You Can’t See the Real Me: Attachment Avoidance,Self-Verification,and Self-Concept Clarity

Abstract

Attachment shapes people’s experiences in their close relationships and their self-views. Although attachment avoidance and anxiety both undermine relationships, past research has primarily emphasized detrimental effects of anxiety on the self-concept. However, as partners can help people maintain stable self-views, avoidant individuals’ negative views of others might place them at risk for self-concept confusion. We hypothesized that avoidance would predict lower self-concept clarity and that less self-verification from partners would mediate this association. Attachment avoidance was associated with lower self-concept clarity (Studies 1-5), an effect that was mediated by low self-verification (Studies 2-3). The association between avoidance and self-verification was mediated by less self-disclosure and less trust in partner feedback (Study 4). Longitudinally, avoidance predicted changes in self-verification, which in turn predicted changes in self-concept clarity (Study 5). Thus, avoidant individuals’ reluctance to trust or become too close to others may result in hidden costs to the self-concept.

Keywords

self/identity close relationships attachment self-concept clarity self-verification

. . . And then you were face to face, in total darkness, with the knowledge that you didn’t know who you were.

—Richard Yates, Revolutionary Road

Toward the end of Revolutionary Road, which charts the unraveling of a suburban marriage, April Wheeler suddenly realizes that she lacks a clear understanding of herself. The reason for this, it seems, is that she stayed at a distance when she first met her husband and never let him see her for who she really is. The present research examines this phenomenon—do people who tend to keep their distance from others struggle to maintain a clear sense of self? Moreover, could their close others’ lack of knowledge about them help to explain this effect?

Previous research on attachment has emphasized how attachment anxiety harms the self-concept, whereas scarce research has examined how avoidance may be detrimental to the self-concept. However, other people are crucial to understanding the self (Cooley, 1922; James, 1890); thus, avoidant individuals may struggle to form a clear understanding of themselves. We hypothesized that individuals high on attachment avoidance experience lower self-concept clarity, having a less coherent sense of who they are (Campbell et al., 1996). This effect should be mediated by low self-verification—by their partner failing to see them the way they see themselves (Swann, De La Ronde, & Hixon, 1994).

Attachment Avoidance and the Self

Attachment shapes people’s thoughts and behavior in close relationships (Bowlby, 1969; Hazan & Shaver, 1987; Shaver & Mikulincer, 2012). Based on their experiences with close others, people develop sets of working models—mental representations of the self and of other people. Two primary dimensions underlie the attachment system (Bartholomew & Horowitz, 1991). Attachment anxiety stems from a negative working model of the self, whereas attachment avoidance stems from a negative working model of other people (Shaver & Mikulincer, 2012). Individuals high on attachment anxiety desire extreme closeness with other people but fear that others will reject them. Conversely, individuals high on attachment avoidance resist high levels of closeness with others; they distrust other people and are hesitant to become too close to them (Hazan & Shaver, 1987).

Both avoidance and anxiety predict a range of interpersonal outcomes, including lower relationship quality, more negative attributions for a partner’s behavior, and how people cope with breakup (Collins, Ford, Guichard, & Allard, 2006; Davis, Shaver, & Vernon, 2003; Etcheverry, Le, Wu, & Wei, 2013; Simpson, 1990). Extant research on the self-concept, however, tends to focus on the anxiety dimension. Anxiously attached individuals have more negative self-views, lower self-esteem, and less complex self-structures (Bartholomew & Horowitz, 1991; Brennan & Morris, 1997; Mikulincer, 1995).

At first glance, attachment avoidance might seem largely unrelated to self-relevant outcomes. Avoidant individuals and secure individuals, for example, do not substantially differ on self-esteem or overall positivity of self-views (Bartholomew & Horowitz, 1991; Mikulincer, 1995). However, other people are generally instrumental to helping people form a clear sense of self. If avoidant individuals are reluctant to become close to others, they may have difficulty knowing who they are.

Avoidance, Self-Verification, and Self-Concept Clarity

The interdependence that people experience in close relationships benefits the self-concept in multiple ways. Closeness with a romantic partner can help people increase their self-esteem and self-efficacy, make progress toward the person they ideally want to become, pursue personal goals more efficiently, and expand their sense of who they are (Aron, Paris, & Aron, 1995; Drigotas, Rusbult, Wieselquist, & Whitton, 1999; Fitzsimons, Finkel, & vanDellen, 2015; Mattingly, Lewandowski, & McIntyre, 2014).

Beyond facilitating self-change, close others are also crucial in helping people affirm their sense of who they already are, through self-verification. People strive to maintain a stable sense of self, and seeking feedback from close others to verify existing self-views is a central means through which people maintain this stability (Swann & Read, 1981). When people receive feedback that does not match their self-views, they attempt to repudiate it (Swann & Hill, 1982). And although in general, people like to feel good about themselves, people also seek to verify their negative self-views (Swann, Pelham, & Krull, 1989).

Close relationship partners are perhaps best positioned to offer self-verification. In fact, simply being in a relationship context can activate self-verification motivations, particularly for important self-views (Kraus & Chen, 2009). Given the strength of this motivation, when people do not receive self-verification, their relationships tend to suffer. Although in less serious relationships, people want their partners to view them positively, self-verification overrides the positivity motivation when the relationship becomes more serious. If people’s spouses do not confirm their self-views, they feel less close to and less committed to their spouses (Swann et al., 1994; Swann, Hixon, & De La Ronde, 1992). Moreover, in marriages, most people have abstractly positive views of their partners, but they vary in the extent to which they have accurate views of their partner’s specific attributes. Only this specific accuracy predicts relationship persistence (Neff & Karney, 2005).

Self-verification from close others is especially important in maintaining a clear sense of self, or self-concept clarity. Self-concept clarity describes the extent to which people have a clear and coherent sense of who they are—whether they perceive that the traits, goals, preferences, beliefs, and social relationships that make up their self-concept are internally consistent and remain stable over time (Campbell, 1990; Campbell et al., 1996; James, 1890). When people experience a threat to their self-concepts, close others are key to helping people restore self-concept clarity. After learning that they had incorrectly answered Medical College Admission Test (MCAT) questions, aspiring medical students recovered their self-concept clarity when a close friend reminded them why they would make a good doctor. Simply receiving emotional support, however, did not improve self-concept clarity, suggesting that receiving specific validation about the self is crucial for self-concept clarity (Slotter & Gardner, 2014).

Given that avoidant individuals are reluctant to become highly interdependent with their partners, they may miss out on many of these advantages of interdependence for the self-concept. Indeed, although avoidant individuals feel positively about the content of their self-concepts (Bartholomew & Horowitz, 1991; Mikulincer, 1995), they may experience lower self-concept clarity. Avoidant individuals have relatively complex self-structures (Mikulincer, 1995). Although self-complexity is generally distinct from self-concept clarity (Campbell, Assanand, & Di Paula, 2003), at very high levels of self-complexity, people experience lower self-concept clarity (Pilarska, 2016).

We hypothesized that, to the extent that avoidance is associated with lower self-concept clarity, lacking self-verification from close others would account for this association. To our knowledge, no previous research has shown a link between avoidance and self-verification. However, if avoidant individuals are reluctant to self-disclose to their partners (Mikulincer & Nachshon, 1991) or to trust feedback that they do receive from them (Mikulincer, 1998), then they may not experience self-verification. Both of these mechanisms are plausible but capture slightly different reasons why a person might not receive self-verification. A lack of self-disclosure means that a partner would not have the information required to self-verify, whereas not trusting a partner’s feedback suggests a belief that a partner’s feedback may not be honest and lead a person to reject that feedback. Both could contribute to a lack of self-verification. Given that verification from close others is a central means through which people understand themselves, this lack of self-verification may in turn undermine self-concept clarity.

The Present Research

Across five studies, we tested the primary hypotheses that (a) attachment avoidance is associated with lower self-concept clarity, and (b) self-verification mediates the association between attachment avoidance and self-concept clarity.

In Study 1, we examined whether attachment avoidance is associated with lower self-concept clarity. In Studies 2 to 3, we tested whether self-verification mediates this association. Study 4 explored why avoidant individuals experience low self-verification. Two equally reasonable possibilities emerge from the attachment literature. It could be that avoidant individuals do not self-disclose to their partners (Mikulincer & Nachshon, 1991), so their partners lack the information necessary to verify their self-concepts. It could also be the case that avoidant individuals lack self-verification because they do not trust their partners (Mikulincer, 1998), so when their partners attempt to verify their self-concepts, they may not trust this feedback and reject it. We explored both explanations in Study 4. Finally, in Study 5, we tested whether avoidance predicts changes in perceived self-verification and self-concept clarity longitudinally over 9 months.

We assessed self-verification both objectively and subjectively. In Studies 2, 4, and 5, participants self-reported their perception of self-verification, consistent with past self-verification research (Wiesenfeld, Swann, Brockner, & Bartel, 2007). In Study 3, we recruited couples and asked both members of each couple to generate self-aspects describing the self and describing the partner. This method enabled us to code objectively for self-verification—whether people actually saw their partners as their partners saw themselves.

In all studies, we report results with and without controlling for attachment anxiety. As attachment anxiety is related to working models of the self, it seems sensible that it might also predict self-concept clarity, and past research has found that anxiously attached individuals are especially susceptible to experiencing low self-concept clarity after a relationship ends (Slotter & Gardner, 2012). Thus, we hoped to show that effects of avoidance on self-verification and self-concept clarity generally emerge independently from anxiety.

Study 1

Study 1 tested the direct association between attachment avoidance and self-concept clarity. Given that the first author had collected multiple existing datasets containing these variables, we opted to test for this association in all of these datasets.¹

Participants and Measures

These studies consisted of five online samples and two lab samples (one of which featured couples), with both undergraduate and adult participants (total N = 1,265). All participants completed measures of attachment avoidance and anxiety, assessed on the short form of the Experiences in Close Relationships Scale (Wei, Russell, Mallinckrodt, & Vogel, 2007). All participants also completed the Self-Concept Clarity Scale (Campbell et al., 1996). See Appendix S.A in supplemental materials for demographics and scale reliabilities in each sample as well as Appendix S.B in supplemental materials for additional results from one sample.

Results

Prior to analysis, all variables were standardized (M = 0, SD = 1). First, we examined associations between avoidance and self-concept clarity; higher avoidance was associated with lower self-concept clarity in every sample (Table 1). The effect remained when we controlled for attachment anxiety (Table 2). We then conducted a meta-analysis on the effect of avoidance on self-concept clarity across these studies. We weighted the beta from each study by the inverse of its variance to determine the meta-analytic beta. Next, we took the square root of the reciprocal of the sum of the weights to determine the meta-analytic standard error. Finally, we divided the meta-analytic beta by the meta-analytic standard error to create a z score, which enabled significance testing (Borenstein, Hedges, Higgins, & Rothstein, 2009). The meta-analysis revealed that avoidance was significantly associated with self-concept clarity (average β = –.43, average SE = .03, Z = 16.97, p < .001). Given the demographic diversity of our samples, we were also able to explore whether any demographic variables moderated our results. However, we did not find any consistent patterns of moderation (see Appendix S.C in supplemental materials).²

Table 1.

Associations Between Attachment Avoidance and Self-Concept Clarity Across Seven Samples in Study 1.

	r	p	95% CI
Sample 1	−.29	<.001	[–.40, –.17]
Sample 2	−.46	<.001	[–.55, –.37]
Sample 3	−.51	<.001	[–.60, –.41]
Sample 4	−.52	.002	[–.73, –.21]
Sample 5	−.41	<.001	[–.54, –.26]
Sample 6	−.45	<.001	[–.61, –.28]
Sample 7	−.60	<.001	[–.71, –.46]

Note. The estimate for Sample 6 is an unstandardized beta rather than a correlation coefficient, as this sample was analyzed using multilevel modeling to account for nonindependence between members of the couple. CI = confidence interval.

Table 2.

Simultaneous Regression Predicting Self-Concept Clarity From Attachment Avoidance and Anxiety Across Seven Samples in Study 1.

	Avoidance			Anxiety
	β	p	95% CI	β	p	95% CI
Sample 1	−.25	<.001	[–.36, –.14]	−.42	<.001	[–.52, –.31]
Sample 2	−.34	<.001	[–.44, –.23]	−.33	<.001	[–.43, –.22]
Sample 3	−.29	<.001	[–.40, –.18]	−.44	<.001	[–.56, –.33]
Sample 4	−.36	.022	[–.63, –.05]	−.42	.010	[–.68, –.10]
Sample 5	−.32	<.001	[–.46, –.19]	−.48	<.001	[–.62, –.34]
Sample 6	−.38	<.001	[–.52, –.23]	−.41	<.001	[–.55, –.26]
Sample 7	−.37	<.001	[–.51, –.23]	−.52	<.001	[–.66, –.38]

Note. The estimate for Sample 6 is an unstandardized beta coefficient. CI = confidence interval.

Discussion

Across seven samples with 1,265 total participants, with students and non-students, young and older adults, and married and unmarried samples, avoidance was consistently associated with lower self-concept clarity, more than and above anxiety. A meta-analysis revealed a significant and substantial effect of avoidance on self-concept clarity. Thus, Study 1 provided evidence for a direct effect of attachment avoidance on self-concept clarity.

Study 2

Having obtained support for the association between avoidance and self-concept clarity in Study 1, we moved onto our mediational hypothesis in Study 2. We expected that lack of self-verification would mediate the association between avoidance and low self-concept clarity.

Participants

This sample consisted of 218 participants from MTurk (43.1% male, 55.5% female; age M = 33.94, SD = 11.05; 87.6% identified as heterosexual, 2.3% as gay or lesbian, 7.8% as bisexual, .5% as queer, .5% as asexual, and .5% as uncertain or questioning; 2.8% identified as American Indian or Alaska Native, 8.3% as Asian/Asian American, 8.7% as Black/African American, 4.8% as Hispanic, 2.8% as Latino, 76.2% as White, 2.3% as Other³), all of whom were in relationships (44.5% married; relationship duration M = 8.14 years, SD = 9.27). We aimed to collect a large sample, as this study was testing our proposed mechanism.

Measures

All items were assessed on a 7-point scale (1 = strongly disagree, 7 = strongly agree).

Attachment

Participants completed the same measures of attachment avoidance (six items; e.g., “I try to avoid getting too close to my partner”; α = .86, M = 2.46, SD = 1.16) and anxiety (six items; e.g., “I need a lot of reassurance that I am loved by my partner”; α = .83; M = 3.37, SD = 1.34) as in the samples reported in Study 1.

Self-verification

We created a four-item measure of perceived self-verification for this study (adapted from Wiesenfeld et al., 2007 to apply to relationships). To improve reliability, we dropped one item (“Even if I act in ways that are different from my true self, my partner knows that the way I’m acting isn’t who I really am”), leaving three items in the final scale (“My partner sees my true self,” “My partner sees me for the person I really am,” “I feel that my partner understands the person I am deep down”; α = .94, M = 5.81, SD = 1.12).

Self-concept clarity

Participants completed the same measure of self-concept clarity as in the samples reported in Study 1 (12 items; e.g., “In general, I have a clear sense of who I am and what I am”; α = .93, M = 4.66, SD = 1.32).

Results

Prior to analyses, all variables were standardized (M = 0, SD = 1). First, we examined correlations between key variables (Table 3). Replicating the effects from Study 1, avoidance was negatively associated with self-concept clarity. Self-verification was associated with both avoidance and self-concept clarity. We then tested whether self-verification mediates the association between avoidance and self-concept clarity, using the PROCESS macro for SPSS (Hayes, 2013). As expected, the indirect effect was significant, (indirect effect = –.10; 95% confidence interval [CI] = [–.20, –.009]; Figure 1).⁴

Table 3.

Associations Between Variables in Study 2.

	(1)	(2)	(3)	(4)
(1) Avoidance
r	—	—	—	—
p	—	—	—	—
95% CI	—	—	—	—
(2) Anxiety
r	.35	—	—	—
p	<.001	—	—	—
95% CI	[.23, .46]	—	—	—
(3) Self-Verification
r	−.65	−.34	—	—
p	<.001	<.001	—	—
95% CI	[–.72, –.57]	[–.45, –.22]	—	—
(4) Self-Concept Clarity
r	−.45	−.60	.38	—
p	<.001	<.001	<.001	—
95% CI	[–.55, –.34]	[–.68, –.51]	[.26, .49]	—

Note. CI = confidence interval.

Figure 1.

Self-verification mediating the association between avoidance and self-concept clarity in Study 2.

Discussion

In Study 2, we found that avoidance was associated with lower self-verification, assessed through people’s self-reports, which in turn predicted lower self-concept clarity. However, we could not conclude whether avoidant individuals actually do experience less self-verification, or whether their working models of others are biasing their perception of the extent to which their partner verifies them. To that end, in Study 3, we attempted to replicate the findings from this study with an objective measure of self-verification.

Study 3

Study 3 aimed to replicate the mediational effect from Study 2 with an objective measure of self-verification. In this study, we assessed self-verification through objective agreement between partners on the content of each person’s self-concept. This approach differs slightly from the traditional method of assessing self-verification, which asks people to rate themselves and their partners on five positive self-attributes and examines correspondence between how people rate themselves and how their partners rate them (Swann et al., 1994). In this method, self-verification is essentially assessed as the extent to which people’s partners see them as positively or negatively as they see themselves. One strength of original measure is that, by standardizing attributes, it captures important differences in the evaluative tone between self–partner agreement. In the current work, we were interested in specific attributes, rather than evaluative tone. We expected that avoidant people’s partners would be less accurate about the content of that person’s self-concept, that there would be less match between the specific attributes avoidant people use to describe themselves (e.g., artistic, disorganized), and the attributes their partners use to describe them. As such, instead of using preselected attributes, we asked participants to generate their own, capturing verification of core self-concept content rather than valence of a person’s self-concept.

Participants and Procedure

We recruited 66 couples for this study (132 individuals; 48.5% male, 51.5% female; age M = 20.30, SD = 2.64; 37.3% identified as Asian/Asian American, 6.2% as Black/African American, 6.9% as Hispanic, 9.9% as Latino, 60% as White, 2.3% as Other). Participants were recruited through the introductory psychology subject pool, flyers around campus, postings on paid participant listservs, and postings in student groups on Facebook. All participants were currently in a romantic relationship, and both members of the couple were required to come into the lab to participate (3.0% married or in a committed lifelong partnership; relationship duration M = 1.11 years, SD = 1.25; 80.3% identified as heterosexual, 4.5% as gay or lesbian, 9.1% as bisexual, 1.5% as queer, 2.3% as pansexual, 2.3% as other).

After consenting to participate, each partner completed an online questionnaire in separate rooms with measures of attachment, self-verification, and self-concept clarity. Once the study was complete,⁵ participants received either course credit or US$10 as compensation.

Measures

Attachment

Participants completed the 36-item Experiences in Close Relationships Scale (Fraley, Waller, & Brennan, 2000; avoidance: α = .92; M = 2.02, SD = 0.77; anxiety: α = .89; M = 3.57, SD = 0.95).

Self-verification

Participants completed an adapted version of the Twenty Statements Test (Kuhn & McPartland, 1954). First, they were asked to “Please think about who you are as an individual, and write 10 answers to the question “who am I” in the spaces provided. Answer as if you were giving the answers to yourself—not someone else. Write your answers in the order that they occur to you. Don’t worry about logic or importance.” Next, they were instructed to “Please think about your partner is an individual, and write 10 answers to the question “who is my partner” in the spaces provided. Answer as if you were giving the answers to yourself—not someone else. Write your answers in the order that they occur to you. Don’t worry about logic or importance.”

Self-concept clarity

Participants completed the same measure of self-concept clarity as in previous studies (α = .91; M = 4.57, SD = 1.25).

Coding

Following the completion of data collection, two independent coders assessed self-verification in two ways. First, coders read the lists of attributes and counted the number of attributes a person generated about themselves that also appeared on the list of attributes their partner generated about them (self-verification count; intraclass correlation coefficient [ICC] = .73). Second, coders made a gestalt assessment of the extent to which the attributes a person generated for themselves overlapped with the attributes their partner generated about them (general self-verification; 1 = no overlap at all, 5 = high overlap; ICC = .75). After coding was complete, coder responses were averaged.

Results

Prior to analyses, all variables except self-verification count were standardized (M = 0, SD = 1). First, we examined associations between our key variables, using multilevel modeling to account for nonindependence between partners (Table 4). Consistent with previous studies, avoidance was associated with lower self-concept clarity. Avoidance was associated with lower self-verification count and lower general self-verification. Self-verification count was associated with greater self-concept clarity, as was general self-verification.

Table 4.

Associations Between Variables in Study 3.

	(1)	(2)	(3)	(4)	(5)
(1) Avoidance
b	—	—	—	—	—
p	—	—	—	—	—
95% CI	—	—	—	—	—
(2) Anxiety
b	.22	—	—	—	—
p	.012	—	—	—	—
95% CI	[.05, .39]	—	—	—	—
(3) Self-Verification (Count)
b	−.25	−.22	—	—	—
p	.004	.012	—	—	—
95% CI	[–.42, –.08]	[–.38, –.05]	—	—	—
(4) Self-Verification (General)
b	−.23	−.20	.86	—	—
p	.007	.015	<.001	—	—
95% CI	[–.39, –.06]	[–.37, –.04]	[.78, .93]	—	—
(5) Self-Concept Clarity
b	−.35	−.30	.19	.23	—
p	<.001	<.001	.026	.011	—
95% CI	[–.51, –.18]	[–.46, –.13]	[.02, .36]	.05, .40	—

Note. The associations are given in unstandardized betas, as we used multilevel modeling to account for nonindependence between partners. CI = confidence interval.

Next, we examined whether self-verification mediated the association between avoidance and self-concept clarity.⁶ We used the PROCESS macro for SPSS (Hayes, 2013). Self-verification count mediated the association between avoidance and self-concept clarity (indirect effect = –.05; 95% CI = [–.11, –.008]; Figure 2). Likewise, general self-verification mediated the association between avoidance and self-concept clarity (indirect effect = –.03; 95% CI = [–.08, –.0005]; Figure 2).⁷

Figure 2.

Self-verification mediating the association between avoidance and self-concept clarity in Study 3.

Discussion

In Study 3, we conceptually replicated the effect from Study 2, showing that self-verification mediates the association between attachment avoidance and self-concept clarity. We assessed self-verification through an adapted version of the traditional self-verification measure (Swann et al., 1994), with coders objectively rating agreement between the attributes people listed to describe themselves and the attributes their partners generated to describe them. Thus, it appears that avoidant individuals actually do receive less self-verification and are not simply reporting less verification due to attachment biases.

Study 4

Studies 2 and 3 provided preliminary evidence for our mediational hypothesis. In Study 4, we explored why avoidant individuals experience less self-verification. It could be that avoidant individuals receive less self-verification because they do not disclose information about themselves to their romantic partners. It is also possible that avoidant individuals lack self-verification because they reject their partner’s feedback. Both explanations seemed equally reasonable, so we did not advance a priori hypotheses as to which might be driving the link between avoidance and self-verification.

Participants and Procedure

We recruited 278 individuals from MTurk (46.4% male, 53.2% female, 0.4% transgender; age M = 35.61, SD = 11.03; 0.4% identified as American Indian or Alaska Native, 3.3% as Asian/Asian American, 6.9% as Black/African American, 4% as Hispanic, 3.3% as Latino, 83.5% as White, .4% as Other), all of whom were in relationships (58.3% married or in a committed lifelong partnership; relationship duration M = 8.61 years, SD = 8.94; 88.1% identified as heterosexual, 2.9% as gay or lesbian, 7.2% as bisexual, 0.7% as pansexual, 0.4% as asexual, 0.7% as other). Participants completed measures of attachment, self-disclosure, trust in partner feedback, and self-concept clarity.

Measures

Unless otherwise indicated, all items were assessed on 7-point scales (1 = strongly disagree, 7 = strongly agree).

Attachment

Participants completed the same measures of attachment avoidance (α = .88; M = 2.27, SD = 1.18) and attachment anxiety (α = .80; M = 3.27, SD = 1.27) as in Studies 1 and 2.

Self-disclosure

Participants completed a measure of self-disclosure, reporting how often they disclose about various topics to their partner (Miller, Berg, & Archer, 1983; 10 items; e.g., “what makes me the person I am”; α = .93; M = 5.53, SD = 1.15); assessed on a 7-point scale (1 = not at all, 7 = extremely).

Trust in partner feedback

We created a measure of trust in partner feedback for this study (five items; e.g., “When my partner tells me things about myself, I tend to believe him or her”; α = .89; M = 5.96, SD = 1.29; see Appendix S.E in supplemental materials for details about this measure and the self-disclosure measure).

Self-verification

We assessed self-verification with the same self-report measure as in Study 2 (α = .95; M = 5.78, SD = 1.31).

Self-concept clarity

Participants completed the same measure of self-concept clarity as in previous studies (α = .94; M = 4.81, SD = 1.34).

Results

All measures were standardized prior to analysis (M = 0, SD = 1). First, we examined correlations among all variables (Table 5). Replicating previous studies, avoidance was associated with lower self-verification and self-concept clarity; avoidance was also negatively associated with self-disclosure and trust in partner feedback. Self-verification was positively associated with both self-disclosure and trust in partner feedback, and it was positively associated with self-concept clarity.

Table 5.

Associations Between Variables in Study 4.

	(1)	(2)	(3)	(4)	(5)	(6)
(1) Avoidance
r	—	—	—	—	—	—
p	—	—	—	—	—	—
95% CI	—	—	—	—	—	—
(2) Anxiety
r	.42	—	—	—	—	—
p	<.001	—	—	—	—	—
95% CI	[.32, .52]	—	—	—	—	—
(3) Trust in Feedback
r	−.66	−.37	—	—	—	—
p	<.001	<.001	—	—	—	—
95% CI	[–.72, –.59]	[–.47, –.26]	—	—	—	—
(4) Self-Disclosure
r	−.63	−.23	.61	—	—	—
p	<.001	<.001	<.001	—	—	—
95% CI	[–.69, –.55]	[–.34, –.12]	[.53, .68]	—	—	—
(5) Self-Verification
r	−.69	−.34	.71	.68	—	—
p	<.001	<.001	<.001	<.001	—	—
95% CI	[–.75, –.62]	[–.44, –.23]	[.64, .76]	[.61, .74]	—	—
(6) Self-Concept Clarity
r	−.49	−.60	.38	.24	.36	—
p	<.001	<.001	<.001	<.001	<.001	—
95% CI	[–.57, –.39]	[–.67, –.52]	[.28, .48]	[.13, .35]	[.26, .46]	—

Note. CI = confidence interval.

Next, we examined whether self-disclosure or trust in partner feedback could partly explain the association between higher attachment avoidance and lower self-verification. Using the PROCESS macro for SPSS (Model 4; Hayes, 2013), we tested for simultaneous mediation by trust in partner feedback and self-disclosure. We found evidence for mediation by self-disclosure (indirect effect = –.19; 95% CI = [–.29, –.11]) and trust in partner feedback (indirect effect = –.23; 95% CI = [–.34, –.12]; Figure 3). Avoidant individuals tended to self-disclosure less and not to trust their partner’s feedback, and these in turn partly accounted for avoidant individuals’ tendency to report less self-verification from their partners.

Figure 3.

Trust in partner feedback and self-disclosure as parallel mediators of the association between attachment avoidance and self-concept clarity in Study 4.

Finally, we explored two serial mediation models (PROCESS model 6), testing (a) whether avoidance predicts self-disclosure, which in turn predicts self-verification, which in turn predicts self-concept clarity, and (b) whether avoidance predicts trust in partner feedback, which in turn predicts self-verification, which in turn predicts self-concept clarity. However, neither model yielded significant evidence for serial mediation (self-disclosure model: indirect effect = .001; 95% CI = [–.05, .05]; trust in partner feedback model: indirect effect = –.03; 95% CI = [–.09, .01]; see Appendix S.F in supplemental materials).⁸

Discussion

Study 4 explored two possible mechanisms that might account for avoidant individuals’ tendency to perceive less self-verification from their partners. Both trust in partner feedback and self-disclosure mediated this association. Avoidant individuals reported less tendency to self-disclose to their partners and less trust in feedback from their partners about their self-concept; these in turn were associated with perception of self-verification. We also explored two serial mediation models predicting self-concept clarity as the ultimate dependent variable. However, we did not find significant evidence for either of these models. Thus, although Study 4 obtained proximal evidence for both potential mediators of the association between avoidance and perceived self-verification, we did not find support for the more complex model.

Study 5

In Study 5, we return to our primary mediational hypothesis: Self-verification will mediate the association between avoidance and self-concept clarity. Studies 2 and 3 supported this hypothesis; however, the data were cross-sectional, and we believe that the process we are examining is one that likely unfolds over time. Although relationships normatively increase in intimacy and self-disclosure over time (e.g., Altman & Taylor, 1973), avoidant individuals tend to shun closeness and intimacy in their everyday interactions (Tidwell, Reis, & Shaver, 1996). Thus, avoidance should influence changes in self-verification over time, and this in turn should influence self-concept clarity. In Study 5, we examined this process longitudinally, following participants over 9 months. We also aimed to rule out several possible confounds. We aimed to show that these effects emerge when controlling for self-esteem. Self-concept clarity and self-esteem, although theoretically distinct, are typically moderately correlated (Campbell, 1990). Although avoidant individuals typically do not experience low self-esteem, given the link between self-esteem and self-concept clarity, as well as the role of self-esteem in self-verification (Swann et al., 1994), we aimed to show that our hypothesized mediational model remained when controlling for self-esteem. We also controlled for extraversion and agreeableness in this model.

Participants and Procedure

This sample consisted of 120 individuals recruited from a Midwestern university (75% female, 25% male; age M = 21.94, SD = 4.05; 85.0% identified as heterosexual, 7.5% as bisexual, 5.8% as gay or lesbian; 4.2% identified as African American/Black, 21.7% as Asian American/Asian, 67.5% as Caucasian/White, 16.7% as Hispanic/Latino/a, 7.5% as Multiracial, and 1.7% as Other), all of whom had been in a relationship for at least a year (4.2% married; relationship duration M = 2.45 years, SD = 1.60). Participants were recruited through flyers around campus; paid subject pool listservs; classroom-wide and dormitory-wide professor e-mails; classroom announcements; fraternity and sorority announcements; online, university-affiliated Facebook groups; student newspaper and newsletter advertisements; and university-targeted Facebook advertisements, with the goal of recruiting as many participants as possible over an academic quarter.

The study followed participants over 9 months. Participants completed a screening questionnaire to determine eligibility, then completed the intake questionnaire online. About 3 months later, they received a link to the Wave 1 questionnaire. About 3 months after completing the Wave 1 questionnaire, they received a link to the Wave 2 questionnaire. Finally, 3 months after completing the Wave 2 questionnaire, participants came into the lab to complete the Wave 3 questionnaire as well as several additional tasks.⁹ Participants received up to US$60 for completing the study.

Of the 120 individuals who completed the intake questionnaire, 114 completed Wave 1, 111 completed Wave 2, and 110 completed Wave 3. In addition, 17 participants had broken up with their partners byWave 3. In the longitudinal analyses, we focus on the 95 individuals who completed the entire study and did not experience a breakup.

Measures

All items were assessed on 7-point scales (1 = strongly disagree, 7 = strongly agree).

Attachment

Participants completed the same measure as in Study 3 (avoidance: α = .91, M = 2.21, SD = 0.78; anxiety: α = .93, M = 2.90, SD = 1.11).

Self-verification

Participants completed a one-item measure at each wave of the study, taken from the measure of self-verification used in Studies 2 and 4 (“my partner sees me for the person I really am”; intake, M = 6.11, SD = 0.91; Wave 1, M = 6.05, SD = 0.83; Wave 2, M = 6.14, SD = 0.72; Wave 3, M = 6.11, SD = 0.75).

Self-concept clarity

Participants completed a one-item measure of self-concept clarity at each wave (“I have a clear sense of who I am and what I am”; Campbell et al., 1996; intake, M = 5.65, SD = 1.15; Wave 1, M = 5.81, SD = 1.13; Wave 2, M = 5.75, SD = 1.05; Wave 3, M = 5.48, SD = 1.08) as well as the full self-concept clarity at intake (α = .90; M = 4.39, SD = 1.10).

Self-esteem

Participants completed the Rosenberg Self-Esteem Scale (Rosenberg, 1965; α = .92; M = 5.52, SD = 1.03).

Extraversion and agreeableness

Participants completed the extraversion and agreeableness subscales of the Mini-International Personality Item Pool–Five-Factor Model (IPIP; Donnellan, Oswald, Baird, & Lucas, 2006; extraversion: α = .81; M = 4.17, SD = 1.34; agreeableness: α = .81; M = 5.64, SD = 0.92.

Results

All variables were standardized prior to analysis (M = 1, SD = 0). Our primary hypothesis was that changes in perceived self-verification would mediate the association between attachment avoidance and changes in self-concept clarity over time. First, we examined the direct association between avoidance and self-concept clarity at intake.¹⁰ Replicating results from Studies 1 – 4, avoidance was negatively associated with both the one-item measure of self-concept clarity (r = –.23, p = .013; 95% CI = [–.39, –.05]) and the full Self-Concept Clarity Scale (r = –.23, p = .011; 95% CI = [–.39, –.05]). We also examined whether avoidance was associated with self-concept clarity across all waves of the study, using multilevel modeling with wave nested within person. Across waves, avoidance was associated with lower self-concept clarity (b = –.18, p = .006; 95% CI = [–.31, –.05]). Next, we conducted cross-lagged analyses to examine whether avoidance at intake predicts changes in self-concept clarity across 9 months. In this analysis, avoidance at intake (controlling for self-concept clarity at intake) did not predict changes in self-concept clarity from intake to Wave 3 (β = .05, p = .57; 95% CI = [–.11, .20]). Thus, although avoidance was consistently associated with overall self-concept clarity, it does not appear to directly predict changes in self-concept clarity over 9 months.

Next, we turned to our hypothesized mediator. Replicating results from previous studies, avoidance was negatively associated with self-verification at intake (r = –.49, p < .001; 95% CI = [–.61, –.34]). We then examined whether avoidance at intake predicts changes in self-concept clarity at Wave 3 through changes in self-verification Waves 1 and 2. To simplify analyses, we averaged across self-verification at Waves 1 and 2¹¹ and used the PROCESS macro for SPSS (Hayes, 2013). In this model, we controlled for both self-verification and self-concept clarity at intake. Our hypothesis was supported (indirect effect = –.15; 95% CI = [–.29, –.07]; Figure 4). As hypothesized, attachment avoidance predicted changes in the perception of partner self-verification, which accounted for changes in self-concept clarity over 9 months.¹²

Figure 4.

Attachment avoidance at intake predicting changes in perceived self-verification at Waves 1 and 2, in turn predicting changes in self-concept clarity at Wave 3 in Study 5.

Auxiliary analyses

We reran the mediation analysis controlling for three related constructs: self-esteem, extraversion, and agreeableness. In this model, the analysis remained significant (indirect effect = –.13; 95% CI = [–.24, –.05]). We also tested an alternative model in which avoidance predicted self-concept clarity at Waves 1 and 2, which in turn predicted self-verification at Wave 3. This analysis did not yield significant evidence (indirect effect = –.005; 95% CI = [–.05, .008]).¹³

Discussion

In Study 5, we obtained longitudinal evidence for our mediational hypothesis. Over 9 months, attachment avoidance predicted changes in perceived self-verification from their romantic partner, which in turn predicted changes in self-concept clarity. These findings suggest that our hypothesized process unfolds over time.

General Discussion

Although past research has established that attachment anxiety can harm the self-concept, the potential detriments of attachment avoidance to the self have been largely unexplored. In general, avoidant individuals have equivalent self-esteem to that of securely attached individuals (Bartholomew & Horowitz, 1991; Mikulincer, 1995). Yet, given the importance of close others to the self, we hypothesized that avoidance might be costly to the self-concept in other ways. Specifically, close others help people clarify their sense of who they are. Receiving self-verification from another person, or feedback that affirms one’s existing self-views, helps people maintain self-concept clarity, a clear and coherent sense of self (Campbell et al., 1996; Slotter & Gardner, 2014; Swann & Read, 1981). However, avoidant individuals are reluctant to become too close to others or to disclose to them (Mikulincer & Nachshon, 1991). Thus, we expected that avoidant individuals would receive lower self-verification from close others, which in turn should account for their lower self-concept clarity.

We tested these hypotheses across five studies. Attachment avoidance was consistently associated with lower self-concept clarity (Studies 1-5). Avoidant individuals reported less self-verification (Study 2) and objectively received less self-verification from their partners, as they showed a greater discrepancy between the self-attributes they generated to describe themselves and those their partner generated to describe the actor’s self-concept (Study 3). Lower self-verification mediated the association between avoidance and lower self-concept clarity (Studies 2 and 3). Both the tendency not to self-disclose and not to trust their partner’s feedback partly explained the link between avoidance and perceived self-verification (Study 4). Longitudinally, avoidance predicted decreases in self-verification; in turn, higher self-verification predicted increases in self-concept clarity (Study 5).

Implications and Future Directions

These studies are among the first to examine how attachment avoidance might have negative implications for the self-concept. Typically, the anxiety dimension in attachment focuses on the extent to which a person has a negative model of the self, whereas the avoidance dimension focuses on the extent to which a person has a negative model of others (see Shaver & Mikulincer, 2012). Consequently, most research on attachment and the self-concept has focused on attachment anxiety. However, the present research suggests that avoidant individuals also experience self-concept impairment.

The present research also adds to the literature emphasizing the importance of interdependence for the self-concept. Close others can help to shape who we are, facilitate self-growth, enhance self-esteem and self-efficacy, and restore self-concept clarity after a threat (Aron et al., 1995; Drigotas et al., 1999; Fitzsimons et al., 2015; Mattingly et al., 2014; Slotter & Gardner, 2014). However, these studies suggest that resisting interdependence may damage the self-concept. Specifically, in Study 4, we found that avoidant individuals experience lower self-verification partly because they hesitate to self-disclose to their partners and to trust their partner’s feedback. Thus, it is not simply the case that avoidance only predicts poorer relationship quality and behavior that threatens the relationship (DeWall et al., 2011; Etcheverry et al., 2013). The same orientation toward others that may destabilize avoidant individuals’ relationships also appears to destabilize their self-concepts.

Future research might benefit from unpacking the possibility that links between attachment orientation and self-concept might vary across cultures. There is suggestive evidence that avoidance may not have unique effects on self-concept clarity beyond attachment anxiety in Taiwanese samples (Wu, 2009). Moreover, in this sample, lower self-esteem mediated the association between attachment and self-concept clarity, for both avoidance and anxiety (Wu, 2009). However, in more individualistic cultures, avoidance is not typically associated with lower self-esteem (Bartholomew & Horowitz, 1991; Mikulincer, 1995). We suspect that in individualistic samples, self-esteem may mediate the association between attachment anxiety and self-concept clarity, but not the association between attachment avoidance and self-concept clarity. There is little existing cross-cultural research on romantic attachment; thus, future research could help illuminate the ways in which attachment shapes the self-concept in ways that are consistent or different across cultures.

Strengths and Limitations

This research contributes toward filling a gap in the literature on attachment and the self-concept, emphasizing that avoidance has detrimental effects on the self. Moreover, to the best of our knowledge, no previous research has shown a link between attachment avoidance and self-verification from one’s partner. We obtained preliminary evidence that avoidant individuals’ lack of self-verification arises from their tendency to withhold self-disclosure and to distrust their partner’s feedback. Moreover, we used two different measures of self-verification—subjective and objective. Past research has asked participants to self-report the extent to which they feel verified (Wiesenfeld et al., 2007) and has compared actor and partner ratings on scales asking the extent to which various preselected attributes are characteristic of the self and partner (e.g., Swann et al., 1994). To the best of our knowledge, no previous research has adapted the Twenty Statements Test (Kuhn & McPartland, 1954) to compare content of the self-concept spontaneously generated by people about themselves and content that their partners generated about them. We believe this method may be helpful for future research focused on self-verification of self-concept content.

However, the current studies did not distinguish clearly between two possible forms of low self-verification. First, a partner could fail to provide self-verification due to inaccuracy—viewing the person as someone who he or she is not. For example, if Harry thinks that Sally is an excellent chef, when Sally knows that she could burn water, the feedback Harry gives Sally about her cooking is inaccurate. This type of failed self-verification is akin to a distorted image in a mirror. Second, a partner could fail to provide self-verification due to imprecision—because they lack a clear image of the person. For example, Harry may not have a clear idea one way or the other about whether Sally identifies as a cook, or whether she is good at cooking. This type of failed self-verification is akin to a blurry image in a mirror. We suspect that avoidant individuals perceive this blurry image form of low self-verification from their partners. Withholding self-disclosure and rejecting their partner’s feedback would most likely result in a blurry image—avoidant individuals’ partners simply do not have a clear idea of who they are. Future research should aim to untangle these two possibilities more directly.

The effects in this research emerged in both student and adult samples, cross-sectionally and longitudinally. All samples were fairly large (Ns > 100). Furthermore, based on the meta-analytic effect sizes from Study 1, we conducted power analyses using G*Power (Faul, Erdfelder, Buchner, & Lang, 2009), which revealed that we would need at least 60 participants (1 – β > .95) to detect the basic association between avoidance and self-concept clarity, and at least 116 participants to detect the effect of avoidance on self-concept clarity when controlling for anxiety. All of our samples were large enough to detect the basic association between avoidance and self-concept clarity. Studies 2 to 4 had large enough samples to detect the association between avoidance and self-concept clarity when controlling for anxiety. Study 5 had sufficient power in the cross-sectional analysis, but was slightly under the number recommended by the power analysis in the longitudinal analyses.

Our results were generally robust when controlling for attachment anxiety in all five studies and were robust when we controlled for self-esteem, extraversion, and agreeableness in Study 5. Out of 30 analyses, 25 hypothesis tests yielded identical conclusions when controlling for anxiety. There was generally not a consistent pattern in terms of which effects remained and which did not, so it is difficult to conclude why this might be. It appeared that the indirect effects through self-verification were especially prone to fall to nonsignificance; however, a meta-analysis suggested that this effect is generally robust.¹⁴ It would be helpful for future research to replicate these effects and determine more clearly whether there are specific circumstances under which avoidance does and does not show unique effects on self-verification and self-concept clarity.

Although the longitudinal design of Study 5 enabled some causal clues about the direction of these effects, this research is limited in that none of the studies were experimental. An experimental manipulation of attachment could determine if inducing attachment avoidance causally reduces the perception of self-verification from one’s partner. It would also be interesting to manipulate self-verification and examine how attachment influences people’s responses to feeling a lack of verification from their partners. Avoidant individuals should experience especially low self-concept clarity when they believe their partners do not see them as they see themselves. However, more securely attached individuals may find means of maintaining relatively high clarity, perhaps by turning to other sources for self-verification or by attempting to correct their partner’s inaccuracies.

One unexpected effect emerged in Study 5. Although avoidance did not significantly predict changes in self-concept clarity over 9 months, when self-verification from the romantic partner was entered into the mediational model, avoidance was then associated with increases in self-concept clarity (i.e., the total effect became positive and significant). We suspect that this finding may reflect a suppression effect (MacKinnon, Krull, & Lockwood, 2000). Future research should examine whether this puzzling effect replicates; until it does, we are reluctant to speculate on its potential meaning.

The present research only examined attachment avoidance; future research could also explore additional mechanisms related to the components of self-concept clarity for both avoidance and anxiety. Specifically, self-concept clarity includes both people’s perception that the content of their self-concept is clear and fits together as well as the perception that this content is stable over time (Campbell et al., 1996). The mechanism proposed in these studies for avoidance relates to this first element of self-concept clarity. That is, a lack of verification from one’s partner undermines one’s certainty about the content of the self-concept. However, anxious individuals may experience lower self-concept clarity due to the second element of self-concept clarity—the perception that the content that makes up their self-concepts is unstable over time. Anxiously attached individuals do experience greater changes in their self-concepts during relationships (Slotter & Gardner, 2012). Thus, they may experience more fluctuations in the content of their self-concepts, which in turn may undermine their self-concept clarity. Future research should examine how these dual mechanisms related to different elements of self-concept clarity may operate distinctly for both anxious and avoidant individuals.

Conclusion

Across five studies, attachment avoidance predicted lower self-concept clarity, lacking a clear and coherent sense of self. Avoidant individuals both perceived and actually experienced a lack of self-verification from their romantic partners—their partners did not see them as they see themselves. Lack of self-verification mediated the association between avoidance and self-concept clarity both cross-sectionally and longitudinally. Thus, when April Wheeler in Revolutionary Road realizes that she does not know who she is, it may be because she has maintained her distance from her husband, preventing him from really knowing her. In turn, his inability to know her may have eroded her understanding of herself over time.

Supplemental Material

Avoidance_SCC_Supplementals_R3 – Supplemental material for You Can’t See the Real Me: Attachment Avoidance, Self-Verification, and Self-Concept Clarity

Supplemental material, Avoidance_SCC_Supplementals_R3 for You Can’t See the Real Me: Attachment Avoidance, Self-Verification, and Self-Concept Clarity by Lydia F. Emery, Wendi L. Gardner, Kathleen L. Carswell, and Eli J. Finkel in Personality and Social Psychology Bulletin

Supplemental Material

emery_online_appendix – Supplemental material for You Can’t See the Real Me: Attachment Avoidance, Self-Verification, and Self-Concept Clarity

Supplemental material, emery_online_appendix for You Can’t See the Real Me: Attachment Avoidance, Self-Verification, and Self-Concept Clarity by Lydia F. Emery, Wendi L. Gardner, Kathleen L. Carswell, and Eli J. Finkel in Personality and Social Psychology Bulletin

Footnotes

Acknowledgements

The authors wish to thank Amy Muise, Emily Dix, Benjamin Le, and Christopher Petsko.

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by a National Science Foundation Graduate Research Fellowship awarded to Lydia F. Emery, a Social Sciences and Humanities Research Council Doctoral Fellowship awarded to Kathleen L. Carswell, and a Northwestern Graduate Research Grant awarded to Kathleen L. Carswell.

Notes

Supplemental Material

Supplementary material is available online with this article.

References

Altman

Taylor

(1973). Social penetration: The development of relationships. New York, NY: Holt, Rinehart and Winston.

Aron

Paris

Aron

E. N.

(1995). Falling in love: Prospective studies of self-concept change. Journal of Personality and Social Psychology, 69, 1102-1112.

Bartholomew

Horowitz

L. M.

(1991). Attachment styles among young adults: A test of a four-category model. Journal of Personality and Social Psychology, 61, 226-244.

Borenstein

Hedges

L. V.

Higgins

J. P. T.

Rothstein

H. R.

(2009). Introduction to meta-analysis. Chichester, UK: John Wiley.

Bowlby

(1969). Attachment and loss: Vol. 1. Attachment (2nd ed.). New York, NY: Basic Books.

Brennan

K. A.

Morris

K. A.

(1997). Attachment styles, self-esteem, and patterns of seeking feedback from romantic partners. Personality and Social Psychology Bulletin, 23, 23-31.

Campbell

J. D.

(1990). Self-esteem and clarity of the self-concept. Journal of Personality and Social Psychology, 59, 538-549.

Campbell

J. D.

Assanand

Di Paula

(2003). The structure of the self-concept and its relation to psychological adjustment. Journal of Personality, 71, 115-140.

Campbell

J. D.

Trapnell

P. D.

Heine

S. J.

Katz

I. M.

Lavallee

L. F.

Lehman

D. R.

(1996). Self-concept clarity: Measurement, personality correlates, and cultural boundaries. Journal of Personality and Social Psychology, 70, 141-156.

10.

Collins

N. L.

Ford

M. B.

Guichard

A. C.

Allard

L. M.

(2006). Working models of attachment and attribution processes in intimate relationships. Personality and Social Psychology Bulletin, 32, 201-219.

11.

Cooley

C. H.

(1922). Human nature and the social order. New York, NY: Charles Scribner’s Sons.

12.

Davis

Shaver

P. R.

Vernon

M. L.

(2003). Physical, emotional, and behavioral reactions to breaking up: The roles of gender, age, emotional involvement, and attachment style. Personality and Social Psychology Bulletin, 29, 871-884.

13.

DeWall

C. N.

Lambert

N. M.

Slotter

E. B.

Deckman

Pond

R. D.

Finkel

E. J.

. . . Fincham

F. D.

(2011). So far away from one’s partner, yet so close to romantic alternatives: Avoidant attachment, interest in alternatives, and infidelity. Journal of Personality and Social Psychology, 101, 1302-1316.

14.

Donnellan

M. B.

Oswald

F. L.

Baird

B. M.

Lucas

R. E.

(2006). The mini-IPIP scales: Tiny-yet-effective measures of the Big Five factors of personality. Psychological Assessment, 18, 192-203.

15.

Drigotas

S. M.

Rusbult

C. E.

Wieselquist

Whitton

S. W.

(1999). Close partner as sculptor of the ideal self: Behavioral affirmation and the Michelangelo phenomenon. Journal of Personality and Social Psychology, 77, 293-323.

16.

Etcheverry

P. E.

Wei

(2013). Attachment and the investment model: Predictors of relationship commitment, maintenance, and persistence. Personal Relationships, 20, 546-567.

17.

Faul

Erdfelder

Buchner

Lang

A.-G.

(2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41, 1149-1160.

18.

Fitzsimons

G. M.

Finkel

E. J.

vanDellen

M. R.

(2015). Transactive goal dynamics. Psychological Review, 122, 648-673.

19.

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, 350-365.

20.

Hayes

A. F.

(2013). Introduction to mediation, moderation, and conditional process analysis. New York, NY: Guilford Press.

21.

Hazan

Shaver

(1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52, 511-524.

22.

James

(1890). The principles of psychology. Cambridge, MA: Harvard University Press.

23.

Kenny

D. A.

(1998). Couples, gender, and time: Comments on method. In Bradbury

T. N.

(Ed.), The developmental course of marital dysfunction (pp. 410-422). New York, NY: Cambridge University Press.

24.

Kraus

M. W.

Chen

(2009). Striving to be known by significant others: Automatic activation of self-verification goals in relationship contexts. Journal of Personality and Social Psychology, 97, 58-73.

25.

Kuhn

M. H.

McPartland

T. S.

(1954). An empirical investigation of self-attitudes. American Sociological Review, 19, 68-76.

26.

MacKinnon

D. P.

Krull

J. L.

Lockwood

C. M.

(2000). Equivalence of the mediation, confounding, and suppression effect. Prevention Science, 1, 173-181.

27.

Mattingly

B. A.

Lewandowski

G. W.

McIntyre

K. P.

(2014). “You make me a better/worse person”: A two-dimensional model of relationship self-change. Personal Relationships, 21, 176-190.

28.

Mikulincer

(1995). Attachment style and the mental representation of the self. Journal of Personality and Social Psychology, 69, 1203-1215.

29.

Mikulincer

(1998). Attachment working models and the sense of trust: An exploration of interaction goals and affect regulation. Journal of Personality and Social Psychology, 74, 1209-1224.

30.

Mikulincer

Nachshon

(1991). Attachment styles and patterns of self-disclosure. Journal of Personality and Social Psychology, 61, 321-331.

31.

Miller

L. C.

Berg

J. H.

Archer

R. L.

(1983). Openers: Individuals who elicit intimate self-disclosure. Journal of Personality and Social Psychology, 44, 1234-1244.

32.

Neff

L. A.

Karney

B. R.

(2005). To know you is to love you: The implications of global adoration and specific accuracy for marital relationships. Journal of Personality and Social Psychology, 88, 480-497.

33.

Pilarska

(2016). How do self-concept differentiation and self-concept clarity interrelate in predicting sense of personal identity? Personality and Individual Differences, 102, 85-89.

34.

Rosenberg

(1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.

35.

Shaver

P. R.

Mikulincer

(2012). Attachment theory: The mind viewed in its social-relational context. In Van Lange

Kruglanski

Higgins

E. T.

(Eds.), Handbook of theories of social psychology (Vol. 2, pp. 160-179). Newbury Park, CA: SAGE.

36.

Simpson

J. A.

(1990). Influence of attachment styles on romantic relationships. Journal of Personality and Social Psychology, 59, 971-980.

37.

Slotter

E. B.

Gardner

W. L.

(2012). How needing you changes me: The influence of attachment anxiety on self-concept malleability in romantic relationships. Self and Identity, 11, 386-408.

38.

Slotter

E. B.

Gardner

W. L.

(2014). Remind me who I am: Social interaction strategies for maintaining the threatened self-concept. Personality and Social Psychology Bulletin, 40, 1148-1161.

39.

Swann

W. B.

De La Ronde

Hixon

J. G.

(1994). Authenticity and positivity strivings in marriage and courtship. Journal of Personality and Social Psychology, 66, 857-869.

40.

Swann

W. B.

Hill

C. A.

(1982). When our identities are mistaken: Reaffirming self-conceptions through social interaction. Journal of Personality and Social Psychology, 43, 59-66.

41.

Swann

W. B.

Hixon

J. G.

De La Ronde

(1992). Embracing the bitter “truth”: Negative self-concepts and marital commitment. Psychological Science, 3, 118-121.

42.

Swann

W. B.

Pelham

B. W.

Krull

D. S.

(1989). Agreeable fancy or disagreeable truth? Reconciling self-enhancement and self-verification. Journal of Personality and Social Psychology, 57, 782-791.

43.

Swann

W. B.

Read

S. J.

(1981). Self-verification processes: How we sustain our self-conceptions. Journal of Experimental Social Psychology, 17, 351-372.

44.

Tidwell

M. C.

Reis

H. T.

Shaver

P. R.

(1996). Attachment, attractiveness, and social interaction: A diary study. Journal of Personality and Social Psychology, 71, 729-745.

45.

Wei

Russell

D. W.

Mallinckrodt

Vogel

D. L.

(2007). The Experiences in Close Relationship Scale (ECR)-short form: Reliability, validity, and factor structure. Journal of Personality Assessment, 88, 187-204.

46.

Wiesenfeld

B. M.

Swann

W. B.

Brockner

Bartel

C. A.

(2007). Is more fairness always preferred? Self-esteem moderates reactions to procedural justice. Academy of Management Journal, 50, 1235-1253.

47.

(2009). The relationship between attachment style and self-concept clarity: The mediation effect of self-esteem. Personality and Individual Differences, 47, 42-46.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.20 MB

0.15 MB