The Trauma of Divorce: Reducing the Impact of Separation on Children

Dean McKay, Ph.D., B.C.E.T.S., F.A.A.E.T.S.


Any member of the American Academy of Experts in Traumatic Stress would concur that trauma may be defined broadly to include not only physical stress, such as combat involvement, auto accidents, natural disaster, torture, but also vicarious sources of trauma, such as observing others being traumatized, as in witnessing a major accident. These are the types of trauma that we most often consider when discussing the iatrogenic effects of overwhelming experiences. What has been largely ignored in the literature is an explicit consideration of divorce as traumatic. Yet the psychological ramifications of the divorce process are considerable, and one cannot overlook the potential traumatizing effects of divorce on the children involved.

First, let's reconsider how to define trauma. There are some key elements. First, it is usually an uncontrollable event. The nature of the event is beyond the scope of ordinary human experience; that is, the event is a rare or infrequent occurrence. In some instances, the event may not be rare, but is nonetheless beyond the scope of human experience, as noted in a review two issues ago of a book which described how refugee camp survivors experience trauma routinely. Moreover, trauma is usually unpredictable. Further, in an effort to process the event, the person is changed.

These are among the defining aspects of trauma, and several theorists have suggested ways in which we process this information. For example, several issues ago, I indicated that one may consider the etiology of Posttraumatic Stress Disorder (PTSD) from a behavioral perspective using the framework offered by Lang (1977) and Foa and Kozak (1986). These theories are equally useful for all classifications of traumatic events. Briefly, these theories suggest that in the immediate aftermath of a traumatic event, an individual copes with the environment by purposely "forgetting" the material, and slowly re-introduces the information when a state of equilibrium is established (Lang, 1977). Further, since re-introducing the material is often disquieting, there are some individuals who are at greater risk for developing a traumatic reaction (Jones & Barlow, 1990). Risk factors include biological elements and elevated levels of anxiety sensitivity (Reiss, 1991; Taylor, 1995), a psychological condition in which changes in bodily functions are perceived as potentially harmful and items present in the environment are then the source of avoidance. The links to trauma here are the situations in which the traumatic information is re-introduced (as in Lang's theory); then those situations are likely to be avoided in the future when one also has elevated levels of anxiety sensitivity. Foa and Kozak (1986) expanded Lang's theory to include mechanisms at work during treatment, and suggested that fear structures, such as those present in anxiety states following trauma, are hierarchically arranged.

Perhaps it would be instructive to provide an example which illustrates our model. Consider the situation in which there is an only child who is under the (false) impression that her parents are getting along well together. In fact, she has believed this to be the case for as long as she was capable of being aware of interpersonal interactions. Suddenly one day she is informed that her parents are going to be divorced. Then, once the situation is out in the open, the parents feel unburdened of maintaining the charade of being agreeable and begin fighting in their daughter¹s presence. This would fit the description of trauma, and one would not be surprised if this child experienced symptoms of stress. First, the incident was sudden and not predictable. Second, other incongruent stressors followed - the parents fighting. In this scenario, the professionals likely to see this child are not psychologists but law guardians and school guidance counselors. These people are also in a prime position for preventative interventions, details of which are offered shortly.

In light of this background definition and illustrative example, there are other specific features that make the trauma of divorce unique. Specifically, there are several issues to be considered from developmental psychology which interface with the divorce process. First, we will consider links to the definition. For the children (and in some instances, the adults), the process is beyond their control. Further, most would agree that it is not an ordinary process, but one which is atypical. The whole procedure may take months or even two to three years to complete. For example, a recent custody evaluation for which I was consulted had been in progress for two and a half years, with all members of the family residing in the same home. The more appropriate analogy here would be to compare this situation to one with ongoing trauma. The divorce process is unpredictable. This is true from the beginning of litigation‹when children often have little understanding of what is occurring, and changes happen rapidly and episodically‹to the final decision, after several court dates where the whole procedure is expected to end, it finally terminates, and most involved are surprised. Finally, the people involved are changed as a result of the process. This is true of the divorce process, where later adult development is usually strongly affected by divorce processes during childhood. In addition to these issues that are directly related to the divorce process in children, there are aspects to consider as they relate to developmental factors in the litigation process.

Attachment Problems

According to Ainsworth, et al. (1978) and others who have followed (i.e., Fonagy et al., 1996; Main, 1996), children develop what are referred to as attachment styles that are predictive of later psychopathological states. Essentially, attachment styles are broadly defined as falling into three patterns: secure, anxious or insecure, and avoidant. Each attachment style may be demonstrated in relation to the child's interaction with a parent or caregiver. Each style is best illustrated by an example. Imagine a child is in the supermarket with a parent. Securely attached children may wander a finite distance away from the caregiver, usually within visual range, without experiencing anxiety or discomfort. Should that distance be too far, however, the child generally attempts to find the parent. Anxious or insecurely attached children may not wander away from the parent, and the environment may be perceived as hostile, even if the distance from the caregiver is within visual range. By contrast, the avoidant child may wander from the parent and experience no concern whatsoever about distance.

These attachment styles have been found to be predictive of later difficulties in interpersonal relations. For example, Allen, Hauser, and Borman-Spurrell (1996) found that children identified as avoidant or insecure in childhood experience interpersonal difficulties in adolescence. Securely attached children perform best with interpersonal relations. This analysis has some implications for the divorce process. Since attachment is a developmental process, this implies that caregivers have some input into the manner in which children develop these styles. For example, inconsistent parental attention, either in the form of discipline, rewards, or availability, has been shown to be related to the development of attachment difficulties (Fonagy et al., 1996). Contrariwise, secure attachment has been found to act as a protective factor from later psychopathology (Fonagy, et al., 1996).

Divorce is a time of tremendous upheaval in families. Given this disruption, should the divorce process occur at a point where the child's attachment style is not fully developed, it stands to reason that the likelihood of secure attachment developing is reduced. The process whereby this may be problematic is illustrated in figure 2.

In addition to the formative aspects, attachment styles change over time. Let's consider the example given before. Here, the young girl believes her parents got along well and is likely to be a securely attached child. However, one could imagine a change in demeanor after learning that her parents were getting divorced. In fact, a shift in attachment is expected‹in this case, insecure attachment is most likely to result. The child is jolted from her false belief that her parents are not a distressed couple (and that therefore she has nothing with which to be concerned). The sudden and unpredicted information to the contrary would shake her faith in the likelihood of a positive outcome, despite reassurances. Contingencies and other environmental issues

Long the stronghold of behavioral theorists, the environmental contingencies play a pivotal role in the development of normal or abnormal adjustment. In particular, reinforcement history is a large contributor to the manner in which one copes with later problems (Skinner, 1953). Sadly, reinforcement history has either been ignored (Salzinger, 1996), improperly examined as something else (such as learned helplessness, Salzinger, 1994), or seen as unmeasurable phenomena (as in the case of constructs explicated in psychodynamic theory). If we use learning history as the basis of discussion, Skinner (1953) made it clear that consistent administration of contingencies (rewards, punishers) was the best method for maintaining normal functioning. During times of divorce, it has been well documented that the environment becomes anything but consistent (Ackerman, 1995), and may even be disruptive (Ackerman, 1995; Bricklin, 1995). For example, since it is in the best interests of the child that both parents work together in providing rewards and discipline, during divorce this process of cooperation frequently breaks down. Therefore, in the absence of consistent environmental contingencies, later problems in adjustment become more likely.

Again, in order to bring this into clearer focus, let's consider the young girl whose parents are getting divorced. The contingencies which were previously operating suggest that this child received frequent positive reinforcement both directly from her parents and indirectly by observing them behaving in an agreeable fashion. However, once it was announced that a divorce would soon take place, she probably did not receive the same level of reinforcement as previously, and her observations of her parents together were aversive instead of positive. Now, as well as being traumatized, this little girl was also likely to be depressed (Costello, 1978).

The Divorce Process

At this point, it appears well established that we may properly conceptualize divorce as being a form of trauma (uncontrollable, unusual and perhaps persistent, and resulting in long term behavioral change) with special implications for children (depending on attachment style and consistency of environmental contingencies). What are the specific aspects of divorce that create the trauma? And further, what may be done to alleviate or minimize the impact? Finally, what palliative measures may be taken when trauma has been clearly induced as a function of divorce? The balance of this paper is devoted to attempting to provide some working answers to these questions. Specific features of divorce that lead to trauma

In divorce proceedings, the most difficult aspect for children (as well as parents) to overcome is the issue of which parent is going to maintain custody. This thorny issue is the single area where mental health professionals (often psychologists) are called upon for expert testimony. The scenarios which lead to the intervention of a psychologist are multi-faceted, but most common is the situation in which the parents cannot agree upon custody determination. Usually, each parent wants to retain sole custody of the child(ren), with the other getting some level of visitation privileges. However, in some instances, parents may agree to some aspects of custody, but pre-existing conditions (such as mental illness) preclude immediate determination of custody. Here again, psychologists are called upon to test the relevant parties for competence and some determination is then made by the courts for custody and visitation arrangements.

It is at this point that one may view the effects of the divorce process in bold relief. Usually by the time all parties agree to engage in a custody evaluation, the divorce process has been well established and the negative effects have taken hold. Since the parents are not in agreement about either custody outright, or some idiosyncratic aspects of custody/visitation, then it is safe to assume that the participants are in conflict about most other aspects, and are not likely to readily agree to terms such as who pays what percentage to the evaluator; which psychologist should conduct the evaluation; and scheduling of appointments. Essentially, the conflict that goes into scheduling meetings with professionals only serves to exacerbate an already established dysfunctional process that is not well understood by the children involved.

Suppose that the girl described so far is informed about her parents' divorce at age seven. Further, the child is informed that she is going to live with only one of these parents and is asked to provide some preference. This may be further defined as traumatic. It is uncontrollable (the child may feel hopeless or helpless; authority figures are clearly requesting a choice be made, when either choice may result in aversive consequences), unusual (generally children may express preferences for both parents simultaneously), and unpredictable (a child may be posed this question for the first time in the law guardian¹s office, not in the confines of a mental health practitioner¹s office). Therefore, by the time the child reaches the psychologist, it is too late. The damage is done and what remains is an attempt to remedy the traumatic sequelae. It would be useful to know how this process may be shunted off before excessive damage is incurred. This, in fact, is consistent with the position adopted in the American Academy of Experts in Traumatic Stress: a multidisciplinary approach is necessary, and further, some professionals known for dealing with particular aspects of trauma are not in the best position to prevent its onset.

Further traumatizing is the process itself. At early stages of development, children become aware that events surrounding the custody decision are intended to determine with whom the child is going to reside. Common reactions children have in the aftermath of custody decisions (with or without the evaluation of a professional) are guilt for the parent who did not gain custody (Franke, 1983). This is the reaction of "If I had been a better child, my parents would still live together." Although Franke (1983) suggests that this reaction is most common in early school-age children, it has been observed across the age range (Gottman & Fainsilber-Katz, 1989). Other emotional stages have been proposed by Franke (1983) such as ages where sadness, anger, false maturity, and denial predominate. However, researchers have found support for each of these states at all age ranges (Barber & Eccles, 1992; Forehand et al., 1991; Shybunko, 1989). Finally, regardless of attachment style, level of conflict between parents, and consistency of environment, Hetherington (1979) indicates that it is unusual for children to experience divorce without some degree of adjustment difficulties.

If it is the case that adjustment problems are normative, then the issue facing other disciplines is how to prevent these reactions from remaining or worsening, resulting in long-term adjustment problems. Some practical guidelines may be offered that are not unlike those suggested to those initiating custody evaluations. First, provide a clear rationale for what is happening. Since traumatic events are by definition poorly processed, this assists in the integration of distressing information. Provide the information in a structured manner. In other words, do not "wing it." It is important to be well-prepared for questions which may arise during the information-dispensing period, since this further aids in the processing of information. Finally, indicate alternatives. This provides a sense of control that can mitigate the otherwise traumatic effects of the divorce process.

Minimizing the Impact

At this point it is clear that children involved in custody litigation are caught in the middle. If the parties involved are genuinely working in the best interests of their children, then law guardians and individual legal counselors are in an excellent position to provide guidance to the parents about how to minimize the impact of the proceedings. Ackerman (1995) notes that often a child who is conflicted (e.g., wanting to be with both parents, feeling one parent is pressuring him or her) adjusts to the process more poorly. Therefore, parents may be guarded against establishing an arrangement where one parent tries to exercise undue influence over the child regarding decisions about the other parent.

At this point, it seems that the manner in which a law guardian or individual law counselor handles a case is of paramount importance. It is important to note that at this time, the professional who may render an evaluation has not been even contacted! In the absence of any direct mental health services, the lawyer in contact with the family is really the best person to prevent the worsening of trauma to the children. Although only case examples, two recent custody evaluations I was involved in illustrate this point well. In one case, the law guardian was knowledgeable about the psychological effects of divorce, and was able to provide an environment where information was dispensed gradually and in a meaningful way. Further, he encouraged the parties to maintain, to the best of their abilities, normal parenting styles during the process. There were two children involved in the process and although they were clearly struggling with the divorce process, their adjustment could be referred to as adequate. By contrast, another law guardian I worked with had little knowledge of the psychological impact and did not advise his clients accordingly; when the three children were being evaluated, they were experiencing considerable distress. Further, the case was considered onerous by this lawyer, whereas in the first case the lawyer perceived the case fairly positively. So the manner in which it is handled may also reduce stress in handling the case.

Repairing the Damage

Invariably, a number of children experience long-term negative consequences from the divorce process. As the proceedings draw to a close, one may observe "phantom recovery": the child appears to be settling down and ostensibly seems adjusted. This may not actually be the case. In keeping with our theory (illustrated earlier by Lang, 1977), following the termination of a traumatic event, there is often a period of relative adjustment and calm. This, however, lays the groundwork for later difficulties where the fear structures are forming and becoming entrenched. When attempts at processing the fear structures follow, traumatic reactions reliably occur. In order that we may circumvent these problems, it is important for children to discuss their concerns following the divorce in an environment that facilitates exposure in a fashion that is not traumatizing. For example, if the child were to discuss the features of the divorce that were difficult, then it would become less likely that future difficulties would arise since the fear structure has been weakened by repetition, which allows for processing of the information. Hetherington (1979) suggests that interventions are generally indicated for children following a divorce, since it is expected that they will experience difficulties. The mental health worker's job at this point is not to pathologize the difficulties, but to assist in the adjustment process in order to prevent psychopathology.

Divorce is a complex event that is most difficult for children in later adjustment. This paper was intended to provide an overview of what factors influence adjustment difficulties in the divorce process and how the variety of professionals involved in such cases may best reduce the harmful impact of divorce. Characterized as traumatic, divorce reactions apparently follow the same course as other traumatizing events. By relying on a network of professionals such as members of AAETS, the divorce process can be made a less traumatizing event for all parties involved.


Ackerman, M.J. (1995). Clinician¹s guide to child custody evaluations. NY: Wiley.

Ainsworth, M.D.S., Blehar, M.C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the strange situation. Hillsdale, NJ: Erlbaum.

Allen, J.P., Hauser, S.T., & Borman-Spurrell, E. (1996). Attachment theory as a framework for understanding sequelae of severe adolescent psychopathology: An 11-year follow-up study. Journal of Consulting and Clinical Psychology, 64, 254-263.

Barber, B.L. & Eccles, J.S. (1992). Long term influence of divorce in single parenting and adolescent families and work related values, behaviors, and aspirations. Psychological Bulletin, 111, 108-126.

Bricklin, B. (1995). The custody evaluation handbook: Research-based solutions and applications. NY: Brunner-Mazel.

Costello, C.G. (1978). A critical review of Seligman¹s laboratory experiments on learned helplessness and depression in humans. Journal of Abnormal Psychology, 87, 21-31.

Foa, E.B. & Kozak, M. (1986). Emotional processing of fear: Exposure to corrective information. Psychological Bulletin, 99, 20-35. Fonagy, P., Leigh, T., Steele, M., Steele, H., Kennedy, R., Mattoon, G., Target, M., & Gerber, A. (1996). The relation of attachment status, psychiatric classification, and response to psychotherapy. Journal of Consulting and Clinical Psychology, 64, 22-31.

Forehand, R., Wierson, M., McCombs, A., Thomas, R., Fauber, L.A., Kempton, T., & Long, N. (1991). A short-term longitudinal examination of young adolescent functioning following divorce: The role of family factors. Journal of Abnormal Psychology, 100, 97-111.

Franke, L. (1983). Growing up divorced. NY: Linden Press/Simon & Schuster.

Gottman, J.M. & Fainsilber-Katz, L. (1989). Effects of marital discord on young children's peer interaction and health. Developmental Psychology, 25, 373-381.

Hetherington, E.M. (1979). Divorce: A child's perspective. American Psychologist, 34, 851-858.

Jones, J.C. & Barlow, D.H. (1990). The etiology of posttraumatic stress disorder. Clinical Psychology Review, 10, 299-328.

Lang, P.J. (1977). Imagery in therapy: An information processing analysis of fear. Behavior Therapy, 8, 862-886.

Main, M. (1996). Introduction to the special series on attachment and psychopathology. Journal of Consulting and Clinical Psychology, 64, 1-3.

Reiss, S. (1991). Expectancy model of fear, anxiety, and panic. Clinical Psychology Review, 11, 141-153.

Salzinger, K. (1994). Measurement of reinforcement history: A neglected area in psychology. Master psychologist invited paper at the annual meeting of the Association for Behavior Analysis. Washington, DC.

Salzinger, K. (1996). Reinforcement history: A concept underutilized in behavior analysts. Journal of Behavior Therapy and Experimental Psychiatry, 27, 199-207.

Shybunko, D.E. (1989). Effects of post-divorce relationship on child development. In Children of divorce: Developmental and clinical issues (pp. 299-313). NY: Haworth Press.

Skinner, B.F. (1953). Science and human behavior. NY: Appleton-Century Crofts.

Taylor, S. (1995). Issues in the conceptualization and measurement of anxiety sensitivity. Journal of Anxiety Disorders, 9, 163-174.

Published by the American Academy of Experts in Traumatic Stress - 2020

join the membership









crisis management specialities


more about certification in crisis management specialities

diplomate credential

The Diplomate distinction is a prestigious credential awarded to members that recognizes their experience in working with survivors of traumatic events and/or crisis management, knowledge, training and level of education.


fellowship credential

The Fellowship designation is the highest honor the American Academy of Experts in Traumatic Stress and National Center for Crisis Management can bestow upon a member. This designation is awarded to Diplomates who have made significant contributions to the field and to the Academy or the Center.