A Great Attachment Debate?



It seems as though the nature versus nurture argument will go on forever, even though we now know a great deal about how the two of them interact in order to affect human behavior.

In the March/April issue of the Psychotherapy Networker, the cover story is titled "The Great Attachment Debate: How important is early experience?  The "debate" is over the issue of whether or not the quality of the relationship between babies and toddlers and their primary attachment figure has a profound effect on  mental health and relationships when the child grows up.  In particular, "attachment" refers to how secure the child feels and behaves with its primary caretaker.

The  two sides of the "debate" in the issue are represented by Jerome Kagan, Ph.D. on the one hand, and the tandem of Alan Sroufe, Ph.D. and Daniel Siegel, M.D. on the other. Kagan researches the effects of inborn temperament, personality, and neurobiology.  Strouffe is a developmental child psychologist.  Siegel is a UCLA psychiatrist who wrote a highly influential book called The Developing Mind.



Jerome Kagan




Alan Sroufe



Daniel Siegel

Jerome Kagan thinks that the "pro" attachment side downplays the importance of both cultural influences and inborn, genetically determined temperament in creating an adult's personality and vulnerability to psychiatric disorders.  "Temperament refers to an inborn predisposition to experience certain feelings and display particular behavior during the early years," he explains. 

The intial work on innate differences in infants in qualities such as activity levels and reactivity was done by child psychiatrists Stella Chess and Alexander Thomas.  The temperament issue represents the "nature" side of the nature-nurture debate.

The "nurture" side of this debate centers around the thesis that the emotional quality of our earliest attachments is a far most important influence on human development than inborn temperament.  Attachment theorists pay particular attention to something they call attunement, which they believe is more important in creating the quality of the infant's attachment than, say, the mother's general traits such as maternal warmth.

Sroufe and Siegal explain, "Attunement, or sensitivity, requires that the caregiver perceive, make sense of, and respond in a timely and effective manner to the actual moment-to-moment signals sent by the child."  The parent has to figure out, for example, how much emotional stimulation a baby needs at any given time.  Too much or too little can disturb the baby, and the baby's need is not a constant but varies widely over even brief periods of time.

I think this whole debate is somewhat silly and depends for its existence on an assumption about attachment that really does not make a lot of sense to me, as I will describe shortly. 

So what do I think is more important in human development, inborn temperament or attachment relationships?  Well first of all, both of these variables always contribute to development.  Second, inborn temperament itself affects and alters attachment patterns.   For example, a colicky infant with an insecure  and anxious mother is a bad combination, while the same mother with a quiet child may do a lot better parenting job in regards to attunement.

The answer to the question as to whether temperament or attachment patterns has the greater effect on ultimate development is, as with almost any question in psychology, it depends.  If the family is accepting and validates the innate predispositions of the child, you get one result.  If they invalidate and denigrate them, you get an entire different result.  The way the child acts also can elicit invalidating reactions from peers and teachers, leading to a sort of self-fulfilling prophecy, as Kagan points out.

Neither side talks about the importance of the choices a person makes, the reasoning they use, or their problem solving strategies in the determination of how a person ultimately acts.

Furthermore, the attunement of the parent to the baby, and what behaviors the parents validate or invalidate, can be quite different at different times. Plus,  there are literally hundreds of other influences on the child which also vary in time. 

With all these factors at work, there is also an effect from what scientists refer to as chaos, in which small changes in initial conditions can lead to big differences in complex phenomena like human behavior later on. This is known as the butterfly effect: the presence or absence of a butterfly flapping its wings could lead to creation or absence of a hurricane. 

It is interesting that the debaters do seem to agree on some points.  Both agree that serious neglect or abuse of infants during their first year or two of life can harm the child's future psychological development. 

I actually disagree somewhat and think that infants are more resilient than they seem to think.  For example, say the mother had an untreated post-partum depression during a baby's early life, but then got treated and became far more attuned to the child.  Chances are, any ill effects of the child's experience during the first two years of life would then be reversed.

 
In this vein, both sides also agree that human psychology can change depending on later experience.  If it could not, then they would both have to think that psychotherapy would be a complete waste of time.  They agree that neither biology nor parenting experience is destiny.  I should certainly hope so.  If we could not adapt to changing environments, our species would have died out eons ago.

Kagan also says another important thing which sometimes gets lost in nature versus nurture debates. Genetic influences on behavior do not determine later personality variables so much as limit them.  This is easier to see with physical traits.  No matter how much Danny DeVito might have trained as a young man, he would never have been able to swim as fast as Michael Phelps did.  Wrong constitution!  This does not mean, however, that training would not have improved DeVito's lap times.
 
A hidden assumption in the whole debate that drives me bonkers is that the most salient patterns in the primary relationship between children and their parents somehow no longer influence a child past the age of two.  Or if you are a psychoanalyst, past the age of five.  Attunement, invalidation, the interactions between the temperaments of parents and children - in short, all of the most salient aspects of their relationship - often continue on and on in slightly altered forms, almost until somebody dies. 
 
If you study the "persistence" of temperament and the security of attachment from childhood to adulthood, and draw conclusions based on just what happened in the first two years of life, you ignore this fact.  If a two year old has an insecure attachment with an unattuned mother, but is taken away from that mother and raised by someone else, you would get a very different view of this "persistence."  Similarly, if you limit the effects of attachment to only early attachment, you are ignoring the effects of extremely important family dynamics all through childhood and early adulthood. 
 
Both sides in the "debate" define attachment the same way.  Since controlling for all the variables is impossible, findings in the research literature will often conflict with each other.  Depending on how studies were structured, it is easy to find studies that over-estimate the importance of attachment in the first two years, and others that under-estimate it.   If on the other hand you assume that the early influences might continue on in time, and determine what they actually were, the "debate" all but disappears.

Childhood Sexual Abuse Taken out of Context

Albert Einstein was said to have had a sign over his office at Princeton that said, “Everything that counts cannot be counted and everything that can be counted does not count.” When it comes to studying the long-term psychiatric effects of child sexual abuse (CSA), it is wise to keep this in mind.

CSA, particularly within an individual’s family of origin, has been implicated as a risk factor for a wide variety of adult psychiatric diagnoses and behavioral problems, including borderline personality disorder (BPD), dissociative identity disorder, depressive disorders, anxiety disorders, alcoholism, eating disorders, somatization disorder, sexual dysfunction, and suicide attempts.

The nature and effects of interpersonal relationship patterns, such as those that transpire in the families of CSA victims, are so complex and unique that it is almost impossible to quantify them for an “empirical” study or “proof” of the causation of a specific psychiatric problem.

Biological and genetic factors almost certainly play a significant factor in predisposing a victim to one or another psychiatric problem. However, that is not the complete answer to two questions: why do certain individuals develop one disorder while seemingly similar individuals develop a different one, and why do some individuals who appear to have suffered severe abuse develop no psychiatric problems at all, while others who seem to have had relatively minor abuse develop several disorders?

Studies that examine psychological and social variables in CSA tend to focus on factors such as who the perpetrator was, what type of abuse was suffered (penetration vs. fondling, for example), the severity and frequency of the abuse, and whether the social welfare or criminal justice system became involved. Rarely, the response of non-abusive relatives to CSA victims, usually the mother, is examined. Most results of such studies have been disappointing regarding finding links between specific psychiatric disorders and these variables.

Often the backgrounds of CSA victims are also characterized by several other types of adverse childhood experiences or generally chaotic family relationships. Not only is that fact often ignored by those who study CSA, but these investigators usually ignore the entire environmental context in which CSA takes place.

Clearly, most of the victim’s interactions with perpetrators and bystanders alike occur at times when abuse is not occurring, and these other parts of such relationships may also have profound effects on the victim’s later relationships and self image. Again, due to their staggering complexity and intermittent nature, they are difficult to study using statistical techniques.

Contextual factors include the entire history of the relationship between the victim and the perpetrator: what is said during, before, and after the abuse; what the relationship between victim and perpetrators is like when the abuse is not taking place; what other people in the family are doing at the time of the abuse and at other times; how each family member relates to the victim; who if anybody knows what is going on and whether or not they intervene; and a whole host of other characteristics of the interpersonal environment of the victim.

The context of the abuse is made even more complex due to the effect of chaos, the theory of which predicts that even small differences in environmental conditions can have large future effects. The so called butterfly effect would occur when small differences in the interpersonal environment of a child lead to a cascade of events which will be somewhat unique to each individual and greatly impact the exact nature of any psychiatric responses to the CSA.

Even during abuse, a victim’s interactions with a perpetrator is not limited to the sex act alone. Words may be spoken; other activities may occur right before, right after, and even simultaneously. For example, personality theorist Lorna Smith Benjamin discussed how the nature of verbal interactions with the perpetrator at the time of the abuse may influence the production of later BPD symptoms. The tendency of some BPD patients to have unstable and intense interpersonal relationships characterized by alternating extremes of idealization and devaluation, for example, might result from a case in which a father “directly instructed her [the incest victim] in how to shift from idealization to devaluation.

“Early in a night visit, the father may say, ‘You are the light of my life; I live for these times together…’ Then, after the incestual attack, he might say, “It’s your fault. You bitch. You whore. You’re filthy. Go take a shower’” (Interpersonal Diagnosis and Treatment of Personality Disorders, 1st edition, p.119).

In helping to create another BPD symptom, at times (other than those times in which CSA takes place), the BPD patient’s autonomy may be attacked by either or both parents, with an accusation of disloyalty if the victim tries to assert her or his independence. This may lead to the type of self-sabotage often seen in BPD patients.

Other and different types of dysfunctional interactions may lead to the development of still other BPD symptoms, leading not only to the unique clinical picture of each individual CSA victim, but to the different combinations of DSM criteria seen in different patients with BPD.

In getting a full description of patients of the entire family context in which CSA takes place, and also tracing back the family context in which the parents grew up (genograms), it becomes possible to make an educated guess about exactly why a given patient developed particular coping skills. The particulars will be different with every patient and every family, but certain themes do come up again and again in patients who develop certain disorders like BPD.