1. Why is the 'attachment' theory so important to family relationships and therefore to custody and access decisions about children?

2. How do children form 'attachment' bonds to their parents?

3. If the courts describe my relationship as 'insecure' does this mean my children will suffer or that I am a bad parent?

4. Why is there so much information on the internet and in books on the infant-mother relationship?

5. Does this mean that other carers, such as fathers, are unimportant to children?

6. Do mothers bond or 'attach' themselves to their children in a different way to fathers?

7. What are the ways in which mothers and fathers establish family relationships with their children?

8. If the relationship with the child is disrupted should the courts terminate access and custody or is it better for the child to carry on trying to maintain the family relationship or 'attachment'?

9. Do children suffer, between 15 and 24 months of age, when they appear to resist transitions from their mothers' houses to their fathers' after marital separation or is this common? 

10. What evidence is there that having multiple attachments with lots of different people diminishes the strength of attachments to the primary attachment figure or figures in the first 2 years of life?

11. What is the 'Strange Situation' procedure?

12. Are so called 'secure' and 'insecure' attachments fixed and stable over time?

13. When the parents have lived together prior to separation what sort of infant-parent attachments or relationships should the courts encourage after separation?

14. When parents have never lived together, and the infant has had no opportunity to become attached to one of the parents, what sort of access and custody arrangements are in the best interests for the welfare of  the child?

15. What are the consequences for the welfare of the child, if he or she, is deprived of a meaningful access or contact with one of their parents by the courts?

16. Does common practice in custody and access disputes reflect this evidence?

17. Why doesn't 'best practice' in custody and access disputes always follow what is in the best interests of the child?

18. Are there any other effects on the child's welfare of the courts and other welfare professionals focusing exclusively on mothers with their infants and presuming fathers to be quite peripheral and unnecessary to children's development and psychological adjustment?

19. What is the ideal 'parenting schedule' the courts should be encouraging for very young children? 

20. Surely a child can only sleep in one bed. Therefore the child can only have one home and the courts are right to deny parents 'shared responsibility' for their children, for example, with Shared Residence Orders (SRO)?

21. Why do the courts usually oppose 'overnight stays'?

22. What does the research on child development say about the benefits of 'overnight stays' for young children?

23. Everybody knows the benefits of breast feeding. Surely this must be a period when access and custody decisions must be less important?

24. What arrangements should be made by the courts when one parent is uncooperative?

25. Different 'parenting styles' often proves to be a major sticking point when the court has to make decisions about access and contact. What does the research suggest?

26. Often judges talk about giving the child a 'voice' in the decision making process about access and contact decisions. What does the research suggest about this?

Decisions regarding custody and access are most often made without reference to the research on child development, although this literature can be useful in conceptualizing children's needs after separation and divorce. Research on attachment processes, separation from attachment figures, and the roles of mothers and fathers in promoting psychosocial adjustment are reviewed in this article. It concludes with a discussion of the implications for young children's parenting schedules.

Powerful influences shape decisions about custody and access arrangements when parents are separating or divorcing. Regardless of whether parents make their decisions independently or rely on therapists, custody evaluators, or judges for recommendations and decisions, statutory, historical, and cultural forces often determine which care arrangements are deemed to be in the children's best interests (Kelly, 1994). Unfortunately, however, decision makers in family law and mental health fields remain largely ignorant about several decades of research on child development. Child development researchers and child custody decision makers rarely cross paths, and most of the relevant publications intended for academic audiences are inaccessible to casual readers.

In this article, we discuss research that directly helps conceptualize custody and access issues that need to be addressed when parents separate. Because so many questions arise regarding appropriate postseparation arrangements for infants and young children, the focus will be on attachment processes, separation from attachment figures, and the roles of mothers and fathers in promoting children's development. To facilitate readability, we primarily cite review articles; readers can study the cited articles for references to the primary literature.

RESEARCH ON ATTACHMENT PROCESSES

Over the past four decades, our understanding of early social and emotional development has improved enormously. In particular, psychologists have identified many of the factors that influence the formation of attachment relationships between infants and their parents, as well as the adverse effects on children of disrupted and distorted parent-child relationships (Lamb, Bornstein, & Teti, in press; Lamb, Thompson, Gardner, & Charnov, 1995; Thompson, 1998). The essence of our emergent understanding of this phenomena is briefly summarized in the following pages.

The development of attachments to parents and other important caregivers constitutes one of the most critical achievements of the first year of life. These enduring ties play essential formative roles in later social and emotional functioning. Infant-parent attachments promote a sense of security, the beginnings of self-confidence, and the development of trust in other human beings. Concerned with the profoundly negative impact on children's development of prolonged separation from parents, Bowlby (1969) first proposed a theoretical explanation for the importance of continuity in relationships, drawing on psychoanalytic and ethological theory. Subsequent decades of research have focused on the phases and types of attachment: the security of attachments, the stability of attachments over time, the contributions of infants and caregivers to the quality or security of attachments, cultural differences in attachment outcomes, and later personality and cognitive characteristics associated with different types of attachment.

Researchers initially focused exclusively on infant-mother attachment, and that literature is best known in the mental health community.In the past 20 years, however, the meaning and importance of infant-father attachments and of attachments to nonfamily caregivers in day care and preschool settings have been studied extensively as well (for detailed reviews, see Lamb, 1997a, 1998; Thompson, 1998).

PHASES OF ATTACHMENT FORMATION

Attachment formation involves reciprocal interactive processes that foster the infant's growing discrimination of parents or caregivers, as well as the emotional investment in these caregivers. Infants who receive sensitive and responsive care from familiar adults in the course of feeding, holding, talking, playing, soothing, and general proximity become securely attached to them (Thompson, 1998). Even adequate levels of responsive parenting foster the formation of infant-parent attachments, although some of these relationships may be insecure. Children are nonetheless better off with insecure attachments than they are without attachment relationships at all.

Bowlby ( 1969) described four phases of the attachment process, and subsequent research has largely confirmed this delineation: (a) indiscriminate social responsiveness, (b) discriminating sociability, (c) attachment, and (d) goal-corrected partnerships.

Indiscriminate Social Responsiveness

During this phase, which occurs between birth and 2 months, the infant uses an innate repertoire of signals to bring caregivers to him or her, including crying and smiling. The child begins to associate the caregivers with relief of distress (from hunger or pain). Furthermore, adults' vocalizations and animated facial expressions create additional opportunities for social interaction. Although infants are able to recognize their parents by voice or smell within the first weeks of life, they accept care from any caregiver during this phase without distress or anxiety (Lamb et al., in press).

Discriminating Sociability

Discriminating sociability occurs between 2 and 7 months of age. Here the infants begin to recognize certain caregivers and prefer interaction with them. Infants thus coo and soothe more readily in response to these familiar figures, orient their posture toward them, and show more pleasure when interacting with them. This attachment-in-the-making indicates that the caregivers' responses are sufficiently prompt and appropriate. During this phase, infants begin to learn reciprocity, a sense of effectiveness ("I can make things happen"), and trust. They generally do not protest when separated from their parents during this phase, but they become anxious if separated from humans for too long.

Attachment

In the attachment phase, which occurs between 7 and 24 months of age, the child, by actively seeking to remain near to preferred caregivers, gives increasingly clear evidence that attachments have been formed. Behaviors demonstrating attachment include differential following and clinging to parents, especially when tired or sick, and preferences for specific caretakers as secure bases for exploration of the environment.

Somewhere around the middle of the Ist year of life, infants begin to cry or protest when separated from their attachment figures. This transition marks the initial attainment of the ability to recognize that parents continue to exist when they are not present, an ability referred to by Piaget as object constancy. Of course, the understanding of this fundamental concept is quite rudimentary at first and continues to mature in the next year and a half of the child's life. As this comprehension matures, the child's ability to tolerate separation from humans grows, although separation does remain stressful for young children. Infants clearly cope better with separation from one attachment figure when they are with another attachment figure. Nevertheless, it is important to minimize the length of time that infants are separated from their attachment figures; extended separations unduly stress developing attachment relationships. If they are attached to both parents, as most infants are, this means that the length of time with each parent needs to be adjusted to minimize the length of time away from the other parent.

Considerable evidence now exists (for a review, see Lamb, 1997a) that documents that most infants form meaningful attachments to both of their parents at roughly the same age (b to 7 months). This is true even though many fathers in our culture spend less time with their infants than mothers do. This indicates that time spent interacting is not the only factor in the development of attachments, although some threshold of interaction is crucial. Most infants come to "prefer" the parent who takes primary responsibility for their care (typically their mothers), but this does not mean that relationships with the other parent are unimportant. The preference for the primary caretaker appears to diminish with age, and by 18 months, this preference often has disappeared.

In general, the ways in which mothers and fathers establish relationships with and influence their children's development is quite similar. Although much has been made of research showing that mothers and fathers have distinctive styles of interaction with their infants, the differences are actually quite small and do not appear to be formatively significant (Lamb, 1997a). The benefits of maintaining contact with both parents exceed any special need for relationships with male or female parents.The empirical literature also shows that infants and toddlers need regular interaction with both of their parents to foster and maintain their attachments (Lamb et al., in press). Extended separations from either parent are undesirable because they unduly stress developing attachment relationships. In addition, it is necessary for the interactions with both parents to occur in a variety of contexts (feeding, playing, diapering, soothing, putting to bed, etc.) to ensure that the relationships are consolidated and strengthened. In the absence of such opportunities for regular interaction across a broad range of contexts, infant-parent relationships fail to develop and may instead weaken.  It is extremely difficult to re-establish relationships between infants or young children and their parents when the relationships have been disrupted. Instead, it is considerably better for all concerned to avoid such disruptions in the first place.

During this phase, children become more mobile, increase their explorations of the world, initiate more social interactions, and develop more extensive and sophisticated linguistic and cognitive abilities. These achievements increase the child's anxiety about separation from important caregivers, and this anxiety is reflected in vigorous vocal and behavioral displays of resistance to separation, especially until approximately 18 months. Thus,  it is common for children between 15 and 24 months of age to resist transitions from their mothers' houses to their fathers' after marital separation, even when children have good attachment relationships with both parents. However, once removed from their mothers' environments, these youngsters function well with their fathers, and vice versa. If planned separations are announced shortly in advance in a calm, matter-of-fact way, with reassurance that the parent (or child) will return, anxiety can be reduced. By 24 months, the majority of children no longer experience severe separation anxiety, although children with very insecure attachments and those whose primary attachment figures have their own separation difficulties may continue to express anxiety.

Goal-Corrected Partnerships

Finally, the goal-corrected partnership phase occurs between 24 and 36 months of age. It involves children's and parents' beginning to plan jointly; children are increasingly able to compromise and to take their parents' needs into some account. Children can now understand to some extent why parents come and go, and they can predict their return. However, children's primitive sense of time continues to make it difficult for 2-year-olds to comprehend much beyond today or tomorrow, and this has implications for the tolerable duration of separation from important attachment figures.In sum, when given the opportunity, infants form multiple attachments, each with unique emotional meaning and importance. Physical caregiving is critical to survival and health, but social and emotional input from diverse attachment figures is important as well. Children with multiple attachments appear to create a hierarchy of caregivers, seeking out the particular caregivers that suit their needs and moods, although they tend to accept any important attachment figure for comfort and soothing when distressed or anxious in the absence of more preferred caregivers. There is no evidence, however, that having multiple attachments diminishes the strength of attachments to the primary attachment figure or figures in the first 2 years of life.

INDIVIDUAL DIFFERENCES IN THE SECURITY OF ATTACHMENT

Extensive research into controlled separations from and reunions with parents (using  the Strange Situation procedure) has supported the classification of attachment into secure and insecure types. Insecure attachments are further classified into avoidant, resistant, and disorganized types (Ainsworth, Belhar, Waters, & Wall, 1978; Lamb et al., 1985, in press; Thompson, 1998). Babies with secure attachments prefer parents over strangers, may cry at separation, and immediately seek interaction or contact with and reassurance from parents when they return. About two thirds of middle-class American infants are securely attached, presumably because their parents are responsive to infant cries and distress and are psychologically available.About 20% of infant-parent attachments in middle-class American homes are insecure avoidant. These babies seem not to notice when separated, avoid greeting the returning parents in the assessment procedure, but do not resist physical contact. Babies with insecure resistant attachments (10% to 12%) show angry, aggressive behaviors upon reunion and are not easily comforted by their parents after separation. A small number (about 5%) of babies display confused behaviors after separation and have been classified as disorganized/disoriented. Their contradictory behaviors upon reunion include gazing away while being held, odd postures, and dazed facial expressions.

Although secure and insecure attachments were once thought to be fixed and stable over time, this appears to be true only when the infants experience reasonably stable family conditions over the course of the first 2 years (Lamb et al., in press; Thompson,1998). Factors known to influence the security and stability of attachments include poverty; marital violence and high conflict between parents; and major life changes such as divorce, death, or the birth of a sibling, which in each instance are associated with more insecure attachments. Insecure attachments are significantly linked to poor styles of parenting that affect the quality of the child's attachment, such as disturbed family interactions, parental rejection, inattentive or disorganized parenting, neglect, and abuse.

It should be noted that infant-parent attachments often become insecure in response to the parents' separation or divorce, at least for a period of time, and infants who experience a reduction in parental discord become more securely attached over time (Cummings & Davies, 1994). Thus, although infants from very high conflict parental relationships may initially have insecure attachments, their relationships with both parents may become more secure if the level of conflict between the parents declines. It is also clear that crosscultural differences in parenting styles and expectations are associated with different patterns of attachment.

Individual differences in the security of attachment are important. Compared to children who were initially insecure, securely attached children later are more independent, socially competent, inquisitive, and cooperative and empathic with peers; have higher self esteem; and demonstrate more persistence and flexibility on problem-solving tasks. These differences seem to reflect not only the initial differences in attachment security but also continued differences in the quality of parenting experienced (for reviews and analyses of these issues, see Lamb et al., 1985, in press; Thompson, 1998).

IMPLICATIONS OF ATTACHMENT RESEARCH FOR CUSTODY AND ACCESS ARRANGEMENTS

MAINTAINING CHILDREN'S ATTACHMENTS AFTER SEPARATION OR DIVORCE

If the parents lived together prior to separation, and the relationships with both parents were at least of adequate quality and supportiveness, the central challenge is to maintain both infant-parent attachments after separation. When there are concerns about child maltreatment, substance abuse, mental illness, or interparental violence, of course, evaluations of parental adequacy are essential, and supervised or restricted visiting may be required to avoid compromising the child's safety or development. Furthermore,  when parents have never lived together, and the infant has had no opportunity to become attached to one of the parents, as is common while paternity is being established legally, special efforts are needed to foster the development of attachment relationships. These issues are beyond the scope of this article, however.In general, relationships with parents play a crucial role in shaping children's social,emotional, personal, and cognitive development, and there is a substantial literature documenting the adverse effects of disrupted parent-child relationships on children's development and adjustment (Lamb, 1999; Lamb, Hwang, Ketterlinus, & Fracasso, 1999).  The evidence further shows that children who are deprived of meaningful relationships with one of their parents are at greater risk psychosocially, even when they are able to maintain relationships with the other of their parents. Stated differently, there is substantial evidence that children are more likely to attain their psychological potential when they are able to develop and maintain meaningful relationships with both of their parents, whether the two parents live together or not. The most common practice in custody and access decisions has been to emphasize and preserve continuity in the infant-mother relationship, with children living with their mothers and having limited contact with their fathers. Thus, the infant or toddler who was accustomed to seeing both parents each day abruptly began seeing one parent, usually the father, only once a week (or once every 2 weeks) for a few hours.  This arrangement was often represented by professionals as being in the best interests of the child due to the mistaken understanding, based on Bowlby's earliest speculations, that infants had only one significant or primary attachment. As a result, early child development research followed untested psychoanalytic theory in focusing exclusively on mothers and infants, presuming fathers to be quite peripheral and unnecessary to children's development and psychological adjustment.  The resulting custody arrangements sacrificed continuity in infant-father relationships, with long-term socioemotional and economic consequences for children. Very large research literatures now document the adverse effects of severed father-child relationships as well as the positive contributions that fathers make to their children's development (for reviews, see Lamb, 1997b).

The research reviewed by Bowlby ( 1973) indicated that the loss or attenuation of significant relationships in childhood can cause anxiety and a profound sense of loss, particularly in the first 2 years, when children have limited cognitive and communicative resources to help cope with loss. Both marital conflict and the abrupt departure of one parent from the child's daily life may foster insecurity in the child's attachments and should thus be avoided.

To be responsive to the infant's psychological needs, the parenting schedules adopted for children younger than 2 or 3 must involve more transitions, rather than fewer, to ensure the continuity of both relationships and the child's security and comfort during a time of great change. The ideal situation is one in which infants and toddlers have opportunities to interact with both parents every day or every other day in a variety of functional contexts (feeding, play, discipline, basic care, limit setting, putting to bed, etc.). To minimize the deleterious impact of extended separations from either parent, there should be more frequent transitions than would perhaps be desirable with older children. As children reach age 2, their ability to tolerate longer separations increases, so most toddlers can manage 2 consecutive overnights with each parent without stress. Schedules involving alternating longer blocks of time, such as 5 to 7 days, should be avoided, as children this age still become fretful and uncomfortable when separated from either parent too long.

There is ample evidence that infants and toddlers get used to regular transitions, such as those associated with enrollment in alternative care facilities, without there being adverse effects on the quality of the attachments to their parents (Lamb, 1998). The same should be true of separations in the context of parental separation or divorce. Infants and toddlers should thus have multiple contacts each week with both parents to minimize separation anxiety and maintain continuity in the children's attachments. Unfortunately,  the concept of location-engendered stability (one home, one bed) has been incorrectly overemphasized for infants and toddlers, without due consideration for the greater significance to the child of the emotional, social, and cognitive contributions of both parent-child relationships. Living in one location (geographic stability) ensures only one type of stability. Stability is also created for infants (and older children) by the predictable comings and goings of both parents, regular feeding and sleeping schedules, consistent and appropriate care, and affection and acceptance (Kelly, 1997). Furthermore, postseparation access or contact schedules that are predictable and that can be managed without stress or distress by infants or toddlers provide stability after separation.

OVERNIGHTS WITH THE NONRESIDENTIAL PARENT

With the historic focus on preserving the mother-infant attachment while establishing an exclusive home, overnights or extended visits with the other parent (mostly the father) were long forbidden or strongly discouraged by judges, custody evaluators, therapists, mental health professionals, family law attorneys, and not surprisingly, many mothers (e.g., Garrity & Baris, 1992; Goldstein, Freud, & Solnit, 1973; Goldstein, Freud, Solnit, & Goldstein, 1986; Hodges, 1991). Hodges (1991), for example, stated that for infants younger than 6 months, "overnight visits are not likely to be in the child's best interests, because infants' eating and sleeping arrangements should be as stable as possible" (p. 175). For infants 6 to 18 months of age, overnight visits "should be considered less than desirable" (p. 176). Although Hodges noted the importance of several visits per week for older infants who were attached to fathers, he recommends that these be limited to several hours. Hodges stated that children might be able to spend overnights "without harm" only after reaching 3 years of age (p. 177).

Such unnecessarily restrictive and prescriptive guidelines were not based on child development research and, thus, reflected an outdated view of parent-child relationships. Furthermore, such recommendations did not take into account the quality of the father-child or mother-child relationship, the nature of both parents' involvement, or the child's need to maintain and strengthen relationships with both parents after separation (Lamb, Sternberg, & Thompson, 1997). Research and experience with infant day care, early preschool, and other stable caretaking arrangements indicate that infants and toddlers readily adapt to such transitions and also sleep well, once familiarized. Indeed, a child also thrives socially, emotionally, and cognitively if the caretaking arrangements are predictable and if parents are both sensitive to the child's physical and developmental needs and emotionally available (Homer & Guyer, 1993; Lamb, 1998).

The evening and overnight periods (like extended days with nap times) with nonresidential parents are especially important psychologically not only for infants but for toddlers and young children as well. Evening and overnight periods provide opportunities for crucial social interactions and nurturing activities, including bathing, soothing hurts and anxieties, bedtime rituals, comforting in the middle of the night, and the reassurance and security of snuggling in the morning after awakening, that 1- to 2-hour visits cannot provide. These everyday activities promote and maintain trust and confidence in the parents while deepening and strengthening child-parent attachments.

There is absolutely no evidence that children's psychological adjustment or the relationships between children and their parents are harmed when children spend overnight periods with their other parents. An often mis-cited study by Solomon (1997) reported high levels of insecure infant-mother and infant-father attachment when parents lived apart, although toddlers who spent overnights with both their fathers and mothers were not significantly more likely to have insecure relationships than those children who did not have overnight visits with both parents.Indeed, as articulated above, there is substantial evidence regarding the benefits of these regular experiences. Aside from maintaining and deepening attachments, overnights provide children with a diversity of social, emotional, and cognitively stimulating experiences that promote adaptability and healthy development. In addition, meaningful father-child relationships may encourage fathers to remain involved in their children's lives by making them feel enfranchised as parents. Other advantages of overnights are the normal combination of leisure and "real" time that extended parenting affords, the ability to stay abreast of the constant and complex changes in the child's development, opportunities for effective discipline and teaching that are central to good parenting, and opportunities to reconnect with the child in a meaningful way. In contrast, brief, 2-hour visits remind infants that the visiting parents exist but do not provide the broad array of parenting activities that anchor the relationships in their minds.

When mothers are breast-feeding, there is considerable hesitation, indecision, and perhaps strong maternal resistance regarding extended overnight or full-day separations. Breast-feeding is obviously one of the important contexts in which attachments are promoted, although it is by no means an essential context. Indeed, there is no evidence that breast-fed babies form closer or more secure relationships to their parents than do bottle-fed babies. A father can feed an infant with the mother's expressed milk, particularly after nursing routines are well established.When there are overnights, it is not crucial that the two residential beds or environments be the same, as infants adapt quickly to these differences. It may be more important that feeding and sleep routines be similar in each household to ensure stability. Thus, parents should share information about bed times and rituals, night awakenings, food preferences and feeding schedules, effective practices for soothing, illnesses, and changes in routine as the child matures. Parents should be encouraged by attorneys or mediators to communicate directly, either verbally or in writing. If this is not possible due to the intransigence of either or both parents, then the court should order the involvement of co-parenting consultants, special masters, or custody mediators until the normal angers of divorce subside (Emery, 1994, 1999; Kelly, 1991, 1994). It is important as well to recognize that protracted litigation and the specter of winning or losing delay the decline of conflict (Maccoby & Mnookin, 1992), and thus, such disputes should be resolved with speed. Furthermore, communication quality should not be judged from the level of conflict surrounding and encouraged by the litigation.

The challenges of child-focused communication require commitment on the parents' part to their children's well-being but will have long-term positive consequences for children and for each of the parent-child relationships. Although it is clear that a cooperative relationship between parents is beneficial, parenting schedules that promote meaningful child-parent relationships should not be restricted after separation if one or both parents are not able to cooperate. Disengaged parents may function effectively in their parallel domains and, in so doing, enhance their children's adjustment (Lamb et al., 1997; Maccoby & Mnookin, 1992; Whiteside, 1998).Because high conflict is associated with poorer child outcomes following divorce (Johnston, 1994; Kelly, in press; Maccoby & Mnookin, 1992), it is preferable that transitions be accomplished without overt conflict. However, it is important to understand how high conflict is conceptualized in the relevant research, as the findings are often misunderstood. Almost by definition, of course, custody and access disputes involve conflict, but it is clear that such conflict in and of itself is not necessarily harmful. The high conflict found harmful by researchers such as Johnston (1994) typically involved repeated incidents of spousal violence and verbal aggression continued at intense levels for extended periods of time and often in front of the children. Johnston emphasized the importance of continued relationships with both parents except in those relatively uncommon circumstances in which intense, protracted conflict occurs.

High conflict at the time of transition may heighten children's anxiety about separation. Even without conflict, transitions can cause unsettled behavior, fretting, and crying as children move from one set of routines or one parental style to another. As noted above, this is especially true of children 15 to 24 months of age, when it is quite normal. If conflict is difficult to avoid because of one or both parents' hostility, then transitions should be implemented by babysitters or should take place at neutral places such as day care centers, special visiting centers set up for this purpose, or supportive grandparents' homes.

Occasionally, mothers are very hostile to fathers after separation as part of a legal strategy to prevent or diminish the fathers' participation in child rearing and co-parenting. In such instances, fathers should not be denied adequate contact with their children because conflict between the parents exists. Similarly, when fathers berate mothers at transitions or refuse to communicate about the infants' behaviors when with them, they will need to demonstrate more cooperative attitudes to warrant more extended contact.

It should be assumed that parents would have somewhat different parenting styles, which are related to their own upbringing and personalities. Regardless of these differences, children (and parents) benefit from discussions of disciplinary techniques and approaches as well as about the achievement of major developmental tasks such as toilet training. Furthermore, children will typically have different social experiences (and holiday rituals) with each parent and with extended families and friends.

HOW MUCH SEPARATION FROM PRIMARY ATTACHMENT FIGURES IS APPROPRIATE?

The extent to which infants and toddlers can tolerate separation from significant attachment figures is related to their age, temperament, cognitive development, social experience, and the presence of older siblings. Aside from their very immature cognitive capacities, infants have no sense of time to help them understand separations, although their ability to tolerate longer separations from attachment figures increases with age. The goal of any access schedule should be to avoid long separations from both parents to minimize separation anxiety and to have sufficiently frequent and broad contact with each parent to keep the infant secure, trusting, and comfortable in each relationship.

Preschool children can tolerate lengthier separations than toddlers can, and many are comfortable with extended weekends in each parent's home as well as overnights during the week. In general, however, most preschool children become stressed and unnecessarily overburdened by separations from either parent that last more than 3 or 4 days. The exception might be planned vacations, in which parents and siblings are fully available to engage preschool children in novel, stimulating, and pleasurable activities. Even so, most parents would be advised to limit vacations at this age to 7 days and to schedule several vacations rather than one single lengthy vacation.When children reach school age, they have significantly more autonomy and greatly increased cognitive, emotional, and time-keeping abilities, so the duration of separations from both parents becomes less critical. Even so, before the age of 7, and often thereafter, most youngsters still enjoy reunions during the week with each parent rather than extended periods without contact. By 7 or 8 years of age, most youngsters can manage 5- to 7-day separations from parents as part of their regular schedules and 2-week vacations with each parent. Court orders for young children that reflect children's increasing ability to tolerate lengthier separations by building age-based and stepwise increases into vacation schedules are most responsive to children's best interests.

Many discussions of custody decisions have emphasized the need to identify a primary caretaker when attempting to determine where children should spend most of their residential time (for a review, see Kelly, 1994). The expanded world of young school-age children, the greater richness of children's emotional and cognitive abilities, and the increasing importance of children's social and recreational life outside the home lead many to conclude that the concept of the primary caretaker should play little role in determining custody, however, particularly after the age of 5 (Chambers, 1984. As noted throughout this article, children are enriched by regular, diverse, and appropriate interactions with two involved and emotionally supportive parents, and this is no less true of school-age children as they journey toward adolescence. Regardless of who has been the primary caretaker, therefore, children benefit from the extensive contact with both parents that fosters meaningful father-child and mother-child relationships.

REFERENCES

Ainsworth, M.D., Blehar, M., Waters, E., & Wall, S. (1978). Patterns of attachment. Hillsdale, NJ: Lawrence Erlbaum.
Bowlby, J. (1969). Attachment and loss: VoL 1. Attachment. New York: Basic Books.
Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation: Anxiety and anger. New York: Basic Books.
Bowlby, J. (1988). A secure base: Parent-child attachment and healthy human development. New York: BasicBooks.
Chambers, D. L. (1984). Rethinking the substantive rules for custody disputes in divorce. Michigan Law Review, 83, 477-569.
Cummings, E., & Davies, P ( 1994). Children and marital conflict: The impact offamily dispute and resolution. New York: Guilford.
Emery, R. E. (1994). Renegotiating family relationships: Divorce, child custody, and mediation. New York: Guilford.
Emery, R. E. (1999). Marriage, divorce, and children's adjustment (2nd ed.). Thousand Oaks, CA: Sage.
Garrity, C. B., & Boris, M. A. (1992). Caught in the middle: Protecting the children of high conflict divorce. San Francisco: Jossey-Bass.
Goldstein, J., Freud, A., & Solnit, A. J. (1973). Beyond the best interests of the child. New York: Free Press.
Goldstein, J., Freud, A., Solnit, A.1., & Goldstein, S. (1986). In the best interests of the child. New York: Free Press.
Hodges, W. (1991). Interventions for children of divorce: Custody, access, and psychotherapy. New York: John Wiley.
Homer, T. M., & Guyer, M. J. (1993). Infant placement and custody. In C. Zeanah (Ed.), Handbook of infant mental health (pp. 462-479). New York: Guilford.
Johnston, J. (1994). High conflict divorce. The Future of Children: Children and Divorce, 4(l), 165-182.
Kelly, J. B. (1991). Parent interaction after divorce: Comparison of mediated and adversarial divorce processes. Behavioral Sciences and Law, 9, 387-398.
Kelly, J. B. (1994). The determination of child custody. The Future of Children: Children and Divorce,4(1), 121-142.
Kelly, J. B. (1997). The best interests of the child: A concept in search of meaning. Family and Conciliation Courts Review, 35(4), 377-387.
Kelly, J. B. (in press). Children's adjustment in conflicted marriage and divorce: A decade review of research. Journal of Child and Adolescent Psychiatry.
Lamb, M. E. ( 1997a). The development of infant-father attachments. In M. E. Lamb (Ed.), The role of the father in child development (3rd ed., pp. 104-120, 332-342). New York: John Wiley.
Lamb, M. E. (Ed.) (1997b). The role of the father in child development (3rd. ed.). New York: John Wiley.
Lamb, M. E. (1998). Nonparental childcare: Context, quality, correlates, and consequences. In W. Damon, I. E. Sigel, & K. A. Renninger (Eds.), Handbook of child development: Vol. 4. Social, emotional, and personality development (5th ed., pp, 73-133). New York: John Wiley.
Lamb, M. E. (1999). Non-custodial fathers and their impact on the children of divorce. In R. A. Thompson & P. Amato (Eds.), The post-divorce family: Research and policy issues (pp.105125). Thousand Oaks, CA: Sage.
Lamb, M. E., Bornstein, M. H., & Teti, D. M. (in press). Development in infancy (4th ed.). Mahway, NJ: Lawrence Erlbaum.
Lamb, M. E., Hwang, C. P" Ketterlinus, R., & Fracasso, M. P ( 1999). Parent-child relationships. In M. H. Bornstein & M. E. Lamb (Eds.), Developmental psychology: An advanced textbook (4th ed., pp. 411-450). Mahwah, NJ:
Lawrence Erlbaum.
Lamb, M. E" Sternberg, K., & Thompson, R. A. (1997). The effects of divorce and custody arrangements on children's behavior, development, and adjustment. Family and Conciliation Courts Review, 35, 393-404.
Lamb, M. E., Thompson, R. A., Gardner, W. P, & Chamov, E. L. (1985). Infant-mother attachment. Hillsdale, NJ: Lawrence Erlbaum.
Maccoby, E., & Mnookin, R. (1992). Dividing the child. Cambridge, MA: Harvard University ' Press.
Solomon, J. (1997). Parenting schedules for the very young child: Summary of a longitudinal study on the development of attachment in separated and divorced families. Unpublished manuscript, Wallerstein Center for the Family in Transition, Corte Madera, CA.
Thompson, R. A. (1998). Early sociopersonality development. In W, Damon & N. Eisenberg (Eds.), Handbook of child development: Vol. 3. Social, emotional, and personality development (5th ed., pp. 25-104). New York: John Wiley.
Whiteside, M. F ( 1998), The parental alliance following divorce: An overview. Journal of Marital and Family Therapy, 24(1), 3-24.