Journal of Anxiety Disorders, 28(4), 382-389. DOI: 10.1016/j.janxdis.2014.03.005.
Aldao, A., & Nolen-Hoeksema, S. (2012). When are adaptive strategies most predictive of psychopathology?. Journal of Abnormal Psychology, 121(1), 276. DOI: 10.1037/a0023598. Aldao, A., Nolen-Hoeksema, S., & Schweizer, S. (2010). Emotion regulation strategies across
psychopathology: A meta-analysis. Clinical Psychology Review, 30, 217–237. DOI: 10.1016/j.cpr.2009.11.004.
Bradley, S. J. (2003). Affect Regulation and the Development of Psychopathology. Guilford Press.
Berking, M., & Whitley, B. (2013). Affect regulation training. Handbook of Emotion Regulation. New York: Guilford, 350-392. DOI: 0.1007/978-1-4939-1022-9.
Bluth, K., & Eisenlohr-Moul, T. A. (2017). Response to a mindful self-compassion intervention in teens: A within-person association of mindfulness, self-compassion, and emotional well-being outcomes. Journal of Adolescence, 57, 108-118. DOI: 10.1016/j.adolescence.2017.04.001.
Bluth, K., Roberson, P. N., Gaylord, S. A., Faurot, K. R., Grewen, K. M., Arzon, S., & Girdler, S. S. (2016). Does self-compassion protect adolescents from stress?. Journal of Child and Family Studies, 25(4), 1098-1109. DOI: 10.1007/s10826-015-0307-3.
Clarke, G. N., DeBar, L. L., & Lewinsohn, P. M. (2003). Cognitive-behavioral group treatment for adolescent depression. Evidence-Based Psychotherapies for Children and Adolescents, 120–134. DOI: 10.1097/00004583-199903000-00014.
Costa, J., Marôco, J., Pinto-Gouveia, J., Ferreira, C., & Castilho, P. (2016). Validation of the psychometric properties of the self-compassion scale. Testing the factorial validity and factorial invariance of the measure among borderline personality disorder, anxiety disorder, eating disorder and general populations. Clinical Psychology & Psychotherapy, 23(5), 460-468. DOI: 10.1002/cpp.1974. Cox, B. J., Rector, N. A., Bagby, R. M., Swinson, R. P., Levitt, A. J., & Joffe, R. T. (2000). Is
self-criticism unique for depression? A comparison with social phobia. Journal of Affective Disorders, 57(1), 223-228. DOI: 10.1016/S0165-0327(99)00043-9.
Diedrich, A., Grant, M., Hofmann, S. G., Hiller, W., & Berking, M. (2014). Self-compassion as an emotion regulation strategy in major depressive disorder. Behaviour Research and Therapy, 58, 43-51. DOI: 10.1016/j.brat.2014.05.006.
Diedrich, A., Hofmann, S. G., Cuijpers, P., & Berking, M. (2016). Self-compassion enhances the efficacy of explicit cognitive reappraisal as an emotion regulation strategy in individuals with major
24 depressive disorder. Behaviour Research and Therapy, 82, 1-10. DOI: 10.1016/j.brat.2016.04.003.
Dunkley, D. M., Sanislow, C. A., Grilo, C. M., & McGlashan, T. H. (2009). Self-criticism versus neuroticism in predicting depression and psychosocial impairment for 4 years in a clinical sample. Comprehensive Psychiatry, 50(4), 335-346. DOI: 10.1016/j.comppsych.2008.09.004
Dunkley, D. M., Zuroff, D. C., & Blankstein, K. R. (2003). Self-critical perfectionism and daily affect: Dispositional and situational influences on stress and coping. Journal of Personality and Social Psychology, 84(1), 234. DOI: 10.1037/0022-3514.84.1.234.
Ehret, A. M., Joormann, J., & Berking, M. (2015). Examining risk and resilience factors for depression: The role of self-criticism and self-compassion. Cognition and Emotion, 29(8), 1496-1504. DOI: 10.1080/02699931.2014.992394.
Ehret, A. M., Kowalsky, J., Rief, W., Hiller, W., & Berking, M. (2014). Reducing symptoms of major depressive disorder through a systematic training of general emotion regulation skills: protocol of a randomized controlled trial. BMC Psychiatry, 14(1), 20. DOI: 10.1186/1471-244x-14-20.
Feldman, C., & Kuyken, W. (2011). Compassion in the landscape of suffering. Contemporary Buddhism, 12(01), 143-155. DOI: 10.1080/14639947.2011.564831.
Garnefski, N., Kraaij, V., & Spinhoven, P. (2001). Negative life events, cognitive emotion regulation and emotional problems. Personality and Individual Differences, 30(8), 1311-1327. DOI: 10.1016/s0191-8869(00)00113-6.
Garnefski, N., Rieffe, C., Jellesma, F., Terwogt, M. M., & Kraaij, V. (2007). Cognitive emotion
regulation strategies and emotional problems in 9–11-year-old children. European Child & Adolescent Psychiatry, 16(1), 1-9. DOI: 9 DOI 10.1007/s00787-006-0562-3.
Gilbert, P. (2010). An introduction to compassion-focused therapy in cognitive behavior therapy. International Journal of Cognitive Therapy, 3, 97–112. DOI: 10.1521/ijct. 2010.3.2.97.
Gilbert, P., Baldwin, M. W., Irons, C., Baccus, J. R., & Palmer, M. (2006). Self-criticism and self-warmth: An imagery study exploring their relation to depression. Journal of Cognitive Psychotherapy, 20(2), 183-200. DOI: 10.1891/jcop.20.2.183.
Gilbert, P., & Procter, S. (2006). Compassionate mind training for people with high shame and self-criticism: Overview and pilot study of a group therapy approach. Clinical Psychology & Psychotherapy, 13(6), 353-379. DOI: 10.1002/cpp.507.
Goldin, P. R., McRae, K., Ramel, W., & Gross, J. J. (2007). The neural bases of emotion regulation: Reappraisal and suppression of negative emotion. Biological Psychiatry, 63, 577–586. DOI:
10.1016/j.biopsych.2007.05.031.
Gross, J. J. (1998). The emerging field of emotion regulation: An integrative review. Review of General Psychology, 2(3), 271. DOI: 10.1037/1089-2680.2.3.271.
Gross, J. J. (2015). Emotion Regulation: Current status and future prospects. Psychological Inquiry, 26(1), 1-26. DOI: 10.1080/1047840X.2014.940781.
Gyurak, A., Gross, J. J., & Etkin, A. (2011). Explicit and implicit emotion regulation: a dual-process framework. Cognition and Emotion, 25(3), 400-412. DOI: 10.1080/ 02699931. 2010. 544160
25 Hein, G., & Singer, T. (2008). I feel how you feel but not always: the empathic brain and its
modulation. Current Opinion in Neurobiology, 18 (2), 153-158. DOI: 10.1016/j.conb.2008.07.012. Jazaieri, H., McGonigal, K., Jinpa, T., Doty, J. R., Gross, J. J., & Goldin, P. R. (2014). A randomized
controlled trial of compassion cultivation training: Effects on mindfulness, affect, and emotion regulation. Motivation and Emotion, 38(1), 23-35. DOI: 10.1007/s11031-013-9368-z.
Joormann, J., & Gotlib, H. (2007). Selective attention to emotional faces following recovery from depression. Journal of Abnormal Psychology, 116 (1), 80-85. DOI: 10.1037/0021-843X.116.1.80.
Koole, S. L., & Rothermund, K. (2011). “I feel better but I don't know why”: The psychology of implicit emotion regulation. Cognition and Emotion, 25(3), 389-399. DOI: 10.1080/02699931.2010. 550505.
Koole, S. L., Webb, T. L., & Sheeran, P. L. (2015). Implicit emotion regulation: feeling better without knowing why. Current Opinion in Psychology, 3, 6-10. DOI: 10.1016/j.copsyc.2014.12.027.
Koster, E., De Raedt, R., Leyman, L., & Lissnyder, E. (2010). Mood-congruent attention and memory bias in dysphoria: exploring the coherence among information-processing biases. Behaviour Research and Therapy, 48 (3), 219-225. DOI: 10.1016/j.brat.2009.11.004.
Kotsou, I., & Leys, C. (2016). Self-Compassion Scale (SCS): Psychometric properties of the French translation and its relations with psychological well-being, affect and depression. PloS one, 11(4), e0152880. DOI: 10.1371/journal.pone.0152880.
Kuyken, W., Watkins, E., Holden, E., White, K., Taylor, R. S., Byford, S., ... & Dalgleish, T. (2010). How does mindfulness-based cognitive therapy work?. Behaviour Research and Therapy, 48(11), 1105-1112. DOI: 10.1016/j.brat.2010.08.003.
Leary, M., Tate, E., Adams, C., Batts Allen, A., & Hancock, J. (2007). Self-compassion and reactions to unpleasant self-relevant events: the implications of treating oneself kindly. Journal of Personality and Social Psychology, 92(5), 887. DOI: 10.1037/0022-3514.92.5.887.
Liverant, G. I., Brown, T. A., Barlow, D. H., & Roemer, L. (2008). Emotion regulation in unipolar depression: The effects of acceptance and suppression of subjective emotional experience on the intensity and duration of sadness and negative affect. Behaviour Research and Therapy, 46(11), 1201-1209. DOI: 10.1016/j.brat.2008.08.001.
MacBeth, A., & Gumley, A. (2012). Exploring compassion: A meta-analysis of the association between self-compassion and psychopathology. Clinical Psychology Review, 32, 545–552. DOI: 10.1016/j.cpr.2012.06.003.
Mauss, I. B., Cook, C. L., & Gross, J. J. (2007). Automatic emotion regulation during anger provocation. Journal of Experimental Social Psychology, 43(5), 698-711. DOI: 10.1016/ j.jesp.2006.07.003.
Mojtabai, R., Olfson, M., & Han, B. (2016). National Trends in the Prevalence and Treatment of Depression in Adolescents and Young Adults. Pediatrics, 136(6), 1-10. DOI: 10.1542/peds.2016-1878.
