Anomalous maternal behavior at four-months and infant attachment disorganization at one year
Table of Contents
Abstract iv Acknowledgements vi List of Tables xii List of Appendices xiii Chapter One Introduction................................................................................... 1 Chapter Two Literature Review........................................................................ 10 The Mother-Infant Relationship 11 Psychoanalytic Theory – The Early Years 11 Psychoanalytic Theory – The Later Years 11 Attachment Theory and Research 16 Ainsworth 19 Maternal Behavioral Correlates and Antecedents of Infant Attachment 20 Infant Research 23 Emotional Regulation Theories and Research 26 What is Optimal Emotional Regulation 27 What is Being Regulated 28 Infant Capacities for Regulating Emotion 29 The Mother as Psychobiological Regulator 30 Attachment as a Measure of Emotion Regulation 33 Maternal Affect Containment & Regulation 34 Disorganized Attachment 36 Origins and Antecedents 38 Outcomes and Sequelae 40 Anomalous Parenting: Parental Behavioral Correlates 41 Relation between Early Caregiving Relationship, Affect Regulation, and Disorganization 43 Segregated Mental Contents and Defensive Exclusion 44 The AMBIANCE Coding System: Assessing Disturbances in Mother-Infant Affec tive Communication 46 Summary of Literature Review and Statement of the Problem 55
Chapter Three Method 1 - The Development of The Modified AMBIANCE– Selected Affective Errors, 4-Months 59 Chapter Four Method 2 - Methods of the Study 70 Chapter Five Results 85 Chapter Six Discussion 119 Appendices 151 References 164
xiii LIST OF TABLES
1. Multiple Linear Regression Model Examining the Impact of Total Frequency of Anom alous Maternal Behavior and Demographic Control Variables on D-ness
94 2. Multiple Linear Regression Model Examining the Impact of Global Ratings of Matern al Disrupted Affective Communication and Demographic Control Variables on D-ness
95 3. Multiple Logistic Regression Model Examining the Impact of Anomalous Maternal R esponse to Infant Distress and Demographic Control Variables on Infant Attachment Security
96 4. Multiple Logistic Regression Model Examining the Impact of Anomalous Maternal R esponse to Infant Distress and Demographic Control Variables on Infant Attachment Disorganization
98 5. Multiple Logistic Regression Model Examining the Impact of Anomalous Maternal R esponse to Infant Distress and Demographic Control Variables on Infant Attachment Security Versus Disorganization
99 6. Multiple Linear Regression Model Examining the Impact of Anomalous Maternal R esponse to Infant Distress and Demographic Control Variables on D-ness
100 7. Multiple Logistic Regression Model Exam ining the Impact of Aggressive Maternal Response and Demographic Control Variables on Infant Attachment Disorganization
101 8. Multiple Logistic Regression Model Exam ining the Impact of Aggressive Maternal Response and Demographic Control Variables on Infant Attachment Security Versus Disorganization
102 9. Multiple Linear Regression Model Examining the Impact of Aggressive Maternal Response and De mographic Control Variables on D-ness
103 10. Hypotheses Summary of Results Table
113 11. Results Summary Table
LIST OF APPENDICES
A. Derivation of Selected Anomalous Maternal Behaviors: The Modified Am biance–Selected Affective Errors, 4 Months (M-AMBIANCE), Miller, 2010
151 B. Modifications to Global Rating Scale for Measuring Level of Maternal Disrupted A ffective Communication
153 C1. Descriptive Statistics for Total Anomalous Maternal Behavior across Entire Sa mple (N=75) and Infant Attachment Categories
160 C2. Descriptive Statistics for Global Rating (1-7) of Level of Maternal Disrupted Affective C ommunication across Entire Sample and Infant Attachment Categories
161 C3. Crosstabulation of Anomalous Maternal Response to Infant Distress Ever by Attachm ent Classification and Across Entire Sample (N=75)
162 C4. Crosstabulation of Aggressive Maternal Response Ever to Infant Distress, Disengagement or Neutral/Positive Cues by Attachm ent Classification (N=75)
1 Chapter One
“Attachment studies typically assess parental “s ensitivity” as the aspect of parental behavior associated with infant attachment security....What is required from the parent to merit this description is a continuing attempt to apprehend the infant’s current subjective reality (affect state, current desired goal, ...level of understanding) and an attempt to devise a response that acknowledges and comments or elaborates on that state.....[However,]....the difficulty of knowing another’s mind guarantees that communication will be fraught with error and require many procedures for disambiguating messages, detecting and correcting misunderstandings, and repairing serious communicative failures....[Parent-infant dialogues] that are open to the entire array of affective communications; that include both partners’ initiatives in a balanced mutually regulated dialogue; that are characterized by active negotiation and repairing of miscues, misunderstandings, and conflicts of interest; and that are actively scaffolded by the developmentally more advantaged partner toward more flexible and inclusive forms are associated with positive outcomes for the child.” (Lyons-Ruth, 1999, pp. 583-4).
Research has shown that the early years are a c ritical developmental period for the growing child, and that unfavorable circumstances during this time can potentiate considerable and irreparable lifelong consequences. The early relational environment, in particular, is recognized as the primary context out of which development and personality emerges. Over the past several decades, a great deal of research has focused on yielding an ever more fined tuned understanding and articulation of the crucial features and long- term implications of this primary bond, the attachment relationship between mother and child. The concept of attachment was first introduced by John Bowlby to describe the affectional tie an infant form s with his mother, and the range of behaviors an infant displays to ensure proximity with this principal haven of protection and comfort during moments of stress and fear. Bowlby’s thinking about the primacy of the mother-child
2 bond was first described in 1944, when he published his groundbreaking research on 44 juvenile thieves, empirically documenting an association between early relational ruptures and later psychopathology. Among the 44 young delinquents whom Bowlby studied, the great majority had suffered prolonged and devastating early separations from their mothers; close to half of this group were characterized as “affectionless”, exhibiting a lack of responsibility, care, affection or regard for others. This was a significant contrast to Bowlby’s control group of clinic patients, very few of whom had experienced early maternal separations, and none of whom were labeled “affectionless”. Despite some limitations of the study, this data established a verifiable link between deprivations in the early caregiving environment and later social-emotional disturbances, which ranged from antisocial behaviors (such as stealing), to more severe expressions of psychopathology (evidenced by Bowlby’s subset of juveniles labeled “affectionless”). Harlow’s (Harlow & Zimmerman, 1959; Harlow & Harlow, 1962) work with rhesus m onkeys offered further evidence that grave disturbances in the early mother- infant bond can have devastating and enduring social, emotional, and behavioral consequences. Raised amidst surrogate mothers constructed from either wire or cloth, Harlow’s monkeys showed a preference and propensity to attach, only with the cloth figures, irrespective of whether these surrogates offered nourishment. These findings buttressed Bowlby’s claim of an innate and universal predilection to attach to a protective and comforting maternal figure. However, monkeys raised in social isolation went on to develop grossly aberrant social behaviors. Under conditions of stress they could be seen crying, screaming, rocking, frantically moving from one object to the next, huddled on
3 the floor, or staring blankly. Years later, they were aggressive with peers and unable to mate and produce offspring. Mary Ainsworth (Ainsworth et. al., 1978) went on to develop an experimental protocol designed to em pirically examine differences in patterns of infant-mother attachment. Ainsworth’s 20-minute “Strange Situation” procedure exposed 12- month- old infants to a series of mounting stressors, involving separations and reunions with the mother along with interactions with a stranger, and was intended to activate the infant’s attachment behavioral system. Based on the infant’s capacity to use the mother as a secure base from which to explore the environment, and the infant’s reactions to maternal separations and reunions, the quality of infant-mother attachment was determined. Ainsworth initially identified 3 organized patterns of infant-mother attachm ent, which correlated significantly with the quality of maternal caregiving that she and her research team had observed within the home during the infant’s first year of life. Infants with secure attachments to their mother showed a flexible balance of exploratory and attachment seeking behavior, and while often evidencing distress following maternal separations, they were readily soothed by their mother when she returned. Infants with insecure-resistant patterns of attachment displayed a high amount of approach and attachment seeking behaviors, yet reduced exploratory behaviors. During maternal separations these infants showed heightened distress, yet they were not easily soothed by their mother when she returned. Infants with insecure-avoidant attachment patterns exhibited a high level of exploratory behavior, yet very little approach and attachment seeking behavior. These infants showed no outward signs of distress during maternal separations, and avoided their mother completely when she returned. While mothers of
4 secure infants were found to be highly sensitive and responsive to infant cues and signals, mothers of insecure-resistant infants were judged to be highly inconsistent and unpredictable in their interactions, and mothers of avoidant infants were found to be highly rejecting and intrusive in their interactions. Throughout 30 years of cumulative research, the quality of infant attachment, as determined via Ainsworth’s protocol, has become a well-established, reliable, and valid means of assessing the caregiving relationship, and predictive of an wide scope of developmental outcomes. It was noted shortly after Ainsworth described her classificato ry system, however, that not all children could readily be grouped into one of the 3 organized patterns of attachment. In 1986, Main & Solomon described a group of infants who evidenced disturbed attachment behaviors during maternal separations and reunions, including indices of bizarre, contradictory or disoriented behaviors, and the absence of a singular, coherent strategy of coping when the attachment behavioral system was activated. The only unifying feature among this group of infants was their display of disorganized or disoriented behaviors in the face of attachment related stress; as such, Main and Solomon labeled this group insecure-disorganized in relation to attachment (Main & Solomon, 1990). Since the discovery of the disorganized category of infant attachm ent in 1986, the literature has documented a robust link between this pattern and the most detrimental outcomes throughout childhood, adolescence, and into adulthood (e.g., Carlson, 1998; Lyons-Ruth & Jacobvitz, 1999; van IJzendoorn, Schuengel, & Bakermans-Kranenburg, 1999; Fonagy, Gergely, Jurist & Target, 2002, Grossman, Grossman &, Waters, 2005). Attachment disorganization has been found to put a child at risk for a host of social-
5 emotional, developmental, and pathological outcomes, including childhood behavioral problems, severe forms of psychopathology, such as borderline personality disorder, and dissociative symptomology. For Lyons-Ruth, disorganized infant attachment reflects a malfunctioning in the attachment relational system. As Carlson (1998) describes, “disturbed attachment relationships are linked to later psychopathology, not as early disorders of the infant, but as markers of a beginning pathological process, risk factors for later pathology in the context of a complex model of interactive, biological and environmental variables” (Carlson, 1998, p.1108) Considering the detrimental sequelae linked to this classification, the need to ascertain the origins of attachm ent disorganization has been of paramount concern for both clinicians and attachment theorists, and a great deal of research has been devoted to this end throughout the past decade. While it has been shown that severe maltreatment and neglect strongly predicts infant attachment disorganization, (Carlson et al., 1989) 15% of the infants from low-risk, middle class samples, who have not been subjected to gross maltreatment and neglect, also reveal disorganized patterns of attachments (van IJzendoorn, Schuengal, & Bakermans-Kranenburg, 1999). Parental unresolved mental representations of their own early attachment related loss or trauma, as measured by the Adult Attachment Interview (AAI, George, Kaplan, & Main, 1985), have been found to account for additional variance in infant disorganization (Madigan et. al., 2006). However, parental loss or trauma does not account for all instances of attachment disorganization and it is not yet clear exactly how the intergenerational transmission from parental representation to infant coping style transpires. This gap in theoretical and
6 clinical understanding is referred to in the attachment literature as the “transmission gap” (van Ijzendoorn, 1995; Madigan et. al., 2006). It is a common tenet of attachm ent theory that the quality of the early interactive caregiving environment strongly influences the quality of infant-mother attachment. Over the past two decades, numerous researchers have explored the etiology of attachment disorganization, and have made substantial strides in describing the parental interactive features associated with infant disorganization. Nevertheless, there is still more work to do in this area, to hone in, and more fully articulate and establish the precise parental behavioral interactive features linked with infant attachment disorganization. Early on, the primary body of literature look ing at antecedents to attachment, focused on the role of maternal sensitivity and its relation to secure patterns of infant attachment. While numerous empirical studies have documented the link between maternal responsiveness to infant rhythms, cues and signals and infant attachment security (see De Wolff & van IJzendoorn, 1997; Ainsworth, 1978), these studies were conducted prior to the identification of the disorganized category of attachment. Maternal sensitivity, as measured by Ainsworth’s global rating scale (Ainsworth et al., 1978), has not been predictive of disorganized attachment status (van IJzendoorn et al., 1999). This earlier body of research predominantly employed global measures and ratings to assess broad qualities of maternal conduct. Yet, these scales offered little insight into the behavioral specificities inherent in these global codes. As noted by Lyons-Ruth and Jacobvitz (1999), “[Ainsworth’s scale] does not appear to be differentiated and specific enough regarding the affective communication involved in fear-related behavior to predict infant disorganization”.
7 Seeking to clarify the early interactive mechanisms of transmission leading to disorganized attachment, Main & Hesse (1990) and Lyons-Ruth (Lyons, Bronfman, & Parsons, 1999) identified 3 hypothesis that have all gained some empirical support. Main & Hesse (1990) initially proposed that disorganized attachment was related to frightened or frightening parental behavior, which placed an infant in the irresolvable situation of “fright without solution”, in which the caregiver becomes both the source and the solution to distress. Building on this theory, Lyons-Ruth and her associates (Lyons-Ruth, Bronfman & Parsons, 1999) suggested that both contradictory caregiving behavior and “extreme parental misattunement” (Lyons-Ruth and Jacobvitz,1999, p.531) were likely to perpetuate fear, and lead to the breakdown of an organized attachment strategy. According to Lyons-Ruth and others, an organized strategy has to work well enough to elicit some basic comfort and protection from the parent. If the parent becomes unresponsive, unpredictable and unavailable, beyond a critical point, the child experiences no felt protection and loses the capacity to sustain an organized strategy. Attempting to design a comprehensive clinical instrum ent that would identify the full scope of disturbances in mother-infant affective communication associated with disorganized patterns of infant attachment, Karlen Lyons Ruth and colleagues developed The Atypical Maternal Behavior Instrument for Assessment and Classification (AMBIANCE) (Bronfman, Parsons, & Lyons-Ruth, 1999),. The AMBIANCE incorporated Main & Hesse’s (1992) FR scale (Frightening, Frightened, Dissociated, or Disorganized Behavior on the Part of the Parent), which indexed a large array of frightened, frightening, and disoriented maternal behaviors theoretically linked to infant disorganization, as well as incorporating additional behaviors theoretically or empirically
8 linked with disorganization or maltreatment. The original AMBIANCE was designed to be used within the Strange Situation when infants were 12-24 months old. In 2004, Kelly adapted the AMBIANCE (Bronfman, et. al., 1999) to use with infants 6 months and younger. While many studies have established a significant relationship between anomalous parenting behavior and disorganized infant attachm ent using both the AMBIANCE and FR scales (see Madigan et al., 2006), very few studies have assessed anomalous parenting behavior prior to 12 months, and much of the variance in disorganized attachment, according to Madigan et al., has still not yet been accounted for. In a meta-analysis by Madigan & her colleagues (2006), which incorporated 12 studies and 851 mother-infant dyads, a limited proportion of the variance in disorganized attachment was found to have been accounted for by anomalous parenting behavior at 12 months or later. As Madigan and colleagues note, however, the very breadth and extensive nature of these scales may inadvertently weaken the link between anomalous parenting behavior and disorganized attachment. They emphasize the need for an explication of “those components or behavioral indices of anomolous behavior that are most directly implicated in the development of disorganized attachment relationships” (p.106). Lyons-Ruth and colleagues (Lyons-Ruth, Bronfman & Parsons, 1999) reported that affective communication errors (e.g., not attem pting to soothe infant when distressed, or laughing while infant was crying) was the only dimension of atypical maternal behavior that significantly distinguished mothers of organized from disorganized infants in their sample of high risk families, although negative-intrusive behaviors (e.g., pulls
9 infant by the wrist; mocks teasing infant), and disoriented behaviors were correlated with an infant’s level of disorganization. This current study will expand on the work of Lyons-Ruth, Bronfman & Parsons, (1999) and of Kelly (2004) by identifying a select array of parental behaviors projected to be the m ost salient, critical predictors of disorganized infant attachment. Thus, this investigation will strive to address the very gap in attachment research emphasized by Madigan et.al.(2006). In addition to establishing a more refined, selective, and focused coding system for assessing severe disturbances in mother-infant interactive behavior, this study will expand on the work of Kelly (2004), by examining 4-month mother-infant exchanges, and endeavoring to identify earlier clinical indices of disorganized infant attachment, the most vulnerable and at-risk population of children the literature has yet to identify. Earlier detection of at-risk dyads offers the hope of providing earlier intervention, and the prevention of adverse developmental and psychological outcomes.
10 Chapter Two
The goals of this study are twofold. The first aim is to expand on the work of Lyons-Ruth, Bronfm an & Parsons (1999) and of Kelly (2004) by identifying a select array of parental behaviors projected to be the most salient and critical predictors of disorganized infant attachment. The second aim of this study is to determine whether selected anomalous maternal behaviors at 4 months can distinguish attachment styles at one year. If so, then this is a step toward creating a more refined, focused coding system for assessing early disturbances in mother-infant affective communication and for predicting attachment. The literature review will thus set out to exam ine the centrality of the mother- infant relationship as described within the psychoanalytic, attachment, infancy, and neurobiological literatures, addressing developmental implications emphasized by both theoretical and empirical findings. The developmental implications of early caregiving behavior, in particular, will be reviewed, with emphasis on the study of attachment. The role of the mother as a regulator of the infant’s affect and psychobiology will subsequently be examined, as will the consequences when derailments in this process ensue. A discussion of disorganized attachment will follow, including a review of the research illuminating the link between disorganized infant attachment and anomalous parental behaviors. A discussion of findings, theories, and suggested implications for future research will follow.
11 The Mother-Infant Relationship
Psychoanalytic theory- the early years . The early years are commonly understood to be a for mative period, when a child’s personality and habits of responding to the world are developing. It is also a time, according to Freud (1949), when a child is most vulnerable, for the ego, which has not yet fully developed, has limited capacity to cope, and thus remains highly susceptible to traumas and neurosis. For early psychoanalytic thinkers, however, traumas and neurosis were thought to result largely from thwarted biological drives or derailments in psychosexual development. The mother, during this early period, was perceived as significant chiefly for her role in reducing drives and gratifying the infant’s physical and physiological needs. As studies of institutionalized children began to s how (Freud & Burlingham, 1943; Spitz, 1945; Robertson, 1952), however, it was not the mere fulfillment of physical or biological needs that ensured an infant’s well-being and healthy development. This work suggests that without the physical, social, and emotional availability and responsiveness of a primary caregiver, a child’s psychological development is compromised.