Content uploaded by Robbie Duschinsky
Author content
All content in this area was uploaded by Robbie Duschinsky on Feb 13, 2015
Content may be subject to copyright.
Crittenden’s Dynamic–Maturational Model of Attachment and Adaptation
Sophie Landa
Newcastle University
Robbie Duschinsky
Northumbria University
This article undertakes a systematic exposition and analysis of Patricia Crittenden’s dynamic–
maturational model of attachment and adaptation. It traces Crittenden’s information-processing model of
attachment behavior to her work with Mary Ainsworth, and shows how this account came to underpin
her integration of insights from cognitive science with developmental psychology. The article draws
surprising conclusions regarding the differences between the dynamic-maturational model and main-
stream attachment theory, clarifying the meaning of contested concepts and identifying important flaws
in previous interpretations of Crittenden’s work.
Keywords: attachment, Crittenden, dynamic–maturational model, information processing, trauma
Mary Main and Patricia Crittenden were both students of Mary
Ainsworth. Their graduate study occurred almost exactly a decade
apart: Main was with Ainsworth at The Johns Hopkins University
from 1968 to 1973, whereas Crittenden was at The University of
Virginia from 1979 to 1983 (Landa & Duschinsky, in press). The
addition by Main and colleagues of the “disorganized/disoriented”
classification of attachment to Ainsworth’s three-category system
has become widely accepted, especially over the past 15 years.
Solomon and George (2011, p. 3) have described how the con-
struct of disorganization “amounts to a paradigm shift” in attach-
ment research and “is now well integrated into the lexicon of
clinicians, especially those involved in providing infant mental
health.” Yet Crittenden’s dynamic–maturational model of attach-
ment and adaptation has emerged as a competing paradigm and has
been met with some excitement, especially among health and
social care professionals in Europe. Crittenden has been described
as “a radical and a pioneer, and academically speaking very
courageous” (Vetere, 2004), and Pocock (2010, p. 305) has en-
thused that “the Dynamic- Maturational Model (DMM) is beauti-
ful.” Milan, Snow, and Belay (2009, p. 1031) have wondered
whether, compared with mainstream attachment theory, “Crit-
tenden’s system may be better suited to studies of clinical phe-
nomena, such as depression, because of the conceptual model from
which it was developed.”
Yet despite repeated calls for exposition and analysis of the
dynamic–maturational model, identifying its differences from
mainstream attachment theory (e.g., Robson & Wetherell, 2011;
Schuengel, 2001), to date, no article undertaking this work has
emerged. This gap is surprising and significant, given that the
dynamic–maturational model is seeing widespread use as a psy-
chological theory, informing clinical and social interventions as
well as research. In part, we suspect that attempts at neutral survey
or comparison have been stalled by the contention between the
attachment paradigms. Another contributory to the lack of com-
mentary on the dynamic–maturational model has been that the
theory is elaborated across an enormous number of widely diffused
texts. Wilkinson (2012), for example, claims that there is urgent
need for a review of the dynamic–maturational model, but feels
that Crittenden’s dispersed arguments make it difficult to identify
her position on key issues such as the meaning and consequences
of trauma. Our response to this issue has been to collect and draw
upon every text Crittenden has put in the public domain across her
career, considering both the development of her ideas over time
and the intellectual infrastructure that has animated and organized
her approach. We therefore present an original account of the
divisions that led to the emergence of the dynamic–maturational
model and the first integrated analysis of Crittenden’s work as a
psychological theory. However, because we are not trained in the
assessment measures developed by Crittenden or their alternatives
from within mainstream attachment theory, we will not address
their respective utility in this article (see Farnfield et al., 2010;
Spieker & Crittenden, 2010).
We shall argue that the core of the dynamic–maturational model
as a research program is the information-processing model. This
was first developed in Crittenden’s doctoral research under Mary
Ainsworth in the early 1980s, to explain anomalous infant behav-
ior in the “Strange Situation Procedure.” We shall therefore begin
our review by identifying how this model led to disagreements
with Main about the meaning of such anomalous behavior, and
hence the cognitive and behavioral processes and consequences of
child maltreatment. We draw on Ainsworth’s unpublished letters
to Bowlby, available in the Wellcome Trust Archive (London,
United Kingdom), to clarify the theoretical stakes of these dis-
agreements. The argument between Main and Crittenden is widely
and incorrectly believed to lie in Crittenden’s rejection of the idea
of attachment “disorganization” in infants. In the place of “disor-
ganization,” Crittenden is understood to instead consider infants
displaying such behaviors to be showing “organized” combina-
tions of avoidant and resistant attachment strategies (e.g., Van
Sophie Landa, Institute of Neuroscience, Newcastle University, New-
castle upon Tyne, United Kingdom; Robbie Duschinsky, Department of
Social Work and Communities, Northumbria University, Newcastle upon
Tyne, United Kingdom.
Correspondence concerning this article should be addressed to Sophie
Landa, Institute of Neuroscience, Newcastle University, Newcastle upon
Tyne, United Kingdom. E-mail: sophie.landa@ncl.ac.uk
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Review of General Psychology © 2013 American Psychological Association
2013, Vol. 17, No. 3, 326–338 1089-2680/13/$12.00 DOI: 10.1037/a0032102
326
IJzendoorn, Schuengel, & Bakermans-Kranenburg, 1999). In fact,
we show that the divergence is better understood to lie in the
different meanings Main and Crittenden give to the concepts of
“attachment organization” and “adaptation.”
Crittenden’s information-processing model is quickened as a
psychological theory by her integration of insights from cognitive
science and developmental psychology regarding the role of dif-
ferent memory systems and maturation on information processing.
The range of topics that have been addressed by Crittenden and
other researchers affiliated with her International Association for
the Study of Attachment (2012) is vast. We will consider two in
the remainder of the article. First, as an illustration of the acuity
of the dynamic–maturational model, we will examine Crittenden’s
insights into the behavior and information processing of maltreat-
ing parents. Second, we will consider the account of trauma
presented by Crittenden, for the interlinked reasons that there have
been calls for clarification on this topic, it brings into relief key
differences between Main and Crittenden, and it has been an area
of recent conceptual innovation. Our review will close by discuss-
ing concerns that have been raised regarding the evidence base and
exhaustiveness of Crittenden’s account of behavioral strategies,
which are perceived to limit the persuasiveness and effectiveness
of some aspects of the dynamic–maturational model. Although
partially accepting these concerns, we shall show how they do not
invalidate Crittenden’s work as an exciting and sometimes
uniquely insightful contribution to psychological theory.
Disorganization
To understand the forces that have animated and organized the
dynamic–maturational model, it must be understood that two in-
escapable and surprising findings confronted Mary Ainsworth’s
doctoral students. The first was that Ainsworth’s ABC classifica-
tion of infant behavior in the Strange Situation Procedure appeared
to account for the overwhelming majority of middle-class infants.
The Strange Situation Procedure was first used by Ainsworth and
Wittig (1969) to assess individual differences in the responses of
56 middle-class nonclinical infants aged 11 months to the depar-
ture of a caregiver. Infants classified as “secure” (Type B) used the
caregiver as a safe base from which to explore, protested at their
departure, but sought the caregiver upon his or her return. Infants
classified as anxious–avoidant (Type A) did not exhibit distress on
separation and ignored the caregiver on their return. Separation of
an infant from her caregiver was theorized by Bowlby (1960) to
necessarily evoke anxiety as a reaction hardwired by evolution, as
the infant cannot survive without the caregiver. Hence, the appar-
ently unruffled behavior of the Type A infants was understood by
Ainsworth as a mask for distress, a point later evidenced through
studies of heart rate (Sroufe & Waters, 1977). Infants classified as
anxious–ambivalent/resistant (Type C) showed distress on separa-
tion, and were clingy and difficult to comfort on the caregiver’s
return.
A set of protocols for classifying infants into one of these groups
was established by Ainsworth’s influential Patterns of Attachment
(Ainsworth, Blehar, Waters, & Wall, 1978). Main (1990),Crit-
tenden (1995, p. 368), and other students of Ainsworth were
therefore brought to ask, “Why are there only three patterns of
attachment when mothers are highly varied?” The fact that these
three patterns appeared so widely suggested that, on the one hand,
the activation of the attachment system when an infant is anxious
appeared to be an innate psychophysiological mechanism. On the
other hand, this finding implied that the quality of the attachment
behavior elicited by this anxiety differed in systematic ways as a
function of the infant’s caregiving environment. A second recur-
ring finding that confronted Ainsworth’s students, however, was
that not all infants could be classified using Ainsworth et al.,
(1978) protocols for classifying infant behavior in a strange situ-
ation. This was especially the case with children from maltreat-
ment samples, but it also occurred in samples of infants from
middle-class homes.
Ainsworth’s first doctoral student, Sylvia Bell, observed cases
in which the infant showed “signs of disturbance, such as inap-
propriate, stereotyped, repetitive gestures or motions. He may
show some resistance to his mother, and indeed he may avoid her
by drawing back from her or averting his face when held by her”
(Ainsworth et al., 1978, p. 62). Ainsworth advised Bell to code
such infants as a subtype of the Type B (secure) pattern. Ain-
sworth’s second doctoral student, Main, also found several infants
who showed unclassifiable behaviors, including “hand-flapping;
echolalia; inappropriate affect; and other behaviors appearing out
of context” (Main, 1977, p. 70). In particular, several infants
showed reunion behavior that combined an attempt to approach the
caregiver with signs of fear and avoidance. Main relates that, from
graduate school, she “had already been intrigued by odd-appearing
behaviors of animals in conflict situations and—after observing
one ‘unclassifiable’ infant in her doctoral study fling her arms
about her head while in an anomalous position on parent en-
trance—Main continued to pursue the problem of ‘unclassifiable’
infants in this light” (Main, Hesse, & Hesse, 2011, p. 435). On
February 1, 1974, Ainsworth wrote to Bowlby,
We have found plenty of evidence that the mothers of A babies dislike
physical contact, and that it is through behavior relevant to physical
contact that they (at least in large part) express rejection. Mary’s
theory is that this puts babies in a double bind, for they are pro-
grammed to want contact and yet are rebuffed (or at least have
unpleasant experiences) when they seek it. Mary’s hypothesis is that
the avoidance (detachment-like) defensive behavior characteristic of
A babies stems from the double-bind (Ainsworth, 1974).
During these years of Main’s doctoral work, Ainsworth (1972)
deployed a technical definition of attachment “organization” as
behaviors oriented toward proximity with the caregiver when the
attachment system is activated by anxiety. Following this empha-
sis, Main focused on physical proximity with the caregiver as the
set-goal of the activated attachment system (Main, 1977). Based
on this assumption, Main (1977) proposed that avoidance serves
proximity in two ways: It keeps the caregiver relatively near
without alienating him or her through approach behavior; it helps
the infant “gain control over, that is, flexibility in his own behav-
ior, a thing he will not have should he . . .break into disorganised
distress” (p. 55). As such, Type A behavior is “a search for control
when disorganisation threatens” (Main, 1981, p. 685).
Yet Main began to note that avoidant infants can at least direct
their attention away from the conflicting demands of the attach-
ment system to both approach and flee from the caregiver. By
contrast, other infants appeared so overcome by this conflict that
they could not develop any coherent strategy for achieving prox-
imity with their caregiver in the strange situation. Hence Main
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
327
THE DYNAMIC-MATURATIONAL MODEL
began to reconceptualize cases in which the infant would approach
the caregiver with their head averted, not as an extreme form of
avoidance but as an effect of behavioral breakdown: a “disorga-
nized” attachment pattern. This new attachment pattern included a
variety of behaviors, including freezing, rocking, disorientation,
crying at the departure of the stranger, and showing confusion or
fear on the return of the caregiver. These behaviors did not need to
last more than a moment to be an indicator of disorganized/
disoriented attachment, and an infant could be placed in the clas-
sification if they scored 5 or more on a 9-point scale of such
behaviors (Main & Solomon, 1986,1990).
The Effects of Maltreatment
Ainsworth was initially skeptical of Main’s introduction of the
“D” classification, wondering whether it was partly operating as a
residual category (Landa & Duschinsky, in press). Yet Ainsworth
came to be “impressed with the need for adding a new ‘D’ or
disorganised category to the classification system” (Ainsworth,
1985a). As she learned about Main’s theories and assessment
measures over the course of 1985, Ainsworth remarked to Bowlby,
“You were right that I am in a sense a student of Mary Main’s”
(Ainsworth, 1985b). Yet as well as being impressed by Main’s
evidence for the need for a “D” classification, in 1985, Ainsworth
was also giving support to Crittenden’s “excellent research on
maltreated children” (Ainsworth, 1985c, p. 788), despite the fact
that Main and Crittenden were already generating quite different
interpretations of the effect of adverse conditions on attachment
behavior.
In 1979, the same year that Crittenden began her graduate
studies, Ainsworth returned from a trip to England with some of
the early chapters from Bowlby’s forthcoming book Loss, pub-
lished in 1980. She announced to her students, “Here is chapter 4
of the Bible” (Crittenden, 2012). This Chapter 4 was Bowlby’s
account of information processing, and it formed the bedrock of
Crittenden’s subsequent thinking. In this chapter, Bowlby (1980)
argued that, “given certain adverse circumstances during child-
hood, the selective exclusion of information of certain sorts may be
adaptive. Yet, when during adolescence and adult the situation
changes, the persistent exclusion of the same forms of information
may become maladaptive” (p. 45). Though otherwise tending to be
hostile toward what he saw as the imprecision of Kleinian con-
cepts, Bowlby (1980, p. 68) identified that Melanie Klein’s con-
cept of “splitting” could helpfully identify the way in which
motivational dispositions, feelings, and memories could also suffer
exclusion as an adaptive response to adverse circumstances. The
result would be that particular behavioral systems, and their asso-
ciated motivational dispositions, feelings, and memories, could be
“segregated” (Bowlby, 1980, p. 59). For instance, Bowlby ob-
served that a child or adult might keep their anger at someone they
depend upon away from consciousness, and find it expressed out
of context under calmer circumstances.
Crittenden’s dissertation, submitted in May 1983, was a study of
73 infants and toddlers. Most of this sample had experienced
severe maltreatment. Like Ainsworth’s previous doctoral students,
Crittenden (1983) found that “not all infants can be placed easily
into the three categories described above” (pp. 14 –15). Drawing
inspiration from Bowlby’s chapter on information processing,
Crittenden proposed that the A and C responses could be regarded
as excluding “some classes of information” relevant to “the acti-
vation of the attachment system” (p. 18), and proposed that both A
and C behaviors should be seen as strategies for maintaining the
availability of the caregiver by “interfering with one’s ability to
process different kinds of information” (p. 18). For example, some
of Crittenden’s sample, particularly as they approached 18 months,
appeared to display false positive affect in the Strange Situation
Procedure. Although they “tended to be difficult with their mothers
until about 1 and a half years of age” (Crittenden, 1992b, p. 339),
by 18 months, Crittenden (1983) observed that these children were
unusually accommodating and can only be classified as cooperative.
These babies pose some very interesting questions. Why are these
children cooperative when their experience with their mother should
provoke a passive or difficult response? And why do so many of them
seem concurrently ill at ease? (p. 66)
Subsequent scholars arguing both for and against the dynamic–
maturational model have generally presumed that Crittenden re-
jected the “disorganized” pattern and instead proposed that such
infants were displaying extreme forms of “organized” A and C
behavior (e.g., Holmes, 2004;Holmes & Chimera, 2010;Van
IJzendoorn et al., 1999). Yet it is important to note that between
the 1970s and 1980s, Ainsworth had changed the meaning she was
giving the term “organized” (Landa & Duschinsky, in press). She
still maintained that proximity may be the set goal of attachment at
11 months. However, her correspondence with Bowlby led to the
conclusion that this proximity seeking is a special case of the more
general goal of the attachment system, which is to maintain the
availability of the caregiver. Citing these letters, Ainsworth (1990,
p. 474) would write that “Bowlby (personal communication, 1987)
holds that ‘availability of the attachment figure is the set-goal of
the attachment system in older children and adults.’”
Ainsworth, therefore, in this period, changed her usage of the
term “organization” to mean behavior that, under conditions of
perceived threat and the activation of the attachment system, is
oriented toward maintaining the availability of the attachment
figure. Crittenden followed Ainsworth in this changed usage,
applying it also to infancy. Whereas Main saw physical proximity
as the set goal of the attachment system, and hence behavior
oriented toward proximity as “organized,” Crittenden took the
availability of the caregiver as the set goal of the attachment
system when activated by the perception of threat, even in infancy.
Although attachment behaviors would not necessary always suc-
ceed, Crittenden theorized that the behaviors instigated by the
attachment system would aim, when possible, to maintain the
availability of the attachment figure as a source of protection. Only
those that could not be seen as aiming at this possibility were
regarded by Crittenden as “disorganized.”
This conclusion was also shaped by differing accounts of the
term “adaptation.” Main’s (1979) had restricted the term “adap-
tive” to explanations of why behavioral systems might have
evolved for a species. Main emphasized proximity-seeking behav-
ior as adaptive for humans in potentially dangerous situations.
Behavioral breakdown at the level of an individual infant, although
it might indeed have some beneficial effects, was not perceived by
Main to not be an expression of a species-level adaptation to
achieve proximity in conditions of perceived threat. Crittenden, by
contrast, used the term “adaptation” as a heuristic for interpreting
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
328 LANDA AND DUSCHINSKY
what function a behavior may have for maintaining the caregiver’s
availability.
To illustrate this difference between Main and Crittenden: one
of the “direct indices of disorganised attachment” for Main and
Solomon (1990, p. 139) was an infant’s hand-to-mouth gesture on
reunion with the parent. This act was understood by Main and
Solomon to signal behavioral breakdown under conditions of
experienced fearful confusion, as it did not serve to facilitate the
proximity expected of attachment behavior. By contrast, Crit-
tenden’s perspective would conceptualize a child using the hand-
to-mouth behavior to stifle a cry of distress as an adaptation to a
rejecting or insensitive caregiving environment. It could then be
understood as a more extreme form of Type A behavior than noted
in the Ainsworth et al. (1978) coding protocols. Hence, those using
the dynamic–maturational model have perceived such hand-to-
mouth behavior to stifle a scream as “organized” because, in such
cases, the child can be understood as working to maintain the
availability of their attachment figure through an act of self-
regulation, which constricts negative affect (e.g., Svanberg, 2009,
p. 105).
Hence whereas Main saw combinations of A and C behavior as
“disorganized,” because they appeared to evidence behavioral
breakdown instead of a coherent strategy for seeking proximity,
Crittenden considered such combinations “organized” “adapta-
tions,” as a result of her different definition of both terms. Crit-
tenden noted that abused and neglected cases tended to “show an
A/C pattern as do a few who are only abused and also a few who
only neglected” (1983, p. 71): The combination of frightening and
inconsistent caregiving makes both avoidant and clingy behaviors
adaptive for these children as they attempt to maintain their care-
giver’s availability. Yet Crittenden certainly did not argue that all
infants who fell outside the Ainsworth et al. (1978) patterns would
necessarily be A/C. For example, Crittenden (1983) notes that one
abused infant was classed as B by her undergraduate coders
because her strange situation behavior was “without either avoid-
ance or ambivalence, she did show stress-related stereotypic head-
cocking throughout the strange situation. This pervasive behavior,
however, was the only clue to the extent of her stress” (p. 75).
Main and Solomon (1990, p. 143) themselves acknowledge that
Crittenden was the first to identify these headcocking forms of
strange situation behavior, which all agree is not a strategic be-
havior but an “indicator of stress.” To take another example of
nonstrategic behavior, Crittenden (1997a) noted that for the chil-
dren of some “unipolar depressed and neglecting parents, nothing
that the children do changes the probabilities of their parents’
behavior. These children appear to use scraps of the strategies in a
self-soothing strategy” (p. 60).
This review now places us in a position to clarify the distinction
between the paradigms. Social workers in particular have purveyed
a mischaracterization of the “disorganized/disoriented” pattern,
which Main calls “widespread” and “dangerous,” which takes her
to believe that maltreatment will always be associated with behav-
ioral breakdown in an infant’s behavior in the Strange Situation
Procedure and vice versa (Main et al., 2011, p. 441). Crittenden
has likewise suffered a widespread and deep mischaracterization
that suggests that the perception of threat will always cause a child
to adopt an A, B, C, or A/C attachment strategy (McMahon &
Ward, 2001, p. 61, is an exception). A close examination of both
paradigms indicates that Main and Crittenden agree on the fact that
“with extreme threat, everyone panics or freezes. Moderate threat
best displays the strategy” (Crittenden, 2008a, p. 277). Both the-
orists believe that not all maltreated infants show a lack of a
coherent strategy in the Strange Situation Procedure—though
many do momentarily and sometimes as an enduring response.
Both Main and Crittenden have argued that nonstrategic behavior
can be caused by multiple factors, though it is true that, since the
1990s, Main has particularly emphasized frightening or dissocia-
tive parental behavior as “one highly specific and sufficient, but
not necessary, pathway to D attachment status” (Hesse & Main,
2006, pp. 310 –311). Rather, the actual difference between the two
theories lies in the fact that Crittenden’s focus on availability,
rather than proximity, as the set goal of the attachment system
leads her to identify intensified patterns of Type A and Type C
information processing in cases where Main perceived behavioral
breakdown. This led to a divergence in the interpretation of Type
C behavior in infancy, the significance of which cannot be over-
estimated in understanding the split between paradigms and the
emergence of the dynamic–maturational model.
Main’s focus was on finding coherent strategies for proximity
seeking as the criterion for “organization.” Main (1981, p. 681)
theorized, therefore, that “behavior can be called disorganised
when it vacillates between opposites without reference to changes
in the environment” and that it “appears in infants reunited with
their mothers while still in the stages of protest or despair and in
the ambivalent infants in the Strange Situation.” Hence, the be-
havior of many infants that the Ainsworth et al. (1978) protocols
might have classified as C tended to be understood by Main
through the concept of “disorganization.” Looking back, Main,
Hesse, and Kaplan (2005, p. 259) reflect that
following the advent of the disorganized infant attachment category
. . . the ambivalent category and its adult equivalent, insecure-
preoccupied, have become rare. This is because many individuals
previously classified as insecure-ambivalent in infancy or preoccupied
in adulthood have been found to be disorganized.
Crittenden, however, had both a wider and different concept of
“organized” behavior. She was therefore brought to ask, “Through
what process could these ambivalent/resistant attachment behav-
iors be serving to maintain the availability of the caregiver?”
Addressing this question, as we shall now see, would lead Crit-
tenden to interpret the difference between Type A and Type C as
reflecting the fact that the basic components of human experience
of danger are two different kinds of information.
Information Processing
Despite the fact that Main and Crittenden differed in their
interpretation of behaviors discrepant to Ainsworth et al.’s (1978)
classificatory protocols, Ainsworth supported work in both para-
digms. Even while serving as Main’s research assistant on a
project to evidence the association between a parent’s experience
of unresolved loss and “D” behaviors in their infant (see Ain-
sworth & Eichberg, 1991), Ainsworth was urging Bowlby to listen,
in person, to Crittenden’s intriguing and audacious theorizing
about information processing. Following Crittenden’s visit to see
Bowlby in London, Ainsworth wrote that she was pleased “that
you had the opportunity to hear about her ideas. I myself think she
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
329
THE DYNAMIC-MATURATIONAL MODEL
is top-notch, and is making an important contribution” (Ainsworth,
1988).
By 1988, Crittenden had begun to go beyond Bowlby in con-
ceptualizing Type A and Type C attachment behaviors as the effect
of different forms of information processing. Her starting point
was the near universality of Types A, B, and C attachment behav-
ior, alongside the fact that behaviors discrepant to Ainsworth et
al.’s (1978) protocols were especially common in her maltreated
doctoral sample. Crittenden (1992c) worked from “a basic premise
of ethology—that universal behaviors often serve functions that
promote survival” (p. 210). She proposed that the basic compo-
nents of human experience of danger are two kinds of information
(see Strathearn, Fonagy, Amico, & Montague, 2009):
1. Affective information: the emotions provoked by the
potential for danger, such as anger or fear. Crittenden
termed this “affective information.” In childhood, this
information would include emotions provoked by the
unexplained absence of an attachment figure.
2. Causal or other sequentially ordered knowledge about the
potential for safety or danger. (Crittenden termed this
“cognitive information”). In childhood, this would in-
clude knowledge regarding the behaviors that indicate an
attachment figure’s availability as a secure haven.
Crittenden proposes that both kinds of information can be split
off from consciousness or behavioral expression as a “strategy” to
maintain the availability of an attachment figure. The term “strat-
egy” is used by Crittenden (1992b) not in “the narrow sense of a
cognitive plan, that is, a response to an articulated problem pre-
ceded by a conscious analysis of behavioral alternatives,” but as a
transformation of information regarding danger that occurs with-
out conscious thought (p. 330).
Crittenden proposed that Type A strategies split off emotional
information about feeling threatened and Type C strategies split
off temporally sequenced knowledge about how and why the
attachment figure is available, whereas Type B strategies effec-
tively utilize both kinds of information without much distortion.
We shall attend to each of these patterns of information processing
in turn.
Type A behavior in the Strange Situation Procedure is inter-
preted by Crittenden as an effect of a process of “splitting.” When
an infant is faced with insensitive or rejecting parenting, one
strategy for maintaining the availability of their attachment figure
is to try to exclude from consciousness or from expressed behavior
any emotional information that might result in rejection. Under
such conditions, emotional information is then termed by Crit-
tenden “negative affects,” not because emotions like anger or fear
are “negative” in themselves but to highlight that the child splits
them off because they might be seen as “negative” by an attach-
ment figure and thereby reduce their availability. Crittenden argues
that splitting off emotional information allows an infant facing
insensitive caregiving to simplify the complexity of the situation
with the neurological means at their disposal: They avoid express-
ing “negative” emotions when they are anxious and, in doing so,
avoid antagonizing or alienating their attachment figure. However,
the segregated affects may find expression within another behav-
ioral system, such as play, when the infant is not anxious and the
attachment system is therefore idling (Crittenden, Partridge, &
Claussen, 1991, p. 497).
Crittenden perceived Type C behavior as representing the op-
posite strategy. Type C strategies distort causal or other temporally
ordered knowledge about the potential for safety or danger. If
knowledge regarding the behaviors that indicate an attachment
figure’s availability as a secure haven is subject to segregation,
then the infant can try to keep the attention of their caregiver
through clingy or aggressive behavior, or alternating combinations
of the two. Such behavior may increase the availability of an
attachment figure who otherwise displays inconsistent or mislead-
ing responses to the infant’s attachment behaviors, suggesting the
unreliability of protection and safety. Type C behavior in the
Strange Situation Procedure is interpreted by Crittenden as an
effect of this form of information processing. The state of affective
arousal of the Type C infant is interpreted by Crittenden (1992a,p.
581) as distinct from, but by degrees continuous with, disorgani-
zation— because the infant adopting this strategy is both “dis-
tressed and unable to process information” about the causality of
caregiver availability.
From 1988 to 1992, Crittenden worked on the development of
her Preschool Assessment of Attachment. From this close obser-
vation of preschool-aged children, Crittenden (1992c, p. 221)
added a further element to her theorization of Type A and Type C
behavior. She proposed that a child develops the neurological
maturity not only to deny negative emotions, as in infancy, but also
to display “false positive” emotions as a Type A strategy. Such
false displays of emotion might include, for example, overbright
caregiving or solicitous behaviors toward the attachment figure. A
child might display such behaviors with a withdrawn caregiver; the
strategic goal of such behavior would then be to try to ensure that
the attachment figure remains engaged and therefore available as a
source of care and protection. Crittenden also added to her account
of Type C the observation that toddlers increasingly have the
neurological maturity to deploy disarming and/or aggressive be-
haviors as a strategy to coerce their caregiver. The caregiver is
distrusted, and the self treated as wronged. Crittenden observes
that these coercive behaviors, although sometimes deployed by all
preschoolers (the “terrible twos”), will particularly dominate the
behavior of children who experience their caregiver as otherwise
unpredictable or untrustworthy in their availability to answer the
child’s needs.
Whereas Type A strategies split off negative affects and Type C
strategies split off causal knowledge about how and why comfort
or abandonment occur, Type B strategies effectively utilize both
kinds of information without much transformation. Crittenden
therefore calls the Type B strategy “balanced.” In the Strange
Situation, this strategy therefore implies protest on the caregiver’s
unexpected departure, and the capacity to be comforted on his or
her return without clinginess or aggression. Crittenden (1992c,p.
230) proposes that as a child develops, however, the primacy of a
Type B strategy will also permit them to flexibly draw on Type A
and Type C strategies as the situation demands, but without com-
pulsion. Crittenden does not perceive Type A and Type C strate-
gies as in themselves problematic, so long as they are not misap-
plied through too general an application. For example, the
individual “balanced” in terms of affect and cognition can both
respond merely “fine” to being asked “how are you?” in the course
of a passing encounter, and also allow a close friend access to their
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
330 LANDA AND DUSCHINSKY
inner fears, aspirations, and reflections. Crittenden (1992a) con-
cludes, therefore, that “these alternative styles of processing infor-
mation appear to match fairly well the secure, ambivalent, and
avoidant patterns of attachment identified in infancy by Ain-
sworth” (p. 581).
Memory Systems
After information processing, the next crucial component of the
dynamic–maturational model is memory. The reason for this is that
if human experience of danger is comprised of two kinds of
information, the ways in which we store and retrieve this infor-
mation become crucially important. They not only will shape what
information from the past is available to the acting subject but also
will profoundly alter how new information is interpreted:
Storing information of different types (e.g., behavior sequences, gen-
eralisations, episodes) creates the possibility of multiple perspectives
on reality. In a sense, the mind has available a multimethod experi-
ment regarding the nature of reality. The experiment succeeds to the
extent that discrepancies among memory systems are noticed, evalu-
ated and resolved. (Crittenden, 1992a, p. 579)
The best of these multimethod experiments are available to
“balanced children” (Type B), who
automatically and preconsciously use new and discrepant information
to revise existing internal representational models to yield progres-
sively more accurate models. Although, at any given moment, the
changes are likely to be minor and nondisruptive, the sum of this
activity is the ongoing reorganization. (Crittenden, 1995, p. 383)
In deepening her analysis of the effects of the Type A and Type
C distortions on information processing, Crittenden drew upon the
work of Schacter and Tulving (1994) to distinguish between four
memory systems. Though Crittenden later attends to other memory
systems as well, these four are the most significant for understand-
ing the effects of alternate forms of information processing. Two
are implicit forms of memory, primarily available preconsciously
to the acting individual; two are explicit forms of memory, avail-
able consciously to the reflecting individual:
1. Procedural memory— contains information about pre-
conscious sensorimotor behaviors and the patterns of
stimuli that serve as cues for such behavior. Developed
and elaborated from infancy, this procedural knowledge
organizes most of our actions most of the time.
2. Imaged memory— contains the sensory perceptions of
past experiences of affective arousal, especially those
relating to feelings of safety or danger. For example,
imaged memory may encode the sound of the raised,
angry voices of one’s attachment figures in an argument,
or it might encode tacit knowledge of the kinds of things
that evoke disgust. In Crittenden’s account, imaged mem-
ory is therefore primarily a form of emotional informa-
tion.
3. Semantic memory— contains generalized descriptions or
predictions about how life works. Crittenden theorizes
semantic memory as descriptive or predictive rules about
temporal sequences, stored in a verbal form. She gives
the “example, ‘good children obey their parents’ (and,
therefore, are not punished)” (Crittenden, 1995, p. 383).
Whereas procedural and imaged memory operate from
infancy, semantic memory first develops at around 18
months. Crittenden understands this memory system as
of primary importance for the development of accurate
knowledge about what and why attachment figures are
available or unavailable under conditions of perceived
threat.
4. Episodic memory—integrates both causal and emotional
information about past events, making recollections
available to an individual. It is, however, “biased to
reflect experiences that recall strong, unresolved feel-
ings” (Crittenden, 1997a, p. 79). Episodic memory be-
comes functional at about 3 years of age.
Crittenden (1995) noted that the A and C strategies of informa-
tion processing result in different access to these types of memory.
The utilization of Type A strategies encourages individuals to
privilege causal information about the availability of an attachment
figure and to split off emotional information. Crittenden therefore
argued that the mobilization of a Type A strategy by an individual
facilitates a prioritization of the storage and recall of semantic
memory— descriptions and prescriptions about how the world
should operate. (Crittenden occasionally implies that Type A strat-
egies may also facilitate the storage and recall of procedural
memory, in which this information is especially relevant to con-
tingencies regarding the availability of safety or danger e.g., Crit-
tenden, 1992a, p. 291). Yet as well as privileging such causal
information, Type A strategies involve splitting off negative emo-
tions, with the goal of maintaining the availability of protection.
As a result, therefore, the affective intensity of imaged and epi-
sodic memories may mean that these are not stored or available for
recall. This produces the effect that psychodynamic theories con-
ceptualize as “repression,” though Crittenden qualified that some
experiences may not have been stored in a retrievable way in the
first place (Crittenden, 1997c, p. 51). With a reduced ability to
access episodic or imaged memories, or integrate them with se-
mantic knowledge, the price of Type A strategies is that they
reduce the ability of an individual to “provide true explanations for
their behavior” (Crittenden, 1995, p. 385).
By contrast, use of Type C strategies encourages individuals
to privilege the storage and recall of imaged information and the
affective components of episodic information. However, they
exclude causal information and, as a result, their thought pro-
cesses and verbal descriptions of events tend to be messy and
unordered. Episodes they relate therefore tend to be ordered
primarily by affective associations, rather than in terms of
temporal succession or consequences. Hence, Crittenden (1995)
observed that those who are dependent upon Type C strategies
can be observed as showing “a ‘lack of recall’ when asked
to provide (semantic) adjectives regarding their attachment
relationships and when asked to access episodic memory
through these semantic adjectives. Once in episodic memory,
however, there should be little evidence of failure to recall” (p.
386), though these attachment-related episodes will tend to lack
temporal order as a result of the distortions of causal informa-
tion enacted by the Type C strategy.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
331
THE DYNAMIC-MATURATIONAL MODEL
Maturation
Crittenden built upon not only Bowlby’s account of information
processing but also his insight that development is best conceptu-
alized using the metaphor of “pathways.” This metaphor empha-
sizes two points. The first is that a strategy is likely to change over
time as the child matures and circumstances change. As a result, “a
given pathway may continue straight or may branch in ways that
may lead to other pathways” (Crittenden, 1997b, p. 51).
This potential for change in strategic functioning, as a result of
maturation or changing circumstances, has two dimensions for
Crittenden. A first dimension is “reorganization.” On the basis of
her account of Type B as balanced between the alternate forms of
information, and the potential at each stage of development for
reorganization, Crittenden contests Main’s emphasis on continuity
in a Type B classification as necessarily optimal. Although main-
tenance of Type B from infancy to adulthood may well be the most
comfortable state, and may be associated with many positive
outcomes, the capacity to achieve an “earned” B status through
reorganization has its own potential. From late childhood, and
particularly when an individual has been supported in learning to
reflect, deliberate reorganization may occur through metacognitive
reflection on potential discrepancies between memory systems
(Crittenden & Poggioli, 2012, p. 395). In this way, the opportunity
is opened for noticing and correcting distortions in how we process
or store our experiences.
For instance, old relationships may change, and new relation-
ships occur throughout maturation and may be the basis for reor-
ganization if an attachment relationship is formed with this figure
or if they serve to facilitate or provoke reflection. Such a new
relationship may be with a clinician or other professional. Hence,
“observing videotaped parent– child interactions with the parent
and discussing these observations from the parent’s perspective
can be a powerful means of creating communication between
procedural and semantic memory systems” (Crittenden, 1992a,p.
593). Or psychotherapy may reveal the continued role played in an
individual’s present by imaged or episodic memories of danger in
childhood situations in which they were told by their attachment
figures that they should experience comfort. Crittenden (1997a)
identifies that metacognitive reflections on the self can lead to an
increase in the repertoire of strategies available to a person, and, in
the process, can be the basis for wisdom and a tempered creativity
(p. 83).
A second dimension of this potential for change is that infor-
mation processing may become further distorted as neurological
development progresses or as experiences of inadequate caregiv-
ing continue or intensify. In particular, Crittenden emphasizes the
neurological changes associated with the change from infancy to
preschool age and from school age to adolescence are periods of
particular, predicable opportunities for change in an individual’s
strategy. As we have seen, for example, the Type A strategy in
infancy involves the exclusion of negative affects—as an adapta-
tion to a caregiving environment that is unresponsive to or reject-
ing of the infant’s attachment behaviors. However, by preschool
age, a child also has the neurological tools at their disposal to also
use false-positive affects, with the goal of maintaining the avail-
ability of the caregiver. Addressing adolescence as another impor-
tant period of neurological change, Crittenden (1997a, pp. 54 –55)
proposed that “after puberty species-specific patterns of sexual
behavior (mental and physical) become powerful influences on
behavior.” Further distortions of information processing that occur
in adolescence can therefore lead, for example, to dangerous
sexual entanglements as an individual excludes the negative
emotions associated with such encounters (a Type A strategy).
Crittenden thus contrasts the developmental focus of her dynamic–
maturational model of attachment and adaptation with the ap-
proach of Main, which does not specifically integrate the potential
for sexuality and reproduction that develops during maturation.
Crittenden also emphasizes that as information processing be-
comes more distorted, the strategy is applied more uniformly in
response to anxiety-provoking situations. The most distorted forms
of Type A or Type C transformation of perception and behavior
become increasingly generated and maintained by the self rather
than strategic adaptations to circumstances (Crittenden et al., 1991,
p. 495). Yet no matter how intensely a strategy is deployed,
Crittenden asserts that there is always the potential for “reorgani-
zation”—a movement toward balanced (Type B) strategies. This
would mean, as we have seen, both a reduction in the distortion of
information and flexible access to Type A or Type C strategies as
the situation warrants. To illustrate, a toddler may have come to
depend upon a Type C strategy of tantrums in aiming to maintain
the availability of an attachment figure whose inconsistent avail-
ability has led the child to distrust or distort causal information
about their apparent behavior. This may lead their attachment
figure to get a clearer grasp on their needs and the appropriate
response to their attachment behaviors. Experiencing more reliable
and predictable information about the availability of their attach-
ment figure, the toddler then no longer needs to use coercive
behaviors with the goal of maintaining their caregiver’s availabil-
ity. Instead, they can experience and utilize both or either emo-
tional and sequential information about potential dangers to inter-
pret their experiences.
Knowledge and Maltreatment
With Crittenden’s information-processing model integrated with
an account of the effects of maturation and her interpretation of
Schacter and Tulving’s work on memory systems, we now have
the three core components of the dynamic–maturation model in
place. What can it do? Perhaps one of the most distinctive appli-
cations of the dynamic–maturation model has been Crittenden’s
sustained effort to make sense of the experiences of maltreating
parents. In a remarkable study, Crittenden, Lang, Claussen, and
Partridge (2000) explored the semantic knowledge about caregiv-
ing of abusing, neglecting, and adequate mothers. The research
found that abusing and adequate parents were found to have the
same level of explicit knowledge about how to meet the needs of
their child. By contrast, neglecting mothers were found to have
dramatically lower levels of parental reasoning. Furthermore, they
found that “there were no effects for parenting intervention” on
parental reasoning or on the child’s attachment classification;
“although this is disappointing, the results of this study may
shed light on the problem. The mothers were all given the same
parenting intervention, which consisted of information about
child development and parenting strategies” (Crittenden et al.,
2000, p. 233).
Crittenden’s integration of an account of memory systems and
developmental pathways into her foundational information-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
332 LANDA AND DUSCHINSKY
processing model offers predictions about the role played by
distortions of information in causing harmful behavior. She con-
cludes that the abusing parents are not accessing their semantic
memories containing descriptions and prescriptions of how to care
for their children when engaged in caregiving. Instead, “the be-
havior of the abusing mothers appears to be affectively triggered
through imaged and episodic representations of previous affec-
tively similar experiences” (Crittenden, 2000, p. 230). By contrast,
the behavior of the neglecting parents is understood by Crittenden
(1993) as caused by “preconscious exclusion from perception of
information that elicits affect” (p. 33), which may further cause
these parents not to perceive or act upon their semantic informa-
tion regarding caregiving. One reason for this may be “depres-
sion,” a term that Crittenden means in the technical sense of a
“failure of the strategy to resolve to problems to which the strategy
is applied, combined with . . . awareness of this” (Crittenden,
1995, p. 398; Crittenden & Landini, 2011, p. 257).
Hence, increasing the knowledge of either the abusing or the
neglecting parents about child development therefore has no effect,
for different reasons, on the likelihood of abusing and neglecting
behaviors. The striking implication is that “many parents and
almost all troubled parents will be unable to tell accurately why
they did what they did when they did it” and that “therefore, it will
usually be futile—and often be misleading—to ask the parent
directly why they did what they did” (Crittenden, 2008a, p. 124).
The dynamic–maturational model offers three mechanisms that
explain the fact that individuals may act upon information from
other memory systems that in fact run counter to their own se-
mantic descriptive and prescriptive knowledge of themselves and
their lives.
The first can occur with individuals utilizing either Type A or
Type C strategies. Crittenden notes that anxiety can short-circuit
information processing. In doing so, she argues against a tendency
within some branches of attachment research to treat an individ-
ual’s representational models “as ‘things’ that a person ‘has’ and
that ‘contain’ information” (Crittenden & Landini, 2011, p. 64).
Instead, she contends that we have a variety of different informa-
tion about our attachment figures, which takes the form of a
disposition for a certain kind of action or lack of action when we
are anxious. Given time to process our response to a situation, we
can integrate these different sources of information into a consid-
ered response. However, when we are anxious, there is a pressure
to short-circuit this integrative processing and just go with the
behavior that seems most appropriate in the moment. Crittenden
and Landini (2011, p. 64) observe that, without reflection, “the
‘strongest’ dispositional representation active at that moment
would regulate behavior.” This dispositional representation could
come from any of the memory systems, but because higher-level
representations such as those from semantic or episodic memory
“require more extensive processing (and therefore more time),
early termination of processing will result in a bias toward enact-
ment of procedural and imaged dispositional representations”
(Crittenden & Landini, 2011, p. 64). This judgment will be pow-
erfully influenced by the kinds of imaged and procedural memo-
ries available to the individual from their childhood (Crittenden,
2008b), but may also be impacted by other processes that change
information processing, such as drug use (Crittenden & Claussen,
2002).
A second mechanism potentially causing behavior to run coun-
ter to semantic descriptions and prescriptions of how one should
act is intensive use of Type C strategies. The Type C strategy
facilitates an individual to omit semantic information about the
temporal sequences and, in doing so, to experience exaggerated
feelings of anger and/or fear. In adulthood, Crittenden theorizes
that the strategies are concerned not only with self-preservation but
the protection of offspring from potential threats. However, the
distortion of information processing associated with this strategy
can cause an individual to unduly interpret the behaviors of their
offspring as threatening. Hence, applied as an approach to parent-
ing, a Type C strategy may lead an adult to false reasoning about
the meaning of their child’s behaviors. For example, an infant’s
cries may be interpreted as deliberate attempts to annoy the care-
giver. These interpretations may be intensified by the unpredict-
able upsurges in “invulnerable feelings (anger) or the vulnerable
ones (fear and desire for comfort)” (Crittenden, 2008a, p. 210)
characteristic of a Type C strategy, in which one affect or the other
has been split off (see Hautamäki, Hautamäki, Neuvonen, &
Maliniemi-Piispanen, 2010).
A third mechanism that can cause behaviors that run counter to
semantic knowledge is the intensive use of Type A strategies. As
we have seen, Crittenden built her information-processing model
on the foundations laid by Bowlby’s (1980) account of “segregated
systems.” Bowlby predicted that motivational dispositions, feel-
ings, and memories could be “segregated” if they were understood
to interfere with the likelihood of being protected. Although kept
from consciousness or expression in behavior under situations of
anxiety, these split-off elements could reappear another time when
evoked by some aspect of the situation. Anger evoked by rejection
by an attachment figure when a child feels potentially threatened
and is anxious may, on another occasion, be expressed at a play-
mate or sibling. When these occur without the individual’s voli-
tion, such out-of-place reappearances of affects or dispositions to
act are termed by Crittenden as “intrusions” into a Type A strategy.
Crittenden and Landini (2011, p. 269) observe that
children placed in care, especially more than once, often have intru-
sions. In video SSP, they tend to occur when a rejected/neglected
child approaches the stranger in an intrusion of desire for comfort,
then loses muscular control and falls to the floor, overwhelmed by the
intruding fear of the unknown . . . person.
Crittenden and Landini (2011, p. 269) suggest that intrusions are
particularly common in two circumstances. First, intrusions may
occur when the segregated affects are especially powerful and little
integrated, and the ongoing work of keeping them at bay is
disrupted—for example, by social or developmental transitions.
Second, intrusions may function to generate a feeling of vitality in
individuals who have compulsively deadened their feelings. Illus-
trative of both points, Crittenden (2008a, pp. 193–195) gives the
example of a woman, Kate, who, since the birth of her second son,
had been hearing voices telling her to cut herself and dreams of
chopping up babies. Crittenden identifies the Type A strategies
that Kate had used, growing up in a children’s home and in a failed
foster placement, to maintain the availability of someone attentive
to her needs. Looking back across Kate’s history, Crittenden
(2008a, pp. 195–196) notes different forms of intrusion:
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
333
THE DYNAMIC-MATURATIONAL MODEL
In the past, these were excitingly promiscuous and dangerous sexual
encounters, but now, desire for pain (expressed as cutting) fulfilled the
function of generating arousal. The repeated instances of cutting
functioned, however briefly, to increase her arousal, to cause others to
express extreme distress, and to attract others, especially profession-
als, to care for her.
Crittenden (2008a) notes that “dangerous as cutting is, it can
function to reverse equally dangerous low arousal” and “it gave
Kate access to professional caregivers in institutional settings.
These, of course, had been her secure base since infancy” (pp.
195–196). Crittenden and Newman (2010, p. 435) therefore hy-
pothesize that, in some cases they observed, “depression and
intrusions appeared to work in tandem regulating arousal” in the
absence of other functioning forms of emotion regulation.
Trauma
Though Crittenden has long been interested in trauma, the issue
has received further elaboration in recent years. Attending to this
topic can help further draw out the precise conceptual and termi-
nological differences between Main and Crittenden, and Crit-
tenden’s distinctive concern with “how disturbed individuals use
information, rather than simply finding that they are not inte-
grated” (Crittenden & Landini, 2011, p. 7). By “trauma,” Crit-
tenden refers to the psychological experience of emotionally or
physically threatening circumstances that cannot be subjected to
effective information processing. This information-processing per-
spective makes sense of the fact that children are especially vul-
nerable to trauma: They are “less able to understand” the meaning
of experiences of danger than adults and “less able to store,
retrieve, and integrate” the meanings they do derive (Crittenden &
Landini, 2011, p. 250). It also makes sense of the fact that “threats
to safety and reproduction more often yield trauma then other sorts
of threat,” because “the brain has ‘hard-wired’ preferences regard-
ing attention; in all species, these function to identify information
relevant to danger and reproduction” (1997c, p. 36). Trauma
results in entrenched, systematic errors in processing with the
result that “either too much irrelevant information is retained (and
used to organize behavior), or too much relevant information is
discarded, or other errors of thought are made regarding the
dangerous event” (Crittenden & Landini, 2011, p. 236). Any of
these outcomes make maladaptive behavior in the present more
likely.
One possibility is that that the trauma will result in an intensi-
fication of a Type C strategy. If causal descriptive or prescriptive
information is disregarded or imaged, or episodic memories are
overemphasized, then a more intensive Type C strategy may be the
result. Crittenden (1997c, p. 52) calls this outcome “preoccupying”
unresolved trauma, and notes that it may turn out to be adaptive if
the rumination leads to an integrative reorganization. A second
possibility is that trauma will result in an intensification of a Type
A strategy. If affective information is disregarded, imaged or
episodic memories are not processed, or semantic information
overemphasized, then the result will be a more extreme form of
Type A thinking and behavior. This would reinforce tendencies to
exclude the emotional significance of the unresolved trauma, pro-
ducing what Crittenden calls “dismissed trauma”—a possibility
she feels is insufficiently considered in Main’s work (Crittenden &
Landini, 2011, p. 240).
Yet it is also possible for information processing and behavior to
become “nonstrategic for at least a while” (Crittenden & Landini,
2011, p. 254). A key form of “broken” strategy addressed by
Crittenden (2008a, p. 102) is “disorientation,” a state in which an
individual suffers from “confusion of information from different
sources (e.g., the self now, the self in the past, one’s mother, one’s
religious guide, etc.).” Crittenden (2008a, p. 118) proposes that
disorientation is particularly likely as a result of “dismissed child-
hood traumas.” Signs of disorientation indicate that the individual
“is anxious to select an effective strategy and does not know how
to do that (and up-regulates arousal)” (Crittenden & Landini, 2011,
p. 307). It is very important to recognize, however, that Crittenden
means something different by the term “disorientation” than Main.
Main and Solomon (1986) identify “disorientation” as a dimension
of the disorganized/disoriented “D” classification, in which behav-
ior seems to signal “a lack of orientation to the immediate envi-
ronment” (p. 97). Yet many of the behaviors Main and Solomon
identify as indicative of disorientation in young children are con-
sidered to be potentially “organized” by Crittenden— because of
her wider and different definition of the term. Momentary signs of
what Main and Solomon would call “disorientation” may well, for
Crittenden, be strategic if they function to give a child “extra time
for information gathering or processing,” perhaps “enabling the
child to make transitions from one behavior pattern to another”
(Crittenden, 1992c, p. 226).
For example, Crittenden (1992c, p. 226) relates a case in which
in the Strange Situation, the compliant child of a hostile mother was
observed actively approaching the door to seek her absent mother. As
she did so, the door suddenly opened and the mother appeared. The
child stopped very abruptly, froze, recovered her balance, did a cute
little “dance,” then turned to pick up a toy and engaged the stranger.
Crittenden observes that within the Main and Solomon (1986)
classification, this “would probably be considered evidence of
disorganisation,” whereas Crittenden argues that “the child’s be-
havior would be evidence of the highly organized and flexible use
of two different patterns of behavior to implement the defended
strategy of maintaining access without closeness” (1992c, p. 226).
Crittenden suggests that “simply turning away, when she was so
obviously approaching her mother, might be perceived by the
mother as a blatant rebuff, thereby eliciting the very anger the child
was seeking to avoid. Hence the ‘appeasement dance’ before the
substitution of an acceptable alternate engagement with a toy and
the stranger” (1992c, p. 226). Crittenden therefore concludes that
“the appearance of disorganisation occurred because the child
needed both to change her behavior and to cover that change in
order to avoid appearing to snub the mother” (1992c, p. 226).
Hesse and Main (2006, pp. 310 –311), in a later formulation, have
in fact likewise stated that they did not ever intend to imply that
disorientation or disorganization would “necessarily” be the result
of frightened, frightening or dissociative processes and that “some
D behavior appear[s] to be the outcome of a readily comprehen-
sible conflict, or of simple confusion.”
Critiques
Crittenden identifies subtypes of the Types A, B, and C strate-
gies, running from B1 to B5, A1 to A8, and C1 to C8 (see Figure
1). She places these in a circle, with B3 at the top as a state of
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
334 LANDA AND DUSCHINSKY
undistorted information processing, with A strategies and C strat-
egies of ascending numerals running down either side and repre-
senting increasing degrees of distortion.
Crittenden identifies two different patterns at each level of
information-processing distortion. For Type A, “the odd-numbered
patterns increasingly idealize the attachment figure whereas the
even-numbered patters increasingly negate the self” (Crittenden &
Landini, 2011, p. 42); the Type C patterns “involve an even-odd
alternation of displaying angry invulnerability with vulnerable fear
and desire for comfort (i.e., C1–2, C5– 6, etc.)” (p. 43). If odd and
even numbers are taken to be equivalent levels of distortion to
information processing, it can be observed that “as the numeral
increases, these states are increasingly generated and maintained
by the self” and “there is an increase in both the extent of distortion
of information and the uniformity with which the strategy is
applied to all perceived threats, appropriately and inappropriately”
(p. 112). The strategies thus numerically advance, it might be said,
from being “states” that occur under conditions of perceived threat,
to being stable and entrenched “traits.”
A crucial point is almost universally missed, even by thoughtful
commentators on the range of Type A and C strategies presented
in the dynamic–maturational model. Schuengel (2001), for exam-
ple, criticizes Crittenden’s proposed additional subtypes beyond
preschool age (represented by the numerals A5 or C5 and higher)
as lacking evidence. We would somewhat agree with this concern,
but argue that this common criticism of the evidence base and
exhaustiveness of Crittenden’s account of behavioral strategies
does not reduce the acuity of the dynamic–maturational model.
This is because the subtypes do not organize the fundamental
structure of Crittenden’s theory. Crittenden did not propose the
subtypes with the intention that they be exhaustive, as her critics
have generally presumed. For instance, some strategies are under-
stood barely to occur on their own—such as A5 (“compulsive
promiscuity”; Crittenden & Landini, 2011, p. 166). Or again,
Crittenden (2008a, p. 69) fully acknowledges that she does not yet
have studies to support the addition of the A8 category (“externally
assembled self”): the success of the dynamic–maturational model
does not ride on whether this A8 personality structure is the only
possible form of extreme affective distortion.
Ainsworth et al. (1978, p. 235) recall that, in forming the ABC
classifications, “the subgroups were identified first, in the process
of grouping together strange-situation protocols that were maxi-
mally similar . . . it was through examination of the similarities
among members of each subgroup that our attention was first
drawn to those variables.” Although following Ainsworth in em-
phasizing the importance of subgroups, Crittenden does not give
them a foundational status within her theory. The point missed by
most commentators is that “the patterns are not categorical. To the
contrary, they are best described in terms of two dimensions”
(Crittenden, 2000, p. 371). Crittenden (1995) explicitly states that
“the categories identify areas on the dimensional framework; that
is, they ‘tack down’ points defined by the dimensions. Whether
Figure 1. Dynamic maturational model of patterns of attachment in adulthood (adapted from Crittenden,
2008a, p. 70, with permission of the author).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
335
THE DYNAMIC-MATURATIONAL MODEL
any individuals exactly fit these points, especially the extreme
points, is not critical to the argument about the dimensions” (pp.
388 –389). The key dimension, represented by movement down the
circumference of the circle, is the degree of Type A or Type C
distortions of information processing. Evidence for the theoretical
primacy of the dimensions over the particular subtypes identified
within them is that Crittenden herself has used statistical tests (e.g.,
ANOVAs), which presume that the dependent variable is contin-
uously distributed when assessing the effect of particular factors
on the attachment and information-processing strategy used by a
sample. In running these tests, odd and even numbers are put
together with each a step of distortion, and, in doing so, a 4-point
scale is produced. Others using the dynamic–maturational model
have also, “with Crittenden’s input,” treated the severity of distor-
tion of information processing as “a quasi-continuous variable”
(Kwako, Noll, Putnam, & Trickett, 2010, p. 413).
The dynamic–maturational model is an intentionally limited
psychological theory: “Development is multidimensional, with
factors other” than Crittenden’s (1995) “central constructs” of
caregiving, information-processing and maturation “affecting vari-
ation in developmental pathways” (pp. 388 –389). This can be
illustrated with the case of the A5 promiscuous strategy, which
Crittenden has utilized as an explanation for the sexualized behav-
ior of adolescents (Crittenden & Poggioli, 2012). As social psy-
chologists and sociologists have demonstrated, sexualized behav-
ior must be understood in the context of societal norms regarding
the meaning of sexuality. There has been a rise in pressure on
women to achieve an ideal of free, self-possessed desirability—
“compulsory heterosexuality”—to which women need to aspire for
their gender performance to be understood as acceptable. This
compulsory heterosexuality, however, intersects with age norms
regarding the limited readiness of young people for sexual activity
and reproduction (Duschinsky, 2011;Renold & Ringrose, 2011).
The dynamic–maturational model can explain what role distortions
of information processing may play in facilitating heterosexualized
behaviors. Yet the form of such behaviors is indelibly shaped by
cultural norms and expectations—for instance, the historically
unprecedented acceptance of pole dancing or genital hair waxing
as signifiers of ordinary “sexiness” (Cokal, 2007). Crittenden
asserts that the dynamic–maturational model “is a metatheory.
That is, it integrates the best contributions of other theories (as
opposed to competing with them)” (Crittenden & Poggioli, 2012,
p. 410). Looking to the future, will the dynamic–maturational
model integrate further insights from social psychology and soci-
ology, or will it remain as a powerful integration of developmental
psychology with cognitive science?
Conclusions
The dynamic–maturational model brings a rich and systematic
interpretive framework to the study of diverse phenomena. It is
powered by Crittenden’s commitment to the idea that theory that
integrates developmental psychology with cognitive science has
the potential to make sense of and offer guidance on how to
alleviate suffering. The theory at times exceeds the available data
and becomes speculative; yet, despite this, psychiatrists, psychol-
ogists, and social workers are already finding it useful in clinical
practice. We would argue that this is because the strengths of the
dynamic–maturational model lie in more than its current empirical
validity, and rather in the way in which it offers an integrated and
suggestive interpretive framework. We therefore agree with Crit-
tenden (1992c, p. 235) that “tested as specified hypotheses using
sound research methods,” the dynamic–maturational model of
attachment and adaptation “possesses the potential to enlighten the
interpretation of data,” whether this is the data of research findings
or of clinical experience.
References
Ainsworth, M. (1972). Attachment and dependency: A comparison. In J.
Gewirtz (Ed.), Attachment and dependency (pp. 97–137). Washington,
DC: Winston.
Ainsworth, M. (1974). Letter to John Bowlby. Unpublished manuscript.
Ainsworth, M. (1985a). Letter to John Bowlby. Unpublished manuscript.
Ainsworth, M. (1985b). Letter to John Bowlby. Unpublished manuscript.
Ainsworth, M. (1985c). Patterns of mother–infant attachments. Bulletin of
the New York Academy of Medicine, 61, 792– 812. Retrieved from
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1911899/
Ainsworth, M. (1988). Letter to John Bowlby. Unpublished manuscript.
Ainsworth, M. (1990). Epilogue. In M. T. Greenberg, D. Ciccheti, & E. M.
Cummings (Eds.), Attachment in the preschool years (pp. 463– 488).
Chicago, IL: Chicago University Press.
Ainsworth, M., Blehar, M., Waters, E., & Wall, S. (1978). Patterns of
attachment: A psychological study of the strange situation. Mahwah, NJ:
Lawrence Erlbaum.
Ainsworth, M., & Eichberg, C. G. (1991). Effects on infant–mother at-
tachment of mother’s unresolved loss of an attachment figure or other
traumatic experience. In C. M. Parkes, J. Stevenson-Hinde, & P. Marris,
Attachment across the life cycle (pp. 160 –183). London, UK: Routledge.
Ainsworth, M., & Wittig, B. (1969). Attachment and exploratory behavior
of one-year-olds in a strange situation. In B. Foss (Ed.), Determinants of
infant behavior (Vol. 4, pp. 113–136). London, UK: Methuen.
Bowlby, J. (1960). Separation anxiety. The International Journal of Psy-
choanalysis, 41, 89 –113. Retrieved from http://onlinelibrary.wiley.com/
journal/10.1111/(ISSN)1745-8315
Bowlby, J. (1980). Attachment and loss: Loss, sadness and depression
(Vol. 3). New York, NY: Basic Books.
Cokal, S. (2007). Clean porn: The visual aesthetics of hygiene, hot sex, and
hair removal. In A. C. Hall & M. J. Bishop (Eds.), Pop-porn: Pornog-
raphy in American culture (pp. 137–153). Westport, CT: Praeger.
Crittenden, P. M. (1983). Mother and infant patterns of attachment (Un-
published doctoral dissertation). University of Virginia, Charlottesville.
Crittenden, P. M. (1992a). Treatment of anxious attachment in infancy and
early childhood. Development and Psychopathology, 4, 575– 602. doi:
10.1017/S0954579400004880
Crittenden, P. M. (1992b). Children’s strategies for coping with adverse
home environments. Child Abuse & Neglect, 16, 329 –343. doi:10.1016/
0145-2134(92)90043-Q
Crittenden, P. M. (1992c). Quality of attachment in the preschool years.
Development and Psychopathology, 4, 209 –241. doi:10.1017/
S0954579400000110
Crittenden, P. M. (1993). An information-processing perspective on the
behaviour of neglectful parents. Criminal Justice & Behaviour, 20,
27– 48. doi:10.1177/0093854893020001004
Crittenden, P. M. (1995). Attachment and psychopathology. In S. Gold-
berg, R. Muir, & J. Kerr (Eds.), John Bowlby’s attachment theory:
Historical, clinical, and social significance (pp. 367– 406). New York,
NY: Analytical Press.
Crittenden, P. M. (1997a). Patterns of attachment and sexual behaviour. In
L. Atkinson & K. J. Zuckerman (Eds.), Attachment and psychopathology
(pp. 47–93). New York, NY: Guilford Press.
Crittenden, P. M. (1997b). Truth, error, omission, distortion, and decep-
tion: The application of attachment theory to the assessment and treat-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
336 LANDA AND DUSCHINSKY
ment of psychological disorder. In S. M. C. Dollinger & L. F. DiLalla
(Eds.), Assessment and intervention across the lifespan (pp. 35–76).
Hillsdale, NJ: Lawrence Erlbaum.
Crittenden, P. M. (1997c). Toward an integrative theory of trauma: A
dynamic–maturational approach. In D. Cicchetti & S. Toth (Eds.), The
Rochester Symposium on Developmental Psychopathology, Vol. 10.
Risk, trauma, and mental processes (pp. 34 –84). Rochester, NY: Uni-
versity of Rochester Press.
Crittenden, P. M. (2000). A dynamic–maturational exploration of the
meaning of security and adaptation. In P. M. Crittenden & A. Hartl
Claussen (Eds.), The organisation of attachment relationships (pp. 358 –
383). Cambridge, UK: Cambridge University Press.
Crittenden, P. M. (2008a). Raising parents: Attachment, parenting, and
child safety. London, UK: Routledge.
Crittenden, P. M. (2008b). Why do inadequate parents do what they do? In
O. Mayseless (Ed.), Parenting representations (pp. 388 – 433). Cam-
bridge, UK: Cambridge University Press.
Crittenden, P. M. (2012, September). Discussant’s remarks. Paper pre-
sented at the Attachment, Assessment & Treatment Conference, Frank-
furt, Germany.
Crittenden, P. M., & Claussen, A. H. (2002). Developmental psychopa-
thology perspectives on substance abuse and relationship violence. In C.
Wekerle & A. M. Wall (Eds.), The violence and addiction equation (pp.
44 –63). Philadelphia, PA: Brunner-Mazel.
Crittenden, P. M., & Landini, A. (2011). Assessing adult attachment: A
dynamic–maturational approach to discourse analysis. New York, NY:
W. W. Norton.
Crittenden, P. M., Lang, C., Claussen, A. H., & Partridge, M. F. (2000).
Relations among mothers’ dispositional representations of parenting. In
P. M. Crittenden & A. Hartl Claussen (Eds.), The organisation of
attachment relationships (pp. 214 –233). Cambridge, UK: Cambridge
University Press.
Crittenden, P. M., & Newman, L. (2010). Comparing models of borderline
personality disorder. Clinical Child Psychology and Psychiatry, 15,
433– 451. doi:10.1177/1359104510368209
Crittenden, P. M., Partridge, M. F., & Claussen, A. H. (1991). Family
patterns of relationship in normative and dysfunctional families. Devel-
opment and Psychopathology, 3, 491–512. doi:10.1017/
S0954579400007653
Crittenden, P. M., & Poggioli, D. (2012). PTSD in the context of childhood
endangerment: Implications for assessment and treatment. In V. Ardino
(Ed.), Post-traumatic syndromes in childhood and adolescence (pp.
391– 415). New York, NY: Wiley. doi:10.1002/9780470669280.ch20
Duschinsky, R. (2011). Ideal and unsullied: Purity, subjectivity, and social
power. Subjectivity, 4, 147–167. doi:10.1057/sub.2011.2
Farnfield, S., Hautamäki, A., Nørbech, P., & Sahhar, N. (2010). DMM
assessments of attachment and adaptation: Procedures, validity and
utility. Clinical Child Psychology and Psychiatry, 15, 313–328. doi:
10.1177/1359104510364315
Hautamäki, A., Hautamäki, L., Neuvonen, L., & Maliniemi-Piispanen, S.
(2010). Transmission of attachment across three generations. Clinical
Child Psychology and Psychiatry, 15, 347–354. doi:10.1177/
1359104510365451
Hesse, E., & Main, M. (2006). Frightened, threatening, and dissociative
parental behavior in low-risk samples: description, discussion, and in-
terpretations. Development and Psychopathology, 18, 309 –343. doi:
10.1017/S0954579406060172
Holmes, J. (2004). Disorganized attachment and borderline personality
disorder: A clinical perspective. Attachment & Human Development, 6,
181–190. doi:10.1080/14616730410001688202
Holmes, P., & Chimera, C. (2010). Attachment and the court process.
Association of Lawyers for Children, 46, 23–26. Retrieved from http://
www.alc.org.uk/
International Association for the Study of Attachment. (2012). DMM
publications. Retrieved from http://www.iasa-dmm.org/images/uploads/
DMM%20Publications%20Jan%202012.pdf
Kwako, L. E., Noll, J. G., Putnam, F. W., & Trickett, P. K. (2010).
Childhood sexual abuse and attachment: An intergenerational perspec-
tive. Clinical Child Psychology and Psychiatry, 15, 407– 422. doi:
10.1177/1359104510367590
Landa, S., & Duschinsky, R. (in press). Letters from Ainsworth: Contesting
the meaning of “organization” in attachment theory. Journal of the
Canadian Academy of Child and Adolescent Psychiatry.
Main, M. (1977). Analysis of a peculiar form of reunion behaviour seen in
some day-care children. In R. Webb (Ed.), Social development in child-
hood (pp. 33–78). Baltimore, MD: The Johns Hopkins University Press.
Main, M. (1979). The “ultimate” causation of some infant attachment
phenomena. Behavioral and Brain Sciences, 2, 640 – 643. doi:10.1017/
S0140525X00064992
Main, M. (1981). Avoidance in the service of attachment. In K. Immelman,
G. Barlow, L. Petrinovitch, & M. Main (Eds.), Behavioral development
(pp. 651– 693). Cambridge, UK: Cambridge University Press.
Main, M. (1990). Cross-cultural studies of attachment organisation. Human
Development, 33, 48 – 61. doi:10.1159/000276502
Main, M., Hesse, E., & Hesse, S. (2011). Attachment theory and research:
Overview with suggested applications to child custody. Family Court
Review, 49, 426 – 463. doi:10.1111/j.1744-1617.2011.01383.x
Main, M., Hesse, E., & Kaplan, N. (2005). Predictability of behavior and
representational processes related to attachment in the Berkeley Longi-
tudinal Study. In K. E. Grossmann, K. Grossmann, & E. Waters (Eds.),
Attachment from infancy to adulthood (pp. 245–304). New York, NY:
Guilford Press.
Main, M., & Solomon, J. (1986). Discovery of a new, insecure-
disorganized/disoriented attachment pattern. In M. Yogman & T. B.
Brazelton (Eds.), Affective development in infancy (pp. 95–124). Nor-
wood, NJ: Ablex.
Main, M., & Solomon, J. (1990). Procedures for identifying infants as disor-
ganised/disoriented during the Ainsworth Strange Situation. In M. T . Green-
berg, D. Cicchetti, & E. M. Cummings (Eds.), Attachment in the preschool
years (pp. 121–160). Chicago, IL: University of Chicago Press.
McMahon, L., & Ward, A. (2001). Helping families in family centres.
London, UK: Jessica Kingsley.
Milan, S., Snow, S., & Belay, S. (2009). Depressive symptoms in mothers
and children. Developmental Psychology, 45, 1019 –1033. doi:10.1037/
a0016164
Pocock, D. (2010). The DMM—wow! But how to safely handle its
potential strength? Clinical Child Psychology and Psychiatry, 15, 303–
311. doi:10.1177/1359104510369457
Renold, E., & Ringrose, J. (2011). Schizoid subjectivities? Re-theorizing
teen girls’ sexual cultures in an era of “sexualization”. Journal of
Sociology, 47, 389 – 409. doi:10.1177/1440783311420792
Robson, K., & Wetherell, A. (2011). The use of the dynamic maturational
model of attachment and adaptation within a systemic setting in
CAMHS. The Cumbria Partnership Journal of Research Practice and
Learning, 1, 48 –52. Retrieved from www.cumbriapartnership.nhs.uk
Schacter, D., & Tulving, E. (1994). Memory systems. Cambridge, MA:
MIT Press.
Schuengel, C. (2001). The organization of attachment relationships: mat-
uration, culture, and context edited by Patricia McKinsey Crittenden and
Angelika Hartl Claussen. Contemporary Psychology, 46, 579 –581.
Solomon, J., & George, C. (2011). The disorganized attachment-caregiving
system. In J. Solomon, & C. George (Eds.), Disorganized attachment &
caregiving (pp. 1–25). New York, NY: Guilford Press.
Spieker, S., & Crittenden, P. M. (2010). Comparing two attachment clas-
sification methods applied to preschool strange situations. Clinical Child
Psychology & Psychiatry, 15, 97–120. doi:10.1177/1359104509345878
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
337
THE DYNAMIC-MATURATIONAL MODEL
Sroufe, A., & Waters, E. (1977). Attachment as an organizational con-
struct. Child Development, 48, 1184 –1199. Retrieved from http://
onlinelibrary.wiley.com/journal/10.1111/(ISSN)1467-8624
Strathearn, L., Fonagy, P., Amico, J. A., & Montague, P. R. (2009). Adult
attachment predicts mother’s brain and peripheral oxytocin response to infant
cues. Neuropsychopharmacology, 34, 2655–2666. doi:10.1038/npp.2009.103
Svanberg, P. O. (2009). Promoting a secure attachment through early
assessment and interventions. In J. Barlow & P. O. Svanberg (Eds.),
Keeping the baby in mind (pp. 100 –114). London, UK: Routledge.
Van IJzendoorn, M., Schuengel, C., & Bakermans-Kranenburg, M. (1999).
Disorganized attachment in early childhood: Meta-analysis of precur-
sors, concomitants, and sequelae. Development & Psychopathology, 11,
225–250. http://dx.doi.org/10.1017/S0954579499002035
Vetere, A. (2004, October). Citation for the Career Achievement Award of the
European Family Therapy Association, presented at the “Creating Futures” 6th
European Family Therapy Association Conference, Berlin, Germany.
Wilkinson, S. (2012, June). ABC⫹D vs. DMM. Paper presented at the
Centennial Nordic Congress of Psychiatry, Tromsø, Norway.
Received October 20, 2012
Revision received January 22, 2013
Accepted February 5, 2013 䡲
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
338 LANDA AND DUSCHINSKY
