Repairing the Bond in Important Relationships: A Dynamic for Personality Maturation

The understanding of how interpersonal relationships promote individual development has relied on the psychoanalytic concept of internalization. Individuals are believed to overcome the deleterious effects of destructive childhood relationships by taking into the self either admirable qualities of important others or characteristics of relationships with important others. Although initially applied to infancy and childhood, internalization has also been promoted by investigators such as Vaillant as being a factor in adult personality growth: “It is fortunate that we never become too old to internalize those whom we admire” (6, p. 8).

However, we are far from a satisfactory understanding of internalization. Lear (7) described Plato as the first to discuss the internalization of experiences, but little progress has been made in developing a dynamic model of how internalization occurs. Blatt and Behrends, however, have suggested such a model (3, 4). Their model centers on the establishment of a gratifying relationship and disruptions in this relationship as the circumstances that may provoke internalization. When threatened with a disruption in the relationship with a gratifying other, the individual attempts to retain the other by internalizing him or her. Although Blatt and Behrends noted that repair is important, it is the loss or threatened loss that is emphasized in their model. The disruption may be as subtle as the failure of the important other to respond empathically or may be as gross as frank abandonment. Although this emphasis on disruption or loss goes back to Freud’s writings (8), it is not consistent with recent infant-caretaker research and may, in fact, represent the dynamics of insecure attachment patterns (i.e., those characterized by the infant either anxiously clinging or avoiding contact with the parent).

This article argues that adult development most often occurs when strong affective bonds are established with important others and the inevitable ruptures of these bonds are repaired. It is both the rupture and the repair processes that are crucial. To develop this emphasis will require a wide net involving research and clinical data from the three most commonly studied two-person relationships out of which individual growth can occur: infant-mother attachment, the psychotherapeutic alliance, and the marital relationship. To the extent that the affective bond process of establish/rupture/repair can be seen as crucial to positive outcome in all three of these dyadic relationships, we may begin to better understand one mechanism of adult development.

The systematic study of infant-caretaker attachment behaviors by means of Ainsworth et al.’s Strange Situation procedure (9) has revolutionized developmental research. Exposing a year-old infant to a series of brief separations from and reunions with the caretaker and assessing the infant’s responses results in the identification of a small number of attachment types that range from secure to varieties of insecure. Longitudinal studies have documented the considerable predictive power of these attachment types for subsequent childhood and adolescent development (10–14).

Until quite recently, the emphasis of the attachment paradigm had been on interactive synchrony (matching, attunement, reciprocity), or the degree to which infant and caretaker can achieve a shared affective state. During recent decades and spurred by the research of Tronick and his co-workers (15–19), a second important emphasis has been elaborated. This involves the concept of mismatches between infant and mother and their repair. In briefly summarizing this research, I will rely substantially on a recent review by Biringen et al. (5).

The research itself involves the microdissection of films of infant-mother interaction. It is clear that such interactions change as rapidly as every few seconds. It is also clear that both participants are actively involved in the interactions and that it is necessary to examine sequences of their interaction (rather than isolated behaviors of either participant) in order to understand what is occurring.

A finding of great significance was that mothers are correctly attuned to their infant’s emotional state only about one-third of the time. The infant’s responses to the frequent misattunements result in appropriate corrections by the mother in another one-third of the occasions. These findings have led to the conclusion that the successful repair of misattunements may be crucial for the infant’s normal development. Tronick and Gianino (15) emphasized that successful repair turns despair into positive emotions. This can lead to the growth of the infant’s sense of mastery, the elaboration of effective coping mechanisms, and, over time, the internalization of a relationship pattern of great value in later life. There can be adverse consequences in infants for whom successful repair is absent or infrequent. He or she may begin to feel helpless (without mastery) and may turn away from relationships and focus on self-regulation. Internalization of a pattern of unsuccessful repair leads to a limited and often later self-fulfilling relationship style. In addition, a negative affective core to the sense of self may be established. Thus, the authors suggested that “Psychopathology may be an outcome of repeated unsuccessful efforts to repair mismatches” (15, p. 5).

Biringen et al. wrote that dyadic harmony is “better characterized by a series of microdyssynchronies and dynamic resolutions than by any metaphor of harmony or attunement” (5, p. 4). They also make the useful distinction between dyssynchrony and conflict, with the latter describing greater tension, struggle, and derailment of the relationship. They also emphasize that some level of disengagement is necessary for the infant to develop the capacity to be alone.

Developmental neurobiology adds emphasis to the affective bond establish/rupture/repair findings of attachment research. Schore has written about the developing brain as experience-dependent (20, 21). He suggested that the infant’s early affect is changed from despair to joy by the repair process. This change in affect is associated with the secretion of dopamine and endogenous opiates, which then prompt synaptic growth in the prefrontolimbic regions of the right brain. These are the regions involved in the infant’s capacity for self-regulation. Failure to achieve repair in a reasonably consistent manner leads to structural changes in the brain that Schore believes form the biological substrate of later psychiatric disorders.

This intriguing body of attachment research has implications beyond that of early development. If the affective bond process of establish/rupture/repair can be identified in the other two dyadic relationships that facilitate individual growth (the psychotherapeutic alliance and marital relationship), we may be moving closer to a basic understanding of the interpersonal context that promotes internalization and psychological growth.

Kohut wrote that in psychoanalysis, it is the repair of relationship ruptures that leads to transmuting internalizations and personality growth (22). As previously noted, Blatt and Behrends emphasized that, although repair is important, it is the rupture that is crucial (3, 4).

Although the psychotherapeutic alliance has been a focus of much theoretical writing, there have been very few studies that address the moment-to-moment exchanges between patient and therapist. Most of the empirical studies have been at a more global level. Horvath and Luborsky (23) reviewed a large body of this global research and concluded that the therapeutic alliance explains up to 50% of the variance in outcome across a broad spectrum of psychotherapeutic approaches.

Much of the research on the therapeutic alliance relies on rating scales filled out by both patient and therapist following each session. Safran and his colleagues (24) took alliance research to a more microanalytic level. From rating scales that identified problematic sessions, they analyzed the verbatim audiotapes of such sessions. Their findings led them to identify markers of both ruptures and effective responses by the therapists. Ruptures were sometimes marked by the patient’s overt expressions of negative feelings about the therapist or the therapy. More often, however, the expressions were indirect and included avoidance maneuvers and facile compliance.

It is crucial that the therapist attend to the rupture directly, be aware of his or her own feelings, empathize with the patient’s experience, and openly accept responsibility for his or her role in the rupture. Safran et al. also concluded that it is helpful for the therapist to explore the patient’s fears of being more direct with the therapist.

In another article (25), Safran addressed the theoretical importance of the repair of ruptures in the psychotherapeutic alliance. He believes that repairs are particularly important because the rupture-repair process reflects a fundamental dilemma of human existence: the innate desire for relatedness and the reality of separateness. Safran then described the ways in which a wide variety of psychoanalytic writers are in accord with and amplify the role of this fundamental dilemma in health and psychopathology.

This broadening attention paid to the psychotherapeutic alliance, and in particular the affective bond sequence of establish/rupture/repair, grew out of greater understanding of the importance of that sequence in the infant-caretaker attachment relationship.

There is much evidence that the affective bond sequence of establish/rupture/repair is of central importance in marital relationships. Relevant findings have been summarized, and a couples therapy format that specifically addresses the crucial repair process has been described (1). Before describing that approach to dysfunctional relationships—which, almost by definition, are characterized by the relative absence of repair processes—several streams of relevant research should be noted.

One development in marital research involves the use of physiological measurements. Studies have documented alterations in pulse rates when repair processes fail and conflict ensues (26–28). For example, as the confrontation grows, a husband’s pulse rate may increase, and he will then withdraw emotionally from the interaction. As he does so, the wife’s pulse rate then increases. In this case, his cardiovascular response is to the confrontation; hers is to the feeling of being abandoned or shut out.

In a series of intriguing studies, Kielcolt-Glaser and her colleagues have demonstrated alterations in both endocrine function and cellular immune regulation following periods of either marital conflict or marital confiding (29, 30). Their important finding for purposes of this article was that conflict (the failure of repair) was associated with deleterious effects on both endocrine and immune functions and that periods of rapport enhanced the functioning of these important physiologic systems. These physiologic alterations lasted for up to 18 hours after the period of either conflict or confiding.

Marital-family research that has focused on well-functioning marriages and families also speaks to the issue of the relationship characteristics that promote increased probabilities of repair (31–36). The characteristics are presented here in a much-simplified 5-point summary:

1. Power is shared equally by the partners.

2. As a consequence, there is a high level of mutual respect for each other’s subjective reality.

3. Respect for subjective reality increases the likelihood of greater self-disclosure.

4. With greater self-disclosure, there is increased opportunity to appreciate both similarities and differences.

5. Appreciation of similarities and differences leads to both increased closeness and augmented individuation.

Because partners in such marriages are more likely to listen attentively, ask for needed clarifications, and respond empathically, there is greater likelihood of repair rather than conflict on those inevitable occasions wherein one partner is not “there” for the other.

In couples presenting with conflict, it is useful to understand their dilemma as a struggle over the definition of the relationship. Meissner (37) described an object-relations or transference approach to marriage in which each partner attempts to define the relationship in keeping with an internal working model that results from childhood experiences. Lear has emphasized that all such transference interactions are intensely political in that the participants are struggling over whose definition of the relationship is to prevail (7).

When the struggle over the definition of the relationship is intense, each partner may blame the other for the problem. It is necessary for the therapist to moderate the conflict and introduce an interactional perspective (A’s behavior leads to B’s response, which, in turn, increases A’s behavior, etc.). When this is accomplished, direct teaching of repair processes can begin. This can be done through homework assignments in which the couple is instructed to audiotape exploratory conversations for analysis in therapy. The details of this approach to couples therapy have been described elsewhere and will not be repeated in this article (1).

When this approach to couples therapy works, it does so, at least in part, because the spouses have learned to be more open with each other, to listen with greater respect, and to learn to repair at least some of their misattunements. When they do, not only does the quality of their life together improve but they establish the relationship context in which individual growth may occur.

Personality maturation occurs as a result of participation in relationships in which strong affective bonds are established and their inevitable ruptures repaired. The affective bond process of establish/rupture/repair is understood as the underlying dynamic of internalization, an important mechanism of growth across the lifespan.

Presented in part during psychiatry grand rounds at the Medical University of South Carolina at Charleston and the University of Texas Southwestern Medical School at Dallas. Received June 25, 1999; revisions received Oct. 12, 1999, and Feb. 18, 2000; accepted March 3, 2000. From the Timberlawn Research Foundation. Address reprint requests to Dr. Lewis, Timberlawn Research Foundation, 2750 Grove Hill Rd., P.O. Box 270789, Dallas, TX 75227.