More on Overnights, Infants, and Toddlers

A few days ago, at http://childmyths.blogspot.com/2014/03/overnights-divorce-and-new-article-by.html, I commented on a recent article that had been sent to me by Dr. Richard Warshak. (The complete citation for the article is given in that post.) In that paper, Dr. Warshak had examined research that is relevant to the issue of having infants and toddlers pay overnight visits to the estranged parent with whom they do not live most of the time. “Overnights” are a hot-button issue for many divorcing parents with very young children, and for a variety of reasons such visits have become a contentious part of the development of post-divorce parenting plans. Although some psychologists have argued that it is unwise to permit overnight visits for infants, toddlers, and even preschoolers, Dr.Warshak states that the available research evidence does not support that position--  and he is supported in his claim by a great many “famous names” in developmental psychology.

Dr.Warshak would like to establish a general guideline for courts, stating that there is no evidence that young children in most divorcing families will suffer from overnight visits to a parent who does not live with them. In this preference, he opposes the several-decades-old principle of the “best interests of the child” as recommended by the authors Joseph Goldstein , Anna Freud, and Albert Solnit; one aspect of this principle includes the idea that a custodial parent should in fact have all the decision-making power about shared parenting and all other aspects of child care.

For myself, I am uneasy about the formation of a general rule in an area where there are so many individual and family differences. But I concede two points. One is that whether or not a general rule is formulated in an “official” way, courts will employ their own general rules, and past experience with “best interests of the child” suggests that these court-determined general rules may wander far from the information on which they purport to be based. The second point is that if there is going to be a general rule, it had better be based on research evidence rather than on values or traditions or other a priori assumptions.

Without yet being in a position to say what general rule about overnights I would like to see in place, I can point to some considerations that I think should be included in any thinking about this subject:

1. Burden of proof.  In discourse that tries to apply rules of logic to an argument, it is considered that one who states that something exists must demonstrate that evidence supports that claim; he or she cannot challenge an opponent to show that it does not exist, and in fact it would be impossible to show this about anything. The burden of proof concept means, in the case of overnights, that those who argue that such visits are potentially harmful to young children have the burden of demonstrating that  harm has resulted in the past. So far, there has been no clear demonstration of harm.
2. Generalizability. The search for research evidence for or against overnight visits is bound to be a frustrating one, because this kind of research is so difficult and the existing work is so weak in comparison to studies that can control variables. No one can be made to divorce for research purposes, so all studies of post-divorce events are full of the confounded variables related to personality, situational, economic, and other factors that played some part in causing the divorce which the couple “chose” for themselves. These factors also influence children’s developmental outcomes and their effects may be impossible to differentiate from the effects of divorce itself, of overnight visits, or of any other factor. In order to generalize from the results of research to the “real life” of individual families, we need to know that the families closely resemble the group from whom the  research evidence was collected--  but because there are so many variables involved, this may be quite difficult to determine.

      It’s especially important not to assume that research on middle-class, well-educated           divorcing couples can necessarily be generalized to high-risk families, ill-educated and      living in poverty--  or vice-versa. This reality can lead to problems because it is often    much easier to get middle-class people to participate in research than is the case for poor    people, who may feel threatened by the apparent nosiness or imagined disapproval of the             researcher. When a study looks at a group of parents of whom a high percentage are          under restraining orders (cf. Solomon & George, 1999), or who have some other unusual       characteristics, it is questionable whether the results of the study should be generalized to the rest of the population.

      In considering generalizability, it’s also critical to see what comparisons were made in       the research being considered. Do these comparisons reflect the differences we might be   interested in understanding? For example, Dr. Warshak’s paper examines with care the     famous study by Solomon and George which has been used to suggest that overnight  visits were problematic for toddlers. Solomon and George compared young children who were having overnight visits with a combined group who were either in an intact family    or were with a mother alone, but not having overnight visits. But surely the important      question for generalizing to the guidance of parenting plans is the difference between children of separated parents who were having overnight visits, and those who were not.

3. Issues about attachment. One of the serious problems related to creating general guidelines about overnights for young children is the extreme popular focus on attachment and the assumption that attachment quality determines everything from childhood obedience to school performance to later success in marriage. Other motivations and developmental factors of childhood, such as exploration, are generally ignored outside developmental psychology. There are also some common misunderstandings about attachment that have become woven into thinking about parenting plans.

1. Monotropy. The original theory of attachment formulated by John Bowlby made the assumption, based on studies of animal behavior, that infants’ and toddlers’ emotional attachment was to a single caregiver at a time (monotropy). As it became clear that this was not the case, there was a tendency to make only a slight change in the monotropy assumption by designating primary and secondary attachment figures.

      However, it seems clear now that several attachment figures can (and usually do) exist for young children, and attachment to one does not alter attachment to others, although a child may show preferences for one over another in certain circumstances. The obvious example is the maintenance of attachment to a parent in spite of many hours a week spent with a day care provider who is also an attachment figure. A reasonable conclusion is that overnight visits to a parent, if they facilitate attachment to that person, do not also destroy attachment to the live-in parent.

2. Time limits. The original view of attachment was that there was a sensitive period during which attachment relationships had to develop, or else they would not be possible. As Everett Waters pointed out in his interview with Jennifer McIntosh in the 2011 Family Court Review issue, this does not appear to be the case. Attachment is a developing, changing aspect of motivation and personality, and does not have the time limits once assumed to hold. Understanding of this point can suggest, as Waters did, that early overnight visits are not required for eventual attachment to a nonresident parent, but on the other hand it does not particularly suggest that such visits will be beneficial.
3. Security and order. Some discussions of parenting plans focus on security of attachment. However, insecure attachments are by no means outside the normal range or associated with psychopathology. Presumably it is not the purpose of a parenting plan to create the ideal of psychological development, but instead its purpose is to balance the needs of family members in a way that provides for “good enough” development of a child. “Insecurity” sounds unpleasant, and in popular thought is associated with neurotic behavior, but in fact insecure attachment is an orderly and adequate strategy for dealing with attachment and exploratory needs.

Much more serious outcomes are associated with disorganized/disordered attachment, a way of responding to caregivers in which young children may “freeze” on seeing the adult, or may use unusual approaches like backing up to the person or approaching while averting the eyes. Solomon and George’s 1999 paper reported that such behavior was more common among the children having overnight visits, and of course this is concerning. However, work like that of Mary Dozier has shown that this behavior is associated with the caregiver’s behaving in a frightened or frightening manner toward the child, and interventions like Dozier’s Attachment and Biobehavioral Catch-up can change the adult behavior and make the child’s attachment strategies more ordered and less disorganized. Unless it can be shown that indeed overnight visits cause adults to behave in frightened or frightening ways (which is not impossible), it is hard to see how shared parenting is directly causal for the child’s disorganized attachment quality.

4. Results now, or later? As has been noted (e.g. by van IJzendoorn et al, 2005), immediate changes in disorganized attachment behavior can be considered a good measure of the effect of events on a young child. However, it is equally or more important to look at long-term outcomes before concluding that an overnight visit is harmful, or an intervention is helpful. Only a well-designed longitudinal study can show the real impact of overnight visits on development. Infants and toddlers, from about 8 to about 24 months, are well-known for their negative response to changes of any kind. Developing autonomy (“No!”) and neophobia are enough to cause young children to resist change and to display stress reactions when more than a minimum number of changes occurs; even ordinary transitions like getting dressed to go outdoors can be fraught with emotion. It would be surprising if children in this age group did not respond negatively to visit experiences, especially if they pick up cues from the adults that anyone is worried or frightened about what is happening. The question is, how do these experiences alter the trajectory of development and lead to more, or to less, desirable outcomes--  if indeed they do?    

4. Gender differences. As this post is getting very long, I am going to be brief on a point that may be of interest. As it happens, girls are more likely than boys to be assessed as securely attached during the toddler period, and girls may be less vulnerable to developmental disruption due to environmental stress.

Men and women may also bring different relevant characteristics to shared parenting. Women with perinatal mood disorders are more likely to have children whose development is problematic, and it is conceivable that mood disorders can disturb a marital relationship to the point of divorce.

Men, on the other hand, may be more likely to bring antisocial behavior into the picture. In  a study that is only indirectly related to the issue of overnight visits (Jaffee, S.R.Moffitt, T.E.,  Caspi, A., & Taylor, A. [2003]. Life with [or without] father: The benefits of living with two biological parents depend on the father’s antisocial behavior. Child Development, 74, 109-126), the authors showed that the children they studied experienced what they called a “double whammy” of behavioral problems when they received both the father’s genetic material and experienced living with him; children with the genetic tendency to antisocial behavior did better when they did not live with antisocial fathers. Antisocial behavior on the father’s part may also disturb the marriage and lead to divorce; the issue may then be to what extent any contact with an antisocial father is beneficial.

All I have done here is to bring up various factors and research issues that I speculate may be related to the effects of overnight visits for infants and toddlers (although they are probably much less relevant for preschool children). These points make me hesitate to think in terms of any general guidelines about overnights--  although I am well aware that some guidelines will be applied in most cases, however questionable they may be. I would like to suggest that rather than offering guidelines for parenting plans, it would be more constructive to develop a protocol for assessing how well a given plan is going for a child, and for fine-tuning plans that are proving problematic--  for example, a plan for visits that is frequently disrupted by scheduling changes or by the influences of non-parent adults.