Attachment and conflict

communication in adult romantic

relationships
Rachel Domingue & Debra Mollen
Texas Woman’s University

ABSTRACT
This study explored the connections between adult attachment
styles (i.e., secure, preoccupied, fearful-avoidant, dismissing)
and communication patterns during conflict (i.e., mutual
constructive, demand-withdraw, mutual avoidance, and with-
holding). Specifically, this study examined how the combina-
tion of both partners’ attachment styles, or couple type (i.e.,
secure-secure, secure-insecure, insecure-insecure), related to
self-reported conflict communication patterns. Couples had
been together for at least two years (i.e., in a dating, cohabi-
tating, engaged, or marital relationship). Participants included
43 different-sex couples and 10 same-sex couples, who lived
primarily in a large metropolitan area in the southwestern US.
Secure-secure couples reported the most mutually constructive
communication, while the insecure-insecure couples group
reported the most demand-withdraw and mutual avoidance
and withholding communication. Implications for counseling
with couples and families are discussed.

KEY WORDS:
adult romantic relationships attachment style •
communication conflict • demand-withdrawal

Having meaningful, fulfilling relationships with others has been found to be

central to overall life satisfaction; however, high rates of divorce persist
(Twenge & King, 2005). While there are a few unique predictors of rela-
tionship satisfaction for same-sex couples, such as social support from and
acceptance of lesbian, gay or bisexual (LGB) identity by friends (Elizur &

All correspondence concerning this article should be sent to Debra Mollen, PhD,Texas Woman’s
University, P.O. Box 425470, Department of Psychology and Philosophy, Denton, Texas 76204,
USA [e-mail: [email protected]].

Journal of Social and Personal Relationships Copyright © 2009 SAGE Publications

(www.sagepublications.com), Vol. 26(5): 678–696. DOI: 10.1177/0265407509347932

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

Domingue & Mollen: Attachment and communication 679

Mintzer, 2003), there are a multitude of predictors of relationship satisfac-

tion that different-sex and same-sex couples share (i.e., intimacy, mutuality,
quality of communication, humor, shared interests, fidelity, dependability,
constructive conflict resolution skills, and shared power) (Eldridge &
Gilbert, 1990; Kaslow & Robison, 1996; Mackey, Diemer, & O’Brien, 2004).
This finding is encouraging since it points to the possibility that the roots
of romantic relationships of different-sex and same-sex couples are similar.
Although this knowledge base gives us an idea of what characteristics long,
satisfying romantic relationships have in common, determining a concep-
tual framework to understand why these relationships endure remains an
important goal.
One such framework is attachment theory, which has been proposed and
researched by Bowlby (1969, 1973, 1980),Ainsworth, Blehar,Waters, and Wall
(1978), and Hazan and Shaver (1987, 1990, 1994). Attachment theory asserts
that secure attachment in adult romantic relationships is characterized by
mutuality, comfort with closeness and intimacy, and reciprocal caregiving and
support. Emotionally Focused Therapy (EFT) merges attachment theory
and experiential therapy and is “one of only about five empirically supported
(EST) marital and family treatment approaches” (Bradley & Johnson, 2005,
p. 181). Susan Johnson (2004a), founder of EFT, asserted,“. . . it is the absence
of disclosing and responsive interactions that begins the process of relation-
ship distress. Individual attachment needs are then left unsatisfied, and it is
this deprivation and distance that eventually lead to conflict and distress”
(p. 35). Conceptually, EFT therapists work with couples to move from
insecure attachment to secure attachment through communication of
attachment-related emotions, needs, and fears. Moreover, EFT watches for
emotional responses that tend to generate others, especially destructive ones
like withdrawal, which usually leads to criticizing and demanding. In order
to inform the counseling profession’s practice of couples therapy, it is critical
that empirical research explore and refine its knowledge base about which
communication patterns are related to secure and insecure attachment.

Attachment and communication

From an attachment perspective, communication is seen as a vehicle for

fostering intimacy and trust in relationships due to partners feeling under-
stood and accepted (Pietromonaco, Greenwood, & Barrett, 2004). Johnson
(2003) stated that secure adults tend to engage in more self-disclosure and
direct communication than insecure adults, because they are confident
enough to be assertive and trust that their partner will respond support-
ively. Insecure adults are more likely to cling, make demands, stonewall, or
withdraw because they believe their partner will reject them or they are
protesting the unresponsiveness of their partner. Johnson (2003) asserted
that “. . . the ability to disclose and confide in a direct way about needs
and fears and to tune in to the other’s experience is crucial if partners are
to define . . . the relationship as a safe haven and a secure base” (p. 109).

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

680 Journal of Social and Personal Relationships 26(5)

Therefore, intimate and open communication builds and maintains the

attachment bond between adult partners.
Bartholomew’s theoretical conceptualization of adult attachment styles
illuminates how men and women would relate to and communicate with
their partners based on their attachment style. Bartholomew (1990) devised
four types of adult attachment styles according to the combinations of
positive/negative model of self and positive/negative model of others.
Bartholomew asserted that a positive model of self is distinguished by a
self-concept characterized by competence, security, high self-worth indepen-
dent of others’ acceptance, and a perception of the self as basically worthy
and lovable. A negative model of the self is characterized by a sense of self-
worth that is primarily dependent on the approval of others and a percep-
tion of the self as basically unlovable and unworthy. A positive model of
others is characterized by an approach orientation toward intimacy and the
belief that others are caring, trustworthy, and responsive, while a negative
model of others is characterized by an avoidant orientation toward intimacy
and a belief that others are rejecting, unresponsive, and distant. Table 1 lists
characteristics associated with each attachment style group based on em-
pirical study (Bartholomew, 1990; Bartholomew & Horowitz, 1991; Collins,
Guichard, Ford, & Feeney, 2004).
Couples’ communication, especially its interactive processes, has been
examined extensively as a predictor of relationship satisfaction. Gottman
(1999a) found that negative affect reciprocity, which is “. . . the increased
probability that a person’s emotions will be negative (anger, belligerence,
sadness, contempt, and so on) right after his or her partner has exhibited
negativity” (p. 37), is the most consistent negatively correlated predictor of
relationship satisfaction. Gottman (1999a) further identified four specific
negative communication behaviors which are the most harmful to couples:
criticism, defensiveness, contempt, and stonewalling. Gottman (1999a) found
that stable, happy couples exhibit criticism, defensiveness, and stonewalling
less frequently than unsatisfied, unhappy couples; however, contempt was
found to be virtually non-existent in stable, happy couples. Gottman (1999a)
asserted that stable, happy couples also engage in more effective repair
attempts when the conflict discussion begins to turn negative. Repair
attempts (e.g., smiling, expressing appreciation to soften complaints, posi-
tively commenting on the communication itself, expressing commitment)
prevent further negative reciprocity, thereby de-escalating the conflict and
creating a less negative, more positive emotional climate. Happy, stable
couples are more successful in their repair attempts, which is linked to what
Gottman (1999b) terms positive sentiment override, which leads one partner
to interpret the other partner’s negative behaviors as neutral due to their
belief that his/her partner and their relationship are positive in an overall,
global sense. Gottman (1999a) also proposed that the emphasis on problem-
solving skills in couples’ therapy is misplaced and should change to a
concentration on helping couples’ continue to engage in dialogues about
their “perpetual problems,” those that are enduring due to fundamental
differences in personality or personal needs, and avoid gridlock.

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

Domingue & Mollen: Attachment and communication 681

TABLE 1
Characteristics associated with the secure, preoccupied, dismissing, and
fearful-avoidant attachment style groups based on the research of
Bartholomew (1990), Bartholomew and Horowitz (1991), and Collins,
Guichard, Ford, and Feeney (2004)

Model of Self

Positive Negative

SECURE PREOCCUPIED
• Self as competent and lovable • Self as incompetent and unlovable
• High self-esteem (source is positive • Low self-esteem (fluctuates based
• relationships) • on the approval of others)
• High self-acceptance • Sees others as caring, trustworthy,
• Confident in social situations • and responsive
Positive

• Sees others as caring, trustworthy, • Relationship goal: create more

• and responsive • closeness; merge
• Relationship goal: mutual • Clings to relationships and feels
• interdependence • lost, overwhelmed, and unlovable
• Expects positive outcomes when • when not in a relationship
• seeking support from partner • Places high importance on
• Places moderate importance on • receiving comfort from and
Model of Others

• receiving comfort from and • maintaining proximity to partner

• maintaining proximity to partner

DISMISSING FEARFUL-AVOIDANT
• Self as competent and lovable • Self as incompetent and unlovable
• High self-esteem (source is work/ • Low self-esteem (fluctuates based
• autonomy) • on the approval of others)
• Sees others as unresponsive and • Sees others as rejecting, distant,
• distant • and unresponsive
Negative

• Relationship goal: stay self-reliant • Conflicting approach and

• and retain autonomy • avoidance orientations toward
• Views relationships as secondary • intimacy due to looking toward
• Places low importance on receiving • partner to fulfill esteem needs while
• comfort from and maintaining • simultaneously fearing his/her
• proximity to partner • partner’s rejection
• Places high importance on receiving
comfort from and maintaining
proximity to partner

The most consistently found dysfunctional communication pattern is the

demand-withdraw communication pattern (i.e., one partner makes demands
for change and the other partner avoids the conflict by walking away, dis-
engaging emotionally, or changing the topic) that was first identified by
Sullaway and Christensen (1983). Christensen and Heavey (1993) reported
in their research that woman demand/man withdraw was more likely to
occur than man demand/woman withdraw in a variety of specific popula-
tions of married couples. Klinetob and Smith (1996), however, reported that

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

682 Journal of Social and Personal Relationships 26(5)

the role of the demander was taken on by the person who generated the
conflict topic and the other partner took on the role of the withdrawer;
therefore, whoever wanted change to occur became the demander and the
one who wanted to keep things the way they were became the withdrawer.
The demand-withdraw communication pattern has been linked to decreased
current and future relationship satisfaction; however, high expression of
positive affect moderated the strength of this negative association (Caughlin
& Huston, 2002; Heavey, Christensen, & Malamuth, 1995). Julien, Chartrand,
Simard, Bouthillier, and Bégin (2003) found that both negative individual
behaviors (e.g., hostility, tension, negative affect, avoidance, denial, and un-
responsiveness) and positive dyadic behavior (e.g., interactional synchrony
characterized by adaptation, timing, pacing, and shared conversational
control) during the conflict task explained unique variance in relationship
quality of both different-sex and same-sex couples.
Pietromonaco et al. (2004) explored how research on conflict in adult
romantic relationships can be conceptualized in an attachment theory
framework. They theorized that individuals’ attachment styles predict what
expectations and beliefs they have of their partners and their relationships
and what relationship goals they strive to achieve, which in turn influence
what thoughts, feelings, and behavior they engage in during conflict. Since
secure individuals believe their partners are available and responsive and
will not abandon them, they view discussion of conflict as an opportunity
to increase intimacy and adjust to each others’ needs and therefore engage
in self-disclosure and use a variety of conflict-resolution strategies. Because
preoccupied individuals fear their partners will abandon them or not be
available when needed, they are likely to interpret conflict as a threat and
display intense emotions based on those fears, which makes it difficult for
them to communicate constructively with their partners. Dismissing individ-
uals will likely interpret conflict as a threat, because they may feel pressured
to engage in self-disclosure and other behaviors that increase emotional
intimacy, which is counter to their relationship goal of maintaining indepen-
dence; therefore, they will withdraw or minimize the significance of the
conflict. Fearful-avoidant individuals will likely display a mixture of beha-
viors associated with dismissing and preoccupied attachment styles, because
they are constantly torn between an approach and an avoidant orientation
toward intimacy.
Their review of research from 1988 to 2002 demonstrated that empirical
evidence supports the hypothesis that secure individuals, when compared
with insecure individuals, report engaging in more constructive behavior
during conflict that facilitates intimacy (Pietromonaco et al., 2004). Couples
consisting of two secure partners engage in more constructive communica-
tion behaviors than couples where one or both partners are insecure partners.
In turn, couples consisting of one secure partner, especially if that partner
is the husband, engage in more constructive communication behaviors than
couples consisting of two insecure partners. Their theories and findings of
their review are consistent with the theoretical assumptions of Emotionally
Focused Therapy (EFT).

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

Domingue & Mollen: Attachment and communication 683

The present study

The present study focused on the exploration of the theoretical connections

between attachment theory and communication between partners in adult
romantic relationships. The theory and literature suggest that if both, or
even one, partner has a secure attachment style, the couple will engage in
more constructive, intimacy-building communication, which has been linked
to increased relationship satisfaction and longevity. If an empirical link
between combined attachment styles of couples and their communication
patterns exists, clinicians may be able to target their interventions to impact
either or both of those variables to help their clients in couples therapy.
Most of the empirical research linking attachment and communication
has been done with different-sex samples, especially young, newly married
couples, and the ethnic compositions of the samples have been predom-
inantly Caucasian. Very few studies have included dating, cohabitating,
engaged, and married different-sex couples and dating and cohabitating
same-sex couples. Studies have also varied greatly in how they measure
attachment and communication.To address the gaps in the research, we used
more refined measures and attempted to recruit a more diverse (e.g., age,
ethnicity, and sexual orientation) sample. Couple type, which was opera-
tionalized as the combination of each individual’s attachment style for each
couple, was used to predict communication patterns. Such a practice may
expand empirical evidence concerning the relationship between attach-
ment style and communication patterns beyond the individual level.

Hypotheses

Hypotheses were focused both on finding differences among the three couple
types (secure-secure, secure-insecure, insecure-insecure) and on whether
specific combinations of insecure attachment styles within couples better
predict some communication patterns. If one or more partner’s specific
insecure attachment style (e.g., dismissing, preoccupied, and fearful avoidant)
predicted a specific negative communication pattern, then clinicians would
be able to focus on and target that dynamic. The authors did not anticipate
any differences between different-sex and same-sex couples or between
men and women for the following hypotheses.
H1: Participants in the secure-secure couples group will report greater
mutual constructive conflict communication than participants in the secure-
insecure group who, in turn, will report greater mutual constructive com-
munication than participants in the insecure-insecure couples group.
H2: Participants in the insecure-insecure couples group will report greater
demand-withdraw communication than participants in the secure-insecure
group who, in turn, will report greater demand-withdraw communication
than participants in the secure-secure couples group.
H3: Participants in the insecure-insecure couples group will report greater
mutual avoidance and withholding communication than participants in the

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

684 Journal of Social and Personal Relationships 26(5)

secure-insecure couples who, in turn, will report greater mutual avoidance

and withholding communication than participants in the secure-secure
couples.
H4: Couples in which at least one partner is dismissing will report greater
demand-withdraw communication than all other couple types.
H5: Participants in the dismissing-dismissing couples group will report
greater mutual avoidance and withholding communication than all other
couple types.

Method

Participants
The sample consisted of couples who had been together for at least two
years, which means they were either in a dating, cohabitating, engaged, or
married relationship. Participants included 43 different-sex couples and 10
same-sex couples, who either lived in a large metropolitan area in a south-
western state or in other states. Participants’ average age was 36.78 years
(SD = 10.46, Range = 23.0–69.0). Fifty-seven women (53.8%) and 49 men
(46.2%) participated. The sample consisted primarily, though not exclu-
sively, of Caucasians (87.7%). Participants reported their sexual orientation
as heterosexual (78.3%), gay (5.7%), lesbian (12.3%), bisexual (2.8%), or
transgender (0.9%). Nearly three-quarters of participants (71.7%) reported
holding a Bachelor’s degree or higher.
The different-sex couples group mainly consisted of married couples
without children (48.2%). The average duration of their marriages was
106.52 months, and they had dated and been engaged to each other for an
average of 25.67 (SD = 36.97) and 9.69 months (SD = 5.53), respectively.
Of the 10 same-sex couples, eight couples reported being in a committed
relationship, while one reported being married and one couple disagreed
on their relationship status. Only one same-sex couple reported having a
child. The average lengths of their dating and committed relationships
were 18.65 months and 85.94 months, respectively. Nine of the 10 couples
reported cohabitation (M = 43.36 months).

Procedure
The different-sex couple group was recruited from community parent groups,
a local university, and local places of worship of diverse denominations. The
same-sex couple group was recruited from a large gay and lesbian church
and a community center that serves the gay and lesbian population. Snowball
sampling was used to ensure that the maximum number of couples would
participate, which resulted in the participation of some couples living in
other states. The following recruitment methods were used: posting flyers
on bulletin boards, inserting an announcement in a newsletter or bulletin,
distributing flyers to professors, providing flyers to group leaders, providing
flyers and packets to group leaders, posting an e-mail on the group’s listserv,
and sending an e-mail to key informants who helped with the snowball

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

Domingue & Mollen: Attachment and communication 685

sampling. The flyer included eligibility requirements and researcher contact

information.
Couples received a packet with two copies of the consent form, an authors-
generated demographic questionnaire, target measures, and a postage-ready,
addressed envelope. Eighteen couples who initially requested packets were
removed from the study due to one or both partners not returning their
packets of questionnaires.

Measures

Demographic questionnaire. The demographic questionnaire assessed age,

gender, ethnicity, sexual orientation, educational level, religious affiliation,
occupation, relationship duration, number and ages of children, and living
arrangement.

Experiences in Close Relationships – Revised (ECR-Revised). Fraley,Waller,

and Brennan’s (2000; Fraley, 2005) ECR-Revised consists of 36 items that
measure the two attachment dimensions of avoidance and anxiety. High
scores on the avoidance subscale indicate high levels of withdrawal and
discomfort with closeness/intimacy, while high scores on the anxiety subscale
indicate high levels of jealousy, fear of rejection/abandonment, and pre-
occupation with attachment needs. Participants’ responses on the 18 items
that made up each of the two subscales were averaged. The internal relia-
bility estimates for the avoidance subscale and the anxiety subscale were .92
and .90 respectively, which is consistent with prior research (Fraley, 2005).

Communication Patterns Questionnaire (CPQ). Sullaway and Christensen’s

(1983) CPQ is a 23-item self-report measure that assesses a couple’s typical
interaction patterns during three phases of conflict: the emergence of a
relational problem, the discussion of the relational problem, and the post-
discussion period (Noller & White, 1990). Each item is accompanied by a
1–9 Likert scale (ranging from very unlikely to very likely; Noller & White,
1990). We changed the wording of the CPQ from “man/woman” to “you
and your partner” in order to make it applicable to same-sex couples.
Christensen and Shenk (1991) delineated three communication pattern
subscales of the CPQ: mutual constructive (both partners initiate discussion
of problems, express their feelings, and engage in negotiation/compromise),
demand-withdraw (one partner nags and makes demands, while the other
partner withdraws), and mutual avoidance and withholding (both partners
avoid discussing problems, avoid each other, and withhold emotional/physical
contact during the post-discussion period). Heavey, Larson, Christensen,
and Zumtobel (1996) revised the CPQ’s mutual constructive communica-
tion subscale so it reflected the presence of positive, and the absence of
negative, communication behaviors and eliminated items that included the
discussion’s outcome. For the mutual constructive communication scale,
participants’ responses to the three positive items (discussion, expression
of feelings, negotiation, compromise) are summed and then their responses

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

686 Journal of Social and Personal Relationships 26(5)

to the three negative items (blaming, threatening, verbal aggression) are

subtracted. We used Heavey et al.’s (1996) revised mutual constructive
communication subscale along with the original demand-withdraw and
mutual avoidance and withholding communication subscales proposed by
Christensen and Shenk (1991). The internal reliability estimates for the
three subscales (mutual constructive, demand-withdraw, and mutual avoid-
ance and withholding) were .83, .77, and .71 respectively, which is consis-
tent with prior research (Christenson & Shenk, 1991; Heavey et al., 1996).

Results

Table 2 displays the descriptive statistics and reliability statistics for the
ECR-R (Anxiety and Avoidance) and the CPQ (Mutual Constructive,
Demand-Withdraw, and Mutual Avoidance and Withholding). Mean Anxiety
and Avoidance scores for the present sample are lower than those reported
by Fraley (2005), indicating that this sample was more secure. To evaluate
the partner agreement about conflict communication patterns, a dependent
samples t-test was performed for each of the CPQ subscales. Analyses
revealed that the average of partners’ difference scores within each couple
was not significantly different on any of the subscales.

TABLE 2
Descriptive and reliability statistics for subscales of ECR-R and CPQ

Cronbach’s
Subscale M SD Min. Max. Alpha

ECR-R: 2.16 0.89 1.00 4.89 0.90

Anxiety
ECR-R: 2.14 0.86 1.00 5.17 0.92
Avoidance
CPQ: 11.67 9.00 –12.00 23.00 0.83
Mutual Constructive
CPQ: 21.53 9.76 6.00 43.00 0.77
Demand-Withdraw
CPQ: 8.36 4.59 3.00 24.00 0.71
Mutual Avoidance
& Withholding

Notes. N = 106. Possible ECR-R Anxiety subscale range = 1.00–7.00; ECR-R Avoidance sub-
scale range = 1.00–7.00; CPQ Mutual Constructive subscale range = –33.00–23.00; CPQ Demand-
Withdraw subscale range = 6.00–54.00; CPQ Mutual Avoidance and Withholding subscale range
= 3.00–27.00. ECR-R = Experiences in Close Relationships; CPQ = Communication Patterns
Questionnaire.

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

Domingue & Mollen: Attachment and communication 687

Preparation of the data for analysis

Several steps were required to assign couples to couple type groups (secure-
secure, secure-insecure, insecure-insecure). First, participants were classified
into one of four individual attachment style groups (secure, preoccupied,
fearful-avoidant, dismissing) based on their ECR-R scores (avoidance and
anxiety) using K-Means cluster analysis (see Brennan, Clark, & Shaver,
1998). This analysis computed four group centroids that acted as the center
clustering points and then classified each participant according to which
cluster center his/her scores were closest to. Figure 1 presents a visual repre-
sentation of the cluster analysis. Using this procedure, a majority of partic-
ipants were secure (n = 69, 65.1%) with fewer preoccupied (n = 23, 21.7%),
dismissing (n = 8, 7.5%), and fearful-avoidant (n = 6, 5.7%) participants.
The centroids for each group were as follows: Secure (1.64, 1.78), Preoccu-
pied (3.04, 2.22), Dismissing (2.56, 4.13), and Fearful-Avoidant (4.30, 3.25).
Second, individuals were classified as either secure or insecure by coding
the fearful-avoidant, preoccupied, and dismissing attachment styles as in-
secure attachment styles. Finally, each couple was assigned to a couple type
group based on the combination of the individuals’ attachment styles:
secure-secure, insecure-insecure, and secure-insecure.

FIGURE 1
Visual representation of K-Means Cluster Analysis used to determine
participants’ individual attachment styles

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

688 Journal of Social and Personal Relationships 26(5)

Analyses for Hypotheses 1, 2, and 3

A 3 (Couple Type: secure-secure, insecure-insecure, secure-insecure) × 2
(Different-sex vs. Same-sex couple) × 2 (Gender: male, female) MANOVA,
with mutual constructive communication, demand-withdraw communication,
and mutual avoidance and withholding communication as the continuous,
dependent variables, was used to test Hypotheses 1, 2, and 3. The within
couple variance was addressed by grouping individual participants with
their partners into couple type groups. The MANOVA revealed significant
variation in reported communication patterns across couple types [Wilks’
lambda = 0.80, F(6,188) = 3.74, p < 0.01] and different-sex vs. same-sex
couples [Wilks’ lambda = 0.92, F(3, 94) = 2.81, p < 0.05], while there was no
significant variation found across gender [Wilks’ lambda = 0.98, F(3, 94) =
0.57, ns] or any significant interactions.
Since there was a significant main effect for couple type, the authors
examined the univariate F tests for each communication subscale across
couple type. There was significant variation of scores on all three subscales
[Mutual Constructive F(2, 96) = 10.07, p < 0.01; Demand-Withdraw F(2, 96)
= 5.79, p < 0.01; Mutual Avoidance and Withholding F(2, 96) = 6.88, p <
0.01] across couple types. The group means for each couple type (secure-
secure, secure-insecure, insecure-insecure) for each communication subscale
are displayed in Table 3. Using the Tukey HSD test, post-hoc multiple
comparisons were made to ascertain which couple type group means were
significantly different. Participants in the secure-secure couples group
reported greater mutual constructive communication than participants in
both the secure-insecure and insecure-insecure couples groups, which do
not differ from one another. Participants in the secure-secure couples group
reported less demand-withdraw and mutual avoidance and withholding
communication than participants in both the secure-insecure and insecure-
insecure couples groups; however, the group means of the secure-insecure
and insecure-insecure couples groups did not differ significantly.

TABLE 3
Group means for CPQ subscales by couple types

Couple Type

Secure-Secure Secure-Insecure Insecure-Insecure

(N = 54) (N = 30) (N = 22)
Subscale M (SD) M (SD) M (SD)

Mutual Constructive 15.26a (5.69) 8.80b (9.99) 6.77b (10.74)

Demand-Withdraw 18.06a (8.59) 24.07b (9.77) 26.59b (9.55)
Mutual Avoidance & 6.57a (3.03) 9.30b (4.37) 11.45b (6.05)
Withholding

Notes. Possible CPQ Mutual Constructive subscale range = –33.00–23.00; CPQ Demand-With-
draw subscale range = 6.00–54.00; CPQ Mutual Avoidance and Withholding subscale range =
3.00–27.00. Dissimilar subscripts indicate significant difference (p < 0.01).

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

Domingue & Mollen: Attachment and communication 689

There was significant variation of scores on mutual avoidance and with-

holding communication (F (1, 96) = 5.69, p < 0.05) across couple types;
however, there was no significant variation for the other two communica-
tion subscales (mutual constructive F (1, 96) = 0.014, ns; demand-withdraw
F(1, 96) = 0.140, ns). Participants in the different-sex couples group
reported greater mutual avoidance and withholding communication (M =
9.00) than did participants in the same-sex couples group (M = 5.60).

Roy-Bargmann Stepdown Analysis. Since the dependent variables were

highly intercorrelated, a Roy-Bargmann stepdown analysis was conducted
to determine if our findings held when the dependent variables’ covariance
was statistically controlled. First, a univariate ANOVA was conducted with
couple type, different-sex vs. same-sex couples, and gender as independent
variables and mutual constructive communication as the dependent variable.
Mutual constructive communication was entered as the first dependent
variable because it measures both positive and negative communication
patterns. Second, an ANCOVA was conducted with the same independent
variables; however, mutual constructive communication was entered as a
covariate and the dependent variable was demand-withdraw communica-
tion. Finally, a second ANCOVA was conducted with the same independent
variables; however, mutual constructive and demand-withdraw communi-
cation were entered as covariates and the dependent variable was mutual
avoidance and withholding communication. The Bonferroni adjustment
was made to control for familywise error due to carrying out three separate
analyses, which resulted in the alpha decreasing from 0.05 to 0.017.
At the first step, as noted above, there was significant variation of mutual
constructive communication across couple type, F(2, 96) = 10.07, p < 0.01.
In the second step, there was no significant variation of demand-withdraw
communication across couple type once the co-variation of mutual construc-
tive communication was controlled, F(2, 95) = 1.06, ns. In the final step, there
was no significant variation of mutual avoidance and withholding communi-
cation across couple type once the co-variation of mutual constructive and
demand-withdraw communication was controlled, F (2, 94) = 0.62, ns. There
was significant variation, however, of mutual avoidance and withholding
communication between the different-sex and same-sex couples groups
(F(1, 94) = 8.23, p < 0.01), after controlling for covariation of the other two
dependent variables. Different-sex couples reported higher levels of mutual
avoidance and withholding communication than did participants in the
same-sex couples group.

Analyses for Hypotheses 4 and 5

In order to test Hypotheses 4, a 2 (Couple Type: at least 1 partner dismiss-
ing, any other couple type) × 2 (Same-sex vs. Different-sex couple) × 2
(Gender: male, female) MANOVA was used. Contrary to Hypothesis 4, the
MANOVA revealed no significant variation in reported communication
patterns across couples types (Wilks’ lambda = 0.94, F (3, 97) = 1.97, ns).
There was likely an insufficient number of couples in the first couple type

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

690 Journal of Social and Personal Relationships 26(5)

group (seven versus 46 in the second couple type group) to accurately test
this hypothesis. Hypothesis 5 was untestable because there was only 1
dismissing-dismissing couple.

Discussion

The pattern of results provides partial support for Hypotheses 1, 2, and 3.

As predicted by Hypothesis 1, participants in the secure-secure couple group
did report greater mutual constructive communication than participants
in both the secure-insecure and insecure-insecure couple groups. Unlike
previous studies (Pietromonaco et al., 2004), however, participants in the
secure-insecure and insecure-insecure couple groups did not differ in mutual
constructive communication.
Analyses for Hypotheses 2 and 3 show that participants in the secure-
secure couple group reported less demand-withdraw and mutual avoidance
and withholding communication than participants in both secure-insecure
and insecure-insecure couple groups, however, the latter two groups did not
differ from one-another. The insecure-insecure couple group did not report
higher amounts of these negative communication patterns than the secure-
insecure group, but did report more negative communication than the
secure-secure couple group as predicted. These results should be considered
tentative, because they were not significant in the Roy-Bargmann step-
down analysis. Since these two negative communication subscales measure
specific negative communication behaviors, the three conflict communica-
tion dependent variables were strongly intercorrelated. Therefore, after
controlling for the effects of one dimension, there was not enough varia-
tion left in the remaining two to achieve statistical significance.
An unexpected finding was that different-sex couples reported higher
levels of mutual avoidance and withholding communication than did same-
sex couples. Although this finding has not been found in previous research
(Julien et al., 2003), it is possible that adults in same-sex relationships have
similar communication styles due to their similar gender socialization and
therefore would be less likely to shut down or withdraw out of frustration
and misunderstanding. No differences between participants in the different-
sex couples and same-sex couples group were observed for mutual construc-
tive and demand-withdraw communication.All of these findings are tentative
and should be interpreted with caution due to the small number of same-
sex couples in the sample.

Implications for theory

Gottman (1999a) emphasized negative reciprocity and the four horsemen of
intimate relationships (i.e., criticism, defensiveness, contempt, stonewalling).
In the current investigation, insecure-insecure couples and secure-insecure
couples engaged in more verbal aggression, withdrawal, avoidance, and
demanding behaviors than the secure-secure couples. This finding is theor-
etically consistent with Gottman’s work because insecure attachment leads

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

Domingue & Mollen: Attachment and communication 691

one to protest one’s partner’s unresponsiveness (making demands, criticism)

or downplay dependence needs by maintaining separateness (defensive-
ness, contempt, stonewalling). Gottman also pointed to the importance of
remaining engaged and avoiding gridlock when communicating. Secure-
secure couples reported that both partners initiated conversations about
relationship problems and that both were able to express their feelings and
feel understood.
Johnson (2003) theorized that different attachment styles lead to differ-
ent communication styles. She asserted that secure adults engage in direct
and assertive conversation and self-disclosure more than insecure adults,
because they trust their partners to respond supportively and view the
relationship as a safe haven with a secure base. In turn, insecure adults tend
to cling, make demands, stonewall, or withdraw, because they believe their
partners will reject them or are protesting their partners’ unresponsiveness.
In the current study, secure-secure couples reported the highest level of
mutual constructive communication; both partners were able to express
their feelings and needs clearly and assertively and trusted that their
partners would not reject them for their self-disclosures. Insecure-insecure
couples and secure-insecure couples reported higher levels of demand-
withdraw and mutual avoidance and withdrawal communication. Johnson’s
theory does not address the combination of the partners’ attachment styles
within the couple; however, it seems logical to deduce that the secure-secure
couple type would report the highest level of positive communication. This
study’s results suggest that it takes two secure individuals to engage in
positive, constructive communication and only one insecure individual to
increase levels of negative, destructive communication.

Implications for counseling practice

Although attachment theory concepts underlie and guide interventions
within Emotionally Focused Therapy (EFT), they can also be integrated
into couples and families counseling regardless of the clinician’s theoretical
orientation. As many clinicians and researchers have discovered, communi-
cation processes within couples and families can be changed and improved
to increase intimacy and connectedness (Johnson, 2004a, 2004b; see Cobb
& Bradbury, 2003 for a review). Since attachment and communication are
related to relationship satisfaction, therapists may use interventions based
on either or both factors to help couples and families create more positive
interactions and preventing relationship distress and dissolution.

Emotionally Focused Therapy. Bradley and Johnson (2005) reported that

Emotionally Focused Therapy (EFT) has been applied to diverse popula-
tions including older and same-sex couples. The EFT therapist views nega-
tive communication patterns, such as demand-withdraw and avoidance, as
symptoms of underlying attachment insecurity within the relationship. The
results of the current study support the link between attachment and com-
munication patterns; specifically mutual constructive conflict communica-
tion was associated with attachment security and demand-withdraw and

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

692 Journal of Social and Personal Relationships 26(5)

mutual avoidance and withholding communication were associated with

attachment insecurity. The EFT therapist seeks to increase attachment
security within the relationship by using several interventions: (i) clarifying
the process occurring during couples’ negative interactional cycles; (ii) high-
lighting emotions underlying each partner’s position in the cycle (e.g.,
sadness, numbing, fear, shame); (iii) working with each partner to compe-
tently express his/her needs and fears and to accept his/her partner’s newly
disclosed information; and (iv) helping each partner to express his/her needs
and wants. These interventions facilitate bonding events characterized by
engagement, empathy, and responsiveness, which contrast with the couple’s
experience of distress and isolation in past interactions. The couple’s attach-
ment security is increased due to gradually seeing their partner as sup-
portive and responsive and having the self-confidence and self-acceptance
necessary to voice their feelings and needs. The CPQ and ECR-R used in
this study could aid the EFT therapist in the assessment of clients’ attach-
ment styles and couples’ conflict communication patterns.

Couples therapy. Cobb and Bradbury (2003) suggest that therapists help
couples as a dyad to shape their behaviors both in conflict and non-conflict
contexts, which will lead to a revision of the partners’ models of self and
others toward attachment security. They pointed to support-seeking, care-
giving, and communication behaviors as intervention targets. For support-
seeking behaviors, couples could learn how to ask each other for help instead
of making extreme demands or denying the need for help. For caregiving
behaviors, couples could learn to respond with empathy and validation and
avoid minimization and blame. Couples’ could improve their communica-
tion by learning how to clearly communicate feelings and needs without
blaming each other or becoming defensive. Partners could also reflect and
amplify their feelings and statements to demonstrate understanding and vali-
dation. Empathy for the partner’s currently maladaptive ways of seeking
support and comfort can be increased by encouraging a discussion about
how those behaviors were adaptive within their family of origin and parent–
child relationship. Couples could explore the attributions they make about
their partners’ behaviors, which could open up their individual models of
others for revision as those attributions are contradicted.

Implications for same-sex couples. Since previous research has shown that
the roots of same-sex couples’ romantic attachment are the same as for
different-sex couples, attachment theory can be also guide work with this
population (e.g. Elizur & Mintzer, 2003; Josephson, 2003). Participants in
the same-sex couples group reported lower mutual avoidance and with-
holding communication than did participants in the different-sex couples
group. This suggests that they are able to stay engaged while communi-
cating in conflict and retain their ability to give to one another, strengths
that therapists can build upon.
There are contextual issues that may warrant attention, however, when
working with same-sex couples. Josephson (2003) asserted that, due to

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

Domingue & Mollen: Attachment and communication 693

heterosexism and internalized homophobia, adults who self-identify as gay,

lesbian, or bisexual may struggle with issues of self-acceptance, which is
linked to insecure attachment styles. Lesbian, gay, bisexual or transgender
(LBGT) adults may be still exploring and establishing their individual
identity (including self-acceptance and establishing an integrated and inclu-
sive self-concept) when they enter initial intimate relationships, because
“. . . the sense of self is constricted by a premature foreclosure on identity
exploration due to an awareness of one’s self as different from the norm”
(Josephson 2003, p. 312). This suggests that couples therapists may need to
help adults in same-sex relationships accept previously disowned parts of
themselves, and, in turn, be better able to express their feelings and needs
to their partners. Alonzo (2005) added that adults in same-sex relationships
are bombarded with social messages that their relationships are not viable
and will not last, which could lead to premature relationship dissolution.
Future research could clarify whether these additional considerations for
clinical work with LGBT clients and same-sex couples is warranted.

Family therapy. Because we found that within couples attachment security

was related to communication patterns, it seems logical to consider other
attachment relationships (e.g. parent–child) at different points in the life-
span. Especially important might be developmental transitions, such as
launching children and role reversal due to parents aging, when attachment
concerns are likely to be activated. Concepts used in couples therapy may
guide therapists’ work with families as they seek to improve communica-
tion and increase attachment security, which may in turn increase the like-
lihood of a successful life transition.
In “EFFT: Emotionally-focused family therapy”, Johnson (2004b) sug-
gested how attachment concepts can be used within families. She asserted
that the primary goals are: (i) to help the parental subsystem strengthen
the ability to serve as a secure base and safe haven for the children, which
gives them confidence as they explore the wider world before them and
decreases separation distress; and (ii) to help the adolescent seek support
and comfort in constructive ways. Interactional cycles that reveal criticism,
anger, and fears of abandonment are the target of intervention. The parent–
child relationship must be increasingly able to adapt and adjust to the
adolescent’s growing sense of individuality and autonomy, while at the
same time, conveying acceptance and belonging within the family.
Moreover, Bradley and Palmer (2003) suggested that aging triggers the
attachment system of both the middle-aged child and aging parent due to
the realization of the parent’s impending mortality and accompanying fears
of loss and final separation. An aging parent’s avoidance of conflict and
rejection of his/her caregiver as well as clingy, demanding behaviors are
seen as distorted attachment signals. The aging parent may be reenacting
coping strategies that they used earlier in life, such as rugged independence
or making excessive demands. Therapists can work with the caregivers by
educating them on the aging process, teaching them coping skills, helping
them avoid compulsive caregiving, and encouraging them to work together

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

694 Journal of Social and Personal Relationships 26(5)

with their siblings. Therapists can also help the aging parent voice their
fears and needs, which will generate empathy and caregiving from the child
caregiver.

Limitations and directions for future research

Several limitations focused on the sample and research design. The sample
was less ethnically diverse than the authors hoped, had fewer same-sex
couples than anticipated, and the different-sex group was dominated by
married couples. Based on these considerations, the findings are more
tentative and less generalizable. The statistical power of the analyses was
lower than hoped due to the sample size. Given the correlational nature of
the study; causality cannot be inferred. The design was cross-sectional and,
therefore, how changes in couples’ attachment styles would affect conflict
communication cannot be identified.
Combining self-report data with observational data from raters and
therapists could improve our understanding of the relationship between
attachment and communication. Recruiting more insecure adults may enable
future researchers to determine whether certain within-couple insecure
attachment styles relate to negative communication patterns. It is unknown
the extent to which age and relationship duration moderate the relation-
ships between attachment couple type and communication patterns.
One theoretical debate has focused on whether attachment style is an
intrapersonal or interpersonal construct. Supporters of the intrapersonal
view of attachment assert that attachment style depends heavily upon
experiences with the parent or caregiver. This approach predicts continuity
across the lifespan and relationships. Supporters of the interpersonal view
of attachment assert that attachment styles can vary across relationships
(e.g., parent–child, different romantic partners) depending on the individ-
uals’ experiences of intimacy and dependability in that specific relationship.
Collins and her colleagues (2004) found that 21% to 44% of the variability
in adults’ attachment security was explained by individual differences in
general attachment style and the remaining variance was explained by
features of the specific relationship and the couple as a unit (Collins et al.,
2004). Their analyses placed the domain-specific working models of family,
friends, and romantic relationships under an overarching general working
model of attachment relationships (Collins et al., 2004). These findings
suggest that even though general working models of attachment influence
more specific types of intimate relationships (e.g., parent, friends, and
romantic partners), one’s attachment security is largely open to the influ-
ence of specific characteristics of that particular relationship. The current
study suggests that the combination of partners’ attachment styles and
communication patterns influence one another, which suggests that their
attachment style could be modified based on their experiences within their
current relationship, thus supporting an interpersonal view of attachment.
Clinical interventions that target communication patterns and beliefs
about self and others, which underlie one’s attachment styles, have the poten-
tial to help couples and families become more comfortable depending on

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

Domingue & Mollen: Attachment and communication 695

each other, see each other as safe havens in times of distress, and commu-
nicate their feelings and needs directly. This process could hold the key to
facilitating close bonds that would resist relationship dissolution and emo-
tional estrangement.

REFERENCES

Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A
psychological study of the strange situation. Hillsdale, NJ: Lawrence Erlbaum Associates.
Alonzo, D. J. (2005). Working with same-sex couples. In M. Harway (Ed.), Handbook of
couples therapy (pp. 370–385). Hoboken, NJ: John Wiley & Sons.
Bartholomew, K. (1990). Avoidance of intimacy: An attachment perspective. Journal of Social
and Personal Relationships, 7, 147–178.
Bartholomew, K., & 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.
Bowlby, J. (1969). Attachment and loss. Vol. 1. Attachment. London: Hogarth Press.
Bowlby, J. (1973). Attachment and loss. Vol. 2. Separation: Anxiety and anger. London: Hogarth
Press.
Bowlby, J. (1980). Attachment and loss. Vol. 3. Loss: Sadness and depression. New York: Basic
Books.
Bradley, B., & Johnson, S. M. (2005). EFT: An integrative contemporary approach. In
M. Harway (Ed.), Handbook of couples therapy (pp. 179–193). Hoboken, NJ: John Wiley &
Sons.
Bradley, J. M., & Palmer, G. (2003). Attachment in later life: Implications for intervention with
older adults. In S. M. Johnson & V. E. Whiffen (Eds.), Attachment processes in couple and
family therapy (pp. 281–299). New York: Guilford Press.
Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self-report measurement of adult attach-
ment: An integrative overview. In J. A. Simpson & W. S. Rholes (Eds.), Attachment theory
and close relationships (pp. 46–76). New York: Guilford Press.
Caughlin, J. P., & Huston, T. L. (2002). A contextual analysis of the association between
demand/withdraw and marital satisfaction. Personal Relationships, 9, 95–119.
Christensen, A., & Heavey, C. L. (1993). Gender differences in marital conflict: The demand-
withdraw interaction pattern. In S. Oskamp and M. Costanzo (Eds.), Gender issues in
contemporary society (pp. 113–141). Newbury Park, CA: Sage.
Christensen, A., & Shenk, J. L. (1991). Communication, conflict, and psychological distance in
nondistressed, clinic, and divorcing couples. Journal of Consulting and Clinical Psychology,
59, 458–463.
Cobb, R. J., & Bradbury, T. N. (2003). Implications of adult attachment for preventing adverse
marital outcomes. In S. M. Johnson & V. E. Whiffen (Eds.), Attachment processes in couple
and family therapy (pp. 258–280). New York: Guilford Press.
Collins, N. L., Guichard, A. C., Ford, M. B., & Feeney, B. C. (2004). Working models of attach-
ment: New developments and emerging themes. In W. S. Rholes and J. A. Simpson (Eds.),
Adult attachment: Theory, research, and clinical implications (pp. 196–239). New York:
Guilford Press.
Eldridge, N. S., & Gilbert, L. A. (1990). Correlates of relationship satisfaction in lesbian couples.
Psychology of Women Quarterly, 14, 43–62.
Elizur, Y., & Mintzer, A. (2003). Gay males’ intimate relationship quality: The roles of attach-
ment security, gay identity, social support, and income. Personal Relationships, 10, 411–435.
Fraley, R. C. (2005). Information on the Experiences in Close Relationships – Revised
(ECR-R) adult attachment questionnaire. http://www.psych.uiuc.edu/~rcfraley/measures/
ecrr.htm (accessed 11 November 2005).

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015

696 Journal of Social and Personal Relationships 26(5)

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.
Gottman, J. M. (1999a). Repair and the core triad of balance. In J. M. Gottman, The marriage
clinic: A scientifically-based marital therapy (pp. 31–86). New York: W.W. Norton & Company.
Gottman, J. M. (1999b). The sound marital house: A theory of marriage. In J. M. Gottman, The
marriage clinic: A scientifically-based marital therapy (pp. 87–110). New York: W.W. Norton
& Company.
Hazan, C., & Shaver, P. R. (1987). Romantic love conceptualized as an attachment process.
Journal of Personality and Social Psychology, 52, 511–524.
Hazan, C., & Shaver, P. R. (1990). Love and work: An attachment-theoretical perspective.
Journal of Personality and Social Psychology, 59, 270–280.
Hazan, C., & Shaver, P. R. (1994). Attachment as an organizational framework for research on
close relationships. Psychological Inquiry, 5, 1–22.
Heavey, C. L., Christensen, A., & Malamuth, N. M. (1995). The longitudinal impact of demand
and withdrawal during marital conflict. Journal of Consulting and Clinical Psychology, 63,
797–801.
Heavey, C. L., Larson, B. M., Christensen, A., & Zumtobel, D. C. (1996). The communication
patterns questionnaire: The reliability and validity of a constructive communication sub-
scale. Journal of Marriage and the Family, 58, 796–800.
Johnson, S. M. (2003). Attachment theory: A guide for couple therapy. In S. M. Johnson and
V. E. Whiffen (Eds.), Attachment processes in couple and family therapy (pp. 103–123). New
York: Guilford Press.
Johnson, S. M. (2004a). The practice of emotionally focused couple therapy: Creating connec-
tion, 2nd edn. New York: Brunner-Routledge.
Johnson, S. M. (2004b). EFFT: Emotionally-focused family therapy: Restructuring attachment.
In S. M. Johnson, The practice of emotionally focused couple therapy: Creating connection,
2nd edn (pp. 243–265). New York: Brunner-Routledge.
Josephson, G. J. (2003). Using an attachment-based intervention with same-sex couples. In
S. M. Johnson and V. E. Whiffen (Eds.), Attachment processes in couple and family therapy
(pp. 300–317). New York: Guilford Press.
Julien, D., Chartrand, E., Simard, M., Bouthillier, D., & Bégin, J. (2003). Conflict, social support,
and relationship quality: An observational study of different-sex, gay male, and lesbian
couples’ communication. Journal of Family Psychology, 17, 419–428.
Kaslow, F., & Robison, J. A. (1996). Long-term satisfying marriages: Perceptions of contribut-
ing factors. American Journal of Family Therapy, 24, 153–168.
Klinetob, N. A., & Smith, D. A. (1996). Demand-withdraw communication in marital inter-
action: Tests of interspousal contingency and gender role hypotheses. Journal of Marriage
and the Family, 58, 945–957.
Mackey, R. A., Diemer, M. A., & O’Brien, B. A. (2004). Relational factors in understanding
satisfaction in the lasting relationships of same-sex and different-sex couples. Journal of
Homosexuality, 47, 111–136.
Noller, P., & White, A. (1990). The validity of the communication patterns questionnaire.
Psychological Assessment, 2, 478–482.
Pietromonaco, P. R., Greenwood, D., & Barrett, L. F. (2004). Conflict in adult close relationships:
An attachment perspective. In W. S. Rholes and J. A. Simpson (Eds.), Adult attachment:
Theory, research, and clinical implications (pp. 267–299). New York: Guilford Press.
Sullaway, M., & Christensen, A. (1983). Assessment of dysfunctional interaction patterns in
couples. Journal of Marriage and the Family, 45, 653–659.
Twenge, J. M., & King, L. A. (2005). A good life is a personal life: Relationship fulfillment and
work fulfillment in judgments of life quality. Journal of Research in Personality, 39, 336–353.

Downloaded from spr.sagepub.com at GEORGIAN COURT UNIV on March 24, 2015