Graduate Theses and Dissertations
Iowa State University Capstones, Theses and
Dissertations
2017
Considering the Role of Relationship-Contingent
Self-Esteem: Attachment Style, Conflict Behaviors,
and Relationship Satisfaction
Amanda Katherine Buduris
Iowa State University
Follow this and additional works at: https://lib.dr.iastate.edu/etd
Part of the Counseling Psychology Commons, and the Social Psychology Commons
This Thesis is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital
Repository. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Iowa State University Digital
Repository. For more information, please contact digirep@iastate.edu.
Recommended Citation
Buduris, Amanda Katherine, “Considering the Role of Relationship-Contingent Self-Esteem: Attachment Style, Conflict Behaviors,
and Relationship Satisfaction” (2017). Graduate Theses and Dissertations. 15268.
https://lib.dr.iastate.edu/etd/15268
Considering the role of relationship-contingent self-esteem: Attachment style, conflict
behaviors, and relationship satisfaction
by
Amanda K. Buduris
A thesis submitted to the graduate faculty
in partial fulfillment of the requirement for the degree of
MASTER OF SCIENCE
Major: Psychology
Program of Study Committee:
Loreto Prieto, Major Professor
Meifen Wei
Carolyn Cutrona-Russell
The student author and the program study committee are solely responsible for the
content of this thesis. The Graduate College will ensure this thesis is globally accessible
and will not permit alterations after a degree is conferred.
Iowa State University
Ames, Iowa
2017
ii
TABLE OF CONTENTS
Page
LIST OF TABLES
…………………………………………………………………………………………………………. iv
ACKNOWLEDGEMENTS
……………………………………………………………………………………………….v
ABSTRACT
………………………………………………………………………………………………………………….. vi
CHAPTER 1: INTRODUCTION ……………………………………………………………………………………….1
CHAPTER 2: LITERATURE REVIEW ……………………………………………………………………………..9
Attachment
………………………………………………………………………………………………………….10
Relationship-Contingent Self-Esteem
……………………………………………………………………..25
Conflict Behaviors ……………………………………………………………………………………………….33
Relationship Satisfaction ………………………………………………………………………………………39
Attachment, Relationship-Contingent Self-Esteem, Conflict Behaviors,
& Relationship Satisfaction
…………………………………………………………………………….41
The Present Study ………………………………………………………………………………………………..51
Research Hypotheses ……………………………………………………………………………………………54
CHAPTER 3: METHOD …………………………………………………………………………………………………56
Participants
………………………………………………………………………………………………………….56
Procedure ……………………………………………………………………………………………………………58
Measures and Materials ………………………………………………………………………………………..59
CHAPTER 4: RESULTS
…………………………………………………………………………………………………64
CHAPTER 5: DISCUSSION
……………………………………………………………………………………………74
Attachment
………………………………………………………………………………………………………….74
Relationship-Contingent Self-Esteem
……………………………………………………………………..77
Conflict Behaviors ……………………………………………………………………………………………….78
Relationship Satisfaction ………………………………………………………………………………………79
Limitations ………………………………………………………………………………………………………….82
Future Research …………………………………………………………………………………………………..85
Implications for Practice ……………………………………………………………………………………….87
REFERENCES ………………………………………………………………………………………………………………88
APPENDIX A. BARTHOLOMEW & HOROWITZ’S (1991) MODEL OF
SELF AND OTHERS ………………………………………………………………………………………….101
APPENDIX B. IRB APPROVAL …………………………………………………………………………………..102
iii
APPENDIX C. INFORMED CONSENT
…………………………………………………………………………103
APPENDIX D. DEMOGRAPHICS & RELATIONSHIP HISTORY QUESTIONNAIRE …….105
APPENDIX E. CHOOSE YOUR OWN ADVENTURE TASK – EXAMPLES OF
QUESTIONS USED
………………………………………………………………………………………….106
APPENDIX F. EXPERIENCES IN CLOSE RELATIONSHIP – SHORT FORM
………………..107
APPENDIX G. RELATIONSHIP-CONTINGENT SELF-ESTEEM
…………………………………..108
APPENDIX H. RELATIONSHIP SATISFACTION
…………………………………………………………109
iv
LIST OF TABLES
Page
Table 1. Participant Demographics
……………………………………………………………………………………..57
Table 2. Sample Means, Standard Deviations, and Ranges for Study Measures ……………………….66
Table 3. Correlations Among Measures ………………………………………………………………………………67
Table 4. Partial Correlations Among Measures and Insecure Attachment
………………………………..68
Table 5. RCSE as a Moderator of Pure Anxious Attachment and Detrimental
Decision Endorsement ……………………………………………………………………………………………69
Table 6. RCSE as a Moderator of Pure Anxious Attachment and Relationship Satisfaction
……….70
Table 7. RCSE as a Mediator of Pure Anxious Attachment and Detrimental
Decision Endorsement ……………………………………………………………………………………………71
Table 8. RCSE as a Mediator of Pure Anxious Attachment and Relationship Satisfaction
…………71
Table 9. ANCOVA – Sex Differences in Pure Anxious Attachment Scores …………………………….72
Table 10. ANOVA – Sex Differences in RCSE Scores………………………………………………………….72
Table 11. ANOVA – Sex Differences in RS Scores
………………………………………………………………73
Table 12. ANOVA – Sex Differences in CYOA Scores ………………………………………………………..73
v
ACKNOWLEDGMENTS
I would like to express my thanks to Dr. Loreto Prieto, my graduate advisor and committee chair,
for his unending support and patience throughout this research project. His continual
encouragement and belief in my ability to complete my thesis were so appreciated.
I would also like to thank Dr. Meifen Wei and Dr. Carolyn Cutrona, my committee members, for
agreeing to be on my committee and for their excellent advice and willingness to answer
questions and generate solutions.
Lastly, I would like to express my gratitude to my family. To my mother, for always believing
that I could do whatever I set my mind to. To my partner, whose support, love, and reminders for
self-care were endless and helped me make it through this process.
vi
ABSTRACT
Research has empirically demonstrated that adult attachment style directly affects
communication between partners in romantic relationships, in particular, how partners address
and handle conflicts in the relationship. The construct of relationship-contingent self-esteem
(RCSE) suggests that individuals behave differently in romantic relationships based on the
degree to which their personal self-esteem is connected to successes or failures in that
relationship. However, few studies have examined how these two independent constructs are
related to one another in terms of relationship communication behaviors during conflict. The
purpose of my study was to examine the moderating and mediating effect of RCSE on the
relation between attachment styles in romantic relationships and conflict behaviors in romantic
relationships, as well as its moderating and mediating effect on the relation between attachment
styles and relationship satisfaction. Results indicated that anxious attachment scores were
significantly positively correlated with RCSE, anxious and avoidant attachment characteristics
were significantly negatively correlated with relationship satisfaction, and that RCSE neither had
a moderating or mediating effect on the relations between pure anxious attachment and number
of detrimental decisions endorsed during conflict scenarios, or level of relationship satisfaction.
Implications for continued examination of role that RCSE plays in the context of conflict in
romantic relationships, as well as implications for clinical work are discussed.
Keywords: adult attachment; relationship-contingent self-esteem; conflict behaviors in
romantic relationships; relationship satisfaction
1
CHAPTER 1. INTRODUCTION
The way in which we think of ourselves depends on our social relationships with others.
Our sense of identity is constructed within the context of our social roles, particularly our key
interactive relationships (Deaux, Reid, Mizrahi, & Ethier, 1995). Of particular interest to
clinicians and researchers are romantic relationships, as individuals think, feel, and act in
significantly different ways in romantic relationships than they do in other social relationships
(Knee, Canevello, Bush, & Cook, 2008). For example, some individuals in romantic
relationships cling to significant others who are abusive to them (Bartholomew, Henderson, &
Dutton, 2001; Lesser, 1990; Steinmetz, 1977), while they would not do so if treated this way in
other social relationships. Though not all individuals put themselves in these kinds of positions,
such observations demonstrate that our need to connect with others at times leads us to act in
ways that may not be adaptive, and that these needs for connection may in fact be more
important to some individuals than their emotional safety or well being. One well established
way in which behavior in romantic relationships has been conceptualized is through the
theoretical lens of attachment theory; this theory has been cited as a key way in which individual
differences in behavior within familial, platonic, and romantic relationships can be understood
(Bowlby, 1969; 1973; 1979; 1983; 1988; Hazan & Shaver, 1987; 1990; 1994).
In its initial development, attachment theory described relational styles present for infants
in relation to their caregivers (typically mothers). Ainsworth, Blehar, Waters, and Wall (1978)
termed these styles as secure, anxious/ambivalent, and avoidant. A secure attachment is
characterized by minimal distress when separated from the caregiver, as well as low levels of
anxiety when exploring the environment independently. Infants with anxious/ambivalent or
avoidant attachments to caregivers (what are generally known as insecure attachments) are
2
highly distressed or withdrawn when separated from a caregiver, and are hesitant to or fail to
explore environments on their own. Some individuals (especially those with anxious/ambivalent
attachment styles), due to evolving maladaptive beliefs about joining with others and fear of
environments (Barber & Buehler, 1996), seek enmeshment with their attachment figures.
Enmeshment refers to a lack of individuation between those in a relationship. For those whose
relationships are characterized by high levels of enmeshment, removal (real or imagined) from
that relationship serves as a large threat to their sense of self and sense of safety.
Those with more avoidant styles of attachment seek a self-protective distance with
attachment figures and relationships, as a way of coping with the threat of losing the presence of
their caregiver, despite experiencing a strong need for their attachment figure. Avoidance refers
to an underlying resistance to close bonding in a relationship, due to perceptions that regard
caregivers are unreliable in their presence and attention. For those acting from an avoidant
position, behaviorally they appear self-protective and evince a nonplussed presentation under
threat of separation and also appear to overtly devalue their relationship with the attachment
figure (Evraire, Ludmer, & Dozois, 2014). These different attachment styles are the result of
cognitive schemas that individuals have learned throughout their infant and childhood
relationships with their caregivers that continue to shape their beliefs into adulthood regarding
their desire for closeness, self-protection, and coping style to retain partners in relationships.
A second key conceptualization that aids clinicians and researchers in better
understanding the behavior of individuals within romantic relationships concerns relationship-
contingent self-esteem (RCSE; Knee et al., 2008). The literature surrounding general contingent-
self-esteem (Crocker & Wolfe, 2001) has demonstrated that individuals highly invest their sense
of self-esteem and sense of worth into particular domains of their lives (e.g., academics, work,
3
religion, family, romantic relationships) so that these individuals become significantly more
sensitive to successes and failures within these specific domains. Scholars have outlined how
RCSE and the dynamic of self-esteem investment occurs within romantic relationships; that is,
those who have high levels of RCSE are more vulnerable to negative or threatening events
within their romantic relationships, and are more positively impacted by affirming or positive
events (Knee et al., 2008). High levels of RCSE bring individuals to maintain their romantic
relationships because their core self-esteem is contingent upon its success (Crocker & Park,
2004); failure in this core relationship would create a severe blow to the low-self esteem of such
an individual. RCSE likely acts as a contributor to differences in how individuals think, feel, and
act in their romantic relationships, and likely has some interaction with the underlying
attachment style regarding how they behave toward their romantic partners.
Previous studies have provided insight into the ways in which those with insecure
attachments and high levels of RCSE behave maladaptively within their romantic relationships.
For instance, those with insecure attachments typically “shut down” in the face of conflict, and
try to avoid conflict all together (Creasey & Hesson-McInnis, 2001; Shi, 2003; Dominigue &
Mollen, 2009) because of the negative implications conflict might pose to the stability of their
relationship. As well, those with high levels of RCSE are typically more distressed by
relationship conflict; for example, they may be more likely to engage in unwanted pursuit
behaviors after a relationship has ended (DiBello, Rodriguez, Hadden, & Neighbors, 2015; Park,
Sanchez, & Brynildsen, 2011; Rodriguez, Knee, & Neighbors, 2014). These findings illustrate
how individuals with insecure attachment styles and high RCSE do not manage conflict well,
and serves as a point of interest, as these response tendencies give insight into how clinicians can
work with individuals who present with relationship concerns. Examining and working with
4
conflict behaviors, and promoting more adaptive conflict behaviors, serves as a meaningful
therapeutic goal.
Finally, how individuals view the stability of their relationships and how they deal with
conflict may significantly impact the degree to which they are happy or satisfied in those
relationships (Heavey, Christensen, & Malamuth, 1995). Relationship satisfaction serves as
another critical variable to examine within the interaction of attachment style, communication
during conflict in romantic relationships, and RCSE. How satisfied individuals are with their
romantic relationship influences their mental and physical health. For example, those who are
happy with, and satisfied in, their relationships are typically more mentally and physically
healthy and are less severely impacted by stress (Gove, Hughes, & Style, 1983; Holt-Lunstad,
Birmingham, & Jones, 2008; Kolves, Ide, & De Leo, 2012), whereas individuals with lower
satisfaction in their relationships, including those who are separated and divorced, are more
likely to endorse suicidal ideation, hopelessness, and depression, and attempt suicide (Stack,
1990; Wyder, Ward, & De Leo, 2009; Batterham, Fairweather-Schmidt, Butterworth, Calear,
Mackinnon, & Christensen, 2014; Till, Tran, & Niederkrotenthaler, 2016). In fact, the
relationship status of divorce has been found to be a strong predictor of suicide rates (Stack,
1992). Examining individuals’ level of relationship satisfaction in their current relationships is
critical, as it lends insight into what individuals might be doing in the relationship to promote
increased relationship satisfaction, and highlights the differentiation between relationship
satisfaction as a primary goal, versus simply being in a relationship as a primary goal.
In my study, I will examine this cluster of constructs (attachment, RCSE, conflict
behaviors, and relationship satisfaction), and how these variables interact with one another to
influence behaviors within romantic relationships. Previous literature has already demonstrated
5
how attachment style and RCSE are related to maladaptive behaviors within relationships (De
Smet, Uzieblo, Loeys, Buysse, & Onraedt, 2015; Knee et al., 2008; Park et al., 2011). Yet, few
studies have empirically examined how RCSE is related to relationship behaviors above and
beyond the influence of attachment style. My contribution to the literature will be to specifically
examine the moderating role of RCSE on the relation between attachment style and conflict
behaviors as well as the effect these variables have on relationship satisfaction.
Extant Research in the Area
The literature to date regarding attachment styles, relationship-contingent self-esteem,
conflict behaviors, and relationship satisfaction has, to a degree, examined the interplay of these
variables within the dynamics of an adult romantic relationships. For example, these variables
can affect conflict styles employed by those with secure vs. insecure attachment styles (Shi,
2003; Domingue & Mollen, 2009; Creasey & Hesson-McInnis, 2001); the coping behaviors used
by those with high relationship-contingent self-esteem (DiBello et al., 2015); and, adaptive or
maladaptive post-break up behaviors (De Smet et al., 2015), thoughts (Brenner & Vogel, 2015),
and growth (Marshall, Bejanyan, & Ferenczi, 2013), dependent upon attachment tendencies and
level of RCSE.
To date, most studies examining conflict dynamics in romantic relationships have
operated under the assumption that participants can self-report on imagined behaviors (e.g,
Collins, 1996) in response to relationship conflict, but this method may not be the most effective
means by which to accurately capture decisions made during real life conflict, because conflict,
by nature, is not a static entity. In addition, there is likely a qualitative difference between how
research participants imagine they will respond to conflict in a given static conflict situation
versus how they (re)act to an iterative set of changing and response-contingent communication
6
stimuli in an ongoing conflict situation (Axelrod, 1984; Rusbult, Verette, Whitney, Slovik, &
Lipkus, 1991). By utilizing a forward moving and stimulus specific methodology in which
participants find their (re)actions have direct consequences on their partners’ return responses in
a conflict, a more accurate picture of real life conflict dynamics may reveal itself (Vicary &
Fraley, 2007).
In summary, attachment style has frequently been referenced as a primary explanation for
differences in conflict dynamics in relationships (Shi, 2003; Domingue & Mollen, 2009; Creasey
& Hesson-McInnis, 2001; Ben-Naim, Hirschberger, Ein-Dor, & Mikulincer, 2013), but RCSE,
an equally important relationship variable, has not been well studied as to its impact with these
variables. Studying these two constructs in tandem, as well as their influence on the nature of
conflict responses and relationship satisfaction, aids in a more thorough understanding of
individual differences in communication patterns within romantic relationships.
The Present Study
Theoretically and empirically, the existing evidence within the attachment and RCSE
literature suggests that those who are more insecure in their attachments in romantic
relationships will behave in maladaptive ways within their romantic relationships, especially
during the course of conflict. Those who are more secure in their relationships are likely better
able to manage the conflict present within relationships, and generally have more confidence in
themselves and their partners to adaptively resolve conflict. However, the relations among
attachment, responding within relationship conflict, and relationship satisfaction are not yet
known.
The findings within the literature on relationship-contingent self-esteem (RCSE) suggest
that the extent to which individuals’ core self-esteem depends on the successes and failures of
7
their romantic relationships can influence the manner in which they handle conflict and how they
evaluate their relationships overall. Although research findings show an association between
attachment styles and levels of contingent self-esteem (Knee et al., 2008; Park et al., 2011), the
assumption that individuals with an anxious or avoidant attachment style will endorse high levels
RCSE (and that those with a secure attachment style will endorse low levels of RCSE) has not
yet been demonstrated. RCSE could conceivably play a stronger role in negative emotions felt
during relationships, as the potential loss of a relationship partner immediately represents (and
RCSE is conceptually related to) a steep decrease in self-esteem and self-worth if a relationship
is in jeopardy (such as in conflict situations). Given this, I will investigate the potential
mediating and moderating role of relationship-contingent self-esteem on the relation between
attachment style and response decisions made during conflict, and the relations among
attachment style, RCSE, and relationship satisfaction.
The findings in the literature concerning the effects of an avoidant attachment on
behaviors in romantic relationship is inconsistent (Ainsworth & Bell, 1970; Ainsworth et al.,
1978; Main, 1979; Hazan & Shaver, 1987; Bartholomew & Horowitz, 1991; Creasey & Hesson-
McInnis, 2001); however, research findings concerning anxious attachments in this area are more
definitive and consistent. For example, anxiously attached individuals would be expected to
endorse higher levels of RCSE and experience a greater degree of distress during relationship
conflict (Creasey & Hesson-McInnis, 2001; Knee et al., 2008; Domingue & Mollen, 2009; Ben-
Naim et al., 2013). However, those endorsing high levels of an avoidant attachment style, may or
may not possess similar internal reactive states, given that their coping behaviors would seem to
indicate less perceived risk to esteem (e.g., Creasey & Hesson-McInnis, 2001). Given this, I will
8
employ avoidant attachment tendencies in participants as a covariate of anxious attachment
endorsements, generating an overall insecure attachment style score.
Importance of Present Study
The extent to which attachment styles and relationship-contingent self-esteem together
can influence how individuals behave in their relationships, as well as how satisfied they are with
their relationships, has not been well examined to date. Further, no studies to date have examined
RCSE in the context of continuous iterative conflict communications. My study adds a broader
understanding of the interplay of these variables to the attachment literature, and provide further
information that can be used by clinicians and researchers to understanding the process of
conflict in romantic relationships. Assessing the mediating or moderating role played by RCSE
on the relation between attachment style and conflict communication tendencies and relationship
satisfaction, sheds light on why some individuals handle challenges and threats to their
relationships in more productive ways than others, and generates implications for psychotherapy
with clientele dealing with conflict oriented difficulties surrounding romantic relationship issues.
9
CHAPTER 2. LITERATURE REVIEW
One central focus of research in the area of romantic relationships has examined why and
how conflict affects individuals differently (Domingue & Mollen, 2009; Creasey & Hesson-
McInnis, 2001; Shi, 2003). For example, some individuals in romantic relationships embrace
conflict and view it as relationship-enhancing, while others may see the slightest disagreement as
a sign that the relationship will fail (Domingue & Mollen, 2009). In addition, the degree to which
romantic partnership affects individuals’ self-esteem and self-worth is another key factor
affecting behavior in romantic relationships (Knee et al., 2008; Mikulincer & Shaver, 2003;
Mikulincer & Shaver, 2007), such as how they handle conflict (DiBello et al., 2015) or the
termination of a relationship (Park et al., 2011). Finally, how a couple manages conflict (i.e.,
constructively or destructively) can influence how satisfied one or both partners are in the
relationship (Heavey et al., 1995; McGinn, McFarland, & Christensen, 2009). Relationship
satisfaction has been shown to have substantial impact on both physical and mental health (Gove
et al., 1983; Kolves et al., 2012), demonstrating its importance in the conflict and romantic
relationship literature.
These three constructs – how people attach themselves to their romantic partners, the
degree to which people make their self-esteem contingent upon their romantic relationships, and
how these elements of their relationship affect their satisfaction with their romantic relationships
– are important for applied psychologists to understand, especially with respect to how their
clients experience conflict in their romantic relationships. Specifically, examining the relations
of these constructs can help to clarify for psychologists the implications that differential
attachment styles, levels of relationship contingent self-esteem, and pattern of communication in
relationship conflicts have for clients’ adjustment, growth, and future behavior in their romantic
10
relationships. Examining the relations of these three variables is the focus of my study. In the
next sections, I will detail these variables, what past research indicates as far as their effects
within romantic relationships, and what I will be doing specifically to explore the inter-relations
among these variables. Next, I will review the theory of attachment style and models of self and
others, and how those relate to our behaviors in relationships.
Attachment
Attachment as an action within relationships has been defined as a “lasting psychological
connectedness between human beings” (Bowlby, 1969, p.194). The creator of attachment theory,
John Bowlby (1982), referred to attachment behavior as “any form of behavior that results in a
person attaining or maintaining proximity to some other clearly identified individual who is
conceived as better able to cope with the world” (p. 668). Attachment was originally studied
within the context of infants’ connectedness with their caregivers, typically their mothers
(Bowlby, 1969).
Infant attachment. Bowlby (1969) first investigated the attachment construct within
infant-mother relationships. The classic Ainsworth et al. (1978) “strange situation” experiment
helped to clearly differentiate the behaviors of infants with different attachment styles. When
separated from their mothers, infants reacted based on how they currently attached to that
attachment figure, where behaviors ranged from comfortable exploring to constant crying.
Bowlby (1988) later described four characteristics of attachment. Proximity maintenance refers
to the desire to be near the people to whom we are attached. We see our attachment figures as
safe havens, and therefore return to them for comfort and safety in the face of a fear or threat.
The attachment figure also acts as a base of security from which the child can explore the
11
surrounding environment. And finally, separation distress refers to anxiety that occurs in the
absence of the attachment figure.
Based on these characteristics, styles of infant and childhood attachment have been coded
into four different categories: secure, anxious-resistant, anxious-avoidant, and disorganized/
disoriented (Ainsworth & Bell, 1970; Ainsworth et al., 1978). Essentially, a securely attached
infant is able to confidently explore its environment freely and engage with strangers in a non-
fearful way when the attachment figure (caregiver) is present. Although securely attached infants
become visibly upset when their attachment figure leaves, these infants are soothed easily and
become happy again when their attachment figure returns. In general, those infants with a secure
attachment are presumed, given behavioral and emotional evidence, to experience the attachment
figure as a ‘secure base’: a caregiver that is consistently available to them and will be there as an
anchor for them from which they explore the world. Research shows that caring, consistently
responding, and attentive caregivers best promote the development of secure attachments within
infants (Dunst & Kassow, 2008; Aronoff, 2012). Overall, the development of a secure
attachment in children is seen as the most adaptive and mentally healthy attachment style for
them to acquire (Ainsworth et al., 1978). The remaining three styles are classified under the
umbrella of insecure attachments and represent less than optimal resolutions of infants’
experiences with, and trust of, caregivers as a stable base of safety and support from which they
can explore the world.
Anxious-resistant attachment (also referred to as ambivalent or dismissing attachment;
Ainsworth et al., 1978) is an insecure attachment style. In the presence of the attachment figure,
an anxious-resistant infant is significantly hesitant to leave the caregiver and explore the
environment and is uneasy, suspicious and fearful around strangers. When the attachment figure
12
leaves, the anxious-resistant infant becomes highly distressed. Some of the infant’s distress
remains upon the return of the attachment figure, resolving only to a worried, unsure sense of
safety and support (versus the confident and consistent sense of support that the securely
attached child demonstrates). An anxious-resistant attachment is believed to develop in response
to a caretaker that is inconsistent in responding in a safe and supportive way toward the infant.
As such, the infant develops a sense of ambivalence toward the caregiver, who sometimes is a
source of nurturance, yet, simultaneously, is not a stable source of safety, support, and comfort
upon which the child can rely (Crittenden, 1999).
A second insecure attachment style has been labeled as an anxious-avoidant type. This
style of attachment is characterized by a child who, when the caregiver is present, generally
avoids or ignores the attachment figure, showing little emotion at all when the attachment figure
leaves or returns (Ainsworth et al., 1978). Ainsworth and Bell (1970) suggested that those infants
with an anxious-avoidant attachments act in the observed indifferent and unemotional manner in
order to cope with and mask their internal distress surrounding the instability of their bond with
their caregiver. Those who embody an anxious-avoidant attachment likely due to a history of
rejection from caretakers in response to expressing attachment needs (Main, 1979). Avoidant
behavior allows these individuals to maintain proximity to their attachment figures while
avoiding the potential for rejection.
Finally, the fourth category, one which has been less well empirically supported and
accepted, is the disorganized/disoriented attachment style. This attachment is characterized by a
general lack of consistency in infants’ reactions to the attachment figure and their departure and
return (Ainsworth et al., 1978). These infants are described as displaying overtly contradictory
behaviors and emotions, random movements, and even periods of disengagement from
13
environmental stimuli. Scholars have criticized this category, because its descriptors are too
encompassing, it does not possess sufficient unique and discriminating characteristics, and it
overlaps substantially with characteristics associated with the anxious attachment style
(Crittenden, 1999). The disorganized/disoriented style is not considered as valid as the other
attachment styles, and in much of the current literature, including attachment measures, this style
is not taken into account.
Baumeister and Leary (1995) and Park (2006) have referred to our need for interpersonal
attachments as a fundamental human motivation; as we seek to relate to others because of our
social needs, we form psychological and emotional attachments to them, in both real and
symbolic ways. The attachment formed between an infant and a caregiver, and how that
relationship develops and functions, has implications for infants’ future relationships, as these
initial, highly impactful relational patterns form our expectations and reactive behaviors in later
relationships (Sund & Wichstrøm, 2002). Bowlby’s attachment theory (1969) proposed that
working models obtained in childhood would continue across development and relationships. As
children grow and distance from caregivers naturally evolves, future relationships (particularly
romantic ones) begin to serve as our primary source for love and psychological support.
Romantic relationships are particularly important to investigate as individuals (for the
most part) chose those with whom they form romantic relationships. Investigators have shown
that attachment plays a major part in our relationships as we mature, particularly within romantic
relationships (Hazan & Shaver, 1987; 1990; 1994). Much like a caregiver provides love and
security for an infant, a romantic partner provides love and security from which individuals
receive psychological support and emotional comfort. Choice of partners may reveal much about
how an individual attaches psychologically to others (Frazier, Byer, Fischer, Wright, & DeBord,
14
1996; Latty-Mann & Davis, 1996; Chappell & Davis, 1998; Collins, Cooper, Albino, & Allard,
2002). As such, examining the attachment styles of adult individuals with respect to their
romantic relationships can be very informative and has clinical implications for treatment of
potentially maladaptive behaviors within their relationships. Next, I will detail how the early
attachment style of children evolves into those held later in life as an adult.
Adult attachment. Hazan and Shaver (1987; 1990) were among the first researchers to
suggest the attachment framework could be extended from the infant-mother relationship to
romantic relationships. They noted similarities between infant/caregiver relationships and adult
romantic partners. Within both kinds of relationships, partners feel safe when the other is close
and responsive; partners engage in close, intimate, bodily contact; partners have emotional
reactions when separating from and rejoining one another; and, partners even engage in “baby
talk”, mimicking earlier patterns of communication heard from childhood caregivers. Like
Bowlby described the central characteristics of infant-mother attachment (1988), Fraley and
Shaver (2000) described the central tenets of adult attachment as they relate to childhood
attachment. First, the emotional and behavioral dynamics of the two relationships are governed
by the same biological system. Next, the kinds of individual differences expressed via attachment
behavior are similar across the two types of relationships. Further, the individual differences in
adult attachment behavior are reflections of people’s experiences in close relationships (e.g.,
relationship with parents). Finally, a romantic relationship involves an interplay of attachment,
caregiving, and intimacy.
When applied to adult romantic relationships, attachment categories are labeled as
secure, anxious-preoccupied, dismissive-avoidant, and fearful-avoidant (Bartholomew &
Horowitz, 1991; Pietromonaco & Barrett, 1997). In comparison to childhood attachment, this
15
categorization finds childhood anxiety-related styles melded into a single dimension, and
avoidant styles instead broken down into two dimensions. Relatedly, much of the adult
attachment literature, and many common adult attachment scales employed in research, combine
the dismissive-avoidant and fearful-avoidant into a single “avoidant” category given their
underlying commonality surrounding individuals’ psychological needs resulting in two different
behavioral expressions for avoidant behavior. In fact, the main body of adult attachment research
posits two main dimensions on which individual differences can be assessed: attachment anxiety
and attachment avoidance (Brennan, Clark, & Shaver, 1998; Cassidy & Kobak, 1988; Mikulincer
& Shaver, 2003; 2007). For this reason, many of the scales measuring attachment focus on
measuring anxiety and avoidance, where low scores on both scales constitute secure attachment.
The anxiety dimension is characterized by a desire for closeness and protection, intense worries
about partner availability and the value one has to the partner, and use of hyperactivating
strategies (e.g. hyper-vigilance of potential threats to relationship, exaggerated appraisal of
threats, rumination over past threatening experiences, excessive reassurance seeking) in attempts
to avoid or cope with insecurity and distress. The avoidance dimension, on the other hand, is
characterized by a discomfort with closeness, lack of trust in using romantic partners as a secure
base, a desire to be emotionally distant and self-reliant, and use of deactivating strategies (e.g.
lack of monitoring of potential threats, inhibition/suppression of threat-related thoughts) to cope
with insecurity and distress (Mikulincer & Shaver, 2007). Regardless, for clarity of the
development of adult attachment research, I will briefly review the four categories below.
A secure adult attachment is viewed as an extension of a secure childhood attachment.
As with infant attachment, secure attachment is promoted by having a caregiver who is
emotionally available, responsive to need behaviors, and capable of regulating his/her own
16
positive and negative emotions (Sable, 2008). In adulthood, individuals with secure attachments
have largely consistently warm and responsive interactions with their romantic partners. Those
with secure adult attachments typically have a positive view of themselves, their partners, and
their relationships (Bartholomew & Horowitz, 1991). They far less frequently feel or behave in
ways similar to those with insecure attachment styles (Collins & Read, 1990; Feeney & Noller,
1990).
Anxious-preoccupied attachments are characterized by individuals having consistently a
high need to receive intimacy, approval, and responsiveness from their romantic partners
(Bartholomew & Horowitz, 1991), beyond what those who are securely attached would
demonstrate. Because of the strength of these desires, someone with an anxious-preoccupied
attachment may become overly dependent on a partner, primarily driven by the negative self-
views possessed by those who have this style of attachment. Those with preoccupied attachments
strive for acceptance by valued others (Bartholomew & Horowitz, 1991; Hazan & Shaver, 1987;
Main, Kaplan, & Cassidy, 1985). Characteristics of the preoccupied group include high levels of:
depth (vs. superficial) of descriptions of self and other, self-disclosure, emotional expressiveness,
frequency of crying, reliance on others, crying in front of others, high levels of relationship
involvement, and low levels of coherence and balance of control in friendships (Bartholomew &
Horowitz, 1991). Individuals who are attached in an anxious-preoccupied manner tend to see
themselves as less than worthy partners, and blame themselves when their relationships do not
function well (Mikulincer & Shaver, 2007). On the whole, people in this attachment group are
highly emotionally expressive, and spend a lot of time worrying and acting impulsively in their
relationships in order to gain the soothing they require to feel stable and secure in themselves and
their relationships (Bartholomew & Horowitz, 1991; Mikulincer & Shaver, 2007).
17
Adults with avoidant attachments have a fundamentally different perspective and set of
psychological and relational needs. Those with dismissive-avoidant attachments highly value
their personal independence in romantic relationships and do not have an easy time trusting or
forming deep attachments with others (Bartholomew & Horowitz, 1991). Characteristics of the
dismissive group include high self-confidence and low emotional expressiveness; low frequency
of crying and warmth (compared to other styles); low self-disclosure, intimacy, level of
involvement in relationships, and capacity to rely on others; and, being more in control than
counterparts in both friendships and romantic relationships (Bartholomew & Horowitz, 1991). In
their relationships, they tend to exhibit less emotional intimacy, as they view their partners less
positively than they view themselves. Those with a dismissive-avoidant attachment typically
suppress or hide their feelings of normal emotional need and bonding out of a lack of trust and
valuing of their partners (Creasey & Hesson-McInnis, 2001). This lack of trust and valuing
results from their long standing (and previously learned) deep-seeded sense of unworthiness:
when they are faced with rejection from a caregiver, often due to their overwhelming demands
and needs, they tended to distance themselves to cope with the loss (Ainsworth et al., 1978;
Main, 1979).
Individuals with fearful-avoidant attachments, on the other hand, tend to be distant in, or
abstain from, relationships for different reasons than dismissive-avoidant individuals. These
individuals have mixed feelings about close relationships. They want to be emotionally close
with someone, but they are also uncomfortable with feeling emotionally close due to their
negative views of themselves and others (Hazan & Shaver, 1987; Bartholomew & Horowitz,
1991). Characteristics of the fearful group include low self-confidence and balance of control in
friend and romantic relationships, and low self-disclosure, intimacy, level of relationship
18
involvement, and reliance on others (Bartholomew & Horowitz, 1991). Similar to those with a
dismissive-avoidant attachment, individuals with fearful-avoidant attachments do not seek high
levels of intimacy in relationships and do not express affection often. As well, individuals with
fearful-avoidant attachments have core psychological concerns surrounding their own
unworthiness, driven by their anticipation that their partners will ultimately disappoint them, be
inconsistent in their care, and will fail to care for them because of the core belief of being
unworthy of such love. For both dismissive and fearful avoidant attachments, the avoidant
attachment behavior helps protect them against potential disappointment (Bartholomew &
Horowitz, 1991), but brings difficulty with being able to connect with friends or romantic
partners.
Though infant and adult attachment styles may appear similar in many ways across the
four respective categories, there are mixed thoughts about whether individuals’ attachment style
as an infant determines their adult attachment. The basis of attachment theory is that the
attachment system is organized by early experiences with caregivers, which then shape and
sustain working models of self and others (Bowlby, 1973). This is consistent with Piaget’s
(1953) theory of cognitive development, where he argued that people assimilate new information
to existing knowledge structure. If we develop an insecure attachment in infancy, throughout our
lives, we may seek to confirm our negative models of self and others. Bowlby (1973) also argued
that we attract relationship partners who fit our working models as a means to remain congruent
with our attachment experiences and models. However, Bowlby (1969; 1982) also noted that
attachment working models reflect actual experiences in relationships, and so are subject to
revision. People with a secure attachment who find their partners have engaged in infidelity may
develop a measure of insecurity as a new part of their working model, and people with insecure
