Angry Rumination Mediates the Unique Associations Between Self-Compassion and Anger and Aggression

Mindfulness DOI 10.1007/s12671-016-0629-2

ORIGINAL PAPER

Amanda Fresnics1 & Ashley Borders1

# Springer Science+Business Media New York 2016

Abstract Mindfulness is known to decrease anger and ag- be useful for developing clinical interventions targeting anger gression. Self-compassion is a related and relatively new con- and aggressive behavior. struct that may predict other clinical outcomes more strongly than does mindfulness. Little research has focused on whether Keywords Self-compassion . Mindfulness . Angry self-compassion is related to anger and aggression, and no rumination . Anger . Aggression studies have explored mechanisms of these associations. The current survey study explores whether angry rumination mediates the unique associations between self-compassion and Introduction anger and aggression, controlling for trait mindfulness. Two hundred and one undergraduates completed questionnaires Mindfulness has known benefits for individuals with anger and assessing self-compassion, mindfulness, angry rumination, aggression problems (Heppner et al. 2008). Self-compassion is and recent anger and aggression. Supporting our hypotheses, a related construct that has received recent empirical attention angry rumination mediated the associations between self- because of its potential to reduce suffering. This positive attri- compassion—particularly its over-identification subscale— bute consists of extending a loving, non-judgmental attitude and anger and aggression when controlling for mindfulness. towards oneself in the face of failure, rejection, or pain (Neff Mindfulness did not predict angry rumination, recent anger, or 2003). Self-compassion contributes to better processing of aggression when controlling for self-compassion. negative emotions (Leary et al. 2007) and is associated with Furthermore, multiple regression analyses predicting less self-criticism, depression, depressive rumination, and anx- aggression-related variables indicated that angry rumination iety (Barnard and Curry 2011; Neff 2003). Moreover, some uniquely predicted over-identification, one of the six self- research suggests that self-compassion may be a stronger pre- compassion subscales. These findings suggest that self-com- dictor of these outcomes than is mindfulness (e.g., Van Dam passion, particularly a lack of cognitive and emotion fusion, et al. 2011;Woodruffetal.2014). Whereas most of the self- may be a more proximal predictor of clinical outcomes than compassion literature focuses on internalizing symptoms, we mindfulness. Implications for current conceptualizations and know much less about its association with externalizing prob- measures of mindfulness are discussed. Self-compassion may lems. A better understanding of the unique effects and mechanisms of self-compassion on anger and aggression could suggest potential directions for clinical interventions targeting externalizing symptoms. * Ashley Borders Self-compassion is a way of relating to oneself when suf- [email protected] fering. Individuals who are self-compassionate (a) treat them- Amanda Fresnics selves with kindness, (b) understand that suffering is a com- [email protected] mon human experience, and (c) are able to hold painful thoughts and emotions in balanced awareness (Neff 2003). 1 Psychology Department, The College of New Jersey, 2000 As commonly measured, each of these three facets of self- Pennington Rd, PO Box 7718, Ewing, NJ 08628, USA compassion is assessed by two opposing subscales (Neff Mindfulness

2003). For instance, (a1) self-kindness, or self-soothing and equally strong predictors (Baer et al. 2012). It is therefore useful self-comforting when facing pain, contrasts with (a2) self- to examine the unique effect of self-compassion on externaliz- judgment, or harsh condemnation and self-blame. Similarly, ing behaviors, controlling for mindfulness, in order to to sepa- (b1) common humanity assesses an understanding that pain rate the effects of the two constructs. Furthermore, because self- and suffering are shared human experiences, whereas (b2) compassion can be analyzed as both a general factor and by its isolation assesses the belief that individuals are unique in their components (Neff 2016;Neffetal.in press), exploring the suffering. The third pair of opposing subscales are called (c1) unique effects of the six subscales of self-compassion may mindfulness and (c2) over-identification. Neff (2003)de- provide additional understanding of which component(s) con- scribes mindfulness as an attitude of equanimity towards un- tribute the most to externalizing problems. pleasant thoughts and emotions, contributing to a state of Research has already shown that mindfulness is associated mental balance. By contrast, over-identification involves fix- with less anger and aggression. Cross-sectional analyses ation or fusion with one’s internal experiences. found that dispositional mindfulness was associated with less Of course, mindfulness has received substantial attention as aggression and hostile attributions (Heppner et al. 2008). In an its own construct. Researchers generally define it as a unique experimental follow-up, participants who received a mindful- way of paying attention (Bishop et al. 2004; Shapiro et al. 2006). ness induction prior to social rejection responded with less Mindfulness involves bringing one’s full attention to the present aggressive behavior than did participants who only received moment, so that individuals are aware of their moment-to- social rejection (Heppner et al. 2008). In addition, moment internal and external experiences. Furthermore, mindful mindfulness-based interventions led to decreases in hostility individuals intentionally approach their experiences with an at- and anger in prison inmates, cancer patients, and health-care titude of acceptance, curiosity, and non-judgment. Clearly, self- professionals (Galantino et al. 2005; Samuelson et al. 2007; compassion and mindfulness are related constructs, in that they Speca et al. 2000). Researchers have also examined possible both promote experiencing discomfort with an accepting attitude mechanisms between mindfulness and aggression. Past cross- (Neff and Dahm 2014). Not surprisingly, self-compassion and sectional studies found that both general and angry rumination mindfulness measures share between 13 and 48 % of their var- mediated the associations between mindfulness and hostility, iance (e.g., Baer et al. 2006;Neff2003). Moreover, self- anger, and aggression (Borders et al. 2010;Petersetal.2015). compassion interventions increase mindfulness (Neff and Therefore, evidence suggests that mindfulness is associated Germer 2013;Smeetsetal.2014), and mindfulness interven- with aggression-related variables, possibly via a mechanism tions increase self-compassion (Baer 2010; Shapiro et al. 2007). of rumination. However, none of these studies included mea- However, theory and empirical evidence suggest that self- sures of self-compassion, so the unique effect of mindfulness compassion and mindfulness are distinct constructs. versus self-compassion could not be examined. It is possible Mindfulness is best conceptualized as a form of attention reg- that both mindfulness and self-compassion uniquely contrib- ulation and therefore applies to all experiences, whether posi- ute to less anger and aggression. Alternatively, one may tive, negative, or neutral (Neff and Dahm 2014). Not all expe- emerge as the stronger predictor of these particular outcomes. riences that one is mindful of require self-compassion. A few studies have explored whether self-compassion is Conversely, self-compassion is at root an attitude towards suf- associated with anger and aggression. For instance, self- fering, rather than a type of attention (Neff and Dahm 2014). compassion was associated with fewer angry responses to Thus, the mindfulness component of self-compassion, and its hypothetical situations (Neff and Vonk 2009). Similarly, opposing subscale over-identification, differs from the general self-compassion predicted less anger when recalling emo- construct of mindfulness in that it specifically relates to person- tionally upsetting sports situations in female athletes (Reis al suffering rather than general awareness of all experiences. In et al. 2015). In terms of aggression, men and women who addition to mindfulness of suffering, self-compassion also en- lacked self-compassion reported being more verbally ag- compasses an individual’s emotional response (self-kindness gressive towards their romantic partners (Neff and vs. self-judgment) and perspective on suffering (common hu- Beretvas 2013). Self-compassion was also associated with manity vs. isolation). Empirically, self-compassion and mind- less dispositional proactive and reactive aggression in at- fulness account for unique variance in well-being (Baer et al. risk adolescent males (Barry et al. 2015). When analyzing 2012;VanDametal.2011), suggesting that they are not iden- self-compassion subscales separately, self-kindness tical constructs. Researchers have also started to compare their uniquely predicted less proactive aggression, whereas iso- relative predictive strength. Self-compassion was a better pre- lation and over-identification uniquely predicted more re- dictor than mindfulness of depression, anxiety, and quality of active aggression (Barry et al. 2015). Thus, four previous life in both clinical and non-clinical samples (Van Dam et al. studies suggest that less self-compassion is associated with 2011; Woodruff et al. 2014). Self-compassion was also twice as more anger and aggressive behaviors in various popula- strong a predictor of well-being than mindfulness (using full- tions. However, none of these studies controlled for mind- scale scores), although some subscales of both constructs were fulness. Furthermore, understanding mechanisms of the Mindfulness associations between self-compassion and both anger and forgiveness, and a kind-hearted understanding of others’ pain aggression could inform treatment programs. and shortcomings (Neff and Pommier 2013). By contrast, an- Rumination is defined as repetitively and passively focus- gry rumination is associated with less other-focused processes ing on negative internal or external events (Watkins 2008). such as empathy and forgiveness (Barber et al. 2005;Boyraz Researchers have defined different types of rumination (e.g., and Waits 2015). Therefore, theory and circumstantial evi- depressive, angry) based on the predominant negative content dence suggest that self-compassion may decrease angry rumi- of the repetitive thoughts (Nolen-Hoeksema et al. 2008; nation, which should in turn decrease subsequent anger and Sukhodolsky et al. 2001). Depressive rumination involves a aggressive behavior. repetitive focus on symptoms of depression and contributes to The current survey study explored the unique associations worse depression and anxiety (for reviews, see Nolen- between trait self-compassion, angry rumination, and recent Hoeksema et al. 2008; Watkins 2008). By contrast, angry anger and aggression in a sample of undergraduate students. rumination involves perseverating about angering events, We hypothesized that self-compassion (measured as a single brooding over perceived injustices, and fantasizing about re- factor) would be associated with less recent anger and aggres- venge (Sukhodolsky et al. 2001). Angry rumination prolongs sion, even after controlling for trait mindfulness. We also ex- angry mood and leads to aggressive behavior (for a review, see amined which of the six self-compassion components unique- Denson 2013). For instance, a past study found that trait angry ly predicted angry rumination, anger, and aggression. Third, rumination predicted aggressive behavior in response to prov- although cross-sectional survey research cannot prove a caus- ocation (Collins and Bell 1997). Moreover, experimentally al mediational model, we expected to find a pattern of corre- induced rumination about a prior provocation increased self- lations consistent with less angry rumination statistically me- reported anger and aggressive behavior (Bushman 2002; diating the unique associations between greater self- Denson et al. 2011). Thus, angry rumination is a robust pre- compassion and less anger and aggression. To further test dictor of anger and aggressive behavior. the relative predictive strength of self-compassion versus Although self-compassion has never been studied in asso- mindfulness, we ran a mediational model with mindfulness ciation with angry rumination, it is associated with less de- as the predictor and self-compassion as the covariate. pressive rumination. For instance, self-compassion predicted less depressive rumination in an undergraduate population (Neff et al. 2007). Furthermore, a self-compassion interven- Method tion for college women led to a decline in depressive rumination, compared to a time management control group (Smeets Participants et al. 2014). A cross-sectional study of emerging adults found that brooding, one component of depressive rumination, me- The sample consisted of 201 participants (M =19.48,SD= diated the relationship between self-compassion and both de- age 1.47, range = 18–29) from a small liberal arts college in the pressive and anxious symptoms (Raes 2010). Depressive ru- Northeast. Approximately 78 % of the sample identified as mination also mediated the association between self- female, 21 % as male, and 1 % as transgender. Most partici- compassion and depressive symptoms in a clinically de- pants (96 %) self-identified as heterosexual, with 1.5 % iden- pressed population (Krieger et al. 2013). These findings sug- tifying as bisexual, 1.5 % as lesbian/gay, and 1 % as gest that greater self-compassion is associated with less de- questioning. The ethnic breakdown was 5 % African pressive rumination, which in turn predicts less internalizing American/Black, 9 % Asian/Asian American, 73 % symptoms. European American/White, 8 % Hispanic or Latino/a, 1 % In the same way that depressive rumination explains the PacificIslander,3%AsianIndian,and1%other. link between self-compassion and depression, angry rumination may explain the associations between self-compassion and aggression-related variables. Theoretically, rumination Procedure functions to solve perceived discrepancies between individuals’ current states and desired goals (Watkins 2008). Thus, Participants were recruited from the psychology participant when people are angry about insults or frustrations, they may pool and received course credit for their participation. They ruminate in an attempt to resolve the perceived discrepancy. It completed online surveys. Informed consent was obtained is likely that the ability to self-comfort and decenter from from all individual participants in the study. The order of mea- angry emotions and an appreciation that all individuals suffer sures and items within each measure were randomized. These will reduce individuals’ perceived discrepancy in response to data were taken from a week-long experience sampling study an angering event. This in turn should lead to less angry ru- on the association between state rumination and anger. The mination. There is evidence that self-compassion promotes measures used in this study were completed at the initial meet- other-focused processes such as perspective-taking, ing. All procedures performed were in accordance with the Mindfulness

APA Ethical Principles and the standards established by The while^. Items were measured on a 4-point scale ranging from College of New Jersey’s Institutional Review Board. 1(not at all)to4(almost always) and were averaged together. Higher total scores indicate more angry rumination. This mea- Measures sure is highly correlated with anger experience, anger expression, negative affectivity, and emotional attention, suggesting Self-Compassion The 26-item Self-Compassion Scale (Neff construct validity (Sukhodolsky et al. 2001). Internal consis- 2003) assesses how participants typically react when facing a tency in our sample was high (α =0.92). struggle or stressor. It consists of six subscales: self-kindness (e.g., BI’m kind to myself when I’m experiencing suffering^), Recent Anger We measured anger in the past week using the self-judgment (e.g., BI’m disapproving and judgmental about anger subscale of the State Trait Anger Expression Inventory my own flaws and inadequacies^), common humanity (e.g., BI 2 (STAXI-2; Spielberger 1999). Example items include BIam try to see my failings as part of the human condition^), isola- irritated^, BI feel like hitting someone^ and BI feel like yelling tion (e.g., BWhen I think about my inadequacies, it tends to at somebody^. Participants answered on a 4-point scale rang- make me feel more separate and cut off from the rest of the ing from 1 (Not at All)to4(Very much so), and higher aver- world^), mindfulness (e.g., BWhen I fail at something impor- aged total scores indicate greater recent anger. The STAXI-2 is tant to me I try to keep things in perspective^), and over- a widely known scale that has been used to measure anger in identification (e.g., BWhen something upsets me I get carried youth, college students, and peoples across various ethnic cul- away with my feelings^). Items were assessed on a 5-point tures (Maxwell et al. 2009; Wongtongkam et al. 2013). scale from 1 (almost never)to5(almost always), and higher Internal consistency in this sample was good (α =0.88). averaged total scores indicate greater self-compassion. The self-judgment, isolation, and over-identification subscales Recent Aggression We measured recent aggression with an are reversed scored. This measure is correlated with positive adapted version of The Explicit Aggression Scale (Borders mental health outcomes, self-esteem, life satisfaction, suggest- et al. 2007). This 17-item measure asks participants to report ing construct validity (Neff 2003). Cronbach’s α for the total the amount of times (e.g., 1, 2, 3–5, 6–10, 11–19, 20+) they measure was 0.93. Reliabilities for the six subscales in our engaged in different types of aggressive behavior (e.g., BIhave sample were also adequate: self-kindness (α = 0.86), self- slapped or hit someone^) or felt hostile (e.g., BI have been judgment (α =0.85),commonhumanity(α =0.83),isolation eaten up with jealousy^) during the past 6 months. Items (α = 0.72), mindfulness (α = 0.77), and over-identification assessing alcohol-related aggression were excluded in this (α =0.80). study. For the purpose of analyses, we employed the lowest number from each interval to indicate the participants’ actual Mindfulness We measured trait mindfulness with the short number of experiences with alcohol-related aggression. This form of the Freiburg Mindfulness Inventory (Walach et al. approach has proven useful in previous work (Borders et al. 2006). This 14-item measure includes questions that are un- 2007). Higher summed scores indicate greater recent aggres- derstood by individuals unfamiliar with the Buddhist tradition. sion. The measure correlates with dispositional aggression It assesses openness to experience (e.g., BI am open to the (Barnwelletal.2006), demonstrating construct validity. experience of the present moment^), mindful presence (e.g., Cronbach’s α in this sample was 0.86. BI sense my body whether cooking, cleaning, or eating^), non- judgmental acceptance (e.g., BI see my mistakes and difficul- Data Analyses ties without judging them^), and insight (e.g., BI pay attention to what’sbehindmyactions^). Items were assessed on a 4- We examined mediation by calculating indirect effects, which point scale ranging from 1 (rarely)to4(almost always)and are defined as the product of the path coefficients for the (a) averaged together. Higher total scores indicate greater mind- independent variable to the mediator and the (b) mediator to fulness. This measure correlates with other measures of mind- the dependent variable after controlling for the independent fulness (Baer et al. 2006; Feldman et al. 2007), suggesting variable (Preacher and Hayes 2008). An indirect effect that is convergent validity. Cronbach’s α in this sample was 0.81. significantly different from zero suggests that the mediator is We will refer to this measure as FMI-mindfulness below. accounting for a significant proportion of the variance between the independent and dependent variables. Using the Angry Rumination The 19-item Anger Rumination Scale SPSS macro PROCESS (Hayes 2012), we tested significance (Sukhodolsky et al. 2001) assesses anger afterthoughts, angry of indirect effects using bootstrapping methodology (Preacher memories, fantasies about revenge, and understanding of and Hayes 2008). This procedure creates an empirical sam- causes. Examples of items were BI ruminate about my past pling distribution of 5,000 bootstrap samples, from which in- anger experiences^, BI analyze events that make me angry^, direct effects and their 95 % confidence intervals are estimat- and BMemories of even minor annoyances bother me for a ed. Recent simulation studies suggest that the bias-corrected Mindfulness bootstrap produces the most accurate confidence intervals First, we examined whether angry rumination mediated the (Cheung and Lau 2008; MacKinnon et al. 2004). Therefore, associations between total self-compassion and aggression- we calculated the bootstrap estimates and bias-corrected con- related variables (see Fig. 1). We ran separate analyses fidence intervals for each indirect effect. Any indirect effect predicting anger and aggression, both with and without con- whose confidence interval does not include zero indicates a trolling for FMI-mindfulness. As seen in Table 2, total self- significant effect of mediation. compassion predicted aggression and angry rumination, whether or not we controlled for FMI-mindfulness. Total self-compassion significantly predicted anger only when not Results controlling for FMI-mindfulness. When angry rumination was included in the models predicting anger, the relationship be- Variables were first assessed for deviations from normality. tween self-compassion and anger was non-significant. Scores for anger and aggression were positively skewed Moreover, the confidence intervals for the indirect effects (skewness values = 1.75 and 2.14, respectively), whereas the did not contain 0, suggesting significant mediation. When other variables were normally distributed. Log transforma- angry rumination was included in the models predicting ag- tions were performed on aggression and anger to achieve ac- gression, the association between self-compassion and aggres- ceptable normality (Tabachnick and Fidell 2006). We used the sion was non-significant, and the indirect effects were again transformed variables to calculate all statistics. Men and wom- significantly different from 0. Thus, angry rumination statisti- en did not differ significantly on any study variables. cally mediated the associations between total self-compassion Descriptive statistics and zero-order intercorrelations for all and less anger and aggression, whether or not we controlled variables are presented in Table 1. As hypothesized, greater for FMI-mindfulness. total self-compassion was associated with more FMI- We next examined the unique effects of the six self- mindfulness and less angry rumination, anger, and aggression. compassion subscales. We first conducted multiple regression Regarding the self-compassion subscales, angry rumination analyses with the six self-compassion subscales and FMI- and aggression were significantly associated with greater mindfulness predicting angry rumination, anger, and aggres- self-judgment, isolation, and over-identification and with less sion (see Table 3). Only the over-identification subscale self-kindness, common humanity, and mindfulness; anger was uniquely predicted angry rumination. None of the self- only significantly correlated with more self-judgment and compassion subscales uniquely predicted anger or aggression. over-identification. In line with previous research, greater These results remained the same even if we did not include FMI-mindfulness was associated with less angry rumination, FMI-mindfulness in the analyses (results not shown). Because anger, and aggression. Furthermore, greater angry rumination of the unique effect of over-identification on angry rumina- was associated with more anger and aggression. tion, we re-ran the above mediation analyses, replacing total

Table 1 Descriptive statistics and correlations among study variables

Variables 1 2345678 9 1011

1. Total self-compassion – 2. Self-kindness subscale 0.84** – 3. Self-judgment subscale −0.84** −0.65** – 4. Common humanity subscale 0.65** 0.55** –0.36** – 5. Isolation subscale −0.72** −0.42** 0.64** −0.34** – 6. Mindfulness subscale 0.70** 0.65** −0.37** 0.55** −0.30** – 7. Over-identification subscale −0.78** −0.50** 0.68** −0.29** 0.61** −0.46** – 8. FMI-Mindfulness 0.67** 0.65** −0.45** 0.48** −0.36** 0.65** −0.46** – 9. Angry rumination −0.50** −0.34** 0.50** −0.25** 0.40** −0.27** 0.50** −0.30** – 10. Angera −0.16* −0.13 0.19** −0.02 0.09 −0.04 0.16* −0.16** 0.33** – 11. Aggressiona −0.28** −0.16* 0.26** −0.13 0.20** −0.18* 0.30** −0.22** 0.38** 0.32** – Mean 2.97 3.02 3.20 3.10 3.26 3.34 3.27 2.62 2.13 1.57 37.70 SD 0.65 0.82 0.90 0.90 0.84 0.75 0.95 0.46 0.59 0.50 44.33 Range 1.23–4.54 1.2–51–51–51–51–51–51.5–3.79 1–3.74 1–3.90 0–261 a For ease of interpretation, we report raw means and standard deviations. We used the transformed variables to calculate all statistics Notes: FMI Freiburg Mindfulness Inventory *p <0.05;**p <0.01 Mindfulness

Fig. 1 Representation of mediation pathways (coefficients presented in Table 2): a Total effect model of self-compassion predicting aggression-related variables. b Mediation model with angry rumination mediating the association between self- compassion and aggression- related variables

self-compassion with the over-identification subscale. when we used the over-identification subscale instead of total Specifically, we examined whether angry rumination mediat- self-compassion (results not shown). Thus, after controlling ed the relationship between over-identification and for self-compassion, we no longer found support for angry aggression-related variables, with and without controlling for rumination mediating the associations between FMI- FMI-mindfulness (see Table 2). Once again, the indirect ef- mindfulness and anger and aggression. fects were significantly different from 0, suggesting mediation. Finally, we switched the roles of self-compassion and FMI- Discussion mindfulness. In other words, we tested whether angry rumination mediated the associations between FMI-mindfulness As hypothesized, self-compassion was negatively associated and aggression-related variables. We again ran separate anal- with recent anger and aggression. These results are in line with yses predicting anger and aggression, both with and without past evidence (Barry et al. 2015; Neff and Beretvas 2013). controlling for self-compassion (see Table 4). When we did Unlike previous studies that examined aggression in at-risk not control for total self-compassion, FMI-mindfulness signif- adolescent males or romantic relationships, we assessed anger icantly predicted anger, aggression, and angry rumination. and aggressive behavior in undergraduates, a population in Moreover, both indirect effects were significant. However, which engagement in verbal and physical aggression is com- when controlling for total self-compassion, FMI-mindfulness mon (Leonard et al. 2002). We also measured the frequency of did not significantly predict angry rumination or the anger and aggressive behavior in the past six months, whereas aggression-related variables. Accordingly, both indirect ef- previous studies assessed dispositional aggression or aggres- fects were non-significant. This pattern of results was the same sion in response to hypothetical situations (Barry et al. 2015;

Table 2 Mediation coefficients with angry rumination as a mediator, without and with controlling for FMI-mindfulness

Variables a path b path c path c` path Indirect effect [CI]

Total self-compassion (IV) Anger (DV) −0.50**/−0.54** 0.35**/0.35** −0.16*/−0.06 0.01/0.13 −0.02 [−0.01 to −0.04] / −0.02 [−0.01 to −0.04] Aggression (DV) −0.50**/−0.54** 0.33**/0.35** −0.28**/−0.21** −0.12/−0.01 −0.16 [−0.09 to −0.25] / −0.19 [−0.10 to −0.29] Over-identification (IV) Anger (DV) 0.50**/0.46** 0.34**/0.34** 0.16*/0.10 −0.01/ −0.06 0.01 [0.007 to 0.02] / 0.01 [0.006 to 0.02] Aggression (DV) 0.50**/0.46** 0.33*/0.34** 0.30**/0.23* 0.13/ 0.08 0.11 [0.06 to 0.18] / 0.10 [0.06 to 0.17]

Note. FMI Freiburg Mindfulness Inventory. Paths correspond with Fig. 1, and coefficients are standardized. Values following the dash indicate models that controlled for FMI-Mindfulness. Unstandardized bias-corrected 95 % confidence intervals (CI) that do not contain 0 indicate a significant indirect effect. *p <0.05,**p <0.01 Mindfulness

Table 3 Multiple regression analysis predicting angry Predictors Angry Rumination Anger Aggression rumination, anger, and aggression β t β t β t

1. Self-kindness subscale 0.014 0.13 0.05 0.37 0.11 0.86 2. Self-judgment subscale 0.17 1.62 0.19 1.51 0.17 1.39 3. Common humanity subscale −0.10 −1.28 0.05 0.54 −0.01 −0.09 4. Isolation subscale 0.06 0.67 −0.08 −0.79 −0.02 −0.25 5. Mindfulness subscale −0.004 −0.046 0.12 1.06 0.03 0.23 6. Over-identification subscale 0.32 3.36* 0.09 0.84 0.18 1.65 7. FMI-Mindfulness −0.03 −0.37 −0.22 −2.16* −0.17 −1.65

Note. FMI Freiburg Mindfulness Inventory *p < 0.05, **p <0.01

Neff and Vonk 2009). Thus, evidence is building that self- our findings with over-identification, cognitive fusion and ru- compassion predicts less anger and aggression measured in mination are highly correlated yet distinct constructs different ways in various populations. (Gillanders et al. 2014). Taken together, these findings suggest Moreover, angry rumination statistically mediated the that merging with one’s thoughts and internal experiences, unique associations between self-compassion and anger and whether it is called over-identification or cognitive fusion, aggression. These findings parallel self-compassion’s nega- engenders maladaptive repetitive thinking. Researchers next tive association with depressive rumination, which in turn need to understand potential mechanisms of this association. predicts less depression (Neff et al. 2007; Neff and Vonk For instance, over-identification with one’s internal experi- 2009). Although we cannot make causal conclusions from ence may discourage perspective taking and forgiveness of cross-sectional data, these results suggest that one possible others (Neff and Pommier 2013), constructs that are also as- way self-compassion reduces anger and aggression is by de- sociated with angry rumination (Barber et al. 2005). Future creasing angry rumination. Our subscale analysis provided research might examine these and other potential mediators of some insight into the association between self-compassion the association between over-identification and angry and angry rumination. Although no self-compassion subscales rumination. uniquely predicted anger or aggression, over-identification In order to compare the relative predictive strength of self- uniquely predicted angry rumination. Interestingly, previous compassion and mindfulness, we examined the unique contri- research found that the over-identification subscale uniquely butions of self-compassion and mindfulness to angry rumina- predicted worry (Van Dam et al. 2011), another form of repet- tion, anger, and aggression. Across both full-scale and sub- itive thought that focuses on future threats, imagined risks, scale analyses, self-compassion predicted aggression-related and uncertainties (Watkins 2008). Both worry and rumination variables, whether or not we controlled for mindfulness. By involve analytical reflection (Trapnell and Campbell 1999; contrast, mindfulness did not predict these variables after con- Watkins 2008), and over-identification may create an internal trolling for self-compassion. Thus, it is possible that self- climate that encourages this kind of self-focused thinking. compassion was a stronger predictor of angry rumination Additionally, over-identification is similar to another construct and subsequent anger and aggression than was mindfulness. called cognitive fusion, which refers to a preoccupation with These findings fit with previous research showing that self- thoughts to such a degree that behavior is dictated by thoughts compassion more strongly predicted internalizing problems rather than by direct experience (Hayes et al. 2011). Similar to like depression and anxiety than did mindfulness (Van Dam

Table 4 Mediation coefficients (pathways represented in Fig. 1) with FMI-mindfulness as the independent variable and angry rumination as the mediator, without and with controlling for self-compassion

Variables a path b path c path c` path Indirect effect [CI]

Anger (DV) −0.30**/0.04 0.31**/0.35** −0.16*/−0.15 −0.07/−0.16 −0.02 [−0.007 to −0.03] / 0.003 [−0.01 to 0.02] Aggression (DV) −0.23**/−0.20** 0.36**/0.35** −0.22**/−0.10 −0.12/−0.12 −0.15 [−0.08 to −0.25] / 0.02 [−0.06 to 0.13]

Note. FMI Freiburg Mindfulness Inventory. Coefficients are standardized. Values following the dash indicate models that controlled for total self- compassion. Unstandardized bias-corrected 95 % confidence intervals (CI) that do not contain 0 indicate a significant indirect effect *p <0.05;**p <0.01 Mindfulness et al. 2011; Woodruff et al. 2014). However, it must be re- mindfulness measure with multiple subscales to the six self- membered that over-identification is considered the flip-side compassion subscales. Finally, we did not assess participants’ of the mindfulness component of self-compassion. Thus, the exposure to meditation and mindfulness-related ideas. Past driving force behind angry rumination could be a lack of research controlling for meditation experience still found that mindfulness. However, mindfulness (and the lack thereof) as self-compassion was a stronger predictor of psychological assessed by the Self-Compassion Scale is distinct from the well-being than was mindfulness (Baer et al. 2012). construct of general mindfulness as measured by mindfulness Nevertheless, future research might examine whether medita- questionnaires. Specifically, mindfulness and over- tion experience moderates the effect of self-compassion. identification in the context of self-compassion relate to hold- Despite these limitations, there are several potential impli- ing one’s suffering in a balanced awareness without being cations of our findings. First, our results suggest a reinterpre- fused or over-identified with it (Neff 2003). By contrast, com- tation of previous findings that rumination mediated the asso- mon mindfulness measures (including the one we used) assess ciations between mindfulness and anger and aggression a general awareness of all experiences, rather than mindful- (Borders et al. 2010; Peters et al. 2015). These previous find- ness of or fusion with personal distress in times of suffering ings may have reflected the effect of self-compassion, partic- (Neff and Dahm 2014). Therefore, we do not conclude that ularly awareness of and nonattachment to suffering that are self-compassion has an effect on angry rumination that mind- not specifically assessed in most mindfulness measures. fulness does not. Rather, we suggest that lack of mindfulness Second, self-compassion and its over-identification compo- of and fusion with one’s suffering (i.e., over-identification), nent, or fusion with internal experiences, may be more closely not mindfulness of one’s general experience, predicts angry related to mental health outcomes than is general mindfulness. rumination. Mindfulness research might benefit from having Whereas mindfulness involves more general processes such the ability to separate the effects of general awareness of ex- as attention regulation and non-judgment of all experiences, perience from mindfulness specifically of internal distress. self-compassion explicitly focuses on how one relates to one’s Furthermore, because one benefit of mindfulness is thought own suffering. Because suffering arguably underlies most to be a decrease of cognitive fusion, measures of mindfulness forms of psychological distress, individuals’ attitude towards should include items that explicitly focus on lack of over- their suffering might be a more proximal predictor and there- identification, or cognitive fusion. fore contribute more variance to clinical outcomes. This study has some notable limitations. Due to the cross- Relatedly, ours and others ’ results may shed light on im- sectional design, we cannot explicitly test a causal mediational portant mechanism of mindfulness-based interventions. model. More specifically, we cannot know the temporal asso- Although it is not the focus of these programs, self- ciations between our constructs. For instance, although we compassion may be a key mechanism by which suggest that self-compassion decreases angry rumination, the mindfulness-based interventions improve well-being (e.g., opposite direction is also possible. Engaging in angry rumina- Baer 2010). As mentioned above, these treatments increase tion could limit the ability to extend kindness to the self be- self-compassion (Baer 2010; Shapiro et al. 2005). In fact, cause rumination encourages preoccupation with negative previous studies found that increased self-compassion medi- thoughts and judgments involving others. Additionally, we ated the effect of mindfulness interventions on decreases in cannot know the directional association between self- stress and depressive symptoms (Kuyken et al. 2010; Shapiro compassion and mindfulness. Although our results suggest et al. 2005). Our results suggest that decreasing over- that mindfulness precedes self-compassion, some evidence identification with internal experiences may be the driving suggests that self-compassion interventions increase mindful- force behind these associations. Relatedly, Acceptance and ness (Neff and Germer 2013;Smeetsetal.2014). Clearly, the Commitment Therapy (ACT) decreases cognitive fusion and directionality of the association between self-compassion and subsequent clinical outcomes (Arch et al. 2012;Ruiz2010). mindfulness remains an area for further investigation, and a Perhaps mindfulness programs that specifically combat cog- bidirectional relationship is likely. Future studies should ex- nitive fusion and promote mindful awareness and acceptance amine these associations using experimental or longitudinal of suffering may also decrease clinical symptoms. Also, be- designs. cause there is initial evidence that mindfulness-based thera- Another study limitation was our measurement of mindful- pies and ACT lead to less aggressive behavior (Fix and Fix ness. We assessed mindfulness as a unidimensional construct. 2013; Zarling et al. 2015), future research might examine Past research suggests that the association between mindful- whether self-compassion and over-identification are mecha- ness and mental health is stronger when multi-faceted mea- nisms of this relationship. sures of mindfulness are used (Woodruff et al. 2014). Finally, our findings suggest novel treatments for anger and Furthermore, using a unidimensional construct may not pro- aggression problems. Self-compassion interventions are al- vide a comprehensive measure of mindfulness. Future studies ready known to reduce depressive rumination and internaliz- predicting externalizing problems should compare a ing problems like depression and anxiety (Neff and Germer Mindfulness

2013;Smeetsetal.2014). These may also decrease tendencies Baer, R. A., Lykins, E. L., & Peters, J. R. (2012). Mindfulness and self- for angry rumination and subsequent anger and aggressive compassion as predictors of psychological wellbeing in long-term meditators and matched non-meditators. The Journal of Positive behavior. For example, compassion-focused therapy (Gilbert Psychology, 7,230–238. doi:10.1080/17439760.2012.674548. 2010) uses a variety of exercises (e.g., visualizations, self- Barber, L., Maltby, J., & Macaskill, A. (2005). Angry memories and compassionate behaviors) to help clinical patients cultivate thoughts of revenge: the relationship between forgiveness and anger – self-kindness and decrease self-criticism. The mindful self- rumination. Personality and Individual Differences, 39,253262. doi:10.1016/j.paid.2005.01.006. compassion program (Neff and Germer 2013)isgearedto- Barnard, L. K., & Curry, J. F. (2011). Self-compassion: conceptualiza- ward non-clinical populations and employs education, group tions, correlates, & interventions. Review of General Psychology, discussion, loving-kindness practice, and other experiential 15(4), 289. doi:10.1037/a0025754. exercises (e.g., placing one’shandonone’s heart during times Barnwell, S. S., Borders, A., & Earleywine, M. (2006). Alcohol- of stress) to develop self-compassion. Loving kindness (fo- aggression expectancies and dispositional aggression moderate the relationship between alcohol consumption and alcohol-related vio- cused on unconditional kindness to all beings) and compas- lence. Aggressive Behavior, 32,517–527. doi:10.1002/ab.20152. sion (focused on alleviating the suffering of all beings) med- Barry, C. T., Loflin, D. C., & Doucette, H. (2015). Adolescent self-com- itations might be particularly useful for treating anger and passion: associations with narcissism, self-esteem, aggression, and aggression problems, because the very focus of these practices internalizing symptoms in at-risk males. Personality and Individual Differences, 77,118–123. doi:10.1016/j.paid.2014.12.036. is incompatible with anger, hatred, envy, and jealousy Bishop, S. R., Lau, M., Shapiro, S., Carlson, L., Anderson, N. D., (Hofmann et al. 2011). These and other self-compassion pro- Carmody, J., & Devins, G. (2004). Mindfulness: a proposed opera- grams could serve individuals who are prone to aggressive tional definition. Clinical Psychology: Science and Practice, 11, tendencies, such as people with borderline personality disor- 230–241. doi:10.1093/clipsy/bph077. der, incarcerated populations, or at-risk youth. Future self- Borders, A., Barnwell, S. S., & Earleywine, M. (2007). Alcohol- aggression expectancies and dispositional rumination moderate the compassion interventions might also focus specifically on de- effect of alcohol consumption on alcohol-related aggression and veloping awareness of and decreasing over-identification with hostility. Aggressive Behavior, 33,327–338. doi:10.1002/ab.20187. angry thoughts and emotions, in order to reduce unproductive Borders, A., Earleywine, M., & Jajodia, A. (2010). Could mindfulness repetitive thinking. decrease anger, hostility, and aggression by decreasing rumination? Aggressive Behavior, 36,28–44. doi:10.1002/ab.20327. Boyraz, G., & Waits, J. B. (2015). Reciprocal associations among self- Compliance with Ethical Standards focused attention, self-acceptance, and empathy: a two-wave panel study. Personality and Individual Differences, 74.,84–89 Conflict of Interest Amanda Fresnics declares that she has no conflict Bushman, B. J. (2002). Does venting anger feed or extinguish the flame? of interest. Dr. Ashley Borders declares that she has no conflict of interest. Catharsis, rumination, distraction, anger, and aggressive responding. Personality and Social Psychology Bulletin, 28, 724–731. doi:10.1177/0146167202289002. Ethical Approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the The Cheung,G.,&Lau,R.(2008).Testing mediation and suppression College of New Jersey’s Institutional Review Board and with the 1964 effects of latent variables: bootstrapping with structural equa- – Helsinki declaration and its later amendments or comparable ethical tion models. Organizational Research Methods, 11,296325. standards. doi:10.1177/1094428107300343. Collins, K., & Bell, R. (1997). Personality and aggression: the dissipation-rumination scale. Personality and Individual Informed Consent Informed consent was obtained from all individual Differences, 22,751–755. doi:10.1016/S0191-8869(96)00248-6. participants included in the study. Denson, T. F. (2013). The multiple systems model of angry rumination. Personality and Social Psychology Review, 17, 103–123. doi:10.1177/1088868312467086. Denson, T. F., Pedersen, W. C., Friese, M., Hahm, A., & Roberts, L. (2011). Understanding impulsive aggression: angry rumination and reduced References self-control capacity are mechanisms underlying the provocation- aggression relationship. Personality and Social Psychology Bulletin, 37,850–862. doi:10.1177/0146167211401420. Arch, J. J., Wolitzky-Taylor, K. B., Eifert, G. H., & Craske, M. G. (2012). Feldman, G., Hayes, A., Kumar, S., Greeson, J., & Laurenceau, J. P. Longitudinal treatment mediation of traditional cognitive behavioral (2007). Mindfulness and emotion regulation: the development and therapy and acceptance and commitment therapy for anxiety disor- initial validation of the Cognitive and Affective Mindfulness Scale- ders. Behaviour Research and Therapy, 50(7), 469–478. Revised (CAMS-R). Journal of Psychopathology and Behavioral doi:10.1016/j.brat.2012.04.007. Assessment, 29,177–190. doi:10.1007/s10862-006-9035-8. Baer, R. A. (2010). Self-compassion as a mechanism of change in Fix, R. L., & Fix, S. T. (2013). The effects of mindfulness-based treat- mindfulness-and acceptance-based treatments. In R. A. Baer (Ed.), ments for aggression: a critical review. Aggression and Violent Assessing mindfulness and acceptance processes in clients: illumi- Behavior, 18(2), 219–227. nating the theory and practice of change (pp. 135–153). Oakland: Galantino, M. L., Baime, M., Maguire, M., Szapary, P. O., & Farrar, J. T. New Harbinger Publications, Inc. (2005). Association of psychological and physiological measures of Baer, R. A., Smith, G. T., Hopkins, J., Krietemeyer, J., & Toney, L. (2006). stress in health-care professionals during an 8-week mindfulness Using self-report assessment methods to explore facets of mindful- meditation program: mindfulness in practice. Stress and Health, ness. Assessment, 13,27–45. doi:10.1177/1073191105283504. 21(4), 255–261. doi:10.1002/smi.1062. Mindfulness

Gilbert, P. (2010). An introduction to compassion focused therapy in Neff, K. D., & Vonk, R. (2009). Self-compassion versus global self-es- cognitive behavior therapy. International Journal of Cognitive teem: two different ways of relating to oneself. Journal of Therapy, 3(2), 97–112. Personality, 77,23–50. doi:10.1111/j.1467-6494.2008.00537.x. Gillanders, D. T., Bolderston, H., Bond, F. W., Dempster, M., Flaxman, P. Neff, K. D., Kirkpatrick, K. L., & Rude, S. S. (2007). Self-compassion E., Campbell, L., … & Masley, S. (2014). The development and and adaptive psychological functioning. Journal of Research in initial validation of the Cognitive Fusion Questionnaire. Behavior Personality, 41,139–154. doi:10.1016/j.jrp.2006.03.004. Therapy, 45(1), 83–101. doi: 10.1016/j.beth.2013.09.001. Neff, K. D., Whittaker, T. & Karl, A. (in press). Examining the factor Hayes, A. F. (2012). PROCESS: a versatile computational tool for ob- structure of the self-compassion scale in four distinct populations: is served variable mediation, moderation, and conditional process the use of a total scale score justified? Journal of Personality modeling Assessment. Hayes, S. C., Strosahl, K. D., & Wilson, K. G. (2011). Acceptance and Nolen-Hoeksema, S., Wisco, B. E., & Lyubomirsky, S. (2008). commitment therapy: the process and practice of mindful change Rethinking rumination. Perspectives on Psychological Science, 3, (2nd ed.). New York: Guilford Press. 400–424. doi:10.1111/j.1745-6924.2008.00088.x. Heppner, W. L., Kernis, M. H., Lakey, C. E., Campbell, W. K., Goldman, Peters, J. R., Smart, L. M., Eisenlohr-Moul, T. A., Geiger, P. J., Smith, G. B. M., Davis, P. J., & Cascio, E. V. (2008). Mindfulness as a means T., & Baer, R. A. (2015). Anger rumination as a mediator of the of reducing aggressive behavior: dispositional and situational evi- relationship between mindfulness and aggression: the utility of a dence. Aggressive Behavior, 34,486–496. doi:10.1002/ab.20258. multidimensional mindfulness model. Journal of Clinical Hofmann, S. G., Grossman, P., & Hinton, D. E. (2011). Loving-kindness Psychology.doi:10.1002/jclp.22189. and compassion meditation: potential for psychological interven- Preacher, K. J., & Hayes, A. F. (2008). Assessing mediation in commu- tions. Clinical Psychology Review, 31, 1126–1132. doi:10.1016/j. nication research. In A. F. Hayes, M. D. Slater, & L. B. Snyder cpr.2011.07.003. (Eds.), The sage sourcebook of advanced data analysis methods – Krieger, T., Altenstein, D., Baettig, I., Doerig, N., & Holtforth, M. G. for communication research (pp. 13 54). Thousand Oaks: SAGE (2013). Self-compassion in depression: associations with depressive Publications, Inc. symptoms, rumination, and avoidance in depressed outpatients. Raes, F. (2010). Rumination and worry as mediators of the relationship Behavior Therapy, 44,501–513. doi:10.1016/j.beth.2013.04.004. between self-compassion and depression and anxiety. Personality – Kuyken, W., Watkins, E., Holden, E., White, K., Taylor, R. S., Byford, S., and Individual Differences, 48,757761. doi:10.1016/j. & Dalgleish, T. (2010). How does mindfulness-based cognitive ther- paid.2010.01.023. apy work? Behaviour Research and Therapy, 48, 1105–1112. Reis, N. A., Kowalski, K. C., Ferguson, L. J., Sabiston, C. M., Sedgwick, doi:10.1016/j.brat.2010.08.003. W. A., & Crocker, P. R. (2015). Self-compassion and women athletes’ responses to emotionally difficult sport situations: an evalua- Leary, M. R., Tate, E. B., Adams, C. E., Batts Allen, A., & tion of a brief induction. Psychology of Sport and Exercise, 16,18– Hancock, J. (2007). Self-compassion and reactions to unpleas- 25. doi:10.1016/j.psychsport.2014.08.011. ant self-relevant events: the implications of treating oneself Ruiz, F. J. (2010). A review of acceptance and commitment therapy kindly. Journal of Personality and Social Psychology, 92, (ACT) empirical evidence: correlational, experimental psychopa- 887–904. doi:10.1037/0022-3514.92.5.887. thology, component and outcome studies. International Journal of Leonard, K. E., Quigley, B. M., & Collins, R. L. (2002). Physical aggres- Psychology and Psychological Therapy, 10(1), 125–162. sion in the lives of young adults prevalence, location, and severity Samuelson, M., Carmody, J., Kabat-Zinn, J., & Bratt, M. A. (2007). among college and community samples. Journal of Interpersonal Mindfulness-based stress reduction in Massachusetts correc- Violence, 17(5), 533–550. doi:10.1177/0886260502017005004. tional facilities. The Prison Journal, 87(2), 254–268. MacKinnon, D. P., Lockwood, C. M., & Williams, J. (2004). Confidence doi:10.1177/0032885507303753. limits for the indirect effect: distribution of the product and resam- – Shapiro, S. L., Astin, J. A., Bishop, S. R., & Cordova, M. (2005). pling methods. Multivariate Behavioral Research, 39,99128. Mindfulness-based stress reduction for health care professionals: doi:10.1207/s15327906mbr3901_4. results from a randomized trial. International Journal of Stress Maxwell, J. P., Sukhodolsky, D. G., & Sit, C. H. (2009). Preliminary Management, 12, 164. doi:10.1037/1072-5245.12.2.164. validation of a Chinese version of the state-trait anger expression Shapiro, S. L., Carlson, L. E., Astin, J. A., & Freedman, B. (2006). – inventory-2. Asian Journal of Social Psychology, 12,1. 11 Mechanisms of mindfulness. Journal of Clinical Psychology, 62, doi:10.1111/j.1467-839X.2008.01264.x. 373–386. doi:10.1002/jclp.20237. Neff, K. D. (2003). The development and validation of a scale to Shapiro, S. L., Brown, K. W., & Biegel, G. M. (2007). Teaching self-care – measure self-compassion. Self and Identity, 2,223250. to caregivers: effects of mindfulness-based stress reduction on the doi:10.1080/15298860390209035. mental health of therapists in training. Training and Education in Neff, K. D. (2016). The self-compassion scale is a valid and theoretically Professional Psychology, 1,105.doi:10.1037/1931-3918.1.2.105. coherent measure of self-compassion. Mindfulness, 7(1), 264–274. Smeets, E., Neff, K., Alberts, H., & Peters, M. (2014). Meeting suffering doi:10.1007/s12671-015-0479-3. with kindness: effects of a brief self-compassion intervention for Neff, K. D., & Beretvas, S. N. (2013). The role of self-compassion female college students. JournalofClinicalPsychology,70,794– in romantic relationships. Self and Identity, 12,78–98. 807. doi:10.1002/jclp.22076. doi:10.1080/15298868.2011.639548. Speca, M., Carlson, L. E., Goodey, E., & Angen, M. (2000). A random- Neff, K. D., & Dahm, K. A. (2014). Self-compassion: what it is, what it ized, wait-list controlled clinical trial: the effect of a mindfulness does, and how it relates to mindfulness. Mindfulness and self- meditation-based stress reduction program on mood and symptoms regulation.NewYork:Springer. of stress in cancer outpatients. Psychosomatic Medicine, 62,613– Neff, K. D., & Germer, C. K. (2013). A pilot study and randomized 622. doi:10.1097/00006842-200009000-00004. controlled trial of the mindful self-compassion program. Journal Spielberger, C. D. (1999). Staxi-2: state-trait anger expression inventory- of Clinical Psychology, 69,28–44. doi:10.1002/jclp.21923. 2; professional manual. PAR, Psychological Assessment Resources. Neff, K. D., & Pommier, E. (2013). The relationship between self- Sukhodolsky, D. G., Golub, A., & Cromwell, E. N. (2001). compassion and other-focused concern among college undergradu- Development and validation of the anger rumination scale. ates, community adults, and practicing meditators. Self and Identity, Personality and Individual Differences, 31,689–700. 12,160–176. doi:10.1080/15298868.2011.649546. doi:10.1016/S0191-8869(00)00171-9. Mindfulness

Tabachnick, B. G., & Fidell, L. S. (2006). Using multivariate statistics, Watkins, E. R. (2008). Constructive and unconstructive repetitive 5th edition. Needham Heights, MA: Allyn and Bacon. thought. Psychological Bulletin, 134,163–206. doi:10.1037/0033- Trapnell, P. D., & Campbell, J. D. (1999). Private self-consciousness and the 2909.134.2.163. five-factor model of personality: distinguishing rumination from reflec- Wongtongkam, N., Ward, P. R., Day, A., & Winefield, H. (2013). tion. Journal of Personality and Social Psychology, 76,284–304. Reliability and validity of self-reported questionnaires related VanDam,N.T.,Sheppard,S.C.,Forsyth,J.P.,&Earleywine,M. to adolescent violence and consequences, Thailand. (2011). Self-compassion is a better predictor than mindfulness International Journal of Social Science Studies, 1,82–92. of symptom severity and quality of life in mixed anxiety and doi:10.11114/ijsss.v1i2.110 depression. Journal of Anxiety Disorders, 25, 123–130. Woodruff, S. C., Glass, C. R., Arnkoff, D. B., Crowley, K. J., doi:10.1016/j.janxdis.2010.08.011. Hindman, R. K., & Hirschhorn, E. W. (2014). Comparing Walach, H., Buchheld, N., Buttenmüller, V., Kleinknecht, N., & Schmidt, self-compassion, mindfulness, and psychological inflexibility – S. (2006). Measuring mindfulness—the Freiburg Mindfulness as predictors of psychological health. Mindfulness, 5,410 Inventory (FMI). Personality and Individual Differences, 40, 421. doi:10.1007/s12671-013-0195. 1543–1555. doi:10.1016/j.paid.2005.11.025. Zarling, A., Lawrence, E., & Marchman, J. (2015). A randomized controlled trial of acceptance and commitment therapy for aggressive behavior. Journal of Consulting and Clinical Psychology, 83(1), 199.