Kenneth Kaye's Publications           go to index


The Genesis of Mother-Infant Interaction: How Parents Create Persons

Introductory chapter to Drug Use in Pregnancy: Mother and Child, edited by Ira J. Chasnoff, MD. MTP Press, UK, 1986, pp. 1-6.

In the past 10-15 years, drug use in our society has increased among women of childbearing age. Concurrently, our understanding of the dangers posed by any extreme disruption of fetal life, childbirth, or postnatal care has deepened by an order of magnitude. We have a more comprehensive view of the effects babies and mothers and fathers normally have upon one another. Therefore the task of pediatricians, nurses, and psychologists concerned with helping addicted neonatal families is more complex than it was thought to be 15 years ago. The intricacies of the social interactions through which the human mind is acquired, the turn-taking structure which characterizes these interactions from the beginning, the fact that all human parents constantly provide a framework of experiences for their infants' encounters with the physical as well as the social world, and the sensitive use that a normal mother or father makes of those innate properties of infant behavior to gradually construct meaningful communicative exchanges long before the baby even understands what meaning is: all these aspects of human development are relevant to concerns regarding the effects of maternal substance abuse on the developing fetus and child.




Normal human development begins within the context of a system of parental and family interaction. What we learned from observations of parent-infant play is that parents instinctively do many things that help an infant emit behaviors that are actually beyond the capability of the infant alone. Infant development continues to unfold and progress within a context in which the infant is an apprentice to the parents. In this way a parent is like a master shoemaker who begins his teaching by setting up small tasks which are slightly beyond his apprentice's competence. The shoemaker closely supervises the apprentice's work and intervenes when necessary, or gives additional details when the apprentice appears ready. Often this help is offered in anticipation of problems in order to assure positive experiences for the apprentice, thereby reinforcing the apprentice's confidence to expand his own competence further. It is within this apprenticeship setting that parents and other caretakers provide for infants. This structuring framework is species-specific and instinctive for humans and, therefore, universally present in human parenting. Such parenting behaviors constitute the environment of every human infant, whether extremely advantaged or disadvantaged. However, the behaviors may be presented with greater or less sensitivity to the child's needs at the moment, with greater or less consistency, and with accurate or inaccurate readings of a baby's intentions and skill. Thus, these behavioral processes are susceptible to impairment.

The most common setting in which mother-infant interaction has been studied has been during face-to-face play. Results of these studies have identified predictable patterns of interaction within normal mother-infant dyads. At 2 months of age it is the mother who initiates greetings when eye contact is made. As the infant matures, such greetings begin to be initiated by the baby. My students and I call this communicative setting the ‘dialogue frame,' a focus of mother and infant on one another with readiness to communicate. The dialogue frame begins with a mother talking to an infant in utero and continues with turn-taking during the pseudo-dialogues in face-to-face play. This dialogue frame will turn into real dialogue as soon as the infant is old enough to talk, but in the meantime it evolves into other kinds of frames which, we believe, are crucial to the infant's cognitive development.

Another common pattern of interactive behavior in normal mother-infant dyads is the ‘instrumental frame.' An illustration of this pattern would be the case of an infant placing shapes into matching holes in a sphere. In early infancy a baby is only successful at such a task if his mother rotates the sphere to present the correctly shaped hole in front of him. This ‘instrumental frame' is also demonstrated in the behavior of a mother who steadies a tower f blocks each time her infant places a new one on the top. Caretakers provide this type of active guidance of an infant's behavior without any awareness that they are doing so. This guidance or ‘instrumental frame' makes it possible for infants to be more effective at performing tasks than they would be by themselves.

Even in the course of normal mother-infant interaction, difficulties may arise that interfere with the smooth ongoing dialogue. A normal infant is not always ready to respond to his mother's invitation to interact each time she gazes at the infant's face. Understandably, this rejection of maternal gaze by the infant can be upsetting to some mothers, especially to those who are already anxious about mothering, or in personal distress. By and large, however, mothers tend to accept the cycles of attention from their infants as normal variation and fit them into their maternal dialogues as opportunity permits. They continue using the dialogue frame and the instrumental frame until the infant learns to become genuine participants in the interaction. For example, a mother with a 2 month old infant may pretend that they are holding hands and talking to each other. Actually, only one of them is holding the other's hand, and microanalysis of such sequences at that age show that babies don't really initiate greetings; mothers structure their own greetings in such a way that there is a high likelihood of eliciting a response. After 3 months of this maternal guidance, when this same infant has reached 5 months of age, the infant is initiating vocal greetings and smiles as often as the mother.




Our theory that early human development proceeds as an apprenticeship is a way of integrating findings from studies of mother-infant interaction. The apprenticeship progresses through a series of stages. The first stage during the first 2 months of life is a period of ‘shared rhythms and regulations' in which the intrinsic self-regulatory processes of the neonate are marshalled by extrinsic functions residing in the adult of the species. The second stage, from 2 to 8 months, is a period od ‘shared intentions,' when adults infer what the baby is trying to accomplish and enable him to begin to fulfill intentions. He is thus able to practice the sensory-motor schemas which are constituent modules of more and more intelligent acts, as Piaget (1952) and Bruner (1972) have shown.

The period of ‘shared memory' begins at about 8 months, when infants and parents begin playing stereotypic games remembered from one day to the next by both partners, yet still depending upon the parent's greater responsibility for simplifying, structuring, and suggesting. Finally, in the second year comes the period of ‘shared language,' by which time the baby is not only comprehending and producing gestures within dyadic relationships but is a self-conscious member of the family system.




The quality of parenting shown by chemically normal human mothers varies widely, with comfortable, confident mothers at one end and anxious, perhaps depressed or neurotic mothers at the other end of a normal spectrum. All along that spectrum we see plenty of variability in the effectiveness of their interactive behaviors and in the smoothness with which they orchestrate the synchrony of turn-taking in their relationship. Therefore any investigation of deviant mother-infant interaction must take into consideration its complexities. It would be convenient if one could simply predict that any baby whose physical system has been affected by his mother's chemical dependence will fail to elicit appropriate reactions from the parents or caretakers, or fail to respond to their stimulation. However, such failings may be only one aspect of many interwoven events that cumulatively create interactive deficiencies.

In other words, it would be a vast oversimplification to conceptualize the problem of substance abuse in terms of effects of drugs upon the developing fetus and child, and consider nothing more than that. Such oversimplification fits a medical or linear cause/effect model. Admittedly, such a model provides a rationale for preventive medicine: keep pregnant women off drugs. It also provides a basis for intervention, but only for the baby. It suggests that our response should be to find the best possible medical treatment for the newborn, as well as psychological interventions to counteract any deficits in the development f the child's intellectual capacities or personality.

A more realistic, if complicated view is that whatever effects drug abuse has upon the fetus, its effects are likely to play out upon the developing child through a multi-layered interaction. The factors in that interaction include at least the following: the infant's neurological system, the parents' sensitivities, the infant's ability to make sense of the cues his parents offer him, and the parents' expectations. All four of those factors must be kept in mind when evaluating chemically dependent newborns.

In regard to the first factor, neurological status, the passively addicted newborn may not present t the mother as alertly or as engagingly as a normal infant. One of the frequent observations by researchers with ‘high-risk infants' is that some of them are not fully ‘there' for their parents to respond to. This problem has been at the center of Brazelton's (1976) work at Harvard for nearly 20 years: evaluating newborns so as to be ale to predict their developmental course over the early weeks of life and thus be able to counsel parents regarding the strengths and weaknesses of their infant.

The second aspect of ‘high-risk infants' is that the baby, and often the mother, requires extra hospital care. This interferes with the normal process of acquaintance for the two members of the dyad by keeping them apart and by labeling the baby as ‘abnormal.' The neonatal and/or maternal care distances an already at-risk mother from her infant.

The remaining factors, the baby's intelligence and the parents' expectations, add a further level of complexity to the notion of risk. The newborn's developmental tasks can be conceptualized in terms of entering the family system as a deficient though stimulating organism, being co-opted by parental frames and only slowly becoming a full member of the system, a person with a mind and a sense of self (Kaye, 1982). This whole process of assimilation into the family system for the purpose of development has been provided for by human evolution, both in the form of neurologic mechanisms present in the neonatal organism, which stimulate adults to act in the way the organism needs them to act, and also in the form of instincts in parents to provide the kinds of frames babies need in order to become parents.

Although parental behaviors are somewhat guided by instincts, as they are in the rest of the animal kingdom, human parental behaviors are often strongly influenced by socially acquired characteristics such as maternal expectations for the baby. It may be these expectations on the part of drug-abusing mothers that guide many of their early interactive behaviors. Therefore, a baby born to a chemically dependent mother might be at risk for starting the interactive process with the handicap of negative maternal expectations, even if the baby has no physiological or neurological effects of in utero drug exposure. Mothers who abuse drugs during pregnancy are often told that their baby will be at risk; or they assume that such a risk exists because they have seen warnings about using drugs during pregnancy. Such an expectation could place the mother-infant interaction at risk of dysfunction even in the absence of any physical factors.

In reality, of course, drug-using mothers not only have those negative expectations, but their infants often confirm their fears, being sleepy, jittery, withdrawing and irritable. Repeated instances of under- and over-responsiveness from her infant may prematurely convince even a patient mother that her infant is impaired, retarded or in pain, and that it is her fault. Worse, she may begin to believe that her infant is intentionally rejecting her.

This downward spiral of negative effects can be intensified if the mother herself is deficient in self-esteem. Such a case would be missed in the linear or medical model. Simplistic cause/effect thinking omits the factors leading to or complicating the mother's drug dependence: her addictive personality, poverty, emotional deprivation, ignorance and a whole host of indicators or poor readiness to be a parent. Whichever of those factors operates in a pregnant woman, and in her spouse or nuclear family, will be vital aspects of the social system into which the baby is born, even if that baby is medically protected from the mother's substance abuse; indeed, even if the mother's chemical dependence was already in the past when she conceived.




It appears that the frequenc ad severity of drug abuse in pregnancy continue to to grow. Researchers face a number of critical issues: Which developmental processes in the exposed infant are mosr affected? Are there critical periods for the fetus or the embryo? What are the subtle effects that combine with maternal characteristics to affect such complex processes as mother-infant interaction? The authors of the following chapters ask those questions in hopes of stimulating more productive investigations and eventually more effective treatment plans. Protecting an infant from the effects of illicit substance use by its mother protects it from only one aspect of a multidetermined pathological system. It is that system, not the drug use alone, that any therapeutic endeavor has to address.





Brazelton, T.B. (1976) Neonatal Behavioral Assessment Scale. (Philadelphia: Spastics International Medical Publications)

Bruner, J. (1972) The nature and uses of immaturity. American Psychologist, 27, 688-704.

Kaye, K. (1982) The Mental and Social Life of Babies: How Parents Create Persons. (Chicago: University of Chicago Press)

Piaget, J. (1952) The Origins of Intelligence in Children (New York: International Universities Press)