Attachment Theory

Full text

A
Attachment Theory
Elaine Scharfe
Trent University, Peterborough, ON, Canada
Definition
Bowlby’s theory of attachment includes several
important foundational constructs. First and fore-
most, attachment relationships are clearly pre-
sented as a biological predisposition evolved to
ensure survival. Individuals are proposed to
develop attachment relationships with caregivers
and seek and maintain proximity to these care-
givers when stressed, ill, or afraid. Differences in
sensitivity of care were proposed to be associated
with individual differences in attachment.
Although Bowlby was particularly interested in
the parent-child relationship, he was clear that
attachment representations would be important
for relationship functioning from “cradle to
grave.”Infants tend to develop a primary attach-
ment with their predominant caregiver; however,
infants can develop multiple attachment relation-
ships, and, throughout childhood and adulthood,
we organize these attachment relationships into a
hierarchy. And finally, Bowlby proposed that once
formed, attachment representations would be
stable –in particular in adulthood –although
attachment representations could change in
response to changes in caregiving during
childhood or traumatic or salient events during
adulthood.
Early Years
The roots of attachment theory were first
established in the 1930s and 1940s when a num-
ber of clinicians observed the negative effects of
maternal separations early in life. John Bowlby
was one of many who observed this effect; for
example, in 1944 he outlined how poor parenting
influenced the behavior of a group of juvenile
thieves. It was not until 1950, when he was
offered a short-term contract to work at the
World Health Organization (WHO), that he had
the opportunity to organize his and others’work in
the area. The resulting report from his WHO con-
tract, published in 1951, summarized what was
known to date about the negative effect of poor
caregiving and maternal separation on infants’
health and well-being. Bowlby later reported that
several reviewers pointed out that an overarching
theory to explain these observations was missing.
In Bowlby’s quest to develop such a theory, he
was greatly influenced by both psychoanalytic
theory and ethology. Perhaps, it would be more
accurate to state that his development of attach-
ment theory, rooted in his psychoanalytic training,
was further strengthened as he acquired knowl-
edge about evolutionary theory and ethological
principals, in particular work of Robert Hinde,
Konrad Lorenz, Harry Harlow, and Charles
#Springer International Publishing AG 2017
T.K. Shackelford, V.A. Weekes-Shackelford (eds.), Encyclopedia of Evolutionary Psychological Science,
DOI 10.1007/978-3-319-16999-6_3823-1

Darwin. It was not until the late 1950s that
Bowlby presented his theory of attachment to the
British Psychoanalytic Society in London, and the
initial reactions were quite negative. In Bowlby’s
words “many psychoanalysts... remained
unconvinced and sometimes very critical.”
Despite this criticism, Bowlby’s research group
at Tavistock continued to amass support for the
theory of attachment which culminated in his
trilogy –attachment and loss
(volume 1, Attachment; volume 2, Separation:
Anxiety and Anger; volume 3, Loss: Sadness
and Depression).
Bowlby consulted and collaborated with sev-
eral notable scientists while developing his theory
of attachment including Harry Harlow, Konrad
Lorenz, and Robert Hinde, but it was, however,
his collaborations with Mary Ainsworth that
helped to put attachment research on the psycho-
logical map. In 1950, Mary Ainsworth traveled to
London and, by chance, applied for a research
position advertised in the London Times and
joined Bowlby’s research team at Tavistock.
With Bowlby, she began studying the effects of
maternal separation on child development; how-
ever, she was well prepared to contribute to his
work. Previous to this move, Ainsworth had com-
pleted her PhD at the University of Toronto, and
there had begun exploring the concept of security
with William Blatz in her dissertation work –it
was Ainsworth who defined the concept of the
“secure base”building upon her research initiated
in Canada. Ainsworth worked at Tavistock until
1954 when she traveled to Uganda, but she, of
course, continued both her work in attachment
and her lifelong collaboration with Bowlby. Fol-
lowing her time in Uganda, she moved to Balti-
more, and finally in 1958 she was given a
permanent position at Johns Hopkins all the
while continuing her collaborations with Bowlby
and her operationalization of attachment.
Use of Attachment Behaviors
Attachment relationships, and the use of attach-
ment behaviors, are clearly presented as a biolog-
ical predisposition evolved to ensure survival.
Bowlby (1969/1991) spent considerable time
describing the nature and function of behaviors
that lead to attachment. Infants use several behav-
iors to seek proximity to their attachment figures
(e.g., crying, vocalizing, following) and also use
behaviors to maintain proximity (e.g., smiling,
clinging). Once infants are mobile and confident
in the care of their attachment figure, they tend to
use the attachment figure as a “secure base from
which to explore,”returning to the caregiver’s
safe haven (i.e., proximity to the attachment fig-
ure) for comfort and reassurance when needed. He
proposed that infants are biologically predisposed
to seek and gain proximity to caregivers and in
these relationships develop a sense of whether
they are worthy of love and support and a sense
of whether their caregivers can be trusted and
relied upon to provide care and support. Over
the course of infancy and childhood, these views
of the self and the other develop into sophisticated
internal working models of what to expect from
close others, and these internal working models
guide behavior over the lifespan.
His observations and work with John Robertson
on the destructive impact of early parent-infant
separations lead him to conclude that children
learn to tolerate longer parental separations; how-
ever, both children and adults find separations from
attachment figures to be distressing and will go
through a predictable sequence of
behaviors –distress, despair, and detachment (see
below). In fact, even predominately secure individ-
uals report distress when separated for prolonged
periods of time with little or no contact (Bowlby
1973/1991). This focus on separations is not sur-
prising given his tendency toward ethology and the
importance of parental protection on survival.
When separated from caregivers, infants were
observed to react with a predictable sequence of
behaviors which Bowlby suggested would have
ensured their survival. First, when initially sepa-
rated, infants will protest to attract the attention of
caregivers and, if possible, attempt to search for
caregivers. Sensitive and responsive caregivers
respond to these initial vocalizations by returning
and comforting infants; infants with less sensitive
or unresponsive caregivers (in our evolutionary
past and the present) develop representations that
2 Attachment Theory

care will not be given or inconsistently received.
Over time, the individuals’experience supports
the development of internal working models of
attachment –individuals’belief that the self is
worthy of care (or not) and that the care needed
when they express distress will be forthcoming
(or not).
From Bowlby’s observations of children who
experienced prolonged separations from their pri-
mary caregivers, he proposed that the next phase
of separation –if protests and searching are not
successful –would be a period of despair. In this
phase, infants stop their protests as well as their
search; this reaction is proposed to have devel-
oped to protect infants from predators when their
initial vocalizations to seek proximity to care-
givers were not successful.
Finally, Bowlby observed that infants who
experienced prolonged separations –and follow-
ing a period of distress and despair were not
reunited with their caregivers –moved to a
phase he described as detachment. He believed
that these infants were proposed to experience
detachment from their separated caregiver,
thereby allowing the opportunity to develop
attachment relationships with new caregivers
who may be better suited to provide the necessary
care and support needed to ensure survival.
Separations and Reunions
Ideally, children should feel secure in the presence
of caregivers, and when threatened they should
seek proximity to caregivers for protection. Chil-
dren, who are separated from insensitive or
rejecting parents, may have developed insecure
representations of their relationship with these
parents, and therefore they may struggle when
coping with the stress of the separation and the
associated effects on the family. Furthermore,
regardless of degree of security, both children
and adults find separations from attachment fig-
ures to be distressing –even predominately secure
children and adults go through the phases of dis-
tress, despair, and detachment when separated for
prolonged periods of time with little or no contact
(Bowlby 1973/1991).
Ainsworth continued to provide support for
Bowlby’s theory and expanded upon the idea
that although the goal of attachment (i.e., seeking
a secure base) is similar for all children, their
mechanisms for seeking proximity differ
depending on caregiving experiences, in particu-
lar the sensitivity and responsiveness of the care-
giver (Ainsworth et al. 1978). It was Ainsworth
and colleagues who first suggested that the
reunion, as well as the separation, is informative
in understanding attachment. In a series of articles
in the late 1960s and early 1970s, Ainsworth and
colleagues introduced the Strange Situation (SS).
For example, Ainsworth and Bell (1970) pre-
sented one of the first studies to describe their
methodology. Interestingly, although perhaps
deliberate on their part but clearly overshadowed
by subsequent work, their initial focus was on the
attachment behaviors during the separations and
reunions and not attachment categories: Their
article clearly summarized the SS procedure and
behind-the-scenes observation techniques. In par-
ticular, they highlighted the importance of infants’
signaling of the parent, actively approaching the
parent, and the aversive behaviors of the infants
that illustrated distress. They outlined that the
experimenters were trained to record observations
of infants’locomotion and crying as well as
manipulation and visual exploration of available
toys. Furthermore, the importance of coding
infants’behaviors such as proximity and contact
seeking, contact maintenance, interaction
avoiding, and interaction resisting as well as
search behaviors during the separation episodes
was highlighted. Their findings hint at individual
differences and, perhaps more importantly, pro-
vide empirical support for Bowlby’s earlier asser-
tions that infants separated from their caregivers
will go through a predictable sequence of distress.
Sensitivity of Care
Attachment theory provides an interesting frame-
work work to explore the effects of parenting and
caregiving for several reasons. First, Bowlby
(1973/1991,1980/1991) proposed that our attach-
ment representations developed from sensitivity
Attachment Theory 3

of care received from our primary caregiver.
Bowlby proposed that attachment security
resulted from responsive, appropriate caregiving
and that as a result of this care, individuals devel-
oped a sense of the self as worthy of care and a
belief that others would be responsive and sensi-
tive when caring (Bowlby 1980/1991)–it was
Bowlby’s assertion that personality development
was influenced consciously by these experiences
that caused criticism from his psychoanalytic col-
leagues. He proposed, and considerable research
has supported, that children should feel secure and
contented when safely in the presence of care-
givers (secure base), and when threatened they
should seek proximity to caregivers as a safe
haven.
Measurement of Individual Differences
in Infant Attachment
Ainsworth’s most prolific contribution to attach-
ment theory and research is without a doubt the
operationalization of infant attachment categories
(Ainsworth et al. 1978): secure (B), avoidant (A),
and resistant (C) –as well as Main’s fourth cate-
gory disorganized (D). Ainsworth’s work not only
provided the theoretical and empirical foundation
for decades of research in infant attachment; her
earlier operationalization of infant attachment also
provided the foundation for research in adult
attachment. Based on their responses to the SS,
infants are classified into one of the following four
categories:
1. Children who are categorized as secure
(B) have developed a trust in their caregivers’
availability and responsiveness and react to the
stress of the SS in ways that highlight their
positive expectations. In particular, secure
infants seek proximity when reunited with
caregivers and can return to play when
comforted.
2. Children who are categorized as avoidant
(A) have developed a belief that they cannot
turn to their caregivers for comfort, and during
the SS, avoidant infants typically avoid the
caregiver upon reunion and control –but do
not regulate –their negative emotions.
3. Children who are categorized as resistant
(C) have learned that care will be
unpredictable, and during the SS they seek
comfort inconsistently often switching from
clingy and sobbing to withdrawing and angry.
They cannot be comforted upon reunion and
do not return to play with the available toys.
4. The disorganized category was added by Main
and colleagues who expanded the examination
of parent-child attachment to high-risk sam-
ples. Children who are categorized as disorga-
nized (D) are often in the care of parents who
are abusive (physically, emotionally, and/or
sexually), and in response to the extreme stress
of their home life, these children do not present
a coherent attachment strategy during the SS
(e.g., freezing rather than proximity seeking).
The SS procedure provided the impetus for
considerable research (well over 7,000 cita-
tions for Ainsworth et al. 1978), and over the
past 50 years, it has been well established that
one of the negative consequences of poor par-
enting is the development of insecure attach-
ment. Insecurely attached individuals, by
definition, tend to have childhood experiences
that are characterized by lack of care and high
control. Insecure children develop a sense of
the world as inconsistent (resistant) or rejecting
(avoidant), and it is well established that inse-
cure attachment negatively influences child
development. Considerable work has explored
attachment beyond infancy into childhood. For
example, research has demonstrated that inse-
cure children are less socially competent, are
more likely to have emotional and behavioral
problems, are more likely to have medical
problems, and score lower on tests of achieve-
ment than secure children. Secure children
report pleasurable interactions with their par-
ents; avoidant children, although often non-
confrontational, will minimize interactions
with parents; anxious-ambivalent children
describe difficult relationships mixed with hos-
tility, sadness, and immature proximity-
seeking attempts; and disorganized children
display negative, punitive behaviors, rejection
4 Attachment Theory

and/or embarrassment of the parent, and some-
times overly bright, although inappropriate,
affect.
Measurement and Development of
Adult Attachment
From the beginning, Bowlby asserted that attach-
ment representations were important “from the
cradle to the grave”(1969/1991, p. 208) and
noted that in adolescence and adulthood, there
would be a “change of the figures towards whom
the [attachment] behavior is directed”(1969/
1991, p. 179). Although several scholars wrote
about the importance of attachment across the
lifespan, it was not until the mid- to late 1980s
that two groups of researchers independently
began to explore the measurement of adult attach-
ment. Their work focused on somewhat different
approaches to assessing adult attachment, and the
two areas of adult attachment research remain,
somewhat, at odds with each other. Main and
colleagues developed an interview-based assess-
ment of adults’attachment of their family of ori-
gin which focused on the coherence of their
representations (Main et al. 1985), whereas
Hazan and Shaver (1987) introduced a simple
three-paragraph forced-choice self-report ques-
tionnaire that focused on adult romantic relation-
ships. Not surprisingly, since they both modeled
their assessment from Ainsworth’s work, both
originally proposed three categories and, it is
well known that both measures of attachment
have proved to be powerful predictors of adult
behavior.
Main et al.’s(1985) adult attachment interview
(AAI) continues to be the primary method of
assessing adult attachment in developmental and
clinical fields despite the time-consuming training
and coding of attachment interviews (see Hesse
2016 for a summary). The interview coding pro-
tocol which assesses individuals’coherence of
their current state of mind with respect to their
families of origin (past-focused) results in one of
four attachment categories: autonomous (secure),
dismissing, preoccupied, and unresolved/disorga-
nized. Bakersman-Kranenburg and van
IJzendoorn (2009) summarized the data from
over 10,000 AAIs addressing issues of baseline
proportions in clinical and nonclinical samples,
gender distributions, as well as differences in dis-
tributions for adolescents and individuals from
low SES samples, ethnic minorities, and
non-Western countries. Using the four-category
classification of a sample of nonclinical mothers
(n=700) as their baseline, their findings indi-
cated that nonclinical fathers, adolescents and stu-
dents, individuals from at-risk samples, and
individuals from clinical samples were more
likely to be classified as dismissing than mothers
from nonclinical samples. Individuals from at-risk
samples were also more likely to be classified as
unresolved, and individuals from clinical samples
were also more likely to be classified as
unresolved or preoccupied.
Hazan and Shaver introduced a simple three-
paragraph forced-choice self-report measure of
adult attachment modeled from Ainsworth’s
three infant categories focusing on adult romantic
relationships. Their three categories –secure,
avoidant, and anxious-ambivalent –proved to be
quite productive, but the categorical measurement
was met with some concern by personal relation-
ship researchers. Although the three-category
measure did not stand the test of time, this seminal
article provided the impetus for the next few
decades of work on adult attachment.
Although the self-report methodology
assessing attachment of close peer relationships
is controversial –some would say that peer rela-
tionships are not attachment relationships (see
below for discussion and van IJzendoorn and
Bakersmans-Kranenburg 2010 for a recent
example) –research exploring the importance of
adult peer attachment relationships continues to
be prolific, and the effects of adult peer relation-
ships on adult social behavior are well
documented.
In 1990, Bartholomew published a paper
which merged the work of Main et al. (1985)
and Hazan and Shaver (1987) as well as Bowlby’s
original descriptions of the self- and other-
models. Specifically, Bowlby had proposed that
throughout childhood, individuals develop a
sense of whether they are worthy of love and
Attachment Theory 5

support (or not) and a sense of whether others can
be trusted and relied upon to provide care and
support (or not). By adulthood, these views of
the self and the other are well established and
have developed into sophisticated internal work-
ing models of what to expect from close others,
and these internal working models guide behavior
over the lifespan. Bartholomew suggested that the
intersection Bowlby’s proposed dimensions of the
self-model and other-model resulted in a four-
category model of attachment. Secure individuals
were defined to have developed a positive model
of both the self and others, and considerable
research has supported that secure individuals
have high self-esteem and self-confidence as
well as high trust and support with others. Preoc-
cupied individuals were defined to have devel-
oped a negative model of the self and a positive
model of others; similarly, considerable research
has supported that preoccupied individuals tend to
have conflicting views of the self and others which
tends to negatively impact their relationships.
Bartholomew noted that her definitions of secure
and preoccupied prototypes were consistent with
both Main et al. and Hazan and Shaver. Although
the previous researchers had each proposed one
type of avoidance, Bartholomew expanded this
definition of avoidance by proposing that avoid-
ance could be either fearful (similar to Hazan and
Shaver’s description of avoidance) or dismissing
(similar to Main et al.’s description of avoidance).
Fearful individuals were defined to have devel-
oped a negative model of both the self and others
and tend to consistently report higher levels of
depression, neuroticism, marital conflict, and
interpersonal sensitivity. Dismissing individuals
were defined to have developed a positive model
of the self and a negative model of others and
reported high levels of self-esteem and self-
confidence but low levels of trust and warmth in
their relationships. Researchers have indicated
that both types of insecure-avoidant adults deny
symptoms of distress (dismissing) or are afraid to
ask for help (fearful) as they believe that others
will reject their attempts at proximity seeking or
challenge their feelings of distress.
In her early work, Bartholomew presented both
interview and survey measures to assess the four-
category model (see Bartholomew and Horowitz
1991; Griffin and Bartholomew 1994). Although
the interview proved to be the most reliable
method, researchers tended to gravitate toward
the less time-consuming, albeit less reliable, self-
report surveys. The four-paragraph measure (RQ,
Relationship Questionnaire) was modeled from
Hazan and Shaver’s three-category measure with
an additional paragraph describing the dismissing
category (see below). Participants were typically
asked to rate each of the four categories on a
Likert scale from “not at all like me”to “very
much like me”and then choose the one category
from the four that was “most like them.”The
Relationship Scale Questionnaire (RSQ) was sim-
ply a list of the 17 statements from the RQ, and
participants were asked to rate each of the items
on a Likert scale from “not at all like me”to “very
much like me.”The four scales were computed by
averaging the items. The paragraphs from the RQ
categories (Griffin and Bartholomew 1994) are as
follows:
Secure: It is easy for me to become emotionally
close to others. I am comfortable depending on
others and having others depend on me. I don’t
worry about being alone or having others not
accept me.
Fearful: I am somewhat uncomfortable getting
close to others. I want emotionally close rela-
tionships, but I find it difficult to trust others
completely or to depend on them. I sometimes
worry that I will be hurt if I allow myself to
become too close to others.
Preoccupied: I want to be completely emotionally
intimate with others, but I often find that others
are reluctant to get as close as I would like. I am
uncomfortable being without close relation-
ships, but I sometimes worry that others don’t
value me as much as I value them.
Dismissing: I am comfortable without close emo-
tional relationships. It is very important to me
to feel independent and self-sufficient, and
I prefer not to depend on others or have others
depend on me.
6 Attachment Theory

Personal relationship researchers were
concerned with the reliability of early measures,
but they were also drawn to the concept of adult
attachment, and the early 1990s saw a wave of
studies proposing new measures of attachment.
Brennan et al. (1998) collated the items from
these attachment measures (323 items which
assessed 60 attachment constructs from 14 differ-
ent scales) and proposed a new measure the Expe-
riences of Close Relationship scale (ECR). The
ECR has two orthogonal dimensions (attachment
anxiety and attachment avoidance) and improved
reliability over the RSQ. The ECR-R (Fraley et al.
2000) attempted to correct some of the limitations
of the ECR; however, both the ECR and the
ECR-R proved to be inadequate measures of secu-
rity and imprecise assessments of Bartholomew’s
four-category model (see Mikulincer and Shaver
2007)–both scales are imperfect due to the lim-
itations of the existing item pool. Although these
measures –Bartholomew’s RSQ, the ECR and
ECR-R –continue to be well used in the literature,
researchers continue to work on improving ques-
tionnaire measures of adult attachment (see
Scharfe 2016).
Are Adult Close Relationship
Attachment Relationships?
Despite the fact that Bowlby asserted that attach-
ment representations were important across the
lifespan and that a change of attachment figures
would be observed in adolescence and adulthood,
there is some controversy about who serves as
attachment figures in adulthood. Personal rela-
tionship researchers have insisted that adult
romantic partners and friends as well as family
members may serve as attachment figures. Con-
siderable work has provided empirical support
that lovers, friends, and family members serve as
attachment figures for adults. This line of research
began with a listing of reasons why adult love
relationships may be attachment relationships in
Hazan and Shaver’s(1987) seminal article. Since
that time, personal relationship researchers have
demonstrated that one can observe attachment
behaviors in adult relationships (i.e., proximity
seeking, proximity maintenance, safe haven,
secure base), similar to infants and children there
are consistent, measureable individual differences
in adult attachment; that adults express anxiety if
their attachment figure is not accessible; and that
adults experience distress if separations from their
attachment figures are prolonged.
Attachment Internal Working Models
and Hierarchies
Bowlby (1969/1991) highlighted specific devel-
opmental changes in attachment relationships
over the lifespan, and the various measures of
adult attachment have allowed researchers to
begin to explore this feature of his theory more
fully. For example, Bowlby proposed that attach-
ment representations started from one primary
attachment relationship in infancy, but he
acknowledged that early on, often soon after
attachment to the primary caregiver was evident,
infants would begin to establish the development
of attachment relationships to other caregivers.
Specifically, infants were proposed to develop a
primary attachment with the caregiver who satis-
fied their basic needs –this primary attachment
relationship was the most intense regardless of the
number of secondary attachment relationships.
Infants and children, however, do receive care
from other caregivers, within and outside of their
family of origin, and they may also develop sec-
ondary attachment bonds to these individuals.
Considerable work has provided support that
infants and children will develop attachment rela-
tionships with adults who are not their primary
caregiver.
During these early years, attachment relation-
ships are typically unidirectional –the caregiver
fulfills the attachment needs of the child. By the
time we reach adolescence and adulthood, most
individuals have multiple attachment relationships
which are organized into a “state of mind”or
internal working model, and these relationships
can be reciprocal (i.e., both individuals fulfill and
provide attachment needs). Furthermore, Bowlby
proposed that in adulthood, there would be a
“change of the figures towards whom the
Attachment Theory 7

[attachment] behavior is directed”(1969/1991,
p. 179) and that adolescents and adults tend to
organize their attachment relationships into a
hierarchy.
Bowlby (1969/1991) proposed that during
adolescence, the child’s attachment to their par-
ents would change due to the importance of other
adults in the child’s life as well as the sexual
attraction to their peers. There has been consider-
able research exploring the change of attachment
from parents to peers initiated by Hazan and
Zeifman (1994). They reported that, between the
ages of 8 and 14 years, adolescents reported that
they approached peers for proximity and safe
haven functions and parents for secure base func-
tions (see also Nickerson and Nagle 2005). They
also noted that during late adolescence (15–17
years), those individuals who had formed peer
romantic relationships were less likely to
approach their parents for attachment functions
(see also Mayseless 2004; Nickerson and Nagle
2005), although research findings have not been
consistent on this point. Consistent with Bowlby’s
suggestion that, for most individuals, the bond
with parents would continue throughout life,
researchers have found that one’s mother, in par-
ticular, continues to be an important attachment
figure throughout the lifespan (Pitman and
Scharfe 2010).
Bowlby (1969/1997) also proposed that individ-
uals, regardless of age, would organize their attach-
ment relationships into a hierarchy and would
demonstrate a preference for a primary attachment
figure. Over the past two decades, several
researchers have found support that individuals do
tend to organize their attachment relationships into
hierarchies, and these findings have been found
across the lifespan in several diverse samples (e.g.,
Doherty and Feeney 2004; Pitman and Scharfe
2010; Trinke and Bartholomew 1997). Interest-
ingly, mothers tend to be listed at or near the top
of the attachment hierarchy across the lifespan.
Attachment Stability
Paradoxically, Bowlby (1980/1991) proposed that
once formed, attachment representations would
remain stable but could also change over the
lifespan. Specifically, he proposed that once
formed, internal working models of attachment
would remain relatively stable in adulthood; how-
ever, he highlighted that changes may occur dur-
ing development (see Del Giudice 2009; Del
Giudice and Belsky 2010 for an evolutionary
explanation for these changes) but may also
change in adulthood in reaction to particularly
traumatic events (see Scharfe 2003 for a summary
of the lifespan research). Researchers have dem-
onstrated that parent-infant attachment shows
moderate-to-high stability (e.g., Waters 1978)
and that change is likely when infants’caregiving
environments change (e.g., Thompson et al. 1982;
Vaughn et al. 1979). In particular, there is some
evidence that it is important to determine how
sensitively changes were managed (e.g., NICHD
Early Child Care Research Network 2001). Sim-
ilar findings have been reported exploring stabil-
ity of attachment throughout childhood (e.g.,
Howes and Hamilton 1992).
Bowlby (1973/1991) proposed that by adult-
hood, attachment representations would be well
developed and more “constrained”and less adap-
tive to change. Over the past few decades,
researchers have reported moderate-to-high sta-
bility of adult attachment representations regard-
less of the method of assessment (e.g., Scharfe
and Bartholomew 1994). To date, several studies
have examined the stability of attachment from
infancy to adulthood and have found moderate
stability when social environments remain rela-
tively stable (e.g., Waters et al. 2000).
Conclusion
Bowlby’s theory of attachment includes several
important foundational constructs including the
proposal that attachment behavior was instinctual
and important across the lifespan. He also
described characteristic attachment behaviors
and individual differences in attachment as well
as both adaptive and maladaptive care environ-
ments. Finally, he outlined the development of
multiple attachments and hierarchies of attach-
ment as well as the development of internal
8 Attachment Theory

working models of attachment and proposed con-
ditions to expect both stability and change of
attachment. Considerable research over the past
40 years has provided empirical evidence for
these foundational constructs.
Cross-References
▶Attachment in Adulthood
▶John Bowlby: Pioneer of Attachment Theory
▶Measurement: Categorical Vs Continuous
▶Sex Differences in Attachment
References
Ainsworth, M. D., & Bell, S. M. (1970). Attachment,
exploration, and separation: Illustrated by the behavior
of one-year-olds in a strange situation. Child Develop-
ment, 41,49–67.
Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall,
S. (1978). Patterns of attachment: A psychological
study of the strange situation. Hillsdale: Erlbaum.
Bakermans-Kranenburg, M. J., & van IJzendoorn, M. H.
(2009). The first 10,000 adult attachment interviews:
Distributions of adult attachment representations in
clinical and non-clinical groups. Attachment &
Human Development, 11, 223–263.
Bartholomew, K. (1990). Avoidance of intimacy: An
attachment perspective. Journal of Social and Personal
Relationships, 7, 147–178.
Bartholomew, K., & Horowitz, L. (1991). Attachment
styles among young adults: A test of a four-category
model. Journal of Personality and Social Psychology,
61, 226–244.
Bowlby, J. (1944). Forty-four juvenile thieves; their char-
acters and home-life. Oxford: Bailliere.
Bowlby, J. (1951). Maternal care and mental health. World
Health Organization Monograph Series, 2, 179.
Bowlby, J. (1969/1991). Attachment and loss: Vol.
1 Attachment. Toronto: Penguin Books.
Bowlby, J. (1973/1991). Attachment and loss: Vol. 2. Sep-
aration: Anxiety and Anger. Toronto: Penguin Books.
Bowlby, J. (1980/1991). Attachment and loss Vol. 3: Loss:
Sadness and Depression. Toronto: Penguin Books.
Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self-
report measurement of adult attachment: An integrative
overview. In J. A. Simpson & W. S. Rholes (Eds.),
Attachment theory and close relationships
(pp. 46–76). New York: The Guilford Press.
Del Giudice, M. (2009). Sex, attachment, and the develop-
ment of reproductive strategies. Behavioral and Brain
Sciences, 32,1–21.
Del Giudice, M., & Belsky, J. (2010). Sex differences in
attachment emerge in middle childhood: An evolution-
ary hypothesis. Child Development Perspectives, 4,
97–105.
Doherty, N. A., & Feeney, J. A. (2004). The composition of
attachment networks throughout the adult years. Per-
sonal Relationships, 11, 469–488.
Fraley, R. C., Waller, N. G., & Brennan, K. A. (2000). An
item response theory analysis of self-report measures of
adult attachment. Journal of Personality and Social
Psychology, 78, 350–365.
Griffin, D. W., & Bartholomew, K. (1994). The metaphys-
ics of measurement: The case of adult attachment. In
K. Bartholomew & D. Perlman (Eds.), Advances in
personal relationships vol 5: Attachment processes in
adulthood (pp. 17–52). London: Jessica Kingsley
Publishers.
Hazan, C., & Shaver, P. R. (1987). Romantic love concep-
tualized as an attachment process. Journal of Person-
ality and Social Psychology, 52,511–524.
Hazan, C., & Zeifman, D. (1994). Sex and the psycholog-
ical tether. In D. Perlman & K. Bartholomew (Eds.),
Advances in personal relationships (Vol.
5, pp. 151–178). London: Jessica Kingsley.
Hesse, E. (2016). The adult attachment interview: Protocol,
method of analysis, and selected empirical studies:
1985–2015. In J. Cassidy & P. R. Shaver (Eds.), Hand-
book of attachment: Theory, research, and clinical
applications (3rd ed., pp. 553–597). New York: The
Guildford Press.
Howes, C., & Hamilton, C. (1992). Children’s relationship
with child care teachers: Stability and concordance
with parental attachments. Child Development, 63,
867–878.
Main, M., Kaplan, N., & Cassidy, J. (1985). Security in
infancy, childhood, and adulthood: A move to the level
of representation. In I. Bretherton & E. Waters (Eds.),
Growing points in attachment theory and research.
Monographs of the Society for Research in Child
Development, 50, 66–104.
Mayseless, O. (2004). Home leaving to military service:
Attachment concerns, transfer of attachment functions
from parent to peers, and adjustment. Journal of Ado-
lescent Research, 19, 533–558.
Mikulincer, M., & Shaver, P. R. (2007). Attachment in
adulthood: Structure, dynamics, and change.
New York: Guilford Press.
NICHD Early Child Care Research Network. (2001). Child
care and family predictors of preschool attachment and
stability from infancy. Developmental Psychology, 17,
847–862.
Nickerson, A. B., & Nagle, R. J. (2005). Parent and peer
attachment in late childhood and early adolescence.
Journal of Early Adolescence, 25, 223–249.
Pitman, R., & Scharfe, E. (2010). Are parents better than
peers?: Exploring the function of attachment hierar-
chies during emerging adulthood. Personal Relation-
ships, 17, 201–216.
Attachment Theory 9

Scharfe, E. (2003). Stability and change of attachment
representations from cradle to grave. In S. M. Johnson
& V. Whiffen (Eds.), Attachment processes in couple
and family therapy (pp. 64–84). New York: Guilford
Publications.
Scharfe, E. (2016). Measuring what counts: Development
of a new 4-category measure of adult attachment. Per-
sonal Relationships, 23,4–22.
Scharfe, E., & Bartholomew, K. (1994). Reliability and
stability of adult attachment patterns. Personal Rela-
tionships, 1,23–43.
Thompson, R. A., Lamb, M. E., & Estes, D. (1982). Sta-
bility of infant-mother attachment and its relationship
to changing life circumstances in an unselected middle-
class ample. Child Development, 53, 144–148.
Trinke, S. J., & Bartholomew, K. (1997). Hierarchies of
attachment relationships in young adulthood. Journal
of Social and Personal Relationships, 14, 603–625.
van IJzendoorn, M. H., & Bakermans-Kranenburg, M. J.
(2010). Stretched until it snaps: Attachment and close
relationships. Child Development Perspectives, 4,
109–111.
Vaughn, B., Egeland, B., Sroufe, A., & Waters, E. (1979).
Individual differences in infant-mother attachment at
twelve and eighteen months: Stability and change in
families under stress. Child Development, 50, 971–975.
Waters, E. (1978). The reliability and stability of individual
differences in infant-mother attachment. Child Devel-
opment, 49, 483–494.
Waters, E., Merrick, S., Treboux, D., Crowell, J., &
Albersheim, L. (2000). Attachment security in infancy
and early adulthood: A twenty-year longitudinal study.
Child Development, 71, 684–689.
10 Attachment Theory