A LITERATURE REVIEW OF CHILD - PARENT/CAREGIVER ATTACHMENT THEORY AND CROSS-CULTURAL PRACTICES INFLUENCING ATTACHMENT.

By P.N. Reebye M.D. FRCPC, S.E. Ross M.Sc. RDN, K. Jamieson M.A.
With the Assistance of Jason M. Clark B.A. (Hons.)

Supplement

THE HISTORY OF ATTACHMENT THEORY

Attachment has always formed the basis for human relationships across cultures. The concept of attachment has more recently become a focus of scientific study. The historical development of attachment has been outlined well by Robert Karen (1994) in his book Becoming Attached. The following summarizes this work.

Infant Development – Nature or Nurture

Early in this century, there were three streams of thought about infant development in North America. These views encompassed what has been widely known as the "nature-nurture" debate. In the 1940s and 1950s, heredity tended to be considered the primary factor mitigating child development. It was believed that children would grow up to become whatever they were designed to become, regardless of how they were raised. Arnold Gesell (cited in LaBarba, 1981), a Yale psychologist and paediatrician, was one of the pioneers of developmental psychology and an important advocate of this viewpoint.

A second view held that children began life with a ‘blank slate’ (i.e., without hereditary or spiritual predispositions); how they developed was entirely a product of their environment. Supporters of this environmentalist perspective, such as the psychologist John Watson (1928), believed that a mother’s affection was potentially damaging to a child’s character, that children should be treated objectively as if they were young adults, and should not be hugged, kissed or held in a parent’s lap.

A third stream of thought was that of philosophers and psychoanalysts, who recognized the influences of both heredity and environment, but saw ‘relationships’, particularly the infant-caregiver relationship, as critical to healthy emotional development.

John Bowlby

John Bowlby (1958; 1960a; 1960b; 1969/1982; 1973; 1980; 1988), a British psychoanalyst and a research scientist who is considered to be the founder of attachment theory, developed this third approach that infants need and actively seek loving relationships. He held that this first relationship, usually with the mother, determines much of our future well-being. He identified two environmental factors of primary significance – the death or prolonged absence of the mother, and a mother’s emotional attitude toward her child. He is best known for writing a trilogy on attachment in the 70s (1969/1982; 1973; 1980), but he began his work in the late 1930s studying the effects of maternal-deprivation. His aim was to create a science of the early environment and its effects on a developing child.

His approach was noteworthy not only for its complexity and scope but also for the controversy it raised. Refining and validating the theory took many years, and included a broad network of colleagues and researchers.

While a medical student, Bowlby was a volunteer at two British residential schools providing alternative educational approaches for maladjusted children. From this experience, he drew a connection between the disturbed behaviour he was observing and the troubled early histories of the children. He also began to openly question the then typical British style of parenting that was unresponsive to emotional demands and imposed strict discipline, advocating for more tolerance and warmth in parenting.

Bowlby and Klein

As a graduate student just prior to the war, Bowlby was influenced by his supervisor, Melanie Klein, who originated psychoanalytic play therapy. Klein’s work with children had led her to believe that a child’s relationship with his or her mother lived within the child and became a template for future relationships. Her beliefs that infants were capable of forming early relationships, that childhood fantasies (including love-hate conflicts) were powerful, and her focus on loss, mourning and depression, shaped a part of Bowlby’s developing attachment theory.

Bowlby also worked part time at the London Child Guidance Clinic. At the Clinic, staff worked therapeutically with troubled children and their mothers based on their belief that unresolved conflicts from the parents’ own early experiences led them to mistreat their children. This furthered Bowlby’s interest in studying the way parents treated their children and the quality of relationships within a family.

His subsequent study of "44 juvenile thieves" (1944) linked the psychopathic personalities of a sub-group of these juveniles (described as ‘affectionless’ detached children) with a history of early prolonged separation of the children from their mothers, and maltreatment by their caregivers (described as disturbed mothers). The study had a significant impact on child psychiatry, and led to Bowlby being commissioned by the World Health Organization (1951) to report on the psychiatric issues of homeless children – a post war concern being studied by the United Nations. This enabled Bowlby to gather data from social workers, child psychiatrists, paediatricians and researchers in France, Sweden, Switzerland, Holland and the United States who had studied maternal deprivation.

The WHO report, entitled Maternal Care and Mental Health (Bowlby, 1951), highlighted the consistency of clinical data, and findings from international studies that used a wide range of different methods to reach the same conclusions. Children who had been separated from their mothers and placed in institutions failed to thrive, learn, or develop relationships, and did not recover upon later adoption. Foster care showed clear benefits over institutions, but even so, children between the ages of one and a half and two and a half, moved to foster care situations due to the war, were unable to adjust and deteriorated severely. The report called for major changes to public policy (social work, hospital practices, and adoption) and an increase in public awareness of the essential value of maternal care.

Bowlby and Ethologists

A critical problem faced by those working in the area of maternal deprivation was the lack of a clear understanding about why a child’s separation from his or her mother had such devastating effects. Many different rationales were presented (breastfeeding, skin-to-skin physical contact, adequate stimulation, for example) which did not seem to Bowlby to hold true to his clinical observations.

Bowlby was influenced by the work of an ethologist, Konrad Lorenz (1957), on imprinting (the instinct of newly born birds and mammals to develop a strong early bond unrelated to feeding), and the general understanding in this scientific field that all animals demonstrate instinctual species-specific behaviour patterns requiring the right environment for normal expression. He posited that human beings must also have instinctual bonding behaviours, intergenerational cues, and pre-determinations for relationship patterns that could go amiss if not supported by the proper environment. The recognition that mother-child separation could cause such serious developmental problems because they were contrary to an infant or child’s instinctive needs became a key component of Bowlby’s attachment theory.

Bowlby and the Robertsons

After the war and prior to completing the WHO study, Bowlby joined the Tavistock Clinic in London, taking responsibility for the development of child psychiatry services and training programs. By 1948 he had a two-year grant to study the effects of early separation. He hired James Robertson, a social worker, to observe children sent to hospital, observe their reactions as they were being admitted and during their stay, and to follow-up on children aged one to three who had been hospitalized. In short term wards, children were observed to go through two stages of emotional reaction to their separation – protest (crying, screaming, frightened, anxiously searching for their lost mother) and despair (listless, losing interest in their surroundings, not eating, seldom crying, losing hope of seeing their mother again). Upon returning home, children clung to their mother and typical behaviour included bed-wetting, temper tantrums, and aggression. For children on long-stay wards, which in the case of TB could be up to four years, a third stage – detachment – was observed and described. The child starts to interact more, smiles and eats, but has become a very different child who seems more and more indifferent to the coming or going or ‘caring’ of his or her mother. Robertson developed a number of different films to document the disturbing changes in hospitalized children over time.

These observations and identified symptoms were angrily rejected by the majority of medical and hospital staff and administrators - it was another ten to thirty years before British, North American and European hospitals consistently encouraged "rooming-in" by parents during their child’s hospitalization. As change began to occur, the husband and wife Robertson team studied and documented on film the emotional reaction of children to hospitalization with their mother always present, and to separation from their mother but with skilled care from a familiar substitute. The Robertsons concluded that even with the best substitute mothering, separation from mother is a trauma for young children because of the discontinuity of their relationship – however, a substitute mother sustains a child’s experience of sensitive maternal-like care and provides an alternate relationship that reduces a child’s stress.

Bowlby and Ainsworth

In 1950, Mary Ainsworth, a Canadian psychologist, moved to London, England to work with Bowlby. For three and a half years, they examined his study data, including the observational work of Robertson, to try to understand why a baby needs a mother figure to develop normally, and why babies are affected so profoundly by a long separation from their mother even after re-uniting.

Mary Ainsworth completed her Ph.D. at the University of Toronto in the late 1930s where she worked under the tutelage of William Blatz, and conducted research on his security theory. Blatz believed that children derived security from being near their parents, and used their parent as a secure base from which to explore the world. Ainsworth’s background as a personality and developmental psychologist was in psychological assessment, and her interest was in the progressive cognitive and emotional development of children. She was drawn to the observational methods used by Robertson for learning about families before, during, and after separation, and by approaches used in ethology for observing family relationships in other species.

Ainsworth and the Attachment Process

In 1954, when her husband took a job at the University in Uganda, Ainsworth (1963) began a study of twenty-eight unweaned babies from twenty-three families in six local villages. With her interpreter, Ainsworth visited each family in their own home for two hours every two weeks for nine months detailing the way each mother interacted with and responded to her infant. Looking for a natural experiment of mother-infant separation, she had understood from European reports that the Ganda separated infants from their mothers when they were weaned, and grandmothers then took on the responsibility of child rearing. These reports were not confirmed however. She observed that weaning was gradual and few children were separated from their mothers after weaning.

The observational information Ainsworth gathered, however, supported Blatz’s security theory (mother as the secure base) and provided her with data on ‘attachment in progress’. She found sixteen common patterns of ‘secure attachment’ behaviour that she listed in chronological order of appearance and divided into five phases of development. She also noted other (insecure) types of attachment behaviour, and the effects of different methods of infant care and cultural factors. Five of the twenty-eight children did not appear to be attached by the end of the year of study, and the causes appeared to be linked to less responsive, available parenting styles. Seven others, who were attached, could not tolerate any separation from their mother, and seemed very insecure. The amount of care provided by the mother herself was a distinguishing factor between the three styles of attachment; the quality of the interaction seemed important but not easily measurable. Ainsworth’s (1963; 1967) observations also suggested the presence of many care providers was not a factor unless, among them, there was no single person that became the primary attachment figure. At the time, her work was the only ‘external’ empirical support to Bowlby’s theory. Her book, Infancy in Uganda, was belatedly published in 1967.

Ainsworth and the Strange Situation Experiments

In 1961, at Johns Hopkins University in Baltimore, Ainsworth began a longitudinal study of twenty-six families expecting new babies. Her aim was replicate the Uganda study in order to answer questions raised by the earlier research, and clarify whether the attachment patterns seen there were universal, as anticipated by Bowlby’s theory.

In her study, families were visited at home for four hours at eighteen times over the first year. Her findings demonstrated the cross-cultural validity of attachment development (Ainsworth, 1964). Of the sixteen attachment behaviours seen in each study, only two differed and seemed culturally based – when the attachment figure (usually mother) ‘returned’ and was greeted by her young child, Ganda children clapped their hands (American children did not) and American children hugged and kissed (Ganda children did not). These greetings matched the different behaviours of Ganda and American adults.

Ainsworth was unable to make a clear comparison between the Ganda and American babies’ use of their mother as a secure base from which to explore the world. It had been very obvious in Uganda, where children were accustomed to having their mother with them all the time and reacted strongly when she left them. In Baltimore, children were accustomed to seeing their mothers come and go on a regular basis, were less likely to cry when she left (thus displaying less secure base activity) and were less anxious about strangers visiting. In an attempt to find a comparably stressful situation for the American children, Ainsworth drew upon an earlier lab experiment by Jean Arsenian (1943) and upon work by Harry Harlowe on rhesus monkeys (1958, 1966).

Ainsworth’s strange situation laboratory assessment (Ainsworth & Wittig, 1969; Ainsworth et al., 1971) for secure base activity, which used a one-way mirror for observation, three chairs and many toys in the observation room, became one of the most widely used procedures in developmental psychology. It included eight episodes: a brief introduction; the 12 month old baby alone within the playroom with his or her mother; a stranger (female) entering the room; the mother leaving her baby with the stranger present; mother returning and stranger leaving; mother leaving her baby totally alone in the room; the stranger coming back to try to comfort the baby; and finally the return of the mother. Ainsworth had identified in advance the way she expected babies to respond during each episode, and although they reacted in a variety of different ways, their general behaviour was highly consistent with these expected patterns.

Once more, she found that there were three main categories of attachment: secure, ambivalent and avoidant. Because of the variation within the groups, Ainsworth subdivided each group, and ended up with four secure sub-groups, and four insecure sub-groups, which consisted of two groups each for the ambivalent and avoidant categories. From her observations, Ainsworth could determine which babies were more or less securely attached, but the differences between ambivalent and avoidant attachment were not obvious without imposing separation stress. Under these situations, the avoidant children showed characteristics of the short stay hospitalized children in the maternal deprivation studies of Bowlby and Robertson. It was another way to look at separation and re-union, but now it applied to the emotional impact of everyday parenting.

Ainsworth developed four scales to rate a mother’s way of being with her baby. The first, sensitivity, measured how often a mother was sensitive to her baby’s signals. The second, acceptance, measured how much acceptance versus rejection she demonstrated. The third, cooperation, measured whether or not she cooperated with the baby’s desires and patterns or interfered by imposing her own schedule and pace to feeding or play. The fourth, emotional accessibility, measured how available she was and how often she ignored her baby. This permitted a more specific comparison between mothers of securely and insecurely attached children and enabled researchers to recognize and measure the quality of the mother-infant behaviours and relationships.

Mothers of securely attached infants/children were significantly more responsive to their childrens cues, were faster at picking them up when they cried, held them longer (when they wanted to be held) and with more noticeable pleasure. At one year of age, their babies cried less and demanded less physical contact than babies whose mothers had responded less frequently or consistently. Mothers of avoidant children were found to be much less emotionally expressive, less affectionate when holding their babies, and substantially more rejecting. All mothers of insecurely attached children exhibited a difficulty in responding to their baby’s attachment needs and signals in a consistent, loving, and sensitive way.

Ainsworth’s conclusion that caregivers needed to respond to an infant’s attachment behaviours if the child was to develop emotional security, and her findings that babies cried less at one year if their cries were quickly and consistently responded to in early months, flew in the face of contemporary behaviorist thinking and were harshly criticized. The opinion of the day was that "fussing over a baby whenever it cried would spoil the child and produce a crybaby."

However, Ainsworth’s careful methodology allowed other researchers using the "strange situation" experiment to correlate attachment styles and mother’s parenting style, with the important characteristics of a child’s ongoing development. As these findings accumulated, attitudes began to change.

The Minnesota Institute of Child Development Studies

The first attachment study conducted at the Institute in the late 70s by Waters et al. (1979), found that forty-eight of fifty ‘middle class’ babies, assessed at twelve months and at eighteen months, fitted the same Ainsworth attachment classification at each time period.

Later, Alan Stroufe (1983) and colleagues began a series of studies with the aim of determining what benefits or disadvantages attachment styles could have for developing children. He was looking for an underlying pattern over time (rather than specific behaviours) that might affect a child’s adaptation at each stage of life with such things as their relationships with others, ways of dealing with stress, resiliency, expectations and approach to life.

Stroufe recruited forty-eight of the mothers and babies, now two-year-olds, studied by Waters. He hypothesized, based on the earlier work of Erik Erickson (1950) and others, that two-year-olds should display autonomy, flexibility, resourcefulness, and use, but not be dependent upon, the assistance of their mother.

A set of activities was devised to test for these characteristics. It was predicted that secure children would easily involve themselves in the play tasks, share their enthusiasm, persist when frustrated, resist self-defeating behaviours, use their mother’s help when they couldn’t figure something out themselves, and generally cooperate with their mother although not immediately comply. In almost every way, the securely attached children behaved as predicted, showed more enthusiasm and persistence, and less frustration. Insecurely attached children were more likely to experience difficulties under tress, and the avoidant children within this group were more likely to pout, whine or lash-out in frustration. Most securely attached children smiled spontaneously at their mothers during the play tasks, while less than half of the insecurely attached children did this.

The mothers appeared to be consistent in their relative ability to be as sensitive to the needs of their children at two years as they were at one year. For example, mothers of securely attached children rated significantly higher in ‘supportive presence’ and ‘quality of assistance,’ giving just enough support to keep their child engaged. Mothers of insecurely attached children were more likely to over-or-under support their child’s experience and/or to misjudge the type of support required for their child’s enjoyment and success in the play tasks.

In 1974, Byron Egeland, a psychologist who was studying risk factors in child abuse, and Stroufe collected a longitudinal sample of two hundred sixty-seven expectant mothers who were mainly single, young, with few resources, little income, and had not planned the pregnancy. The study predicted security of attachment at one year by assessing the mother’s prenatal emotional health and attitude toward the baby. Depressed mothers and those with little interest in the baby during their pregnancy, were more likely to have insecurely attached children at one year.

When the children in this study were turning five years old, forty were enrolled in a specially created nursery school on campus. Two groups of children were closely observed for fifteen to twenty weeks. Predictions about typical preschool behaviour (e.g., having an active engagement with peers, in need of less adult assistance, able to follow rules, able to contain their impulses) were made.

Children with secure attachment histories scored higher in all areas – resiliency, self-esteem, independence, positive relationships, ability to enjoy themselves – and were seen to have more friends, superior social skills, and more empathy for peers in distress. Insecurely attached ambivalent children seemed too preoccupied with their own needs to deal with the needs of others displayed unpredictable, somewhat disruptive behaviour that antagonized other children. In addition, when they encountered aggressive behaviour, they were unsure how to respond, and became vulnerable targets for bullying. Two ambivalent patterns were identified by the researchers: the tense, impulsive child with poor concentration, easily upset by failures, and the fearful, oversensitive child, with little initiative or persistence.

Three types of insecurely attached avoidant children – the bully, the loner, the daydreamer - were identified. Their personalities elicited more negative and controlling responses from their teachers. By all measurements the avoidant children, like the ambivalent children, were highly dependent (seeking attention in negative ways, making more frequent contact with teachers, observer and teacher ratings) except when injured or disappointed they withdrew.

Later studies of preschool children confirmed and added to these results (Sroufe, 1988; Sroufe et al., 1990) with positive aspects of growth and development more likely to be present in the group of securely attached youngsters. For individual children, however, secure attachment was not an automatic guarantee of future emotional well being – particularly in the high-risk poverty sample. Sroufe and his team felt that a child’s understanding of relationships could only come from the relationships he or she had encountered – that the behaviour of the securely attached children who reacted sensitively to their companions, supported their vulnerable peers, and did not allow themselves to be bullied, was reflective of their home experience.

As they reached early adolescence, the securely attached children continued to demonstrate greater relationship abilities, being more likely to have and maintain friends, function well in groups (dealing with status, roles, conflict resolution), and without worry of sharing friends with others. Insecurely attached children demonstrated some gender differences as they grew older, with ambivalent boys likely to be shy and withdrawn, avoidant boys likely to be angrily aggressive, and girls coping by being socially appropriate, smiling, and internalizing feelings.

Fathers

Michael Lamb (Bornstein & Lamb, 1992) was one of the first attachment researchers to look at the role of fathers. He found in-home assessment of babies at 7, 8, 12 and 13 month of age that showed no preference for mothers or fathers for most attachment measures, and when in distress would be comforted by either depending on who was present. However when both parents were present, distressed infants regardless of gender, selected their mother.

Mary Main and colleagues (1981) conducted the first study of the quality of the attachment to father and found the same proportion of children were securely attached to fathers as to mothers but that children could be attached to none, one or both. For confidence and competence, one secure attachment provided higher measures than none, and two secure attachments provided higher measures than one. For empathy, two secure attachments were associated with significantly higher measures, and has been consistently noted in later studies (Main & Weston, 1982; Easterbrooks & Goldberg, 1990; Suess, Grossman & Sroufe, 1992).

While the fundamental relationship established over the first two years between the infant and primary caregiver (usually mother) overshadows the influence of any other attachment figure, the impact of the second parent (present or absent, involved or not, supportive or harsh) is of well known clinical significance.

Fathers’ growing involvement with their young children is seen as supportive of the child’s ability to move outside their mother’s sphere to the external world (Sroufe, 1992). They provide role models for their sons, and can provide daughters with a sense of personal value in relations with the opposite sex. Of additional importance, the supportive nature of husbands to their wives and marital satisfaction have both been associated with an enhanced quality of the relationship between mother and child (Belsky, 1984).

Mary Main and the Berkeley Studies

In the mid-70s, Mary Main, a student of Mary Ainsworth, in association with graduate student Donna Weston at Berkeley (University of California), and later with Judith Solomon, began a longitudinal study of middle-class families. In the process of assessing the security of attachment of children to both mother and father at 12 and 18 months, Main identified a group of children more disturbed than others, and not fitting Ainsworth’s three original categories. This group of children, categorized as having disorganized or disoriented attachment pattern, demonstrated behaviours associated with both ambivalent and avoidant children in unpredictable ways, and sought proximity to their mothers in disoriented manners. Main found that many of the mothers of disoriented children had suffered traumatic, unresolved losses in early childhood which she believed could be transmitted to the infants as a result of the mother’s own anxiety and fearfulness.

In 1982, forty of the six year old study children and their parents participated in a two hour videotaped session to look at reunion behaviour for this age group (compared to one year olds); the child’s level of functioning; the parent’s recollection of their own childhood; and any connections in parenting experience. Using pictures of children involved in separations from their parents, such as going to bed, to their first day at school, or parents leaving for a weekend and for two weeks, study children were asked to describe how the child in the photograph would be feeling.

Securely attached children could relate the experience of the child in the picture to their own, felt sad but could provide realistic responses about what the child could do (e.g. ask for a phone number so contact could be maintained, ask to go with them), and could talk about how they felt or would feel. Ambivalent and avoidant children also felt sadness for the pictured child but responded differently. For ambivalent children the separation seemed to mean a permanent separation from parents, which sometimes made them deeply angry. For avoidant children the distress of separation seemed overwhelming; their ability to become detached from the situation was no longer in evidence (as it had been at one year), and they could not think of anything the pictured child might do that could make a difference. For disoriented children, the separation provoked significant fear, such that they might imagine that the pictured child, they or their parents would be seriously hurt or killed.

When the children were shown a photograph of their arrival at the ‘lab’ with their parents to confirm all was well, the children in each of the attachment categories reacted differently. Securely attached children generally smiled and showed interest in the photo, while insecurely attached children tended to avoid the photo or actively walk away. The reunion session was completed with the return of one parent, and then within 3 minutes the other.

Upon the initial viewing of the videotaped reunions, the researchers were unable to find signals that were meaningful except at the extremes of security and insecurity. Eventually they were able to read subtle signs, and found in 85% of the cases the classification matched the assessment

at one year. The secure six year olds were demonstrably comfortable and relaxed in their verbal and physical relationships with their parents, and close to half brought themselves into brief contact with one or both – a touch in passing, a lean toward or against them, an arm around them for a moment. Insecurely attached children limited their dialogue, their range of subject matter and emotional response in a seemingly rigid manner of communication. The disoriented children were more likely to try to control their parent and the situation.

Challenges to the Assumptions about Avoidant Children

The interpretation of the behaviour of avoidant children as insecure brought some of the greatest challenges to the attachment theory and Strange Situation studies. Critics suggested that such behaviour could be a demonstration of healthy self-reliance and control of fear for example.

Adult Attachment Classifications – The Berkeley Studies

A component of attachment theory links the internalizing of a parent’s own experience of being parented, with the quality of the parent-child relationship they are able to develop with their own child.

From the interviews of parents of six-year-olds in the Berkeley study which were designed to assess their early attachment experiences, feelings, and states of mind, Mary Main was able to identify patterns of adult attachment. The three categories of adult attachment patterns matched those that Ainsworth developed for early childhood: secure-autonomous, dismissing of attachment pre-occupied with early attachment. The majority of children of the secure parents had been classified as securely attached at one year, three quarters of the children of dismissing parents had been classified as avoidant, and the majority of children of the preoccupied parents were seen as ambivalent. A key difference seen between ‘secure’ and ‘insecure’ adults was the secure adults’ ability to understand and talk reflectively about insights they have of themselves and their own parents, which in turn is seen to be an important link to being able to help their child gain understanding of his or her feelings and impulses.

Parenting or Temperament

In the mid-50s, Stella Chess and her husband, Alexander Thomas, both psychiatrists, began the New York Longitudinal Study of infants and mothers (136 from educated middle class families, 95 from working class Puerto Rican families) to explore the influence of infant temperament on later clinical problems. The mothers in the study provided details about all aspects of their children’s behaviour and their responses. Families participated on a regular basis from the time their children were three months old until they reached adulthood.

Chess came to the study believing from her own clinical and mothering experience that infants were born with certain temperaments, and that mothers were being overwhelmed with advice (and blame) about parenting practices. Nine variables were used to assess children’s temperament: activity level, rhythmicity (e.g. sleeping, waking, eating cycles), approach or withdrawal (when introduced to new things/people), adaptability, intensity of reaction, threshold of responsiveness (e.g. to visual, hearing, social stimuli), quality of mood (positive or negative), distractibility, attention span and persistence.

Chess then created four broad categories: easy babies (4 in 10), slow to warm up babies (1-2 in 10), difficult babies (1in 10 of her sample), and other babies with mixed characteristics (3-4 in 10). The mothers of difficult children did not appear to be different from the other mothers, although their attitude became less positive over time. The difficult babies were a trial for any adult and only a few could remain patient, consistent and good-humoured. "Slow to warm up’ children could also be difficult to parent. A number of children had emotional or behavioural problems requiring attention – including 70% of those with difficult temperaments.

Chess found that although temperament had a significant effect on personality, and that having a difficult temperament might predict later problems, the environment always played a key role. Parents, family, peers and teachers could positively or negatively contribute to a child’s adaptation, as could cultural expectations. Chess saw temperament and environment as continually interacting and modifying the other, and she encouraged tolerance of differences. In advocating sensitivity to infant’s and children’s differences Chess confirmed the idea of sensitive care promoted by Bowlby and Ainsworth, but at the time the researchers saw themselves on opposite sides of the nature-nurture (and the so-called mother blaming-mother supporting) debate. Research and writings on mother’s attunement with her infant and child provide similar support for quality of care (see, for example, Winnecott, 1965; Stern, 1985; Tronick, 1978).

More recently, Dymphna van den Boom (1988, 1995?, 1997) studied 100 children who were highly irritable at birth by dividing them into 2 groups of 50 pairs of mothers and babies. One group was provided six hours (3 visits) of counseling when the babies were 6-9 months old to enhance the mother’s sensitivity and effectiveness. In the group that received the intervention, 68 % of the fussy babies were classified as securely attached at one year. In the control group, 28% were classified as securely attached. This left little doubt that quality of care was an important variable in quality of attachment.

previous page