Attachment Security in Couple Relationships:
A Systemic Model and Its Implications
for Family Dynamics
Theory and research on adult attachment
style emphasize the crucial role that the
sense of attachment security plays in the
formation and maintenance of couple re-
lationships. In the present article, we review
studies that have examined the contribu-
tion of adult attachment style to relational
cognitions, emotions, and behaviors as
well as to the formation, quality, and sta-
bility of dating and marital relationships.
We discuss some of the measurement and
design issues raised by this research.
Based on the reviewed findings, we pro-
vide an integrative, systemic theoretical
model delineating how the links between
partners attachment security and the
quality of their couple relationship occurs.
Finally, we discuss the implications of this
model for the understanding of how at-
Both authors are Professors of Psychology, Dept.
of Psychology, Bar-Ilan University and Co-Directors
of the Peleg-Bilig Center. Send correspondence
to Dr. Mikulincer, Department of Psychology, Bar-
Ilan University, Ramat Gan 52900, Israel; e-mail:
mikulm@mail.biu.ac.il // Dr. Florians e-mail: floriav@
mail.biu.ac.il
Philip A. Cowan, Ph.D., Professor, e-mail:
pcowan@Socrates.berkeley.edu and Carolyn Pape
Cowan, Adjunct Professor, e-mail: ccowan@uclink4.
berkeley.edu are affiliated with The Institute of Hu-
man Development, Department of Psychology, Uni-
versity of California, Berkeley CA.
405
Family Process, Vol. 41, No. 3, 2002 FPI, Inc.
MARIO MIKULINCER, Ph.D.
VICTOR FLORIAN, Ph.D.
PHILIP A. COWAN, Ph.D.
CAROLYN PAPE COWAN, Ph.D.
tachment style and couple relationships
combine to affect the family system in gen-
eral, and parent-child relationships and
childrens developmental outcomes, in
particular.
Fam Proc 41:405 434, 2002
I N the last two decades, the study of
couple relationships and marital sat-
isfaction has received ample attention in
the professional literature. Several stud-
ies have attempted to identify the major
factors that may contribute to the quality
of these relationships. One of the most
promising and examined factors is the
pattern of the individuals attachment or-
ganization. Specifically, the sense of at-
tachment security has been identified as a
major variable explaining variations in
the quality of dating and marital relation-
ships (see Feeney, 1999a, Shaver &
Hazan, 1993, for reviews). In this article,
we review relevant published material on
the association between the sense of at-
tachment security and quality of couple
relationships. Then we present a systemic
model, delineating the intervening paths
between these two constructs and the re-
ciprocal influences between the two part-
406 / FAMILY PROCESS
ners, and extending it to other aspects of
the family system.
THE SENSE OF ATTACHMENT SECURITY
According to Bowlbys (1973) theory, in-
teractions with significant others who are
available and supportive in times of
stress facilitate the formation of a sense of
attachment security. Waters, Rodrigues,
and Ridgeway (1998) viewed this sense as
a set of expectations about others avail-
ability and responsiveness in times of
stress, which are organized around a ba-
sic prototype or script. This script seems
to include the following if-then proposi-
tions: If I encounter an obstacle and/or
become distressed, I can approach a sig-
nificant other for help; I am a person wor-
thy of receiving help; he or she is likely
to be available and supportive; I will ex-
perience relief and comfort as a result of
proximity to this person; I can then return
to other activities. In Bowlbys (1973)
terms, the sense of attachment security
provides an individual with a framework
for maintaining wellbeing, formulating
effective emotion-regulation devices, de-
veloping positive models of self and oth-
ers, and engaging in exploration, affilia-
tion, and caregiving activities.
Although the sense of attachment secu-
rity may be formed during early interac-
tions with primary caregivers, Bowlby
(1988) contended that every meaning-
ful interaction with significant others
throughout life may affect beliefs about
others availability and supportiveness.
Moreover, although the sense of attach-
ment security may be quite general, it is
also common for people to develop rela-
tionship-specific beliefs organized around
actual experiences with a specific partner.
These beliefs do not necessarily fit with
the global sense of attachment security
and may be influenced by the quality of
the specific relationship (Collins & Read,
1994). In fact, like every mental represen-
tation, the sense of attachment security
can be contextually activated by actual or
imagined encounters with available fig-
ures, even in persons who have global
doubts about others goodwill (Baldwin,
Keelan, Fehr, et al., 1996).
In the last 15 years, numerous studies
have examined the sense of attachment
security in adulthood. The most fre-
quently used strategy is to examine as-
sociations between the global sense of
attachment security and theoretically rel-
evant constructs. Specifically, these stud-
ies have focused on a persons attachment
style and compared persons whose re-
ports suggest a secure style with those
whose reports suggest more insecure
styles. This line of research has been
guided by two different conceptual and
methodological approaches (Hazan &
Shaver, 1987; Main, Kaplan, & Cassidy,
1985) that use different assessment tech-
niques and tap different aspects of attach-
ment style. Main et al.s (1985) approach
is based on a developmental perspective
and assesses early attachment to parents
through an intensive, reliable, and well-
validated interview (the Adult Attach-
ment InterviewAAI; George, Kaplan, &
Main, 1985) that demands complex and
skillful content and stylistic interpreta-
tion of narrative accounts of relationship
qualities. Hazan and Shavers (1987) ap-
proach is based on a personality and so-
cial psychology perspective and assesses
current attachment orientations to signif-
icant others (not only parents but also
romantic partners) through self-report
measures that have been found to be par-
simonious and psychometrically sound.
Recent advances in the conceptualiza-
tion and assessment of adult attachment
style indicate that this relational con-
struct seems to be organized around two
underlying dimensions (Brennan, Clark,
& Shaver, 1998). The first dimension, typ-
ically called avoidance, reflects the ex-
tent to which people distrust others good-
will and strive to maintain emotional dis-
MIKULINCER, FLORIAN, P. COWAN, & C. COWAN
tance and remain independent from a
relationship partner. The second dimen-
sion, typically called anxiety, reflects
the degree to which people worry that a
partner might not be available or support-
ive in times of need. Persons scoring low
on these two dimensions exhibit the se-
cure style and are characterized by a pos-
itive history of attachment interactions
and a global sense of attachment security.
Studies using both the AAI and self-
report measures of attachment style have
generally supported Bowlbys hypotheses
about the psychological correlates of the
sense of attachment security. First, per-
sons having a sense of attachment secu-
rity tend to react to stressful events with
lower levels of distress than persons who
score high on avoidance or anxiety dimen-
sions (B.C. Feeney & Kirkpatrick, 1996;
Mikulincer & Florian, 2001). Second, per-
sons who hold a sense of attachment se-
curity are more likely to cope with stress
by relying on support-seeking than do
persons who score high on avoidance or
anxiety dimensions (Fraley & Shaver,
1998; Simpson, Rholes, & Nelligan, 1992).
Third, securely attached persons hold
more positive expectations about relation-
ship partners than persons who score
high on the avoidance dimension (Collins,
1996; Collins & Read, 1990). Fourth, se-
curely attached persons hold more posi-
tive self-views than persons who score
high on the anxiety dimension (Bar-
tholomew & Horowitz, 1991; Mikulincer,
1998). Fifth, persons who hold a sense of
attachment security are more likely to en-
gage in exploration and affiliation activi-
ties, and to be more sensitive and respon-
sive to their partners needs than persons
scoring high on avoidance or anxiety dimen-
sions (Feeney, 1996; Mikulincer, 1997;
Mikulincer & Selinger, 2001).
In his writings, Bowlby (1979) also em-
phasized the possible implications of at-
tachment security for couple relation-
ships. In his own words, there is a strong
Fam. Proc., Vol. 41, Fall, 2002
/ 407
causal relationship between an individu-
als experiences with his parents and his
later capacity to make affectional bonds
(p. 135). In particular, Bowlby (1979)
highlighted marriage as the affectional
bond in which the influence of attachment
history is most likely to be manifested.
Following this theoretical formulation,
studies have attempted to test empiri-
cally whether relatively enduring differ-
ences in attachment style would be man-
ifested in the quality of adult couple rela-
tionships and marriage.
In the next two sections, we (1) outline
briefly some of the methodological issues
involved in examining links between at-
tachment style and couple relationship
qualities, and (2) review the existing rel-
evant studies. Based on the findings, we
then present a systemic model to suggest
how the links between attachment and
marital quality occur. Moreover, we dis-
cuss the implications of our formulation
for the understanding of how adult at-
tachment and marital relationships com-
bine to affect parent-child relationships
and childrens developmental outcomes
an endeavor that places attachment in
the context of the family system and gives
a central role to couple relationships as a
potential mechanism in the transmission
of attachment relationships across the
generations.
METHODOLOGICAL ISSUES
The Assessment of Attachment
As we noted above, two major methods
of measuring attachment style in adult-
hood were developed in the 1980s. Main
and her colleagues (Main & Goldwyn,
1994; George, Kaplan, & Main, 1985)
have relied on a narrative approach to
elicit working models of attachment.
The 60- to 90-minute AAI (see Hesse,
1999, for a description) asks interviewees
to choose 5 adjectives to describe their
relationships with mother and father, to
408 / FAMILY PROCESS
supply anecdotes illustrating why they
characterized each relationship with those
adjectives, to speculate about why their
parents behaved as they did, and to de-
scribe change over time in the quality of
their relationships with parents.
Tracing their assumptions back to Ains-
worths early formulations describing in-
fants attachments (Ainsworth, Blehar,
Waters, & Wall, 1978), the creators of the
AAI assumed that attachment patterns
were best conceptualized as categories or
types. Coding of the AAI is based on an
analysis of 5 continuous scales intended
to capture the quality of early experi-
ences, separately with mother and with
father (e.g., loving, rejecting), and on 12
scales that describe a persons current
state of mind regarding those experiences
(e.g., derogation of attachment, coherence
of the narrative). Based on a configura-
tional analysis of these scales, which are
thought to represent dominant discourse
strategies (Main & Goldwyn, 1996), AAI
narratives are coded as indicative of ei-
ther secure, insecure-dismissing, or inse-
cure-preoccupied working models of at-
tachment.*
The second main method of measuring
adult attachment differs in three impor-
tant ways from the AAI. First, the data
come primarily from questionnaires (but
see Scharfe & Bartholomew, 1995, for a
comparison of questionnaire and inter-
view methods). Second, the aim of these
questionnaires is not to examine working
models of early parent-child relation-
ships, but rather to assess styles of at-
tachment in adult close relationships.
Third, items in these questionnaires focus
explicitly on whether the self is worthy of
love and whether the other will be avail-
* Two additional categoriesUnresolved as a re-
sult of loss and Cannot Classifyare part of the
system, but the three primary categories represent
the major alternative strategies for dealing with
threats to attachment relationships.
able for support when the need arises.
Initially, these questionnaires adopted a
categorical approach (Hazan & Shaver,
1987), asking individuals to choose among
three brief prototypical descriptions of Se-
cure, Avoidant, and Preoccupied attach-
ment in adult intimate relationships.
However, subsequent versions decon-
structed the paragraphs into sentences
(Collins & Read, 1990) and computed con-
tinuous attachment scores (e.g., Fraley,
Waller, & Brennan, 2000).
An enduring concern, which has not
been given the attention it deserves, is
that there is not always correspondence
between the narrative and questionnaire
methods, or the categorical versus contin-
uous scoring of questionnaires. Because the
more intensive Adult Attachment Inter-
view focuses on early parent-child rela-
tionships and the various questionnaire
attachment style measures focus on adult-
adult intimate relationships, it should not
be surprising to find that the overlap is
quite low at best (Shaver, Belsky, & Bren-
nan, 2000; Cowan & Cowan, 2001). Com-
paring the self-classification of attachment
categories using the paragraph method
and the classification based on dimen-
sional analysis of questionnaire responses,
Brennan et al. (1998) found highly statis-
tically significant results; yet, nearly half
of the participants classified as secure on
one measure were classified as insecure
on the other. As we shall see, despite the
fact that there is often low agreement be-
tween methods of measuring attachment,
different studies using different methods
tend to produce similar trends concerning
the connections between partners attach-
ment and marital quality.
A little-noted but important fact is that
most of the items used to assess attach-
ment are phrased in general terms (I am
somewhat uncomfortable being close to
others), but some investigators re-write
the items to focus on specific relationships
(I am somewhat uncomfortable being
MIKULINCER, FLORIAN, P. COWAN, & C. COWAN
close to my partner). The two versions
tend to be significantly correlated in the .3
to .5 range (Cowan & Cowan, 2001). This
modest correspondence leaves room for
the suggestive findings described below
revealing that self-reported attachment
style with ones partner shows higher cor-
relations with marital satisfaction than
self-reported general attachment style in
relation to unspecified others.
Measurement of Marital Quality
Marital quality is measured in different
ways in different studies. It is not prob-
lematic that different studies use differ-
ent questionnaires assessing marital sat-
isfaction, since they all tend to be very
highly correlated (Gottman, 1993). What
is more at issue is that most researchers
rely on an individuals self-report of the
quality of his or her intimate relationship,
while a few base their conclusions on ob-
servations of marital interaction, often in
laboratory settings. Although significant
correlations between self-reports and ob-
servations are consistently found, it is
rare that one measure explains more than
25% of the variance in the other (cf. Lev-
enson & Gottman, 1983).
Determining Sequence and
Causal Connections
The question at the heart of this article
is whether attachment patterns can be
described as antecedent to marital qual-
ity, or, more strongly, as playing a causal
role in partners ability to establish a pos-
itive couple relationship. As we shall see,
this question is difficult to answer from
the data that are presently available. One
obstacle to making causal inferences is
that most studies of attachment style and
couple relationship not only assess both
constructs at the same time, but, as we
have indicated, both kinds of data are ob-
tained from the same person. Conclusions
about the linkage between the two are
then confounded by the information
Fam. Proc., Vol. 41, Fall, 2002
/ 409
sourcea problem that Bank, Dishion,
Skinner, and Patterson (1990) describe
colorfully as glop. Our review will show
that this does not pose an insurmountable
problem, because studies with indepen-
dent sources of attachment and couple re-
lationship data support the hypothesis
that the two are functionally related.
More difficult to deal with is the nature
of the functional relationship. Many in-
vestigators assume that longitudinal de-
signs will solve the problem, reasoning
that if attachment measured at Time 1
predicts couple relationship quality as-
sessed at Time 2, we can determine the
direction of effects. But as two of us have
noted elsewhere (Cowan & Cowan, 2002),
it is possible that earlier couple relation-
ship qualities produce the Time 1 attach-
ment results. That is, causal hypotheses
can be supported but not proven by lon-
gitudinal designs. What we need, then,
are experimental and quasi-experimental
longitudinal designs in which earlier in-
tervention-induced changes in attachment
style result in later changes in relation-
ship quality, or vice versa. Having raised
a number of methodological concerns, we
now review findings on attachment and
couple relationship.
A REVIEW OF EMPIRICAL FINDINGS
Relationship Expectations and Beliefs
Adult attachment studies have pro-
vided important information on the asso-
ciation between the sense of attachment
security and positive beliefs about couple
relationships. For example, Hazan and
Shaver (1987) found that participants
who classified themselves as securely at-
tached scored higher than insecure per-
sons in beliefs about (a) the existence of
romantic love and (b) the possibility that,
although romantic feelings wax and
wane, they may reach the intensity expe-
rienced at the start of the relationship,
and in some cases, never fade. Subse-
410 / FAMILY PROCESS
quent studies have reported that, as com-
pared to more insecure people, securely
attached people held more optimistic be-
liefs about love relationships and mar-
riage (Carnelley & Janoff-Bulman, 1992;
Pietromonaco & Carnelley, 1994; Whit-
aker, Beach, Etherton, et al., 1999), were
more likely to use a positive frame in
thinking about couple relationships (Boon
& Griffin, 1996), and were less likely to
evaluate negative relational outcomes
(Feeney & Noller, 1992) and to endorse
dysfunctional beliefs about couple rela-
tionships (Whisman & Allan, 1996).
The formation and maintenance of
long-term romantic relationships: Attach-
ment studies have consistently reported
that persons differing in attachment style
vary in (a) the likelihood of being involved
in long-term couple relationships, and (b)
the vulnerability of these relationships to
disruption. For example, more securely
attached persons have been found among
seriously committed dating couples or
married couples than in samples of single
individuals (e.g., Kirkpatrick & Davis,
1994; Kobak & Hazan, 1991; Senchak &
Leonard, 1992). Accordingly, Hill, Young,
and Nord (1994) found that persons who
reported a secure attachment style were
more likely to attain marriage/cohabita-
tion and less likely to experience divorce
than insecure persons.
There is also extensive evidence that
secure persons have more stable dating
relationships than insecure persons
(Feeney & Noller, 1990, 1992; Hazan &
Shaver, 1987; Kirkpatrick & Davis, 1994;
Kirkpatrick & Hazan, 1994; Shaver &
Brennan, 1992). This finding was repli-
cated in both cross-sectional and longitu-
dinal studies. For example, Kirkpatrick
and Hazan (1994) found that secure per-
sons relationships were more likely to be
intact after 4 years than insecure persons
relationships. In contrast, avoidant per-
sons were more likely 4 years later to be
seeing more than one person or to be
not seeing anyone and not looking,
whereas anxiously attached persons were
most likely to indicate that they were not
seeing anyone and looking for a partner.
Conceptually similar findings were
found by Klohnen and Bera (1998) among
women who participated in a 31-year lon-
gitudinal personality study. At ages 21,
27, 43, and 52, information was collected
about their marital status. Participants
also provided information about their
commitment to marriage at age 21, mar-
ital tensions at age 27, relationship satis-
faction at age 43 and 52, and attachment
style at age 52. Women who endorsed a
secure attachment style at age 52 showed
a different relationship trajectory from
women with an insecure attachment style
beginning as early as age 21. First, se-
curely attached women were more likely
to be married at age 52 and reported
higher relationship satisfaction than
women who endorsed an insecure style at
the same age. Second, securely attached
mid-life women had reported higher com-
mitment to getting married and starting a
family at age 21 than insecurely attached
women and this early difference seemed
to have come true 6 years later at age
27, when secure women were more likely
to be married and report fewer marital
tensions than women who endorsed an
insecure style. However, one should be
aware of the retrospective nature of this
study as well as of the possibility that
variations in relationship trajectory
might have affected womens attachment
style at age 52.
In one of the only studies of attachment
and marital stability using the narrative
approach, Crowell and Treboux (2001) as-
sessed 146 premarital dating couples with
the AAI and with another interview fo-
cused specifically on their relationship as
a couplethe Couple Relationship Inter-
view (CRI). Five years later, they found
that the AAI did not predict marital
breakup, but couples in which both part-
MIKULINCER, FLORIAN, P. COWAN, & C. COWAN
ners were categorized as insecurely at-
tached on the CRI were more likely to
have separated or divorced.
The sense of attachment security has
also been found to be inversely associated
with problems in relationship formation
and maintenance (Bartholomew & Horo-
witz, 1991; Bookwala & Zdaniuk, 1998;
Doi & Thelen, 1993; McCarthy & Taylor,
1999; Thelen, Sherman, & Borst, 1998).
For example, Bartholomew and Horowitz
(1991) assessed maladaptive interper-
sonal behavior, as measured by the In-
ventory of Interpersonal Problems, and
found that securely attached persons did
not score extremely high in any of the
problem scales. In contrast, whereas at-
tachment anxiety was positively related
to scores of hard to be sociable, hard to
be submissive, too responsible, and too
controlling, attachment avoidance was
positively correlated with scores of hard
to be intimate. These findings were rep-
licated in self-reports and friend-reports.
Overall, the findings consistently show
that secure persons, as compared to inse-
cure persons, (a) are more likely to be
involved in long-term couple relation-
ships, (b) have more stable couple rela-
tionships, and (c) suffer from fewer diffi-
culties and/or disruptions in the relation-
ship. The few longitudinal studies suggest
that attachment security antedates indi-
ces of marital stability, but it is too early
to claim that individuals attachment se-
curity plays a direct role in whether cou-
ples stay together or break up.
Attachment security as a mate selection
standard: Another relevant line of re-
search has focused on mating preference
and claimed that a person seeking to form
a long-term couple relationship would
prefer to mate with securely attached
partners, because they hold a positive ori-
entation towards this type of relationship.
In support of this view, Pietromonaco and
Carnelley (1994) and Chappell and Davis
(1998) found that participants, regardless
Fam. Proc., Vol. 41, Fall, 2002
/ 411
of their own attachment style, reported
more positive emotions and less negative
emotions when imagining a relationship
with a secure rather than an insecure
partner. Accordingly, Baldwin et al.
(1996, Study 3), Frazier, Byer, Fischer, et
al. (1996), and Latty-Mann and Davis
(1996) constructed vignettes of potential
partners differing in their attachment ori-
entations and found that secure partners
were preferred over insecure partners.
The Quality of Dating Relationships
The bulk of relevant data has been re-
ported by studies that have focused on the
association between attachment security
and quality of dating relationships. Spe-
cifically, these studies have tested the hy-
pothesis that secure attachment would be
linked to the formation of satisfactory
dating relationships, which are character-
ized by emotional involvement, intimacy,
commitment, trust, friendly communica-
tion patterns, and caring.
The hypothesized positive association
between the sense of attachment security
and satisfaction with dating relationships
has been consistently documented in sev-
eral cross-sectional studies using differ-
ent measures of attachment style (e.g.,
forced-choice tripartite categorization,
Adult Attachment Interview, Relation-
ship Questionnaire) and different scales
of relationship satisfaction (e.g., Relation-
ship Rating Form, Dyadic Adjustment
Scale, Marital Satisfaction Index). All
these studies have found that attachment
security is significantly correlated with
relationship satisfaction, with securely
attached persons reporting the highest
level of satisfaction and anxiously at-
tached persons reporting the lowest level
(for details of these studies in Table 1, see
Appendix). Generally, this association
was found in both men and women and
has been replicated in prospective longi-
tudinal studies (see Table 1). Moreover,
some studies have found that the associ-
412 / FAMILY PROCESS
ation between secure attachment and sat-
isfaction with dating relationships cannot
be explained by other personality factors,
such as the big five factors, depression,
dysfunctional beliefs, self-esteem, and
sex-role orientation (e.g., Carnelley, Pie-
tromonaco, & Jaffe, 1994; Jones & Cun-
ningham, 1996; Shaver & Brennan, 1992;
Whisman & Allan, 1996).
Using both global and relationship-spe-
cific measures of attachment orientation,
Cozzarelli, Hoekstra, and Bylsma (2000)
found a significant positive association
between reports of secure attachment
within a specific current dating relation-
ship and reports of satisfaction with that
relationship. Unexpectedly, reports of
global attachment style in close relation-
ships were not significantly related to re-
ported satisfaction with a specific dating
relationship. The same pattern has also
been found in a study of married couples
(Cowan & Cowan, 2001). It seems that
relationship-specific secure attachment is
more relevant to explain satisfaction with
a couple relationship than is a global mea-
sure of attachment security.
The sense of attachment security has
also been found to contribute to other ba-
sic characteristics of dating relationships.
For example, significant positive associa-
tions have been found between reports of
secure attachment and several measures
of involvement and interdependence in
dating relationships (e.g., Rubins Love
scale, Dependency scale, Self-disclosure
scale, Relationship Rating Form) in a
number of cross-sectional (Collins &
Read, 1990; Feeney, 1999b; Feeney & Nol-
ler, 1991; Feeney, Noller, & Patty, 1993;
Hazan & Shaver, 1987; Hendrick & Hen-
drick, 1989; Levy & Davis, 1988; Miku-
lincer & Erev, 1991) and longitudinal
studies (e.g., Kirkpatrick & Davis, 1994;
Kirkpatrick & Hazan, 1994; Simpson,
1990). Accordingly, ratings of attachment
security were significantly associated
with greater commitment to a dating re-
lationship, and ratings of attachment
avoidance were significantly associated
with lower levels of commitment (Hen-
drick & Hendrick, 1989; Kirkpatrick &
Davis, 1994; Kirkpatrick & Hazan, 1994;
Levy & Davis, 1988; Mikulincer & Erev,
1991; Pistole, Clark, & Tubbs, 1995; Pis-
tole & Vocaturo, 1999; Shaver & Bren-
nan, 1992; Simpson, 1990; Tucker &
Anders, 1999).
In a study of the quality of dating rela-
tionships over a period of 4 months,
Keelan, Dion, and Dion (1994) found that
securely attached persons maintained
high stable levels of commitment and
trust in a dating relationship during the
followup period. In contrast, insecure per-
sons exhibited a decrease of commitment
and trust over the same period. Moreover,
secure persons reported a constant low
level of perceived relationship costs (how
much one invested in the relationship)
over the 4-month period, whereas inse-
cure persons showed increases in such
perceived costs over time. The findings
imply that the relationship commitment
of insecure persons may deteriorate over
time and that time may exacerbate initial
differences in relationship commitment
between attachment groups.
Persons differing in attachment style
have been also found to differ in the qual-
ity of their communication pattern with a
dating partner. For example, Fitzpatrick,
Fey, Segrin, and Schiff (1993) found that
self-reports of secure attachment style
were related to higher reported levels of
positive mutual patterns of communica-
tion and lower levels of demanding and
withdrawal patterns. Accordingly, Guer-
rero (1996) videotaped dating couples while
discussing important personal problems
and found that securely attached persons
scored higher than avoidant persons in
measures of trust-receptivity, gaze, facial
pleasantness, vocal pleasantness, general
interest in the conversation, and atten-
tiveness to partners speech while dis-
MIKULINCER, FLORIAN, P. COWAN, & C. COWAN
cussing problems with their partners. In
addition, secure people scored lower in
vocal and physical signs of distress than
anxiously attached people.
In the same vein, Tucker and Anders
(1998) videotaped dating couples while
discussing positive aspects of their rela-
tionships and found that persons with a
more secure attachment style tended to
laugh more, touch their partner more,
gaze more, and smile more during the
interaction than insecure persons. Ac-
cordingly, secure persons were rated as
significantly more nonverbally expressive
and appeared to be experiencing more en-
joyment than insecure persons. At a dy-
adic level, couples in which both partners
were securely attached were rated as ex-
periencing more enjoyment during the
conversation than couples in which at
least one partner was insecurely at-
tached.
Importantly, similar patterns of find-
ings have been found in dyadic studies
that examined the effects of a persons
secure attachment on his or her dating
partners reports of relationship satisfac-
tion and quality (Collins & Read, 1990;
Jones & Cunningham, 1996; Kirkpatrick
& Davis, 1994; Mikulincer & Erev, 1991;
Scharfe & Bartholomew, 1995; Shaver &
Brennan, 1992; Simpson, 1990). In these
studies, the two partners of dating cou-
ples completed adult attachment scales
and reported on their satisfaction with,
and appraisal of the dating relationship.
Generally, a persons secure attachment
style was significantly associated with the
partners reports of relationship satisfac-
tion and quality (e.g., intimacy, commit-
ment). However, this dyadic effect was
stronger and more consistent for womens
than mens secure attachment. In addi-
tion, both partners sense of attachment
security made a significant contribution
to their joint relationship satisfaction. In
fact, both partners were dissatisfied when
at least one of the partners scored high on
Fam. Proc., Vol. 41, Fall, 2002
/ 413
attachment anxiety or avoidance. Only
one study (Whisman & Allan, 1996) found
that a persons attachment style did not
significantly predict the partners satis-
faction.
Despite the strong evidence of associa-
tion between security of attachment and
relationship quality in dating couples, our
cautionary notes at the beginning of this
article indicate that we cannot infer cau-
sality from correlational data. For exam-
ple, the finding that secure persons have
partners who report high levels of satis-
faction may equally reflect the possibility
that (a) the behavior and attitudes of se-
cure persons reinforce their partners sat-
isfaction, (b) their partners high levels of
satisfaction lead participants to feel more
securely attached in the relationship, and
(c) secure persons choose partners who
are able or willing to maintain long last-
ing satisfactory relationships. Given the
ambiguity here, there are two alternative
courses of action with regard to the for-
mation of theoretical models. One is to
wait until intervention studies establish
the direction of effects. A second alterna-
tive is to question the linear causal
premise and wonder instead whether the
linkage is bidirectional, with attachment
and relationship quality involved in circu-
lar patterns of influence, as family system
theories suggest (e.g., Wagner & Reiss,
1995).
The Quality of Marital Relationships
Studies of married couples have also
provided strong supportive evidence on
the link between attachment security and
relationship satisfaction (see Table 2 in
the Appendix). In a study of newlywed
couples, Senchak and Leonard (1992)
found that secure couples (both partners
described themselves as securely at-
tached) reported higher marital satisfac-
tion and intimacy than mixed (one spouse
chose the secure description and the other
defined himself or herself as insecure)
414 / FAMILY PROCESS
and insecure couples (both partners de-
scribed themselves as insecurely at-
tached). No significant difference was
found between mixed and insecure cou-
ples, implying that the attachment inse-
curity of one spouse may have an overrid-
ing influence on the quality of the mar-
riage. However, because the couple was
used as the unit of analysis in this study,
the actual effects of each partners attach-
ment style on relationship quality re-
mained unassessed.
In dealing with this problem, Feeney
(1994) and Feeney, Noller, and Callan
(1994) analyzed the effects of a persons
attachment security on his or her own
reports of marital satisfaction as well as
on his or her partners reports after 12
and 21 months of marriage. Findings in-
dicated that a persons attachment secu-
rity was significantly associated with both
partners reports of high marital satisfac-
tion. These findings have been replicated
and extended in a number of subsequent
studies (e.g., Davila, Bradbury, & Fincham,
1998; Feeney, 1999c; Lussier, Sabourin, &
Turgeon, 1997; Mikulincer, Horesh, Levy-
Shiff, et al., 1998). Importantly, Davila,
Karney, and Bradbury (1999) replicated
these findings at five points of measure-
ments during 3 years (every 6 months) in
a sample of newlywed couples. In addi-
tion, they reported that changes in hus-
bands and wives reports of secure at-
tachment predicted concurrent changes
in the persons own and partners reports
of marital satisfaction.
Studies of marriage have also linked
the sense of attachment security with more
marital intimacy (Mayseless, Sharabany,
& Sagi, 1997), less marital ambivalence
(Volling, Notaro, & Larsen, 1998), and
more positive climate within the marriage
(Diehl, Elnick, Bourbeau, & Labouvie-
Vief, 1998). Furthermore, Mikulincer and
Florian (1999) found significant associa-
tions between spouses attachment style
and their reports of marital cohesion and
adaptability (FACES III). Whereas
spouses who endorsed a secure style re-
ported relatively high family cohesion and
adaptability, spouses who endorsed an
avoidant style reported relatively low lev-
els in these two dimensions, and spouses
who endorsed an anxious attachment
style reported high family cohesion but
low family adaptability. Attachment secu-
rity has been also found to be associated
with positive and constructive marital
patterns of communication (Feeney, 1994;
Feeney et al., 1994). Specifically, both
wives and husbands reports of secure at-
tachment were related to more satisfac-
tion, disclosure, and involvement in vid-
eotaped marital interactions as well as to
more mutual and less coercive patterns of
communication during these interactions.
In addition, secure spouses were more ac-
curate than insecure spouses in the non-
verbal communication of neutral and neg-
ative message. Importantly, these findings
were also found when communication pat-
terns and communication accuracy were
measured nine months after the assess-
ment of attachment style.
Two studies found positive associations
between attachment security and quality
of videotaped married couple interac-
tions, one using the Adult Attachment In-
terview, the other using a self-report Q-
sort method. With attachment coded from
the AAI, Cohn, Silver, C.P. Cowan, et al.
(1992) found that, although the attach-
ment style classification was not signifi-
cantly related to self-reports of marital
satisfaction, it was significantly associ-
ated with observers ratings of couple in-
teractions in a laboratory setting. Specif-
ically, husbands classified as secure on
the AAI showed more positive and harmo-
nious interactions with their wives than
husbands classified as insecure. Though
wives attachment classification was not
directly related to the quality of marital
interaction, Cohn, Silver, et al. (1992)
concluded that the potential detrimental
MIKULINCER, FLORIAN, P. COWAN, & C. COWAN
effect of wives attachment insecurity was
buffered by husbands attachment secu-
rity. Insecurely attached women married
to securely attached men had more har-
monious interactions than did insecure
women married to insecure men. In an-
other study, using an 84-item Q sort of
attachment completed by each spouse,
Kobak and Hazan (1991) examined the
role that relationship-specific attachment
representations play during problem-
solving and confiding (sharing a mar-
riage-related disappointment with the
partner) interactions. Findings revealed
that husbands who held a more secure
representation of marriage were less re-
jecting and more supportive during a
problem-solving interaction than insecure
husbands. Secure wives were less likely
than insecure wives to be rejected by their
husbands in a confiding task.
A SYSTEMIC THEORETICAL MODEL
The reviewed data clearly indicate that
the sense of attachment security is asso-
ciated with (a) positive beliefs about cou-
ple relationships, (b) the formation of
more stable couple relationships, (c) sat-
isfaction with dating relationships and
marriage, (d) high levels of intimacy, com-
mitment, and emotional involvement
within the relationship, and (e) positive
patterns of communication and interac-
tions in both dating and married couples.
On this basis, one may wonder why and
how this relational construct is so rele-
vant to couple relationships. In the next
paragraphs, we provide a systemic model
that delineates the role of secure attach-
ment in couple relationships (see the Fig-
ure).
Our analysis indicates that three main
paths may underlie the association be-
tween a sense of attachment security and
the formation and maintenance of stable
and satisfying couple relationships. First,
the affective consequences of secure at-
tachment interactions with a significant
Fam. Proc., Vol. 41, Fall, 2002
/ 415
other distress alleviation due to the
maintenance of proximity to attachment
figureswould lead to a positive orienta-
tion toward togetherness and foster the
organization of interaction goals around
the pursuit of intimacy and closeness,
which, in turn, would encourage involve-
ment in long-lasting couple relationships.
Second, the positive mental representa-
tions of self and others that characterize
the sense of attachment security would
foster the development of a cognitive-af-
fective framework for the management of
conflict and thus for maintaining satisfy-
ing couple relationships. Third, the sense
of attachment security would facilitate
the satisfaction of other basic psychologi-
cal needs (e.g., exploration, affiliation,
caregiving) within the couple relation-
ship, which, in turn, would further in-
crease relationship satisfaction.
As can be seen in the Figure, our pro-
posed model is derived from a systemic
theoretical framework and fulfills the four
criteria delineated by family system the-
orists (Wagner & Reiss, 1995). First, in-
teractions between people are seen by ob-
servers and also by family members as
patterned, with regularities that permit
rules to be inferred. In our model, marital
interactions are patterned along intrapsy-
chic and interpersonal regularities re-
lated to each partners sense of attach-
ment security. The arrows that connect
the different components of the model
represent these regularities.
Second, the whole is greater than the
sum of its parts. This means that the
structure or organization of the intrapsy-
chic and interpersonal elements in the
whole system affects how any one element
interacts with any other element. In our
model, the association between attach-
ment security in one individual and his or
her marital cognitions and behaviors de-
pends in part on the attachment security
of both partners. Accordingly, we propose
that the partners influence one another in
416 / FAMILY PROCESS

FIGURE. A Systemic model of attachment security in couple relationships.

complex, reciprocal and even cross-con-
struct ways. For example, sense of attach-
ment security in one partner can facilitate
the acceptance of autonomy needs in the
other, which, in turn, can foster that part-
ners sense of attachment security. The
bi-directional arrows connecting the wifes
and the husbands diagrams represent
this reciprocal influence (see Figure).
Third, intrapsychic and interpersonal
influences are circular rather than linear;
it is difficult, if not impossible to tell
where the first cause of any behavior
lies. In the exposition of the model pre-
MIKULINCER, FLORIAN, P. COWAN, & C. COWAN
sented in the Figure, we mainly discuss
the path from attachment to intervening
mechanisms, to the quality of couple in-
teraction. This line seems reasonable in
the developmental lifespan sense that
children develop secure or insecure at-
tachments long before they become in-
volved in intimate couple relationships.
But as Bowlby (1973) implied early on,
the formation and maintenance of attach-
ment is a lifelong dynamic process in
which real relationships (marriage, ther-
apy) sometimes alter peoples schemas
and expectations. Thus, we have drawn
two-headed arrows in the Figure to ac-
knowledge that the connection between
attachment and couple relationship qual-
ity has a chicken and egg quality be-
cause we do not always know which
comes first.
Fourth, family systems are self-regulat-
ing. The domains of the family are dy-
namically interconnected in the sense
that changes in any aspect of the system
can lead to changes in other aspects. At-
tachment insecurity in one family mem-
ber is likely to have ripple effects through-
out the entire family system. Accordingly,
changes that occur in some other aspects
of the family system (e.g., parent-child
relationship) can alter some aspect of the
association between secure attachment
and couple relationship quality. We view
this property as an extension of our model
to the family system and it is represented
by the insertion of the partners diagrams
within a larger family system framework.
Affect Regulation and Interaction Goals
Bowlby (1988) assumed that having a
sense of attachment security reflects a
history of interactions with supportive
and loving others who bring comfort and
relief in times of stress. One implication of
this assumption is that during these pos-
itive interactions, securely attached per-
sons might have learned that proximity
maintenance is rewarding, that they can
Fam. Proc., Vol. 41, Fall, 2002
/ 417
rely on attachment behaviors as an effec-
tive means of affect regulation, and that
they could organize interpersonal behav-
iors around the basic goal of the attach-
ment systemproximity maintenance. As
a result, these persons would be prone to
forming close relationships, and would be
particularly ready to search for intimacy
and interdependence in such relation-
ships. Accordingly, they would put em-
phasis on the benefits of being together
with a romantic partner, be more likely to
dismiss potential relationship threats and
wounds, and organize their interaction
goals around the attainment and mainte-
nance of intimacy and closeness. In this
way, attachment security would enhance
a persons motivation to be involved in
long-lasting stable couple relationships.
This path is in line with Kirpatricks
(1998) contention that attachment secu-
rity reflects a positive orientation toward
long-term mating strategies.
Insecure persons experiences with non-
responsive others teach them that attach-
ment behaviors are painful and that other
interaction goals and behaviors should be
developed as defenses against the distress
caused by attachment experiences (Bowlby,
1988). In response to this distress, anx-
iously attached persons seem to construe
their interaction goals around the hyper-
activation of the attachment system and
the unfulfilled need for security. There-
fore, these persons would view couple re-
lationships and partners as a means for
achieving felt security via clinging and
hypervigilant responses. Accordingly, al-
though they would desire intimate rela-
tionships, their tendency to hyperactivate
the attachment system may lead them to
feel a chronic sense of frustration and dis-
satisfaction due to their unfulfilled needs
for demonstrations of love and security.
That is, this anxious form of insecurity
would also motivate a move toward an
attachment figure, but with a different
approach than that of a person with a
418 / FAMILY PROCESS
secure working model of attachment. In
contrast, avoidant persons seem to react
to distress by organizing their interac-
tion goals around the deactivation of the
attachment system and a search for au-
tonomy and control. As a result, when
distressed, these persons would take
distance from partners and be reluctant
to form interdependent relationships
(Bowlby, 1988).
These hypothesized attachment-style
differences in interaction goals received
strong empirical support in Mikulincers
(1998, Studies 2 and 4) studies of the
sense of trust in close relationships. In
these studies, participants who classified
themselves as securely attached tended to
emphasize intimacy enhancement as the
most important trust-related gain and
to show relatively high accessibility of
thoughts about intimacy in a trust-re-
lated context. For these persons, episodes
that validate their sense of trust may con-
tribute to the formation and maintenance
of intimate close relationships, while be-
trayal of trust may raise concerns about
closeness and intimacy.
Whereas anxiously attached persons
tended to emphasize security enhance-
ment as the most important trust-related
gain and to show relatively high accessi-
bility of thoughts about security in a
trust-related context, avoidant persons
tended to emphasize control goals and to
show high accessibility of thoughts about
control in trust-related contexts (Miku-
lincer, 1998). For anxiously attached per-
sons, security seeking seems to be a cen-
tral component of their sense of trust. Ep-
isodes in which partners behave in a
responsive way may be appraised as con-
tributing to security feelings, while be-
trayal of trust may be appraised as a
threat to these feelings. Avoidant persons
seem to organize their sense of trust
around concerns about control. For these
persons, this pursuit of control seems to
be necessary to validate their sense of
self-reliance and to insure the attainment
of desired outcomes in the absence of con-
fidence that the partner will voluntarily
respond to their needs. Avoidant persons
may perceive their partners responsive-
ness as a validation of the control they
exert over partners behaviors, whereas
betrayal of trust may raise doubts about
the control they have in the relationship.
These findings highlight the fact that
the sense of attachment security is more
related to relational goals of closeness and
intimacy than to intrapersonal, egoistic
needs of security and control. It is possible
that the internalization of the sense of
attachment security satisfies these in-
trapersonal needs and frees cognitive and
emotional resources for the regulation of
relationship quality. On this basis, se-
curely attached persons could develop a
more open, selfless, and caring attitude
toward their close relationship partners.
They could become active agents respon-
sible for their partners welfare and rela-
tionship quality rather than passive re-
cipients of caring and comfort, and thus
could move from egocentric to more recip-
rocal relationships. This is particularly
important in the realm of dating relation-
ships and marriage, in which the attach-
ment and caregiving systems should
maintain a balanced and dynamic equilib-
rium, and both partners are equally re-
sponsible for the maintenance of a satis-
fying, stable relationship.
Securely attached persons desire for in-
timate relationships is also manifested in
their proneness to disclose and share per-
sonal information and feelings one of
the most basic means for the formation
of intimate relationships. In a series of
studies, Mikulincer and Nachshon (1991)
found that participants who classified
themselves as securely attached persons
reported that they tended to disclose more
personal information to relationship part-
ners than avoidant persons, and showed
more disclosure flexibility and reciprocity
MIKULINCER, FLORIAN, P. COWAN, & C. COWAN
than insecure persons. Moreover, secure
persons were found to disclose more per-
sonal information and to feel better inter-
acting with a high than low discloser part-
ner. In contrast, avoidant persons self-
disclosure and emotional reactions were
not affected by their partners disclosures.
These findings have been replicated in
subsequent studies (Keelan, Dion, & Dion,
1998; Pistole, 1993; Tucker & Anders,
1999).
It seems that secure persons are not
only able to disclose personal information
but they are also highly responsive to oth-
ers disclosure. In our view, self-disclosure
is a necessary but not sufficient behavior
for creating intimacy and closeness. A
partner moving toward intimacy should
be responsive to the partners communi-
cation, reinforce the partners confidence
in their good intentions, and promote
more intimate disclosure. On this basis,
we can conclude that secure persons re-
sponsiveness to a partners disclosure
seems to be a suitable strategy for devel-
oping stable and satisfactory relation-
ships. Overall, there is consistent evi-
dence that the sense of attachment secu-
rity is related to a positive orientation
toward togetherness and to the organiza-
tion of interaction goals around the pur-
suit of intimacy and closeness. In our
view, this relational orientation would
motivate people to engage in intimate in-
teractions and to invest efforts in re-
lationships that would be characterized
by intimacy, commitment, emotional in-
volvement, trust, and supportiveness. On
this basis, secure persons positive orien-
tation toward togetherness would contrib-
ute to the stability and quality of couple
relationships.
Mental Representations of Self
and Others
Bowlby (1973) asserted that the attain-
ment of a sense of attachment security
would be manifested in the development
Fam. Proc., Vol. 41, Fall, 2002
/ 419
of a positive self-image. In Bowlbys
terms, . . . the model of the attachment
figure and the model of the self are likely
to develop so as to be complementary and
mutually confirming (Bowlby, 1973, p.
205). When a person interacts with non-
responsive and unavailable others, he or
she will likely experience himself or her-
self as incompetent and unlovable. By
contrast, when a person anticipates oth-
ers availability and responsiveness, he or
she will consequently experience himself
or herself as competent and valuable. Em-
pirical research has consistently found
that the sense of attachment security is
related to high self-esteem (e.g., Bar-
tholomew & Horowitz, 1991) and to the
ability to manage stressors while main-
taining a sense of optimism and self-effi-
cacy (see Mikulincer & Florian, 2001, for
a review).
Self-representations and Couple Rela-
tionships: In our view, these positive mod-
els of self would also contribute to the
formation and maintenance of stable and
satisfying couple relationships. Securely
attached persons feel accepted and loved
by their partners, which, in turn, would
encourage them to reciprocate this love
and further strengthen their willingness
to care for the partner in times of need. In
addition, these positive representations
include a sense of self-efficacy in dealing
with threats and life problems, which
may lead to the adoption of a more confi-
dent attitude toward relationship obsta-
cles as well as to the adoption of more
constructive interpersonal problem solv-
ing strategies. On this basis, secure per-
sons would be able to deal with interper-
sonal conflicts without appraising them in
a catastrophic way and letting them lead
to conflict escalation. This constructive
conflict management strategy would di-
rectly contribute to the stability of couple
relationships. There is extensive evidence
of an association between secure attach-
ment and the adoption of constructive
420 / FAMILY PROCESS
strategies as couples attempt to resolve
problems (Gaines, Reis, Summers, et al.,
1997; Gaines, Granrose, Rios, et al., 1999;
Levy and Davis, 1988; Lussier et al.,
1997; Pistole, 1989; Scharfe & Bar-
tholomew, 1995; Simpson, Rholes, & Phil-
lips, 1996).
We are not aware of research that re-
lates attachment styles or working mod-
els directly to the emotion regulation
strategies that centrally affect both the
quality and stability of marital relation-
ships (Gottman, 1993). Cowan, Cohn,
Cowan, and Pearsons (1996) finding that
AAI-assessed attachment security in men
and women was connected with low mar-
ital conflict suggests that such a link
might exist. Secure attachment itself can
be viewed as an emotion regulation strat-
egy in which a person experiencing threat
or loss seeks out another for soothing and
support. Also, the ability to maintain a
coherent narrative during the AAI when
discussing personal and often highly emo-
tional issues is indicative of an ability to
regulate negative emotion in the service
of problem solving. We speculate that in
individuals or pairs with secure models of
attachment, the ability to regulate nega-
tive emotion helps the partners avoid es-
calating negative interchanges in a way
that leads to loss of controla pattern
that is one of the major risks for both
decline in marital satisfaction and mari-
tal dissolution (Gottman & Levenson,
2000).
Representations of Others and Couple
Relationships: The sense of attachment
security also includes positive models of
others (Bartholomew & Horowitz, 1991)
expectations that others will behave in a
caring and benevolent manner. These
positive representations may contribute
directly to several positive aspects of close
relationships that could maintain and en-
hance relationship satisfaction over time.
First, these representations would be
manifested in the sense of trust toward a
partner. According to Rempel, Holmes,
and Zanna (1985), trust involves (a) the
appraisal of a partner as reliable and pre-
dictable, (b) the belief that a partner is
concerned with ones needs and can be
counted on in times of need, and (c) feel-
ings of confidence in the strength of the
relationship. Data from several studies
support this view, in that secure persons
have been found to report higher levels of
trust toward their partner than insecure
persons (e.g., Collins & Read, 1990; Miku-
lincer, 1998; Simpson, 1990).
Second, the readiness to disclose one-
self and share personal feelings with a
partner may be fostered by the positive
models of others that characterize se-
curely attached persons. It demands a be-
lief that the partner can be trusted and
would not abuse the disclosed emotions,
thoughts, and information. Therefore, se-
cure persons who hold such positive be-
liefs about others intentions and goodwill
would be those who would be particularly
ready to disclose personal feelings and
thoughts to their partner. As a result,
these persons would tend to form rela-
tionships characterized by openness and
emotional involvement, which, in turn,
may contribute to relationship satisfac-
tion.
Third, the belief of the supportiveness
of the partner would promote good feel-
ings toward the partner, such as grati-
tude, warmth, and love. Moreover, due to
their positive representations of others,
secure individuals would interpret unex-
pected behaviors by partners in less rela-
tionship-threatening terms. In this way,
they would avoid unnecessary conflicts
and prevent escalation of negative emo-
tions toward partners.
These relationship-enhancing apprais-
als of partners behaviors were docu-
mented by Collins (1996), who asked par-
ticipants to explain negative partner be-
haviors and to report on the causes of
these behaviors. Secure participants gave
MIKULINCER, FLORIAN, P. COWAN, & C. COWAN
generally more positive and relationship-
enhancing explanations of negative rela-
tional events. Compared to insecure per-
sons, secure persons were less likely to
view a partners behavior as intentional,
negatively motivated, stable over time,
and global across relationship areas. Im-
portantly, Collins also found that secure
persons relationship-enhancing explana-
tions of negative partner behaviors buff-
ered negative affect toward a partner as
well as the arousal of relationship con-
flicts.
More direct support for the role of rep-
resentations of others in the association
between attachment security and rela-
tionship satisfaction is provided by Mor-
rison, Urquiza, and Goodlin-Jones (1997)
and Whisman and Allan (1996) who found
that positive attributions for partner be-
haviors appeared to mediate the associa-
tion between the sense of attachment se-
curity and relationship satisfaction in
dating couples.
Satisfaction of Other Basic Needs
Beyond emphasizing the psycho-evolu-
tionary nature of the attachment system,
Bowlby (1969) identified other needs that
are also placed in an evolutionary context
and maintain a dynamic interplay with
attachment needs. In Bowlbys (1969)
terms, the experience of inner distress
and the disruption of ones sense of at-
tachment security may activate prox-
imity-related cognitions and behaviors,
which, in turn, may inhibit the activation
of cognitions and behaviors related to
other basic needs (e.g., exploration, af-
filiation, caregiving). Moreover, because
they are preoccupied with regulating
their own distress, they may have fewer
available resources for engaging in affili-
ation, exploration, and/or caregiving ac-
tivities.
Persons who hold a sense of attachment
security would have more available re-
sources to engage in affiliation, explora-
Fam. Proc., Vol. 41, Fall, 2002
/ 421
tion, and caregiving behaviors with their
partners. Accordingly, these persons
would hold a positive and accepting atti-
tude toward these behaviors in their part-
ner. Since they feel confident in a part-
ners supportiveness, they would be sen-
sitive to the partners exploration and
affiliation needs and tolerate his or her
explorative and affiliative behaviors even
if these activities imply a momentary ab-
sence of the partner as an attachment
figure. Moreover, since they rely on sup-
port-seeking as an affect regulation de-
vice, they would accept and even encour-
age a partners caregiving behaviors.
Overall, secure persons would feel that a
close relationship is an adequate inter-
personal setting for satisfying not only
attachment needs but also for accomplish-
ing other important life tasks. As a result,
the fulfillment of these basic needs would
further contribute to both partners satis-
faction with their couple relationship.
Exploration Needs: In human develop-
ment, one of the basic evolutionary needs
is to explore the environment and to learn
new abilities and skills (Bowlby, 1969). In
adulthood, this explorative activity is
manifested in career development, work-
related activities, and the learning of new
adaptive skills, which, in turn, may fur-
ther develop a personal sense of auton-
omy, mastery, achievement, and control.
These activities demand energy and time
that otherwise might be spent with a close
relationship partner.
In couple relationships, the need for ex-
ploration can be encouraged or frustrated
by the relational attitudes and behaviors
of both partners. When one feels confident
in the partners availability and respon-
siveness in times of need and/or the part-
ner is attentive and supportive of ones
need for exploration, even when this
comes at expense of time spent together,
the satisfaction of this need may be facil-
itated. As a result, the individual would
feel free to develop his or her own poten-
422 / FAMILY PROCESS
tialities and be more willing to reciprocate
his or her partners need for exploration.
Mutual respect, understanding, and mar-
ital satisfaction would thus be enhanced.
Accordingly, partners would appraise
their couple relationship as promoting
rather than inhibiting their own sense of
autonomy and their personal develop-
ment. In the long run, the satisfaction of
exploration needs could contribute to re-
lationship stability and diminish unnec-
essary conflicts related to possible differ-
ences in the partners trajectory of per-
sonal development.
In contrast, the satisfaction of explora-
tion needs would be frustrated when one
partner restrains the others attempts to
spend energy and time outside the rela-
tionship and/or threatens with separation
and divorce if his or her wishes and de-
mands are not being fulfilled. Accord-
ingly, when a person is afraid of the part-
ners reactions and is not secure about his
or her availability and responsiveness, he
or she may inhibit exploration in advance
and give up any attempt to develop an
autonomous personality in order to please
the partner. In most of these cases, one
may expect that many nuclei of frustra-
tion, tension, conflict, and dispute may
arise within the couple. Furthermore, the
individual may develop negative feeling
toward the partner and the relationship,
such as a sense of suffocation and coer-
cion, a sense of limitation of personal
choices and activities, or a sense of self-
derogation in order to satisfy a partners
egoistic needs. As a result, marital satis-
faction would decrease and the likelihood
of separation may increase.
Indeed, adult attachment studies have
found that securely attached persons
were more likely than insecure persons to
engage in exploration activities and to
open their cognitive structures to new ev-
idence (e.g., Green-Henessy & Reis, 1998;
Mikulincer, 1997; Mikulincer & Arad,
1999). Second, a persons sense of attach-
ment security would also encourage a
partners exploration attempts. On this
basis, the sense of attachment security
would contribute to the satisfaction of
both partners exploration needs.
Affiliation Needs: Affiliation needs refer
to the phylogenetically evolved tendency
to be sociable with others (Bowlby, 1969).
They drive people to spend time in the
company of others, to be involved in
friendships, and to engage in a wide vari-
ety of social activities, such as play, alli-
ance against outsiders, and squabbles
(Weiss, 1998). In his taxonomy of social
interactions, Weiss suggested that at-
tachment and affiliation behaviors differ
in the level of perceived exclusiveness,
permanence, and emotionality. First, at-
tachment behaviors demand exclusivity
from the provider of a secure base, be-
cause relationships between this provider
and a third person may reduce his or her
availability when needed. In contrast, af-
filiation behaviors may not necessarily
demand exclusiveness. Rather, the incor-
poration of other persons in an affiliation
relationship may be perceived as a posi-
tive outcome because these persons can
advance common interests, facilitate
learning, and strengthen potential alli-
ances (Weiss, 1998). Second, whereas at-
tachment relationships may persist over
time, affiliation relationships may be
ended with relative ease (Weiss, 1998).
Third, attachment behaviors may involve
stronger emotions than affiliation behav-
iors. In attachment behaviors, people ex-
perience a cycle of tension and relief, ac-
companied by feelings of anxiety, fear,
and gratitude (Bowlby, 1988). These emo-
tions may be weaker or even irrelevant
when people seek others for companion-
ship or play (Weiss, 1998).
In couple relationships, affiliation needs
could be manifested in two different ways:
(1) a persons motivation to develop a
friendship relationship with his or her
partner by attempting to engage in com-
MIKULINCER, FLORIAN, P. COWAN, & C. COWAN
mon activities and recreations and to
spend time together with the same friends;
(2) a persons motivation to maintain his
or her own separate network of friends,
which sometimes may come at expense of
the energy and time devoted to the part-
ner. In this way, the satisfaction of affili-
ation needs may contribute to both the
sense of couples togetherness and the
sense of individuation and personal
freedom within the couple. The issue of
balancing these two issues is at the heart
of most theories of what makes for satis-
fying couple relationships (Gottman, 1993).
We believe that the sense of attachment
security would directly contribute to the
satisfaction of both partners affiliation
needs. First, the sense of attachment se-
curity would encourage affiliative activi-
ties within the couple relationship be-
cause secure persons put strong emphasis
on togetherness. Moreover, the sense of
attachment security would facilitate af-
filiative activities outside the couple rela-
tionship because the person is confident
that the partner would continue to love
him or her even if energy and time is
spent with other friends. Indeed, Miku-
lincer and Selinger (2001) found that se-
cure persons were more likely than inse-
cure persons to engage in affiliative activ-
ities and to hold a flexible balance
between attachment and affiliation goals
in close relationships.
Caregiving Needs: Caregiving needs are
designed to provide protection and sup-
port to others who are either chronically
dependent or temporarily in need, and
they are guided by an altruistic orienta-
tionthe alleviation of others distress
(Bowlby, 1969). These needs drive us to
help, assist, and comfort significant oth-
ers, and motivate us to protect these per-
sons from any threat or danger. These
caregiving activities often entail personal
sacrifice in terms of time, resources, and
mental efforts. Although caregiving needs
in one individual are very responsive to
Fam. Proc., Vol. 41, Fall, 2002
/ 423
the arousal of attachment needs in his or
her partner, Bowlby (1969) viewed them
as two separate motivational systems.
Whereas attachment needs imply that
the person seeks support and protection,
caregiving needs imply that the person
seeks to be an active provider of support
and protection.
The satisfaction of caregiving needs
seems to play an important role in couple
relationships. We hypothesize that in
long-lasting, satisfying relationships, peo-
ple will be attentive to partners needs
and to learning when, how, and in which
areas the partner wishes or expects sup-
port and protection. In addition, persons
would be expected to learn to accept the
partners offer of support and comfort and
to appraise it as a sign of love and caring,
not as a sign of a patronizing or unequal
relationship. In other words, persons
would be expected to learn to accept their
partners caregiving efforts without feel-
ing a loss of personal control or tension
due to a power struggle.
In contrast, caregiving needs would be
frustrated when the partner is not able or
willing to accept the others assistance
and support or when he or she reacts with
hostility, suspicions, or even rejection to
the partners caring efforts. These needs
can be also frustrated when the person,
due to his or her personality or relational
history, is not attentive to the partners
needs and is not able to relieve success-
fully the partners suffering and pain.
These reactions can elicit interpersonal
conflicts around issues of trust, coopera-
tion, reciprocity, and the provision and
receipt of support. Furthermore, they
may lead to feelings of personal alien-
ation, low self-esteem, neglect, and inferi-
ority. Specifically, the person may develop
a sense of being stuck in the unequal po-
sition of the weak, needy, and/or eter-
nal infant. The most probable outcome of
this kind of interaction is dissatisfaction
424 / FAMILY PROCESS
and a desire to leave the frustrating rela-
tionship.
In our view, the sense of attachment
security would contribute to the satisfac-
tion of both partners caregiving needs.
First, a persons sense of attachment se-
curity would encourage his or her own
caregiving attempts, because he or she
would have available resources to attend
to a partners needs and provide adequate
care for alleviating distress. Moreover, se-
cure persons positive models of others
would be likely to foster the perception of
others as deserving help, and motivate
people to provide the necessary support to
restore or maintain a partners welfare.
Indeed, adult attachment studies have
found that self-reports of attachment se-
curity are associated with relatively high
levels of reported responsiveness to a ro-
mantic partners needs (e.g., Carnelley,
Pietromonaco, & Jaffe, 1996; Feeney,
1996; Kunce & Shaver, 1994). Accord-
ingly, secure persons have been found to
offer spontaneously more comfort and re-
assurance to a romantic partner in times
of need than insecure persons (e.g., Col-
lins & Feeney, 2000; Fraley & Shaver,
1998; Simpson et al., 1992). Second, be-
cause of their reliance on support seeking
as an affect regulation device and their
high sense of self-esteem, secure persons
would hold a positive and accepting atti-
tude toward their partners supportive
and caring behaviors, thereby leading to
the satisfaction of the partners caregiv-
ing needs.
EXTENSION OF MODEL TO THE
FAMILY SYSTEM
We have focused on couple relation-
ships and only indirectly on the intergen-
erational transmission of attachment.
Given the space limitations here, we can
offer only a sketch of possible intergen-
erational linkages in the model we have
presented and how the dynamics might
play out in the life of a family. A growing
body of research finds relatively high con-
cordance between mothers working mod-
els of attachment, based on their re-
sponses to the Adult Attachment Inter-
view, their infants security of attachment
after a brief separation (Ainsworth et al.,
1978; van IJzendoorn, 1995), and mea-
sures of childrens adaptation with peers
in later years (Sroufe, Carlson, & Shul-
man, 1993). These findings have been
used to support hypotheses about the con-
tinuity of relationship quality across the
generations. The high degree of concor-
dance between adults attachment styles
and their childrens attachment styles
and peer relationship quality is especially
impressive because it occurs across meth-
ods (adult narratives, observations of par-
ent-child interaction, peer reports), across
measurement contexts, and across time.
And yet, we know surprisingly little about
the mechanisms that underlie this conti-
nuity or how to explain what appears to
be a strong tendency to repeat relation-
ship patterns from one generation to an-
other.
It has been widely assumed that the
quality of parent-child relationships is
the linking mechanismthat adults who
are themselves securely attached tend to
provide a secure base for their children
and there is substantial evidence for this
kind of association in studies of infants
(Haft & Slade, 1989), toddlers (Crowell &
Feldman, 1988), and preschoolers (Cohn,
Cowan, Cowan, & Pearson, 1992). For ex-
ample, Cohn and colleagues found that
fathers who gave coherent narratives of
their early family relationships provided
more warmth and structure to their pre-
schoolers during challenging tasks than
fathers whose narratives were not coher-
ent.
We offer here the idea that the quality
of the relationship between the parents
plays a central role in the generational
transmission of working models of attach-
ment. Two studies of different samples of
MIKULINCER, FLORIAN, P. COWAN, & C. COWAN
fathers and mothers with preschoolers
and kindergartners (Cohn et al., 1992;
Cowan, Bradburn, & Cowan, in press)
support this hypothesis. Based on the AAI
continuous ratings of (a) whether the par-
ents in the study described their own par-
ents as loving, and (b) whether the par-
ents were still angry with their parents
(the childrens grandparents) in ways
that disrupted their AAI narratives, less
securely-attached men and women were
in marriages that tended to be more con-
flictful (observational data) and in parent-
child relationships with their own chil-
dren that were less effective (observation-
al data). In turn, when parents were
assessed as less securely attached when
their children were preschoolers, the chil-
dren were significantly more likely to
be seen by their kindergarten teachers
as having internalizing or externalizing
problems in school one and two years
later. Over and above measures of parent-
ing style, both attachment and marital
data from the parents contributed signif-
icantly to predicting the childrens adap-
tation to school. Other analyses of data
from the second Cowan et al. study
(Cowan, Cowan, & Heming, in press) re-
vealed that preventive interventions in
the form of couples groups designed to
foster more effective marital and parent-
child relationships resulted in positive
outcomes for the parents and for their
children.
These findings lead us to conclude ten-
tatively that marital quality may play a
causal role in affecting parenting style
and childrens adaptation. The question
remains: how does this occur? One possi-
bility is that conflicted parent-child rela-
tionships spill over to interfere with the
relationship between one or both parents
and the child. The second possibility, sug-
gested by Davies and Cummings (1998),
is that marital conflict has a direct effect
on the child, disrupting attachment rela-
tionships and creating emotional insecu-
Fam. Proc., Vol. 41, Fall, 2002
/ 425
rity that plays out in the childs outside-
the-family relationships.
Extrapolating from these findings, and
subject to replication and extension of the
results, we conclude that the transmis-
sion of attachment relationships from
grandparents to parents to children is not
simply a matter of parenting. When a per-
son learns early on that he or she is wor-
thy of love, and that adults will be respon-
sive and available in times of need, he or
she is more likely to establish satisfying
relationships with other partners, and to
have the inclination and ability to work
toward solving relationship problems and
regulating emotions so that they do not
escalate out of control. The family envi-
ronment established by couples who can
regulate emotions and solve problems ef-
fectively facilitates both mother-child and
father-child relationships that, in turn,
foster a childs ability to explore new ideas
and relationships (Byng-Hall, 1999). All
of these processes appear to foster the
childrens cognitive, social, and emotional
development.
CONCLUDING REMARKS
The field of family psychology still
struggles with the eternal question of why
some couples succeed in maintaining a
long-lasting, satisfactory relationship
while others fail in this relational task.
From our point of view, attachment the-
ory is one of the main promising concep-
tual frameworks for raising and testing
useful hypotheses concerning the psycho-
logical and ecological factors that contrib-
ute to positive relational outcomes. This
theoretical framework allows the exami-
nation of the role that inner resources,
such as the sense of attachment security,
may play in the dynamic relational pro-
cesses that characterize the different
stages of marriage and family develop-
ment. New research is beginning to show
that the links between attachment and
couple relationships have consequences
426 / FAMILY PROCESS
for childrens development. Because of the
universality of attachment processes, this
theoretical framework could be used in
examining marital and family processes
across different ethnic, cultural, and reli-
gious groups. Despite important differ-
ences in methods of measuring attach-
ment, and issues in the conceptualization
of attachment as a categorical or contin-
uous phenomenon, the accumulation of
knowledge that has been achieved in the
last two decades provides a clear, coher-
ent picture of the consistent connections
between a sense of attachment security
and the formation and maintenance of
stable and satisfactory couple relation-
ships. Furthermore, research has pro-
vided relevant empirical data on the psy-
chological processes that help explain the
positive relational outcomes associated
with a sense of security of attachment.
The model presented in this article sug-
gests some of the mechanisms that may
underlie these processes. Nevertheless,
some important issues that have not been
addressed here could be examined in fu-
ture studies. First, future studies should
be designed to expand and deepen our
understanding of (a) the interplay be-
tween attachment processes and the sat-
isfaction of other basic needs within cou-
ple relationships, and (b) how this dy-
namic interplay may contribute to
relationship satisfaction and quality. Sec-
ond, there is a need for additional empir-
ical efforts that attempt to integrate at-
tachment theory and research with data
on family dynamic processes and mecha-
nisms. Third, most of the existing re-
search has focused on the early stages of
couple relationships and marriage. For a
fuller understanding of the links between
security of attachment and couple rela-
tionship quality, it will be important to
examine the role that attachment pro-
cesses play in later stages of marriage and
family development (e.g., in midlife and
aging couples). Fourth, the emphasis in
this article has been on consistency and
coherence across domains, as if attach-
ment operates as a template for the de-
velopment of other intimate relation-
ships. The correlations we reported are
far from perfect. More research is needed
on the possible effects that specific couple
and family relationships may have on
each partners sense of attachment secu-
rity.
More theoretical and empirical efforts
should be invested in applying the cumu-
lative knowledge gleaned from adult at-
tachment research to enhancing and im-
proving interventions in marital and fam-
ily therapyto find modifiable aspects of
attachment that can facilitate family re-
lationships and modifiable aspects of fam-
ily relationships that can enhance the se-
curity of each individuals attachment.
This endeavor is necessary not only to
improve the life of men, women, and chil-
dren living in families. It is essential in
that it can provide crucial theoretical in-
formation about the nature of the links
between attachment and marriage, and
about the causal connections between the
two domains. We hope that our article
will act as a stimulus and serve as a
guideline for further theoretical and clin-
ical debate, as well as for empirical stud-
ies in attachment and couple relation-
ships across the lifespan.
REFERENCES
Ainsworth, M.D., Blehar, M.C., Waters, E., &
Wall, S. (1978). Patterns of attachment: A
psychological study of the strange situation.
Hillsdale NJ: Lawrence Erlbaum Associ-
ates.
Baldwin, M.W., Keelan, J.P.R., Fehr, B., Enns,
V., & Koh Rangarajoo, E. (1996). Social-cog-
nitive conceptualization of attachment work-
ing models: Availability and accessibility ef-
fects. Journal of Personality and Social Psy-
chology 71: 94 109.
Bank, L., Dishion, T.J., Skinner, M., & Patter-
son, G.R. (1990). Method variance in struc-
tural equation modeling: Living with glop
MIKULINCER, FLORIAN, P. COWAN, & C. COWAN
(pp. 247279). In G.R. Patterson (ed.), De-
pression and aggression in family interac-
tion. Hillsdale NJ: Lawrence Erlbaum Asso-
ciates.
Bartholomew, K., & Horowitz, L.M. (1991). At-
tachment styles among young adults: A test
of a four-category model. Journal of Person-
ality and Social Psychology 61: 226 244.
Berman, W.H., Marcus, L., & Berman, E.R.
(1994). Attachments in marital relations
(pp. 204 231). In N.B. Sperlng & W.H. Ber-
man (eds.), Attachment in adults: Clinical
and developmental perspectives. New York:
Academic Press.
Bookwala, J., & Zdaniuk, B. (1998). Adult at-
tachment styles and aggressive behavior
within dating relationships. Journal of So-
cial and Personal Relationships 15: 175190.
Boon, S.D., & Griffin, D.W. (1996). The con-
struction of risk in relationships: The role of
framing in decisions about relationships.
Personal Relationships 3: 293306.
Bowlby, J. (1969). Attachment and loss: Vol. 1.
Attachment. New York: Basic Books.
Bowlby, J. (1973). Attachment and loss: Vol. 2.
Separation: Anxiety and anger. New York:
Basic Books.
Bowlby, J. (1979). The making and breaking of
affectional bonds. London: Tavistock.
Bowlby, J. (1988). A secure base: Clinical ap-
plications of attachment theory. London:
Routledge.
Brennan, K.A., Clark, C.L., & Shaver, P.R.
(1998). Self-report measurement of adult at-
tachment: An integrative overview (pp. 46
76). In J.A. Simpson & W.S. Rholes (eds.),
Attachment theory and close relationships.
New York: Guilford Press.
Brennan, K., & Shaver, P.R. (1995). Dimen-
sions of adult attachment, affect regulation,
and romantic relationship functioning. Per-
sonality and Social Psychology Bulletin 21:
567583.
Byng-Hall, J. (1999). Family couple therapy:
Toward greater security (pp. 625 645). In J.
Cassidy & P.R. Shaver (eds.), Handbook of
attachment: Theory, research, and clinical
applications. New York: Guilford Press.
Carnelley, K.B., & Janoff-Bulman, R. (1992).
Optimism about love relationships: General
vs specific lessons from ones personal expe-
Fam. Proc., Vol. 41, Fall, 2002
/ 427
riences. Journal of Social and Personal Re-
lationships 9: 520.
Carnelley, K.B., Pietromonaco, P.R., & Jaffe,
K. (1994). Depression, working models of
others, and relationship functioning. Jour-
nal of Personality and Social Psychology 66:
127140.
Carnelley, K.B., Pietromonaco, P.R., & Jaffe,
K. (1996). Attachment, caregiving, and rela-
tionship functioning in couples. Personal Re-
lationships 3: 257277.
Chappell, K.D., & Davis, K.E. (1998). Attach-
ment, partner choice, and perception of ro-
mantic partners: An experimental test of the
attachment-security hypothesis. Personal
Relationships 5: 327342.
Cohn, D.A., Cowan, P.A., Cowan, C.P., & Pear-
son, J. (1992). Mothers and fathers working
models of childhood attachment relation-
ships, parenting styles, and child behavior.
Development and Psychopathology 4: 417
431.
Cohn, D.A., Silver, D.H., Cowan, C.P., Cowan,
P.A., & Pearson, J. (1992). Working models
of childhood attachment and couple rela-
tionships. Journal of Family Issues 13: 432
449.
Collins, N.L. (1996). Working models of attach-
ment: Implications for explanation, emotion,
and behavior. Journal of Personality and
Social Psychology 71: 810 832.
Collins, N.L., & Feeney, B.C. (2000). A safe
haven: An attachment theory perspective on
support seeking and caregiving in intimate
relationships. Journal of Personality and
Social Psychology 78: 10531073.
Collins, N.L., & Read, S.J. (1990). Adult at-
tachment, working models, and relationship
quality in dating couples. Journal of Person-
ality and Social Psychology 58: 644 663.
Collins, N.L., & Read, S.J. (1994). Cognitive
representations of attachment: The struc-
ture and function of working models (pp.
5392). In K. Bartholomew & D. Perlman
(eds.), Attachment processes in adulthood.
London: Jessica Kingsley.
Cowan, P.A., Bradburn, I., & Cowan, C.P. (in
press). Parents working models of attach-
ment: the intergenerational context of prob-
lem behavior in kindergarten. In P.A.
Cowan, C.P. Cowan, J. Ablow, V.K. Johnson,
& J. Measelle (eds.). The family context of
428 / FAMILY PROCESS
parenting in childrens adaptation to ele-
mentary school. Monographs in Parenting.
Cowan, P.A., & Cowan, C.P. (2001). A couple
perspective on the transmission of attach-
ment patterns. (pp. 62 82). In C. Clulow
(ed.). Adult attachment and couple psycho-
therapy: The secure base in practice and
research. London: Brunner-Routledge.
Cowan, P.A., Cowan, C.P., & Heming, T. (in
press). Two variations of a preventive inter-
vention for couples: effects on parents and
children during the transition to elementary
school. In P.A. Cowan, C.P. Cowan, J. Ab-
low, V.K. Johnson, & J. Measelle (eds.). The
family context of parenting in childrens ad-
aptation to elementary school. Monographs
in Parenting.
Cowan, P.A., Cohn, D., Cowan, C.P., & Pear-
son, J.L. (1996). Parents attachment histo-
ries and childrens internalizing and exter-
nalizing behavior: Exploring family systems
models of linkage. Journal of Consulting
and Clinical Psychology 64: 53 63.
Cowan, P.A., & Cowan, C.P. (2002). What an
intervention design reveals about how par-
ents affect their childrens academic
achievement and behavior problems (pp.
7598). In J.G. Borkowski, S. Ramey, & M.
Bristol-Power (eds.), Parenting and the
childs world: Influences on intellectual, ac-
ademic, and social-emotional development.
Mahwah NJ: Lawrence Erlbaum Associates.
Cozzarelli, C., Hoekstra, S.J., & Bylsma, W.H.
(2000). General versus specific mental mod-
els of attachment: Are they associated with
different outcomes? Personality and Social
Psychology Bulletin 26: 605 618.
Crowell, J.A., & Feldman, S.S. (1989). Moth-
ers working models of attachment relation-
ships and mother and child behavior during
separation and reunion. Developmental Psy-
chology 27: 597 605.
Crowell, J., & Treboux, D. (2001). Attachment
security in adult partnerships (pp. 28 42).
In C. Clulow (ed.), Adult attachment and
couple psychotherapy: The secure base in
practice and research. London: Brunner-
Routledge.
Davies, P.T., & Cummings, E.M. (1998). Ex-
ploring childrens emotional security as a
mediator of the link between marital rela-
tions and child adjustment. Child Develop-
ment 69: 124 139.
Davila, J., Bradbury, T.N., & Fincham, F.
(1998). Negative affectivity as a mediator of
the association between adult attachment
and marital satisfaction. Personal Relation-
ships 5: 467 484.
Davila, J., Karney, B.R., & Bradbury, T.N.
(1999). Attachment change processes in the
early years of marriage. Journal of Person-
ality and Social Psychology 76: 783 802.
Diehl, M., Elnick, A.B., Bourbeau, L.S., &
Labouvie-Vief, G. (1998). Adult attachment
styles: Their relations to family context and
personality. Journal of Personality and So-
cial Psychology 74: 1656 1669.
Doi, S.C., & Thelen, M.H. (1993). The Fear of
Intimacy Scale: Replication and extension.
Psychological Assessment 5: 377383.
Feeney, B.C., & Kirkpatrick, L.A. (1996). Ef-
fects of adult attachment and presence of
romantic partners on physiological re-
sponses to stress. Journal of Personality and
Social Psychology 70: 255270.
Feeney, J.A. (1994). Attachment style, com-
munication patterns and satisfaction across
the life cycle of marriage. Personal Relation-
ships 1: 333348.
Feeney, J.A. (1996). Attachment, caregiving,
and marital satisfaction. Personal Relation-
ships 3: 401 416.
Feeney, J.A. (1999a). Adult romantic attach-
ment and couple relationships (pp. 355377).
In J. Cassidy & P.R. Shaver (eds.), Hand-
book of attachment: Theory, research, and
clinical applications. New York: Guilford
Press.
Feeney, J. (1999b). Issues of closeness and dis-
tance in dating relationships: Effects of sex
and attachment style. Journal of Social and
Personal Relationships 16: 571590.
Feeney, J.A. (1999c). Adult attachment, emo-
tional control, and marital satisfaction. Per-
sonal Relationships 6: 169 185.
Feeney, J.A., & Noller, P. (1990). Attachment
style as a predictor of adult romantic rela-
tionships. Journal of Personality and Social
Psychology 58: 281291.
Feeney, J.A., & Noller, P. (1991). Attachment
style and verbal descriptions of romantic
partners. Journal of Social and Personal Re-
lationships 8: 187215.
MIKULINCER, FLORIAN, P. COWAN, & C. COWAN
Feeney, J.A., & Noller, P. (1992). Attachment
style and romantic love: Relationship disso-
lution. Australian Journal of Psychology 44:
69 74.
Feeney, J., Noller, P., & Callan, V.J. (1994).
Attachment style, communication and satis-
faction in the early years of marriage (pp.
269 308). In Bartholomew, K., & Perlman,
D. (eds.), Attachment processes in adult-
hood. London: Jessica Kingsley.
Feeney, J.A., Noller, P., & Patty, J. (1993).
Adolescents interactions with opposite sex:
Influence of attachment style and gender.
Journal of Adolescence 16: 169 186.
Fitzpatrick, M.A., Fey, J., Segrin, C., & Schiff,
J.L. (1993). Internal working models of re-
lationships and marital communication.
Journal of Language and Social Psychology,
12: 103131.
Fraley, R.C., & Shaver, P.R. (1998). Airport
separations: A naturalistic study of adult
attachment dynamics in separating couples.
Journal of Personality and Social Psychol-
ogy 75: 1198 1212.
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 Psychol-
ogy 78: 350 365.
Frazier, P.A., Byer, A.L., Fischer, A.R.,
Wright, D.M., & DeBord, K.A. (1996). Adult
attachment style and partner choice: Corre-
lational and experimental findings. Personal
Relationships 3: 117136.
Fuller, T.L., & Fincham, F.D. (1995). Attach-
ment style in married couples: Relation to
current marital functioning, stability over
time, and method of assessment. Personal
Relationships 2: 1734.
Gaines, S.O. Jr., Granrose, C.S., Rios, D.I.,
Garcia, B.F., Young, M.S., Farris, K.R., &
Bledsoe, K.L. (1999). Patterns of attachment
and responses to accommodative dilemmas
among interethnic/interracial couples. Jour-
nal of Social and Personal Relationships 16:
275285.
Gaines, S.O., Jr., Reis, H.T., Summers, S.,
Rusbult, C.E., Cox, C.L., Wexler, M.O.,
Marelich, W.D., & Kurland, G.J. (1997). Im-
pact of attachment style on reactions to ac-
commodative dilemmas in close relation-
ships. Personal Relationships 4: 93113.
Fam. Proc., Vol. 41, Fall, 2002
/ 429
George, C., Kaplan, N., & Main, M. (1985). The
Adult Attachment Interview. Unpublished
manuscript, Department of Psychology,
University of California, Berkeley.
Gerlsma, C., Buunk, B.P., & Mutsaers,
W.C.M. (1996). Correlates of self-reported
adult attachment styles in a Dutch sample
of married men and women. Journal of So-
cial and Personal Relationships 13: 313320.
Green-Hennessy, S., & Reis, H.T. (1998).
Openness in processing social information
among attachment types. Personal Relation-
ships 5: 449 466.
Gottman, J.M. (1993). What predicts divorce?
The relationship between marital processes
and marital outcomes. Hillsdale NJ: Law-
rence Erlbaum Associates.
Gottman, J.M., & Levenson, R.W. (2000). The
timing of divorce: Predicting when a couple
will divorce over a 14-year period. Journal of
Marriage & the Family 62: 737745.
Guerrero, L.K. (1996). Attachment-style dif-
ferences in intimacy and involvement: A test
of the Four-Category Model. Communica-
tion Monographs 63: 269 292.
Haft, W., & Slade, A. (1989). Affect attun-
ement and maternal attachment: A pilot
study. Journal of Infant Mental Health 10:
157172.
Hazan, C., & Shaver, P.R. (1987). Romantic
love conceptualized as an attachment pro-
cess. Journal of Personality and Social Psy-
chology 52: 511524.
Hendrick, C., & Hendrick, S. (1989). Research
on love: Does it measure up? Journal of Per-
sonality and Social Psychology 56: 784 794.
Hesse, E. (1999). The Adult Attachment Inter-
view: Historical and current perspectives
(pp. 395 433). In J. Cassidy & P.R. Shaver
(eds.), Handbook of attachment: Theory, re-
search, and clinical implications. New York:
Guilford Press.
Hill, E.M., Young, J.P., & Nord, J.L. (1994).
Childhood adversity, attachment, security,
and adult relationships: A preliminary
study. Ethology and Sociobiology 15: 323
338.
Jones, J.T., & Cunningham, J.D. (1996). At-
tachment styles and other predictors of re-
lationship satisfaction in dating couples.
Personal Relationships 3: 387399.
Keelan, J.R., Dion, K.L., & Dion, K.K. (1994).
430 / FAMILY PROCESS
Attachment style and heterosexual relation-
ships among young adults: A short-term
panel study. Journal of Social and Personal
Relationships 11: 201214.
Keelan, J.P.R., Dion, K.K., & Dion, K.L.
(1998). Attachment style and relationship
satisfaction: Test of a self-disclosure expla-
nation. Canadian Journal of Behavioural
Science 30: 24 35.
Kirkpatrick, L.A. (1998). Evolution, pair-bond-
ing, and reproductive strategies: A reconcep-
tualization of adult attachment (pp. 353
393). In J.A. Simpson & W.S. Rholes (eds.),
Attachment theory and close relationships.
New York: Guilford Press.
Kirkpatrick, L.A., & Davis, K.E. (1994). At-
tachment style, gender, and relationship
stability: A longitudinal analysis. Journal of
Personality and Social Psychology 66: 502
512.
Kirkpatrick, L.A., & Hazan, C. (1994). Attach-
ment styles and close relationships: A four-
year prospective study. Personal Relation-
ships 1: 123142.
Klohnen, E.C., & Bera, S. (1998). Behavioral
and experiential patterns of avoidantly and
securely attached women across adulthood:
A 31-year longitudinal perspective. Journal
of Personality and Social Psychology 74:
211223.
Kobak, R.R., & Hazan, C. (1991). Attachment
in marriage: Effects of security and accuracy
of working models. Journal of Personality
and Social Psychology 60: 861 869.
Kunce, L.J., & Shaver, P.R. (1994). An attach-
ment-theoretical approach to caregiving in
romantic relationships (pp. 205237). In K.
Bartholomew & D. Perlman (eds.), Advances
in personal relationships (Vol. 5). London:
Jessica Kingsley.
Latty-Mann, H., & Davis, K.E. (1996). Attach-
ment theory and partner choice: Preference
and actuality. Journal of Social and Per-
sonal Relationships 13: 523.
Levenson, R.W., & Gottman, J.G. (1983). Mar-
ital interaction: Physiological linkage and
affective exchange. Journal of Personality
and Social Psychology 45: 587597.
Levy, M.B., & Davis, K.E. (1988). Lovestyles
and attachment styles compared: Their re-
lations to each other and to various relation-
ship characteristics. Journal of Social and
Personal Relationships 5: 439 471.
Lussier, Y., Sabourin, S., & Turgeon, C.
(1997). Coping strategies as moderators of
the relationship between attachment and
marital adjustment. Journal of Social and
Personal Relationships 14: 777791.
Main, M., & Goldwyn, R. (1988). Adult attach-
ment scoring and classification system. Un-
published manuscript, University of Califor-
nia, Berkeley, available from first author.
Main, M., Kaplan, N., & Cassidy, J. (1985).
Security in infancy, childhood, and adult-
hood: A move to the level of representation.
Monographs of the Society for Research in
Child Development 50: 66 104.
Mayseless, O., Sharabany, R., & Sagi, A.
(1997). Attachment concerns of mothers as
manifested in parental, spousal, and friend-
ship relationships. Personal Relationships 4:
255269.
McCarthy, G. (1999). Attachment style and
adult love relationships and friendships: A
study of a group of women at risk of experi-
encing relationship difficulties. British
Journal of Medical Psychology 72: 305321.
McCarthy, G., & Taylor, A. (1999). Avoidant/
ambivalent attachment style as a mediator
between abusive childhood experiences and
adult relationship difficulties. Journal of
Child Psychology and Psychiatry and Allied
Disciplines 40: 465 477.
Mikulincer, M. (1997). Adult attachment style
and information processing: Individual dif-
ferences in curiosity and cognitive closure.
Journal of Personality and Social Psychol-
ogy 72: 12171230.
Mikulincer, M. (1998). Attachment working
models and the sense of trust: An explora-
tion of interaction goals and affect regula-
tion. Journal of Personality and Social Psy-
chology 74: 1209 1224.
Mikulincer, M., & Arad, D. (1999). Attach-
ment, working models, and cognitive open-
ness in close relationships: A test of chronic
and temporary accessibility effects. Journal
of Personality and Social Psychology 77:
710 725.
Mikulincer, M., & Erev, I. (1991). Attachment
style and the structure of romantic love.
British Journal of Social Psychology 30:
273291.
MIKULINCER, FLORIAN, P. COWAN, & C. COWAN
Mikulincer, M., & Florian, V. (1999). The as-
sociation between spouses self-reports of at-
tachment styles and representations of fam-
ily dynamics. Family Process 38: 69 83.
Mikulincer, M., & Florian, V. (2001). Attach-
ment style and affect regulationImplica-
tions for coping with stress and mental
health (pp. 537557). In G. Fletcher & M.
Clark (eds.), Blackwell handbook of social
psychology: Interpersonal Processes. Oxford,
UK: Blackwell Publishers.
Mikulincer, M., Horesh, N., Levy-Shiff, R.,
Manovich, R., & Shalev, J. (1998). The con-
tribution of adult attachment style to the
adjustment to infertility. British Journal of
Medical Psychology 71: 265280.
Mikulincer, M., & Nachshon, O. (1991). At-
tachment style and patterns of self-disclo-
sure. Journal of Personality and Social Psy-
chology 61: 321332.
Mikulincer, M., & Selinger, M. (2001). The
interplay between attachment and affilia-
tion systems in adolescents same-sex
friendship: The role of attachment style.
Journal of Social and Personal Relation-
ships 18: 81106.
Morrison, T.L., Urquiza, A.J., & Goodlin-
Jones, B.L. (1997). Attachment, perceptions
of interaction, and relationship adjustment.
Journal of Social and Personal Relation-
ships 14: 627 642.
Pietromonaco, P., & Carnelley, K. (1994). Gen-
der and working models of attachment: Con-
sequences for perceptions of self and roman-
tic partners. Personal Relationships 1: 63
82.
Pistole, M.C. (1989). Attachment in adult ro-
mantic relationships: Style of conflict reso-
lution and relationship satisfaction. Journal
of Social and Personal Relationships 6: 505
512.
Pistole, M.C. (1993). Attachment relation-
ships: Self-disclosure and trust. Journal of
Mental Health Counseling 15: 94 106.
Pistole, M.C., Clark, E.M., & Tubbs, A.L.
(1995). Love relationships: Attachment style
and the investment model. Journal of Men-
tal Health Counseling 17: 199 209.
Pistole, M.C., & Vocaturo, L.C. (1999). Attach-
ment and commitment in college students
romantic relationships. Journal of College
Student Development 40: 710 720.
Fam. Proc., Vol. 41, Fall, 2002
/ 431
Rempel, J.K., Holmes, J.G., & Zanna, M.P.
(1985). Trust in close relationships. Journal
of Personality and Social Psychology 49: 95
112.
Rholes, W.S., Simpson, J.A., & Blakely, B.S.
(1995). Adult attachment styles and moth-
ers relationships with their young children.
Personal Relationships 2: 3554.
Scharfe, E., & Bartholomew, K. (1995). Accom-
modation and attachment representations
in couples. Journal of Social and Personal
Relationships 12: 389 401.
Senchak, M., & Leonard, K.E. (1992). Attach-
ment styles and marital adjustment among
newlywed couples. Journal of Social and
Personal Relationships 9: 51 64.
Shaver, P.R., Belsky, J., & Brennan, K.A.
(2000). The adult attachment interview and
self-reports of romantic attachment: Associ-
ations across domains and methods. Per-
sonal Relationships 7: 25 43.
Shaver, P.R., & Brennan, K.A. (1992). Attach-
ment styles and the Big Five personality
traits: Their connections with each other
and with romantic relationship outcomes.
Personality and Social Psychology Bulletin
18: 536 545.
Shaver, P.R., & Hazan, C. (1993). Adult ro-
mantic attachment: Theory and evidence
(pp. 29 70). In D. Perlman & W. Jones
(eds.), Advances in personal relationships
(Vol. 4). London: Jessica Kingsley.
Simpson, J.A. (1990). Influence of attachment
styles on romantic relationships. Journal of
Personality and Social Psychology 59: 871
980.
Simpson, J.A., Rholes, W.S., & Nelligan, J.S.
(1992). Support seeking and support giving
within couples in an anxiety-provoking sit-
uation: The role of attachment styles. Jour-
nal of Personality and Social Psychology 62:
434 446.
Simpson, J.A., Rholes, W.S., & Phillips, D.
(1996). Conflict in close relationships: An
attachment perspective. Journal of Person-
ality and Social Psychology 71: 899 914.
Sroufe, L.A., Carlson, E., & Shulman, S.
(1993). Individuals in relationships: Devel-
opment from infancy through adolescence
(pp. 315342). In D.C. Funder, R. Parke, C.
Tomlinson-Keesey, & K. Widaman (eds.),
Studying lives through time: Personality
432 / FAMILY PROCESS
and development. Washington DC: Ameri-
can Psychological Association.
Thelen, M.H., Sherman, M.D., & Borst, T.S.
(1998). Fear of intimacy and attachment
among rape survivors. Behavior Modifica-
tion 22: 108 116.
Tucker, J.S., & Anders, S.L. (1998). Adult at-
tachment style and nonverbal closeness in
dating couples. Journal of Nonverbal Behav-
ior 22: 124 109.
Tucker, J.S., & Anders, S.L. (1999). Attach-
ment style, interpersonal perception accu-
racy, and relationship satisfaction in dating
couples. Personality and Social Psychology
Bulletin 25: 403 412.
van IJzendoorn, M.H. (1995). Adult attach-
ment representations, parental responsive-
ness, and infant attachment: A meta-analy-
sis on the predictive validity of the Adult
Attachment Interview. Psychological Bulle-
tin 117: 387 403.
Volling, B.L., Notaro, P.C., & Larsen, J.J.
(1998). Adult attachment styles: Relations
with emotional well-being, marriage, and
parenting. Family Relations 47: 355367.
Wagner, B., & Reiss, D. (1995). Family sys-
tems and developmental psychopathology:
Courtship, marriage, or divorce? (pp. 696
730). In D. Cicchetti & D.J. Cohen (eds.)
(1995). Developmental psychopathology, Vol.
I: Theory and methods. New York: John
Wiley & Sons.
Waters, H.S., Rodrigues, L.M., & Ridgeway, D.
(1998). Cognitive underpinnings of narra-
tive attachment assessment. Journal of Ex-
perimental Child Psychology 71: 211234.
Weiss, R.S. (1998). A taxonomy of relation-
ships. Journal of Social and Personal Rela-
tionships 15: 671 683.
Whisman, M.A., & Allan, L.E. (1996). Attach-
ment and social cognition theories of roman-
tic relationships: Convergent or complemen-
tary perspectives? Journal of Social and
Personal Relationships 13: 263278.
Whitaker, D.J., Beach, S.R.H., Etherton, J.,
Wakefield, R., & Anderson, P.L. (1999). At-
tachment and expectations about future re-
lationships: Moderation by accessibility.
Personal Relationships 6: 4156.
Manuscript received August 2, 2001; revision
submitted and accepted January 24, 2002.
MIKULINCER, FLORIAN, P. COWAN, & C. COWAN / 433
APPENDIX

TABLE 1
Studies Demonstrating a Significant Attachment-Satisfaction Link in Dating Couples

Satisfaction
Authors Year N Sample Attachment Measure Measure
Levy & Davis 1988 391 Individuals Single Tripartite Items RRF
Hendrick & Hendrick 1989 391 Individuals Single Tripartite Items RRF
Pistole 1989 137 Individuals Tripartite categorization DAS
Collins & Read 1990 71 Couples Adult Attachment Scale DAS
Simpson 1990 144 Couples Adult Attachment 11-items scale
Questionnaire
Feeney & Noller 1991 116 Individuals Tripartite categorization Free reports
Shaver & Brennan* 1992 242 Individuals Tripartite categorization RRF
Feeney et al. 1993 193 Individuals Tripartite categorization Single item
Carnelley et al., Study 1 1994 163 Individuals Relationship Questionnaire; Single item
51-item attachment
measure
Feeney 1994 361 Couples 13-item Avoid/Anxiety Scale QMI
Keelan et al.* 1994 137 Individuals Tripartite categorization Investment Scale
Kirkpatrick & Davis* 1994 354 Couples Tripartite categorization RRF
Brennan & Shaver 1995 242 Individuals Tripartite categorization RRF
and Brennan et al. (1989)
scale
Scharfe & Bartholomew* 1995 64 Couples Peer Attachment Interview 7-items scale
Carnelley et al. 1996 52 Couples 48-item Avoid/Anxiety Scale Single Item
Frazier et al. 1996 83 Couples Adult Attachment Scale MRQ
Jones & Cunningham 1996 186 Couples 13-item Avoid/Anxiety Scale 6-item scale
Whisman & Allan 1996 68 Couples Adult Attachment Scale DAS
Morrison et al. 1997 385 Individuals Adult Attachment Scale MSI
Bookwala & Zdaniuk 1998 85 Individuals Relationship Questionnaire Simpson scale
Diehl et al. 1998 304 Individuals Relationship Questionnaire Family APGAR
Klohnen & Bera* 1998 100 Individuals Tripartite Categorization 3-items scale
Feeney 1999b 72 Couples 13-item Avoid/Anxiety Scale QMI
Feeney 1999c 238 Couples 13-item Avoid/Anxiety Scale QMI
McCarthy 1999 40 Individuals Single Tripartite items APFA
Tucker & Anders 1999 61 Couples Adult Attachment Simpson scale
Questionnaire
Collins & Feeney 2000 93 Couples Adult Attachment Scale 25-item scale
Cozarelli et al. 2000 112 Individuals Relationship Questionnaire Hendrick scale
Notes: * longitudinal design; APFA Adult Personal Functioning Assessment; DAS Dyadic Adjustment
Scale; QMI Quality Marriage Index; MRQ Multidimensional Relationship Questionnaire; MSI
Marital Satisfaction Inventory; RRF Relationship Rating Form.

Fam. Proc., Vol. 41, Fall, 2002

434 / FAMILY PROCESS

TABLE 2
Studies Demonstrating a Significant Attachment-Satisfaction Link in Married Couples

Satisfaction
Authors Year N Sample Attachment Measure Measure
Kobak & Hazan 1991 40 Couples Tripartite categorization DAS
Marital Q Sort
Senchak & Leonard 1992 322 Couples Tripartite categorization FAM
Cohn, Silver, et al. 1992 35 Couples Adult Attachment Interview DAS
Berman et al. 1994 18 Couples Attachment Style Inventory DAS
Carnelley et al., 1994 48 Individuals Relationship Questionnaire; DAS
Study 2 51-item attachment measure
Feeney et al.* 1994 35 Couples 13-item Closeness-comfort and QMI
relationship anxiety scales
Fuller & Fincham* 1995 53 Couples Tripartite categorization and MAT
Hazan & Shaver (1988) scale
Rholes et al. 1995 44 Individuals 13-item Avoid/Anxiety scale DAS
Feeney 1996 229 Couples 13-item Avoid/Anxiety scale QMI
Gerlsma et al. 1996 578 Individuals Tripartite Categorization RISS
Lussier et al. 1997 263 Couples 15-items Tripartite Scale DAS
Davila et al. 1998 289 Couples 13-item Closeness-comfort and MAT; Semantic
relationship anxiety scales Differential
Mikulincer et al. 1998 80 Couples Tripartite Categorization DAS
Volling et al. 1998 62 Couples Tripartite Categorization Single Item
Davila et al.* 1999 172 Couples Adult Attachment Scale MAT
Notes: * longitudinal design. DAS Dyadic Adjustment Scale; FAM Family Assessment Measure;
QMI Quality Marriage Index; MAT Marital Adjustment Test; RISS Relational Interaction Satisfac-
tion Scale.