Attachment and Intervention
Attachment and Intervention
Attachment and Intervention
Everyone knows that prevention and early intervention in cases of child maltreatment can
reduce parents’ and children’s suffering greatly and save money as well. But how should
we intervene - and with whom? In this brief article, I consider these two issues from my 30
years’ experience working in prevention and amelioration of mental retardation, abuse and
neglect, and mental illness.
Attachment theory is a theory about protection from threat. Attachment behavior is infants’
contribution to enabling caregivers to protect and comfort them. Although mothers’
sensitive responsiveness to infant signals is crucial to infants’ safety and security, mothers
don’t “naturally” know what babies want. Moreover, they have many other demands on
their attention and activity. Consequently it is up to babies to signal their needs. Patterns of
attachment are infants’ strategies for shaping mothers’ behavior. When an adaptive strategy
is used, mothers should become more competent and infants more safe and comfortable. In
some cases, however, the threat is such that the infants’ strategy can only accomplish part
of this. In cases of risk, strategies can create discrepancies between appearance and reality
in which infants (or children) appear more or less at risk than they actually are. These
strategies are associated with risk for physical abuse and neglect, sexual abuse, and
psychological distress (Crittenden, 1999).
Attachment behavior (e.g., crying, looking, reaching, clinging, calling) attracts the mother’s
attention. Once she arrives, she picks up the baby and begins trying out remedies for the
problem. She expects her baby to stop crying when she provides the needed solution.
But it’s not that simple. The way that mothers do this has substantial impact on how their
babies will learn to use their minds. Some mothers think their babies must learn to wait,
must learn that they are not the only important thing in life. True enough! We all agree, but
when? When should the baby learn that? When can babies learn that? Not in the first
months of life. Other mothers think that babies can’t wait at all. Others vacillate between
these - depending upon how the mother feels or who is guiding her at the moment (Is her
husband there? Her mother?)
Cognition and affect. Research has shown that for newborns to learn contingencies,
the contingencies must be nearly perfect (Gergely, 2001). Baby cries, mother comes. Each
time, every time- without intervening events. The more predictably responsive the mother is
early on, the more quickly and firmly the infant learns the relation between his crying and
mother’s coming. Predictable interpersonal contingencies are one way that infants - that all
people - feel connected. Babies whose mothers come quickly and predictably learn sooner
what to expect of themselves and their mothers than babies whose mothers are less
predictable.
But what should the mother do? Surely we can’t expect that, without any prior experience
with this baby, she will know already what he or she needs. Mothers everywhere do the
same thing: they pick the baby up and put him against their chest, on the left where their
heartbeat can be felt, and murmur soothing sounds while stoking and rocking the baby
rhythmically. This is the sensitive part of Mary Ainsworth’s notion of “sensitive
responsiveness” (Ainsworth, 1979); in Dan Stern’s terms, it is “attunement” (1985).
Mothers bring their own rhythms into accord with their babies’ arousal and then help to
modulate the baby’s arousal from aroused crying to calmly alert. When that occurs with
regularity, babies feel themselves in synchrony with another human. Shared states of
arousal are the second way that babies - and all humans - know that they are together with
another human.
Type B: Balanced and secure. By three months of age, babies need variability
(Gergely, 2001). Now they can wait. Call out, say you’re coming and the baby will wait -
in eager anticipation of a certain event occurring at an uncertain moment. The intense
negative arousal of crying alone (a bad feeling) is transformed into the excited expectation
(a good feeling) of soon being together. Cognitively and affectively, such favored babies
make meaning out of life and relationships and, in the process, they learn what causes what,
how to regulate their own feeling states, and how to communicate with other humans. Such
a baby is on the way to being interpersonally secure and intra-psychically balanced with
regard to affect and cognition.
Type A: A preference for cognition. Some mothers are highly predictable, but
unattuned to their infants. In some cases, the mother rarely comes or comes after such a
delay that they baby has already given up hope. When she arrives, the infant is unable to
find the contingency. Unsoothed crying escalates quickly until the baby is extremely
distressed. Often mothers of these babies think, “He’s got to learn to control himself!” He
does, but not at this age. Indeed, he can’t at this age. Or maybe their attention is
preoccupied with other things and they just don’t register the baby’s need at all. Either way,
the baby is left to cry until he exhausts himself and falls asleep.
What has this infant learned? That there is predictably no response to his actions.
That when he feels distressed, things get worse and worse and there is no relief except
sleep.
Other mothers respond promptly and consistently, but angrily. They speak sharply,
touching and picking the baby up abruptly or harshly. The baby feels worse, predictably
worse. But if he cries more intensively- as he will - his mother gets more upset. Their
negative feelings will escalate in synchrony.
A third group of mothers come promptly and predictably, but instead of soothing
their babies, they smile and laugh - as if to deny the baby’s feelings. The babies cry more
and the mothers smile harder, with sharp teeth displayed in a face filled with fear.
All these babies learn about contingencies: they cause nothing, they irritate other
people, they cause others to laugh when they feel bad. The babies learn about affect as well.
They learn that displaying aroused negative affect leads to more intense negative affect. By
about three months of age, brain maturation enables them to inhibit behavior - and these
babies inhibit the expression of negative affect. Life gets better; they are less stressed and
so are their caregivers.
Type C: A preference for affect. A third group of mothers responds to their babies’
crying, but unpredictably. Sometimes they respond promptly, sometimes after a delay,
sometimes even before the baby has really signaled! Sometimes they respond with
comforting attunement, but often not. These babies are on a schedule of unpredictable,
intermittent positive reinforcement of negative affect. Such a schedule maintains displays of
negative affect for long periods of time and at high intensity in spite of positive
reinforcement of incompatible behavior, punishment, or attempts to extinguish the behavior.
These babies don’t know how to predictably affect the contingencies on their mothers’
behavior and they feel intensely badly about it. Cognition fails them and affect overwhelms
them. Their increasing arousal distresses their mothers until parent and infant are joined in
their inability to regulate their feelings. They neither communicate reciprocally, nor inhibit
negative affect. Instead, they are filled with rapidly escalating mixed negative feelings of
anger, fear, and desire for comfort.
Child abuse and neglect. Type B babies are generally safe and protected. So are
most Type A and Type C babies. Nevertheless, in extreme cases, Type A babies are
harmed. Mothers whose own concerns overwhelm them may not perceive their babies’
signals. Such mothers protect themselves and neglect basic needs of their infants. Other
mothers are over-vigilant and over-demanding, expecting their babies to respond like older
children; they punish their children’s demands severely, abusively. Mothers who fear
distress and need everything happy, respond incongruently - and thus psychologically
maltreat their infants. In infancy, there is little the children can do to protect themselves
except inhibit the negative affect that leads nowhere and exhausts them. Put another way,
Type A babies organize around predictable contingencies and inhibit displays of negative
affect that elicit undesirable outcomes from their mothers. Cognitive representations
organize - or dispose - their behavior.
Mothers of Type C babies are middling in sensitive responsiveness, falling between the
mothers of Types B and A infants. They are both too sensitive (alerting when there is no
signal) and too insensitive (failing to alert when there is a signal) and also too responsive
(over-reacting) and too unresponsive (giving little response). Their babies become highly
aroused and feel bad; often this is expressed somatically as problems with eating, sleeping,
and attending. Although they spend too little time in the comfort of interpersonal
engagement (and sleep) and too much time in distressed arousal, they are not usually
maltreated. They learn to act on the basis dispositional representations (DRs) of how they
feel.
Type B: We can work it out. Once children can walk, the comforting relationships
of infancy become hierarchical relationships in which parents use authority to restrict
children and teach them self-protective behavior. This produces conflict between children’s
desires and their parents’ protection. Type B toddlers try to negotiate these differences, but
with their limited access to language, they are very dependent upon parents’ predicting and
preventing struggles. Having fewer rules makes their protective function clearer to children;
knowing their function makes children more willing to cooperate. In addition, parents who
are predictably firm in enforcing the rules have toddlers who accept the rules. If, in
addition, the parent prevents problems (by removing forbidden objects or distracting the
child’s attention to safe activities), the child is not overwhelmed by having to remember too
many rules and not frustrated by always being reprimanded. Many parents, however, can’t
manage this and their children shift from Type B in infancy to a more compliant (Type A)
or persuasive (Type C) strategy in toddlerhood.
Type C: It’s about me! Some toddlers learn to manipulate their parents’ feelings by
turning protective rules into personal battles. It’s about me! These toddlers exaggerate their
displays of feelings. Angry omnipotence is alternated with disarming displays of tender
vulnerability. The displays shape and mold their parents’ feelings. As a result, parents are
both coerced to do their toddler’s bidding and, mindful of the importance of protecting their
children, anxious to regain authority. To the extent that the parent forgets the protective
function of the rule and focuses on enforcing authority, they enter the toddler’s dispute on
the toddler’s terms. It’s about me! No, it is about me! The struggle begins and, once begun,
few parents know how to resolve it. Their toddlers’ intense displays of affect leave them
anxiously aroused and with few ways to regulate either their own feelings or those of their
toddlers. In moments of intense arousal, toddlers will sometimes be hurt by parents’ over-
zealous punishment. Rates of physical punishment spike abruptly in the middle of the
second year of life - as do rates of injury from punishment (cf. Crittenden, 2004).
Other parents feel as anxiously aroused as their children. This leaves their children feeling
unsafe. When parents become distressed, children fear lack of competent protection. They
agitate to elicit it, becoming perilously needy. Both groups of toddlers are now more at risk
than in infancy for maltreatment, in the forms of sudden and unpredictable attacks and
negligent failure to enforce safety procedures. A different group of children is at risk for
abuse. Their parents, however, being coerced into being more responsive, appear more
normal than in infancy.
Toddlers whose parents are harshly punitive learn to do exactly as their parents
desire, even before it is requested; their compulsive compliance protects from the parents’
anger. Toddlers whose parents used incongruent positive affect learn to do the same; all
appears happy while, in fact, there is no affective synchrony. For these toddlers, there is no
chance of coercing the parent because, from the parent’s perspective, it’s not about the
child.
Compulsive children refine the Type A strategy of infancy into a tool for eliciting
attentive care from their parents, who now appear less depressed, angry, or insensitively
incongruent than when their toddlers were infants. With compulsive strategies, Type A
toddlers become less at risk for maltreatment than in infancy and more at risk of certain
kinds of psychological distress at later ages.
Type B: When language communicates. Some children are given words that
accurately describe their feelings - even when these are negative feelings that express their
frustration with their parents. Similarly, they are helped to tell the simple episodes of their
daily life - even when these are unpleasant and built around uncertainty. Open and
elaborated verbal communication is typical of Type B children. Parents of such children are
comfortable with mixed feelings and complex causation and are satisfied with a less than
perfect reality.
Type C: When words don’t work. Type C children, on the other hand, diverge in
two directions: constant chatter that keeps nothing discrete or clear and silence that hides
what isn’t understood. In both cases, however, language fails to communicate with clarity.
The chatter functions to keep parents focused on the child while failing to clarify exactly
why the child needs this attention or how events are causally connected. Silence marks the
place where neither the child’s nor the parents’ perspective can be tolerated by the other.
Parents of silent children often have fearful secrets, either in their own endangered past or
in their marriage, from which they wish to protect their child. Unfortunately, instead of
protecting children, all too often they only confuse the child about why things happen as
they do. Ironically, too many and too few words have similar effects: they exacerbate
negative feelings and obscure the causal relations between parent and child. In an effort to
ensure that they will be protected, some Type C children abandon language as a strategic
tool and engage in provocative and risk-taking behavior
The school years: Why did I do that?
Up to about six years of age, children are refining their understanding of the effects of (1)
their behavior on others and (2) their feelings on their behavior. This occurs in infancy in
implicit, non-verbal ways (i.e., procedural and imaged memory) and in toddlerhood is
transformed into explicit, verbal information (i.e., semantic memory and connotative
language). Later, in the preschool years, the experiences that form the basis for these
understandings are encapsulated in episodes, together with language that conveys the affect
associated with the experience (i.e., episodic memory). That is, by age six, children have
many ways of knowing, each of which is a dispositional representation (DR) that can
influence their behavior.
To understand the relation between attachment and maltreatment in the school years, one
must focus on how children explain their own behavior. This is an integrative process that
requires children to examine their own motivations, i.e., their DRs. When all the DRs
suggest the same action, there is nothing to examine. The crucial occasions are those in
which the various DRs motivate incompatible responses. Which type of DR does a child
rely on most often when what he usually does, feels like doing, should do, and recalls
doing are in conflict? When children do what they should do, in spite of not feeling like
doing so, no one questions them - and the discrepancy among DRs is likely to go unnoticed
by all. But when children do what adults think they should not do, they are asked, “Why did
you do what you did when you knew you weren’t supposed to?!! Of course the answer is,
“Because I felt like it and thought you wouldn’t find out.” But many parents will punish a
child who answers honestly like that. So children learn to deceive both others and, more
importantly, themselves about their reasons for behaving as they did. They learn to spout
parent-pleasing platitudes that, in fact, had nothing to do with their behavior.
The irony, in both cases, is that the parents themselves rarely understand why they are
doing what they do, especially when, as in cases of maltreatment, it has become clear that
they should not have done it.
The point is quite simple: Types A and C are psychological opposites that might require
opposite interventions. Giving the same intervention to a mixed group might be helpful to
those using one strategy and harmful to those using its opposite. For example, prescriptive
or information-based approaches might be counter-indicated for Type A parents as might
contingency-based behavioral techniques. On the other hand, Type C parents might
experience greater negative arousal when imagery, somatic enactments, or episodic recall
were emphasized - which, of course, might be very beneficial techniques to use with Type
A parents.
Starting with the most competent parents, needing the least intervention, parent education
in group settings is an appropriate preventive intervention when parents can use and
integrate all sources of information, but lack specific information about young children.
Given the small, single-generation families that exist today, parent education is relevant to
the needs of many first-time mothers. Both the content and the group context can enable
mothers to enlarge their repertoire of possible responses to the babies while helping at-
home mothers to feel less isolated.
Parents who are capable of integrating information, but who are stumped regarding some
particular problem, may benefit from short-term counseling around that problem. In this
case, information may be offered, but more importantly the counselor helps the parents to
reconsider the problem from new perspectives until a new way forward is discovered. For
counseling to be effective, however, parents must have access to both cognitive and
affective information, be able to communicate effectively in words, and be skilled and
comfortable with critical, integrative reasoning processes.
When parents are relatively verbal, but not skilled with integrative processes, infant
intervention may be appropriate. When this is done without the infant being physically
present, but with videotaped interactions of infant and parent, the parent can learn to (1)
observe the baby accurately, (2) explore their own feelings while watching themselves with
their babies, and (3) reflect on what they see and feel. Having other mothers present and
engaged in the same process can give each mother more practice, including less emotionally
arousing practice than with their own baby, as well as enlarging their repertoire of things to
do (through observational learning). Having the babies are present will reduce the reflective
opportunity for the mothers whereas, if interactions are not videotaped, mothers skewed
recall may distort the reflective process.
When parents function primarily on the basis of implicit information and especially if they
themselves have been exposed to danger, either when they were young or currently in their
adult relationships, adult psychotherapy (individual, marital, or family) for the parent might
be needed. The focus of such psychotherapy should be identifying the distortions in
meaning attribution, bringing all forms of transformation to awareness (making them verbal
and conscious), and learning the process of integration of information. Once that is
managed (a long-term process in cases of severe distortions), the other forms of
intervention (listed above) can be used productively.
Clearly far more work is needed to understand how best to apply the plethora of available
treatment approaches to parents and children. At a minimum, however, we should select
interventions that (1) address parents’ needs and skills and (2) carry little risk of increasing
or creating problems. That is, treatment should be sensitively responsive to the unique
characteristics of each parent and should be implemented in ways that reduce the possibility
of aggravating the situation.
In addition, we should not fool ourselves into thinking that early intervention can inoculate
families against future problems. Instead, we should promote services that (1) increase
parents’ awareness of how they generate information and select behavior and (2) foster
reflective, integrative processing. If that is accomplished early on, simple periodic screening
can identify any on-going need for anticipatory guidance, parent education tied to older
children’s needs, or counseling around specific problems.
The hypothesis offered here is that if the parents’ (a) use of biased and limited sources of
information and (b) failure to engage in reflective, integration is not addressed early on, the
family can be expected to respond to each new developmental challenge in skewed ways
that risk escalation of problems. In this case, the family and the treatment services are likely
to remain in frequent contact around solving ever-changing crises. We can do much better
than this! The Dynamic-Maturational Model of attachment (Crittenden, 1995) is an attempt
to integrate information about human adaptation across the life-span and from numerous
theoretical perspectives to meet the needs to troubled children and their parents.
Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of
Attachment: A psychological study of the strange situation. Hillsdale, NJ: Erlbaum
Associates.
Bowlby, J. (1980). Attachment and loss. Vol. III: Loss. New York: Basic Books.
Crittenden, P. M. (in press). Why do inadequate parents do what they do? In Mayseless, O.
(Ed.) Parenting representations: Theory, research, and clinical implications. Cambridge:
Cambridge U. Press.
Crittenden, P.M. (1992). Quality of attachment in the preschool years. Development and
Psychopathology, 4, 209-241.
Damasio, A. R. (1994). Descartes’ error: emotion, reason, and the human brain. New
York: Avon Books.
Fonagy, P., & Target, M. (1997). Attachment and reflective function: Their role in self-
organization. Development and Psychopathology, 9, 679-700.
Gergely, G. (2001). The obscure object of desire: ‘Nearly, but clearly not like me’:
Contingency preference in normal children versus children with autism. Bulletin of the
Menninger Clinic, 65, 411-426.
Schacter, D. L., & Tulving, E. (1994). What are the memory systems of 1994? In D. L.
Schacter & E. Tulving (Eds.), Memory systems 1994 (pp. 1-38). Cambridge, MA:
Bradford.
Stern, D. (1985). The interpersonal world of the infant. New York: Basic Books.
Vygotsky, L.S. (1987). The collected works of L.S. Vygotskii. R.W. Rieber & A.S.
Carlton (Eds.), translated by N. Minick. New York: Plenum Press.