• No results found

Affect Intolerance is Associated with Insecure Attachment and Reduced Self-Esteem in Adults

N/A
N/A
Protected

Academic year: 2022

Share "Affect Intolerance is Associated with Insecure Attachment and Reduced Self-Esteem in Adults"

Copied!
21
0
0

Loading.... (view fulltext now)

Full text

(1)

Affect Intolerance is Associated with Insecure Attachment and Reduced Self-Esteem in Adults

Michael A. Kisley*1,Christina S. Caudle1,Amanda M. Harvey1 Authors’ affiliations:

1Department of Psychology, University of Colorado - Colorado Springs, Colorado Springs, CO, USA

*Corresponding author: Dr. Michael A. Kisley

Department of Psychology, University of Colorado - Colorado Springs, 1420 Austin Bluffs Parkway, Colorado Springs, CO 80918

Email: mkisley@uccs.edu Tel: 719-255-4564

“I am glad to see you can be moved to anger. Anger is good. I get angry all the time.”

-Mahatma Gandhi1

RESEARCH ARTICLE

ABSTRACT

Despite the adaptive relevance of emotion, many individuals view their negative emotional experiences to be un-important, unnecessarily distressing and/or intolerable. Two studies were conducted to investigate whether such de-valuing “meta-emotional” views are associated with attachment insecurity, specifically anxious and avoidant attachment, in adults. Self-report questionnaires of meta-emotional philosophy (including “affect intolerance”), attachment insecurity, and self-esteem were collected from two samples: one through Amazon Mturk (N=96), another from students enrolled at a U.S. university (N=166). Correlation analyses demonstrated that affect intolerance was associated with insecure attachment in both studies. In the second study, self-esteem was shown to mediate that relationship, fully for avoidant attachment, and partially for anxious attachment. These findings contribute to a growing body of literature demonstrating that dismissive, intolerant, or devaluing personal views of one’s own negative emotions is associated with deleterious psychological outcomes.

Keywords: Emotion; Meta-Emotion; Affect; Self-Esteem; Attachment; Evolution

(2)

1. Introduction: The fundamental and adaptive importance of emotion (Izard, 1972; Plutchik, 1980), highlighted especially by an evolutionary perspective (Al-Shawaf, Conroy-Beam, Asao, & Buss, 2014; Nesse, 1990; Tooby & Cosmides, 1990), fails to inform the personal understanding and daily-life application of so-called “negative” emotions for many modern humans (Oatley, 2004). Despite the adaptive value of fear to lead one away from danger, or of guilt to impel one to consider social reparations, many otherwise healthy individuals describe such emotions as un- important, hindering, unnecessarily distressing, and/or intolerable (Karnaze &

Levine, 2018; Mitmansgruber, Beck, Höfer,

& Schüßler, 2009; Norman & Furnes, 2014;

Stapinski, Abbott, & Rapee, 2014).

Although such devaluing views of uncomfortable emotional experiences may lead an individual to attempt regulating (i.e., changing) their emotions or altering their behavior to avoid acute emotional discomfort (Harmon-Jones, Harmon-Jones, Amodio, & Gable, 2011; recently reviewed by Naragon-Gainey, McMahon, & Chacko, 2017), such a view may nevertheless be maladaptive in a broader sense given that emotions have played a vital role in the evolution of Homo sapiens and our ancestral species over millions of years. Indeed, devaluing views and beliefs concerning negative emotion are associated with several deleterious psychological experiences for individuals including reduced well-being and life satisfaction (Karnaze & Levine, 2018; Mitmansgruber et al., 2009), emotional dysregulation (Karnaze & Levine, 2018; Manser, Cooper, & Trefusis, 2012), paradoxically increased experiences of

negative affect (Ford, Lam, John, & Mauss, 2017; Salters-Pedneault, Gentes, & Roemer, 2007) including symptoms of depression and anxiety (Leahy, 2002; Yoon, Dang, Mertz,

& Rottenberg, 2018), and increased levels of worry and stress (Stapinski et al., 2014). The present research was designed to investigate the possibility that an intolerant view of one’s emotions may also be related to difficulties with close relationships by investigating adult attachment insecurity in romantic relationships, specifically individual differences in anxious and avoidant dimensions (Hazan & Shaver, 1987). Further, we aimed to investigate whether one’s view of negative emotion are related to their self-esteem, and whether this could explain any demonstrated relationship between attachment insecurity and an intolerant view of emotions.

An individual’s thoughts and feelings regarding their emotions contribute to their

“meta-emotional philosophy.” Gottman, Katz, and Hooven (1996) employed this concept to contrast parents who engage in emotion “coaching” with those who display

“dismissing” behaviors in response to their child’s emotional experiences. The latter is associated with beliefs that emotions are not important and should be denied or ignored.

By contrast, coaching involves validating a child’s emotional experience and assisting them in understanding their emotions and the situations in which they arise. Outside of parenting, an individual’s thoughts and feelings about their own emotions will impact their psychological experience and function as well (Norman & Furnes, 2014).

Several scales have been developed to assess

“meta-emotion” (Mitmansgruber et al.,

(3)

2009) and closely-related constructs including “affect intolerance” (Stapinski et al., 2014), “distress tolerance” (Simons &

Gaher, 2005), “emotional schema” (Leahy, 2002), “attitudes toward emotion” (Harmon- Jones et al., 2011), “evaluations of emotions” (Netzer, Gutentag, Kim, Solak, &

Tamir, 2018), “help vs. hinder” lay theories of emotion (Karnaze & Levine, 2018), and

“beliefs about emotions” (Manser et al., 2012). In stark contrast to an evolutionary view of emotions as having “fundamental importance” (Plutchik, 1980) and

“inherently adaptive functions” (Izard, 1972), these scales measure the extent to which uncomfortable emotions are considered problematic by individuals, although the different scales vary in their emphases and specific terminologies. We focus here on affect intolerance, for which such emotional experiences are considered to be beyond one’s control, to be distressing, to lead to negative consequences, and/or to impel avoidance or suppression (Stapinski et al., 2014),

Whereas emotions can be viewed as natural and inevitable, meta-emotion is thought to be learned and thus socially- and culturally- determined (Mendonça, 2013; Tsai, Knutson, & Fung, 2006). Parents from civilized cultures throughout the world often engage in suppressive, shaming and punitive responses to their children’s emotional displays, though differentially depending on many variables including each culture’s relative level of collectivism (Friedlmeier, Corapci, & Cole, 2011). Regardless of the specific pattern, openly devaluating a child’s natural experience, in this case their emotions, is likely to harm their self-

acceptance (Glantz & Pearce, 1989). As such, one would expect a relationship between the meta-emotion philosophy an individual learns from their parents and their self-esteem, defined by Rosenberg (1965) as one’s overall appraisal of their self-worth.

However, this had not yet been tested empirically until the present investigation. In contrast to individuals raised in modern civilization, in modern hunter-gatherer bands, psychological functions that arise naturally, such as emotions and emotional displays, are typically indulged (Glantz &

Pearce, 1989; Konner, 1977; Konner, 2010;

Morelli & Tronick, 1991). Importantly, this echoes the critical concepts of sensitive and responsive parenting in the development of secure attachment behaviors (Mikulincer &

Shaver, 2016) including parental responsiveness to emotional displays by a child (Ainsworth, Blehar, Waters, & Wall, 1978). Konner (1977), for example, observed that early indulgence typical of the

!Kung San living in small bands (<40 people) in northwestern Botswana gives way to very little dependency later in childhood, consistent with the development of a secure attachment style.

Like affect intolerance, insecure relationship attachment, specifically anxious and avoidant attachments styles, occurs at relatively high rates in modern societies.

Anxious adult attachment, characterized by elevated need for closeness and heightened worry about relationships (Hazan & Shaver, 1987), was found to describe 18% of otherwise healthy adults in a meta-analysis of studies conducted throughout several Western countries (Van Ijzendoorn &

Bakermans-Kranenburg, 1996). Avoidant

(4)

attachment, describing 24% of adults, is characterized by discomfort with closeness and dismissive behaviors in relationships.

Given the importance of environment on the development of attachment style, it’s not surprising that different cultures exhibit different patterns of attachment insecurity.

For example, anxious attachment has been shown to be relatively more frequent within a sample of East Asians living in the United States (Agishtein & Brumbaugh, 2013), consistent with higher rates of anxious attachment more generally within East Asia (Schmitt et al., 2004). Of relevance to the present investigation, East Asians have also been shown to exhibit significantly higher levels of affect intolerance than Australians and Europeans (Stapinski et al., 2014), and Asian-Americans more strongly endorse a

“hindrance theory” of emotion compared to European-Americans (Karnaze & Levine, 2018). Montague, Magai, Consedine, and Gillespie (2003) demonstrated overall higher rates of attachment insecurity in a sample of African Americans compared to European Americans. This effect was partly explained by “punitive emotion socialization” where, for example, a parent responds to their child’s negative emotional display by ignoring them or shaming them.

There is also more direct evidence to support a link between affect intolerance and insecure attachment. A parent’s meta- emotional philosophy (coaching or dismissing) predicts the security of a child’s attachment to that parent (Chen, Lin, & Li, 2012) and also appears to contribute to a child’s ability to attach to others more generally (Mikulincer & Shaver, 2016). In the context of adult relationships, Manser et

al. (2012) found an association between an individual’s beliefs about emotions (as overwhelming, shameful, invalid, and useless) and their level of attachment insecurity. An “inability to tolerate distress”

has also been shown to be associated with anxious attachment (Norberg, Crone, Kwok,

& Grisham, 2018). Less direct support for a connection between affect intolerance and attachment comes from a meta-analytic study of children up to age 18 demonstrating that greater emotional understanding (i.e.,

“the ability to identify, interpret, and communicate about one’s own and others’

emotions”) is robustly associated with greater attachment security (Cooke, Stuart- Parrigon, Movahed-Abtahi, Koehn, &

Kerns, 2016). These authors concluded their report by suggesting that meta-emotional theory may provide a helpful framework for this area of research. Relatedly, there have been numerous demonstrations of associations between attachment insecurity and difficulties with emotional regulation (Mikulincer & Shaver, 2016), which is likely to be shaped by an individual’s meta- emotional philosophy.

The present research was conducted to directly investigate the relationship between affect intolerance (i.e., a dismissive meta- emotional philosophy) and insecure attachment in adults. Findings from existing empirical studies, as well as attachment theory, suggest an important connection, but little research has been done to directly compare these constructs within individuals.

Drawing from the literature reviewed above, we hypothesize that greater levels of affect intolerance will be associated with reduced attachment security. To better understand

(5)

any found association, a second study was conducted to investigate the potential mediational role of self-esteem, which has previously been shown to be associated with attachment insecurity (Cassidy, 1988;

Feeney & Noller, 1990; Foster, Kernis, &

Goldman, 2007; Mikulincer & Shaver, 2005), but has not yet been studied in the context of affect intolerance. By doing so, we aimed to provide an initial contrast between the following ideas: i) insecure attachment is directly related to affect intolerance; or rather, ii) insecure attachment is indirectly related to affect intolerance through reduced feelings of self-worth.

Findings that conform to the former pattern would support the idea that insecure attachment specifically involves intolerance of one’s negative emotions. By contrast, findings that conform to the latter pattern would suggest that affect intolerance is not necessarily a core feature of attachment insecurity, but rather is more closely related to a lowered sense of self-worth, which is also a typical feature of insecure individuals.

2. Study 1: This study was designed to test the primary hypothesis, that affect intolerance is related to insecure attachment.

To provide an initial assessment of the robustness of any found relationships, and because multiple scales have been published, we utilized two different affect scales here: The Affect Intolerance Scale (AIS; Stapinski et al., 2014) and the Meta- Emotion Scale (MES; Mitmansgruber et al., 2009). Planned analyses included examination of partial correlations to determine the extent to which affect intolerance and insecure attachment (i.e., anxious and avoidant dimensions) are

related over and above demographic variables known to be relevant including age and relationship status (Mikulincer &

Shaver, 2016). Self-esteem was included in a secondary partial correlation matrix as a preliminary assessment of the extent to which this variable might mediate this relationship.

2.1 Methods:

2.1.1 Participants: The sample consisted of 96 individuals (Males = 56, Females = 40) ranging in age from 22-69 years (M = 36.54, SD = 12.27). Forty-one participants reported their race to be European American/White (43%), 30 as Asian American/Pacific Islander (31%), 5 as Hispanic/Latino (5%), 2 as African American (2%), 1 as Native American (1%), and 16 as more than one or another race (17%). Twenty participants reported being “Single, NOT in a relationship, 29 as “Single, but in a committed relationship,” and 46 as being

“Married/partnered.” The latter two categories were consolidated to form one

“partnered” category, ultimately leaving two categories: “single” (n = 20) and “partnered”

(n = 75).

Participants were recruited through Amazon Mechanical Turk (Mturk). Mturk is an online marketplace that enables individuals and businesses to outsource tasks requiring human intelligence. Experiments run on Mturk have been shown to produce results consistent with both laboratory settings and comparable online resources (Paolacci, Chandler, & Ipeirotis, 2010). As such, it is considered a suitable platform for obtaining data through survey administration. Mturk workers, or providers, browse available

(6)

tasks and complete them in return for a set monetary payment. In Study 1, we invited Mturk providers to participate in a task involving a series of questionnaires about their own emotional experiences and behaviors in close relationships.

Compensation for participation was US

$0.50.

2.1.2 Materials and Procedure: After providing informed consent, participants were administered an online survey that consisted of four different questionnaires, followed by demographic questions (age, race/ethnicity, gender identity, and romantic relationship status). The four questionnaires, described below, were presented in randomized order across participants.

The Rosenberg Self-Esteem Scale (Rosenberg, 1965) was used to assess individual self-esteem. Participants indicated their agreement using a 4-point Likert scale ranging from “strongly agree” to

“strongly disagree” on 10 statements regarding their feelings about their own self- worth. Sample statements include, “I feel I do not have much to be proud of” and “I feel that I have a number of good qualities.”

Cronbach’s alpha in the original study was .92, and.87 in the present study.

The second questionnaire used in Study 1 was the MES (Mitmansgruber et al., 2009).

This scale is intended to measure meta- emotional attitudes and experiences. Using a 6-point Likert scale (from “is not at all true for me” to “is completely true for me”), participants were instructed to rate 28 statements such as “I repeatedly get angry about my emotional reactions” from the anger subscale, and “I learn through my

feelings” from the interest subscale. The MES has been shown to assess six different meta-emotional factors: anger, compassionate care, interest, contempt/shame, tough control, and suppression. The original development of this scale included a second-order analysis of the six subscales yielding two overarching factors: negative meta-emotions (consisting of anger, contempt/shame, tough control, and suppression) and positive meta- emotions (consisting of interest and compassionate care). Therefore, negative meta-emotion can be broadly conceptualized as feeling angered by or ashamed of negative emotions and suppressing or controlling them. In contrast, positive meta-emotion is seen as representing self-understanding and active interest in one’s emotional experiences. Because some of the 6 underlying factors are computed from small numbers of items (e.g., suppression is computed from two items) and because some of these alpha’s were relatively low (e.g., contempt/shame was .68 and suppression was .70) we decided to utilize only the overarching negative and positive meta-emotions scales in regression analyses here.

Participants also completed the AIS (Stapinski et al., 2014), which provides an assessment of beliefs about negative emotions considered to be maladaptive. This scale specifically evaluates ideas that emotions are threatening (threat expectancy subscale) and that they ought to be avoided or suppressed (avoid/suppress subscale).

Respective examples: “I fear that my negative feelings won’t go away,” and “I try to distract myself from negative feelings.”

(7)

For this questionnaire, participants are instructed to think about their beliefs with consideration given to “a broad range of negative feelings, such as anxiety, sadness, anger, depression, and stress.” Their level of agreement was indicated with a 7-point Likert scale from “strongly disagree” to

“strongly agree.” Cronbach’s alphas for the threat expectancy and avoid/suppress items were .94 and .88 in the original study, respectively, and .96 and .92 in the present sample.

The final questionnaire evaluated adult attachment style using the Experiences in Close Relationships-Revised (ECR-R) Questionnaire (Fraley, Waller, & Brennan, 2000). This scale measures individual tendencies to experience anxiety and/or avoidance in intimate relationships. Here, participants received instructions to respond according to their general experiences in relationships and not according to the specific dynamics of a current relationship.

Example statements include “I find it difficult to allow myself to depend on romantic partners” (avoidant attachment subscale) and “It makes me mad that I don’t get the affection and support I need from my

partner” (anxious attachment subscale).

Responses were given using a 7-point Likert scale that ranged from “strongly disagree” to

“strongly agree.” Cronbach’s alphas were .96 for the anxious attachment subscale and .95 and for the avoidant attachment subscale.

Fraley, Heffernan, Vicary, & Brumbaugh (2011) reported alphas of .85, anxiety, and .88, avoidance.

2.2 Results: The final dataset contained less than .3% missing data after removal of one participant who failed to complete an entire questionnaire. Twenty-seven (of 104) scale items contained one or two missing items with percentage of missing data for each scale item ranging between 0-2.1%. The pattern of the missing data was determined to be missing completely at random. The missing scores were imputed using stochastic regression. Final scale totals used in analysis were calculated using the completed dataset. All scale totals were normally distributed with skew and kurtosis within the conventional boundary of ± 2.0 (skew range: -.92, .23; kurtosis: -1.19, .81).

Means and standard deviations for variables are presented in Table 1.

Table 1

Descriptive Statistics

M (SD) AIS Threat Expectancy 52.31 (18.89) AIS Avoid/Suppress 62.47 (14.67) Negative MES 3.45 (1.07) Positive MES 3.81 (.96) Self-Esteem 18.84 (5.89) Anxious Attachment 3.38 (1.50) Avoidant Attachment 3.08 (1.30)

(8)

A full correlation matrix was created for all scale totals (see Table 2). Of note, anxious attachment style was positively correlated with threat expectancy (r = .72; Figure 1 top), avoid/suppress (r = .43), negative meta- emotion (r = .70), and positive meta-emotion (r = .22) as well as negatively correlated with self-esteem (r = -.60). Avoidant attachment

style was positively correlated with threat expectancy (r = .52; Figure 1 middle), avoid/suppress (r = .26), and negative meta- emotion (r = .39) while negatively correlated with positive meta-emotion (r = -.22) and self-esteem (r = -.67). Anxious and avoidant attachment styles were positively correlated with one another (r = .48).

Table 2

Correlations Among Scale Scores

1 2 3 4 5 6 7

1. AIS Threat Expectancy --

2. AIS Avoid/Suppress .61*** --

3. Negative MES .79*** .67*** --

4. Positive MES .18 .43*** .37*** --

5. Self-Esteem -.65*** -.37*** -.47*** .17 --

6. ECR Anxious Attachment .72*** .43*** .70*** .22* -.60*** --

7. ECR Avoidant Attachment .52*** .26* .39*** -.22* -.67*** .48*** -- Note. AIS = Affect Intolerance Scale; MES = Meta-Emotion Scale; ECR = Experiences in Close Relationships; *p < .05 ** p < .01 *** p < .001

A partial correlation matrix was created for the scales with age and relationship status partialed out (see Table 3). Anxious attachment style continued to be positively correlated with threat expectancy (r = .66), avoid/suppress (r = .36), and negative meta- emotion (r = .61) as well as negatively correlated with self-esteem (r = -.64).

Avoidant attachment style continued to be positively correlated with threat expectancy

(r = .54), avoid/suppress (r = .23), and negative meta-emotion (r = .43). It was negatively correlated with self-esteem (r = - .66). Anxious and avoidant attachment styles continued to be positively correlated with each other (r = .56).

(9)

Figure 1. Scatter-plots illustrating the relationships between attachment insecurity (anxious and avoidant scales of the ECR-R) and affect intolerance (in this case the threat expectancy scale of the AIS) for the entire study 1 sample.

0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0

0.0 20.0 40.0 60.0 80.0 100.0

Attachment Anxiety

Threat Expectancy

0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0

0.0 20.0 40.0 60.0 80.0 100.0

Attachment Avoidance

Threat Expectancy

0.0 5.0 10.0 15.0 20.0 25.0 30.0

0.0 20.0 40.0 60.0 80.0 100.0

Self-Esteem

Threat Expectancy

(10)

Table 3

Correlations Among Scale Scores With Age and Relationship Status Partialed Out

1 2 3 4 5 6 7

1. AIS Threat Expectancy --

2. AIS Avoid/Suppress .56*** --

3. Negative MES .74*** .63*** --

4. Positive MES .10 .44*** .30** --

5. Self-Esteem -.65*** -.32** -.45*** .18 --

6. ECR Anxious Attachment .66*** .36** .61*** .11 -.64*** --

7. ECR Avoidant Attachment .54*** .23* .43*** -.20 -.66*** .56*** -- Note. AIS = Affect Intolerance Scale; MES = Meta-Emotion Scale; ECR = Experiences in Close Relationships; *p < .05 ** p < .01 *** p < .001

Finally, a partial correlation matrix was created for the attachment and meta-emotion scales with age, relationship status, and self- esteem partialed out (see Table 4) as an initial test of the idea that self-esteem may mediate the relationships between attachment style and meta-emotion. Anxious attachment style demonstrated the same patterns as before with diminished correlations. It was positively correlated with threat expectancy (r = .42),

avoid/suppress (r = .21), and negative meta- emotion (r = .46). Additionally, anxious attachment became positively correlated with positive meta-emotion (r = .30). After partialing out self-esteem, avoidant attachment style no longer demonstrated significant correlations with any meta- emotional factor. Anxious and avoidant attachment styles remained positively correlated with one another (r = .24).

Table 4

Correlations Among Scale Scores With Age, Relationship Status, and Self-Esteem Partialed Out

1 2 3 4 5 6

1. AIS Threat Expectancy --

2. AIS Avoid/Suppress .48*** --

3. Negative MES .66*** .58*** --

4. Positive MES .29** .54*** .44** --

5. ECR Anxious Attachment .42*** .21* .46*** .30** --

6. ECR Avoidant Attachment .21 .03 .20 -.11 .24* --

Note. AIS = Affect Intolerance Scale; MES = Meta-Emotion Scale; ECR = Experiences in Close Relationships; *p < .05 ** p < .01 *** p < .001

(11)

2.2.1 Results Summary: Affect intolerance was found to be related to insecure attachment confirming the primary hypothesis of study 1. This effect was robust in the sense that it did not depend upon which published scale was examined.

Anxious attachment and avoidant attachment were correlated with the threat expectancy and avoid/suppress subscales of the AIS and, separately, with the negative meta-emotion subscale of the MES.

Accounting for the effects of self-esteem also consistently reduced the level of correlation between attachment and meta- emotional factors. In the case of anxious attachment, correlations were reduced. For avoidant attachment, correlations were no longer significant. The potential role for this construct in the relationship between affect intolerance and insecure attachment was investigated more thoroughly in the following study.

3. STUDY 2: This study was designed to more directly investigate whether self- esteem mediates the demonstrated relationship between affect intolerance and insecure attachment. In order to reduce the overall number of analyses conducted (to minimize type I error), and because the results from study 1 were substantively similar regardless of the scale employed, we decided to utilize only one affective scale for study 2. Whereas all AIS items consistently refer to “a broad range of negative feelings, such as anxiety, sadness, anger, depression, and stress” (Stapinski et al., 2014) the individual items from the MES at times refer to specific emotions that differ between items (e.g. “when I feel guilty because I have

made a mistake, I am quite unforgiving with myself”), risk confounding emotions and thoughts (e.g., “Time and again, my thoughts and emotions are fascinating and important to me”), or fail to explicitly mention emotions or feelings within an item (e.g., “Repeatedly, there are situations when I excoriate myself”) (Mitmansgruber et al., 2009). As such, and because the research described here was designed to investigate affect intolerance broadly, we decided to employ the AIS for study 2. Four separate planned mediation analyses were conducted to investigate whether lowered self-esteem would mediate any relationship between each of the two affect intolerance subscales and each of the two attachment insecurity subscales.

3.1 Methods:

3.1.1 Participants: In study 2, 166 participants (Males = 21, Females = 144, Gender variant = 1) were recruited from psychology students enrolled in a U.S.

university, ranging in age from 17-55 years old (M = 21.88, SD = 5.90). Participants primarily reported their race as European American/White (n = 116; 69.5%), 24 reported Hispanic/Latino (14.4%), 12 as Asian American/Pacific Islander (7.2%), 3 as African American (1.8%), and 12 as more than one or another race (7.2%). Sixty-three reported their relationship status as “Single, NOT in a relationship,” 78 as “Single, but in a committed relationship,” and 23 as

“Married/partnered.” Students were invited to participate in an online survey investigating the influence of emotions on a

(12)

person’s relationships. Compensation for participation was course extra credit.

3.1.2 Materials and Procedure: Informed consent was provided. Participants then proceeded to complete the SES, AIS and ECR-R questionnaires, order randomized across participants, and finally the same demographic items presented in Study 1.

Cronbach’s alphas in the present study were as follows: SES, α= .93; AIS, α = .96 for threat expectancy, α= .91 for avoid/suppress; and ECR-R, α= .94 for anxious attachment, and α= .95 for avoidant attachment.

3.2 Results: The final dataset contained .22% missing data after removal of three

participants who failed to complete one or more pages of the survey. Twenty-seven (of 96) scale items contained missing items with percentage of missing data for each scale item ranging between 0-1.80%. The pattern of the missing data was determined to be missing completely at random, and missing scores were imputed using stochastic regression. Final scale totals used in analysis were calculated using the completed dataset.

All scale totals were normally distributed with skew and kurtosis within the conventional boundary of ± 2.0 (skew range:

-.58, .49; kurtosis: -86, .60). Means and standard deviations for variables are presented in Table 5.

Table 5

Descriptive Statistics

M (SD)

AIS Threat Expectancy 48.06 (18.92) AIS Avoid/Suppress 56.65 (14.61) Self-Esteem 19.14 (6.28) Anxious Attachment 3.22 (1.30) Avoidant Attachment 3.06 (1.32)

A correlation matrix was created for all scale totals (see Table 6). Of note, anxious attachment style was positively correlated with threat expectancy (r = .57) and avoid/suppress (r = .51) as well as negatively correlated with self-esteem (r = -.54).

Avoidant attachment style was positively correlated with threat expectancy (r = .24) and avoid/suppress (r = .27), and negatively correlated with self-esteem (r = -.30).

Anxious and avoidant attachment styles

(13)

were once again correlated with one another (r = .37).

Table 6

Correlations Among Scale Scores

1 2 3 4 5

1. AIS Threat Expectancy --

2. AIS Avoid/Suppress .75*** --

3. Self-Esteem -.66*** -.49*** --

4. ECR Anxious Attachment .57*** .51*** -.54*** --

5. ECR Avoidant Attachment .24** .27** -.30*** .37*** -- Note. AIS = Affect Intolerance Scale; ECR = Experiences in Close Relationships;

*p < .05 **p < .01 *** p < .001

To examine self-esteem’s potential to mediate the relationship between affect intolerance (as measured by the threat expectancy and avoid/suppress subscales of the AIS) and insecure attachment (anxious and avoidant subscales of the ECR-R), four separate mediation analyses were performed using the guidelines offered by Baron &

Kenny (1986). The outcome of these analyses are described below.

3.2.1 Threat Expectancy, Self-Esteem, and Attachment Insecurity: Prior to controlling for self-esteem, threat expectancy positively predicted anxious attachment style (β = .57, p < .001). After controlling for self-esteem, threat expectancy continued to negatively predict anxious attachment style but to a lesser degree (β = .38, p < .001) suggesting a partial mediation (see Figure 2a), Sobel Z = 2.89, p < .01, Κ2 = .19 (medium-large effect size). Similarly, threat expectancy negatively predicted avoidant attachment style (β = .24, p < .01) prior to controlling for self-esteem. Threat expectancy no longer predicted avoidant attachment after controlling for self-esteem (β = .07, n.s.),

suggesting complete mediation (Figure 2b), Sobel Z = 2.44, p < .05, Κ2 = .17 (a medium- large effect).

3.2.2 Avoid/Suppress, Self-Esteem, and Attachment Insecurity: Avoid/Suppress also positively predicted anxious attachment (β = .51, p < .001). After controlling for self- esteem, avoid/suppress still positively predicted anxious attachment, though to a lesser degree (β = .32, p < .001), suggesting a partial mediation (Figure 2c), Sobel Z = 3.47, p < .001, Κ2 = .19 (a medium-large effect). Avoid/suppress also positively predicted avoidant attachment (β = .27, p <

.01). After controlling for self-esteem, avoid/suppress no longer predicted avoidant attachment (β = .16, n.s.), suggesting a complete mediation (Figure 2d), Sobel Z = 2.35, p < .05, Κ2 = .11 (medium effect size).

(14)

a)

b)

c)

d)

Figure 2. Mediation effects of self-esteem on relationship between a) threat expectancy and anxious attachment, b) threat expectancy and avoidant attachment, c) avoid/suppress and anxious attachment, d) avoid/suppress and avoidant attachment. Values are standardized beta weights.

Standard errors are in parentheses. *p < .05, **p < .01, ***p < .001.

(15)

3.2.3 Results Summary: Self-esteem mediated the relationship between affect intolerance and insecure attachment.

However, the extent of mediation depended on attachment style. Whereas self-esteem partially mediated the relationship between measures of affect intolerance and the anxious attachment dimension, it fully mediated the predictive relationship for the avoidant attachment dimension.

4. Discussion: As hypothesized, affect intolerance was related to insecure attachment in adults. This was shown with two different metrics of affect intolerance (AIS and MES in study 1) and across two different samples (AIS across studies 1 and 2). Among the most robust findings was that threat expectancy (a subscale of the AIS) was significantly correlated to both anxious and avoidant attachment scores regardless of the inclusion of demographic variables.

Stapinski et al. (2014) described this factor as comprising a “range of beliefs regarding the consequences and threatening nature of negative emotional experiences.” Similarly, negative meta-emotion from the MES (Mitmansgruber et al., 2009) was significantly correlated with both anxious and avoidant attachment scores in study 1.

This factor is computed by combining scores from the anger, contempt/shame, tough control, and suppression subscales. Taken together, these findings provide support for the idea that affect intolerance is broadly and robustly related to insecure attachment, consistent with previous research (Chen, Lin, & Li, 2012; Manser et al. (2012);

Norberg, Crone, Kwok, & Grisham, 2018).

However, both studies 1 and 2 provided evidence that the relationship between affect

intolerance and insecure attention is not identical for anxious and avoidant styles, specifically with regards to the role of self- esteem in mediating that relationship.

Self-esteem significantly mediated the relationship between affect intolerance and insecure attachment, although differentially for anxious and avoidant dimensions.

Whereas reduced self-esteem fully explained the relationship between affect intolerance (both the threat expectancy and avoid/suppress subscales of the AIS) and avoidant attachment, it only partially explained the relationship for anxious attachment. Thus, for avoidant individuals, affect intolerance did not predict difficulties with intimate relationships above and beyond that which could be explained by a lowered sense of self-worth. By contrast, anxious individuals appear to experience difficulties in the context of intimate relationships related to both their lowered sense of self-worth and their intolerant meta- emotional philosophy. Thus, an anxious individual’s preoccupation with a relationship may relate to difficulty trusting that their partner values them (due to low self-esteem) and to their worry that negative emotional experiences which occur in the context of a relationship will be intolerable.

Though relatively intolerant of negative affect, an avoidant individual’s difficulties with intimate relationship seem nevertheless to be more directly related to their lowered sense of self-worth. This is consistent with the observation that avoidant individuals engage in thoughts and behaviors that minimize their vulnerability to emotional pain to begin with (Hazan & Shaver, 1987;

Mikulincer & Shaver, 2016). Regardless, a

(16)

clearer understanding of these issues will require additional research. The present study was not designed to test for differences between anxious and avoidant attachment, but rather as a test of the idea that affect intolerance would be related to insecure attachment broadly, and indirectly through reduced self-esteem.

Correlations between attachment insecurity and lowered self-esteem have been demonstrated before (Cassidy, 1988; Feeney

& Noller, 1990; Foster et al., 2007). Past studies were designed from the perspective of classical attachment theory such that the formation of insecure attachments early in development were conceptualized to lead to reduced self-esteem and additional mal- adaptions including social and emotional difficulties (Mikulincer & Shaver, 2005).

However, we specifically designed the present investigation to remain neutral concerning the direction of causality between insecure attachment and affect intolerance. Parental responsiveness to an infant/child’s emotional displays is known to be important for the development of secure attachments (Ainsworth et al., 1978), but it is also known to be important for the development of an individual’s meta- emotional philosophy (Gottman et al., 1996). Just as confirmatory responses to emotional displays lead to secure attachments, they also provide demonstrable evidence to a child that their emotional experiences are valid and important.

Consequently, it’s not immediately clear which comes first: insecure attachment or affect intolerance. Regardless, reduced self- esteem appears to play an important mediating role between these variables. For

example, dismissive responses to an infant/child’s natural and inevitable emotional experiences and expressions might negatively impact their self-esteem (Glantz & Pearce, 1989) and hence their ability to trust and engage in close relationships. Alternatively, dismissive responses might disrupt the formation of secure attachments, negatively impacting self-esteem, and consequently an individual’s ability to tolerate their negative emotional experiences. The cross-sectional approach employed here does not allow us to directly compare these models. Doing so in future research will be worthwhile, as the models provide conflicting implications for the most effective ways in which to intervene with children and adults in order to improve their attachment security, affect tolerance, and overall psychological well- being.

There are several limitations to the current studies that limit the interpretation of the findings. Both measures of affect intolerance employed here depend upon self-report. It’s possible that individuals who are especially intolerant of negative affect will have little insight into their emotional experiences, and thus have difficulty reporting accurately on their level of tolerance (Mendonça, 2013;

Norman & Furnes, 2014). Nevertheless, we found wide variation in these measures across both of our samples (e.g., Tables 1 and 5, Figure 1), so presumably intolerant individuals can report their difficulties.

Regardless, the development of measures based on behavioral observation would be helpful to provide insight into the relationship between an individual’s meta- emotional philosophy and their actual

(17)

affective experiences and associated behaviors. Also important to note, the sample from study 2 was predominantly female. This limits the potential generalizability of the findings concerning the relationship between affect intolerance, self-esteem, and insecure attachment.

However, the correlations between these measures were comparable to those found in study 1 which had a more balanced distribution of men and women (compares Tables 2 and 6). The extent to which gender contributes to patterns of attachment insecurity is debated (Mikulincer & Shaver, 2016; Simpson & Belsky, 2016). As such, a more cautious interpretation of study 2 is that self-esteem mediates the relationship between affect intolerance and insecure attachment in a predominantly female sample.

It's important to note that we did not screen individuals for mental or emotional disorders. It could be argued that affect intolerance arises primarily in individuals who have disordered affect (e.g., clinical depression or anxiety disorder) because their emotional experience is extreme and, thus, more difficult to tolerate. However, we might then expect discontinuities or non- linearities in the distribution of affect intolerance scores. Instead we found linear, continuous variation across the range of affect intolerance scores and their proportionality to attachment and self- esteem scores (e.g., Figure 1). Regardless, from the present investigation we can’t rule out the possibility that strong negative affective experience leads one to develop an intolerant philosophy, as opposed to affective intolerance being learned, as

theorized in the meta-emotion literature (Mendonça, 2013; Tsai et al., 2006). Of relevance, Hurrell, Houwing, and Hudson (2017) recently demonstrated that parents of children with anxiety disorders exhibit less emotion coaching behaviors than parents of children without anxiety disorders. Although suggestive, this certainly does not by itself demonstrate that parents’ dismissive responses to their children’s emotional displays lead to anxiety difficulties.

Nevertheless, such an interpretation is consistent with the suggestion that parental emotion coaching can buffer some children from the development of psychopathology (Hooven, Gottman, & Katz, 1995).

Concerning insecure attachment, Chen et al.

(2012) found that parental meta-emotional philosophy (coaching vs. dismissing) predicted the security of their child’s attachment, but their child’s level of

“emotionality” did not. Further disentangling the contributions of early experience and temperament on the development of affect intolerance awaits additional research. Regardless of how and when affect intolerance develops, the research presented here provides evidence for a robust association between adult meta- emotional philosophy, self-esteem, and attachment security.

Acknowledgements: Support was provided by the University of Colorado - Colorado Springs (UCCS) College of Letters, Arts, and Sciences, the UCCS Department of Psychology, and a private donor.

(18)

References:

1. Agishtein, P., & Brumbaugh, C. (2013).

Culural variation in adult attachment:

The impact of ethnicity, collectivism, and country of origin. Journal of Social, Evolutionary, and Cultural Psychology, 7(4), 384-405.

2. Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the strange situations. New Jersey:

Lawrence, Erlbaum Associates.

3. Al-Shawaf, L., Conroy-Beam, D., Asao, K., & Buss, D. M. (2014). Human emotions: An evolutionary psychological perspective. Emotion Review, 8(2), 173-186.

doi:10.1177/1754073914565518

4. Baron, R. M., & Kenny, D. A. (1986).

The moderator-mediator variable distinction in social psychological research. conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173-1182. doi:10.1037/0022- 3514.51.6.1173

5. Cassidy, J. (1988). Child-mother attachment and the self in six-year-olds.

Child Development, 59(1), 121-134.

doi:10.1111/j.1467- 8624.1988.tb03200.x

6. Chen, F. M., Lin, H. S., & Li, C. H.

(2012). The role of emotion in parent- child relationships: Children's emotionality, maternal meta-emotion, and children's attachment security.

Journal of Child and Family Studies, 21(3), 403-410. doi:10.1007/s10826- 011-9491-y

7. Cooke, J. E., Stuart-Parrigon, K. L., Movahed-Abtahi, M., Koehn, A. J., &

Kerns, K. A. (2016). Children's emotion

understanding and mother-child attachment: A meta-analysis. Emotion, 16(8), 1102-1106.

doi:10.1037/emo0000221

8. Feeney, J. A., & Noller, P. (1990).

Attachment style as a predictor of adult romantic relationships. Journal of Personality and Social Psychology, 58(2), 281-291.

9. Ford, B. Q., Lam, P., John, O. P., &

Mauss, I. B. (2017). The psychological health benefits of accepting negative emotions and thoughts: Laboratory, diary, and longitudinal evidence.

Journal of Personality and Social Psychology, doi:10.1037/pspp0000157 10. Foster, J. D., Kernis, M. H., & Goldman,

B. M. (2007). Linking adult attachment to self-esteem stability. Self and Identity, 6(1), 64-73.

doi:10.1080/15298860600832139 11. Fraley, R. C., Heffernan, M. E., Vicary,

A. M., & Brumbaugh, C. C. (2011). The experiences in close relationships—

Relationship structures questionnaire: A method for assessing attachment orientations across relationships.

Psychological Assessment, 23(3), 615- 625. doi:10.1037/a0022898

12. Fraley, R. C., Waller, N. G., & Brennan, K. A. (2000). An item response theory analysis of self-report measures of adult attachment. Journal of Personality and Social Psychology, 78(2), 350-365.

doi:10.1037/0022-3514.78.2.350

13. Friedlmeier, W., Corapci, F., & Cole, P.

M. (2011). Emotion socialization in cross-cultural perspective. Social and Personality Psychology Compass, 5(7), 410-427. doi:10.1111/j.1751- 9004.2011.00362.x

14. Gandhi, A. (2017). The Gift of Anger and Other Lessons from My Grandfather

(19)

Mahatma Gandhi. New York: Gallery Books.

15. Glantz, K., & Pearce, J. K. (1989). Exiles from eden: Psychotherapy from an evolutionary perspective. New York: W.

W. Norton & Company.

16. Gottman, J. M., Katz, L. F., & Hooven, C. (1996). Parental meta-emotion philosophy and the emotional life of families: Theoretical models and preliminary data. Journal of Family Psychology, 10(3), 243-268.

doi:10.1037//0893-3200.10.3.243

17. Harmon-Jones, E., Harmon-Jones, C., Amodio, D. M., & Gable, P. A. (2011).

Attitudes toward emotions. Journal of Personality and Social Psychology, 101(6), 1332-1350.

doi:10.1037/a0024951

18. Hazan, C., & Shaver, P. (1987).

Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52(3), 511-524.

19. Hooven, C., Gottman, J. M., & Katz, L.

F. (1995). Parental meta-emotion structure predicts family and child outcomes. Cognition and Emotion, 9(2- 3), 229-264.

doi:10.1080/02699939508409010

20. Hurrell, K. E., Houwing, F. L., &

Hudson, J. L. (2017). Parental meta- emotion philosophy and emotion coaching in families of children and adolescents with an anxiety disorder.

Journal of Abnormal Child Psychology, 45(3), 569-582. doi:10.1007/s10802- 016-0180-6

21. Izard, C. E. (1972). Patterns of emotions: A new analysis of anxiety and depression. New York: Academic Press.

22. Karnaze, M. M., & Levine, L. J. (2018).

Data versus spock: Lay theories about whether emotion helps or hinders.

Cognition and Emotion, 32(3), 549-565.

doi:10.1080/02699931.2017.1326374 23. Konner, M. (1977). Infancy among the

kalahari desert san. In P. H. Leiderman, S. R. Tulkin & A. Rosenfeld (Eds.), Culture and infancy: Variations in the human experience (pp. 287-328). New York: Academic Press.

24. Konner, M. (2010). The evolution of childhood: Relationships, emotion, mind. Cambridge, MA: Belknap Press of Harvard University Press.

25. Leahy, R. L. (2002). A model of emotional schemas. Cognitive and Behavioral Practice, 9(3), 177-190.

doi:10.1016/S1077-7229(02)80048-7 26. Manser, R., Cooper, M., & Trefusis, J.

(2012). Beliefs about emotions as a metacognitive construct: Initial development of a self-report questionnaire measure and preliminary investigation in relation to emotion regulation. Clinical Psychology and Psychotherapy, 19(3), 235-246.

doi:10.1002/cpp.745

27. Mendonça, D. (2013). Emotions about emotions. Emotion Review, 5(4), 390- 396. doi:10.1177/1754073913484373 28. Mikulincer, M., & Shaver, P. R. (2005).

Mental representations of attachment security: Theoretical foundation for a positive social psychology. In M. W.

Baldwin (Ed.), Interpersonal cognition (pp. 233-266). New York: Guilford.

29. Mikulincer, M., & Shaver, P. R. (2016).

Attachment in adulthood: Structure, dynamics, and change (2nd ed.). New York: The Guilford Press.

(20)

30. Mitmansgruber, H., Beck, T. N., Höfer, S., & Schüßler, G. (2009). When you don't like what you feel: Experiential avoidance, mindfulness and meta- emotion in emotion regulation.

Personality and Individual Differences, 46(4), 448-453.

doi:10.1016/j.paid.2008.11.013

31. Montague, D. P. F., Magai, C., Consedine, N. S., & Gillespie, M.

(2003). Attachment in african american and european american older adults: The roles of early life socialization and religiosity. Attachment and Human Development, 5(2), 188-214.

doi:10.1080/1461673031000108487 32. Morelli, A. M., & Tronick, E. Z. (1991).

Efe multiple caretaking and attachment.

In J. L. Gewirtz, & W. M. Kurtines (Eds.), Intersections with attachment (pp. 41-51). New Jersey: Lawrence Erlbaum Associates.

33. Naragon-Gainey, K., McMahon, T. P., &

Chacko, T. P. (2017). The structure of common emotion regulation strategies:

A meta-analytic examination.

Psychological Bulletin, 143(4), 384-427.

doi:10.1037/bul0000093

34. Nesse, R. M. (1990). Evolutionary explanations of emotions. Human Nature, 1(3), 261-289.

doi:10.1007/BF02733986

35. Netzer, L., Gutentag, T., Kim, M. Y., Solak, N., & Tamir, M. (2018).

Evaluations of emotions: Distinguishing between affective, behavioral and cognitive components. Personality and Individual Differences, 135, 13-24.

doi:10.1016/j.paid.2018.06.038

36. Norberg, M. M., Crone, C., Kwok, C., &

Grisham, J. R. (2018). Anxious attachment and excessive acquisition:

The mediating roles of

anthropomorphism and distress intolerance. Journal of Behavioral Addictions, 7(1), 171-180.

doi:10.1556/2006.7.2018.08

37. Norman, E., & Furnes, B. (2014). The concept of "metaemotion": What is there to learn from research on metacognition?

Emotion Review, 8(2), 187-193.

doi:10.1177/1754073914552913

38. Oatley, K. (2004). Emotions: A brief history. Malden, MA: Blackwell.

39. Paolacci, G., Chandler, J., & Ipeirotis, P.

G. (2010). Running experiments on amazon mechanical turk. Judgment and Decision Making, 5(5), 411-419.

40. Plutchik, R. (1980). Emotion: A psychoevolutionary synthesis. New York: Harper & Row.

41. Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, N.J.:

Princeton University Press.

42. Salters-Pedneault, K., Gentes, E., &

Roemer, L. (2007). The role of fear of emotion in distress, arousal, and cognitive interference following an emotional stimulus. Cognitive Behaviour Therapy, 36(1), 12-22.

doi:10.1080/16506070600874281 43. Schmitt, D. P., Alcalay, L., Allensworth,

M., Allik, J., Ault, L., Austers, I., . . . Shackelford, T. K. (2004). Patterns and universals of adult romantic attachment across 62 cultural regions: Are models of self and of other pancultural constructs?

Journal of Cross-Cultural Psychology, 35(4), 367-402.

doi:10.1177/0022022104266105

44. Simons, J., & Gaher, R. (2005). The distress tolerance scale: Development and validation of a self-report measure.

Motivation and Emotion, 29(2), 83-102.

doi:10.1007/s11031-005-7955-3

(21)

45. Simpson, J. A., & Belsky, J. (2016).

Attachment theory within a modern evolutionary framework. In J. Cassidy,

& P. R. Shaver (Eds.), Handbook of attachment (3rd ed., pp. 91-116). New York: The Guilford Press.

46. Stapinski, L. A., Abbott, M. J., & Rapee, R. M. (2014). Development and validation of the affect intolerance scale to assess maladaptive beliefs and avoidance of emotion. Behaviour Change, 31(3), 205-221.

doi:10.1017/bec.2014.13

47. Tooby, J., & Cosmides, L. (1990). The past explains the present. emotional adaptations and the structure of ancestral environments. Ethology and Sociobiology, 11(4-5), 375-424.

doi:10.1016/0162-3095(90)90017-Z 48. Tsai, J. L., Knutson, B., & Fung, H. H.

(2006). Cultural variation in affect valuation. Journal of Personality and Social Psychology, 90(2), 288-307.

doi:10.1037/0022-3514.90.2.288 49. Van Ijzendoorn, M. H., & Bakermans-

Kranenburg, M. J. (1996). Attachment representations in mothers, fathers, adolescents, and clinical groups: A meta- analytic search for normative data.

Journal of Consulting and Clinical Psychology, 64(1), 8-21.

doi:10.1037/0022-006X.64.1.8

50. Yoon, S., Dang, V., Mertz, J., &

Rottenberg, J. (2018). Are attitudes towards emotions associated with depression? A conceptual and meta- analytic review. Journal of Affective Disorders, 232, 329-340.

doi:10.1016/j.jad.2018.02.009

References

Related documents

While it is clear that certain items that are presented net of tax in the financial statements, such as discontinued operations and the cumulative effects of changes in

Assess the quality of Gothmog Ltd’s internal audit department and the extent to which you, the external auditor, may place reliance on its work.. Click to display/hide the solution

We examine the effect of relational constructs, such as satisfaction, trust and commitment on relationship performance (that is, positive word-of-mouth communication and the

As a leader in promoting patient care excellence, safety and quality since 1996, CALNOC is proud to unveil the CALNOC Research Fund—dedicated to supporting clinical

5 Perform a TB symptom screen: type, severity and onset and duration of each symptom. Arrange for immediate evaluation of symptomatic associates. NOTE: Use respiratory

• Data Governance Strategy Planning • Operating Model Transformation • Business Process Reengineering • Accenture Delivery Methods Whether it is a full data governance

Over the first two decades of market reform, substantial educational and economic disparities emerged in China. This study has attempted to address two questions: 1) what have been