U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Committee for the Study of Health Consequences of the Stress of Bereavement; Osterweis M, Solomon F, Green M, editors. Bereavement: Reactions, Consequences, and Care. Washington (DC): National Academies Press (US); 1984.

Cover of Bereavement

Bereavement: Reactions, Consequences, and Care.

Show details

CHAPTER 7Monkeys' Responses to Separation and Loss

This six-month-old monkey who has been separated from its mother exhibits many signs of depression in its posture and behavior


This six-month-old monkey who has been separated from its mother exhibits many signs of depression in its posture and behavior. Because of the many parallels with human responses, researchers are studying nonhuman primates to better understand bereavement (more...)

An infant, having reached an age when independent locomotion is readily accomplished, wanders around a room, eagerly touching and exploring everything he sees. He plays vigorously with two young companions. But when he suddenly discovers that his mother is nowhere in sight, he quickly becomes agitated, moves about the room in rapid, distracted movements, and begins to scream and cry. He no longer approaches the objects in the room, nor does he initiate play with his friends. Indeed, efforts by his friends to initiate play with him result in brief distracted encounters, quickly terminated by the repeated crying and searching of the motherless infant.

Over the next few hours the infant overtly calms somewhat, but any of a variety of stimuli sets off the whole train of emotional responses once again. The time to go to sleep, for example, triggers a striking increase in crying and agitation. In the next day or two, the pattern shifts from one of agitation to one in which the infant gradually withdraws from his environment almost completely, directing much of his activity towards himself. Thumb-sucking, genital manipulation, and self-clasping emerge, as do changes in posture and facial expression. The infant seems lethargic and unresponsive, almost unable to hold his body nor mally upright; his face appears drawn, and he frequently closes his eyes in apparent, but often not true, sleep. This depressed pattern continues, relatively unabated, for a dishearteningly long time.

The pattern described is not of a human youngster's response to loss, but the reaction of a six-month-old monkey. This description reflects a number of aspects of a phenomenon observed in human beings throughout history and studied in detail in other primates during the past two decades. In both human and nonhuman primates, when a strong emotional bond has been established between two individuals, loss of one has important psychological and emotional consequences for the other. Although researchers have been able to learn much about the factors that influence the course and intensity of bereavement in people, the constraints imposed by a number of ethical and practical matters leave current understanding of the role of many factors relatively uncertain. Working with appropriate animal models, however, while bearing in mind the ethical issues that must still be considered, the environment and experiences can be controlled and the impact of important events can be assessed at preselected times under the rigorous light of experimental scrutiny.


Like people, nonhuman primates are genetically heterogeneous and intelligent enough to be influenced in complex ways by their past experiences and current circumstances. Thus, it is not surprising that their responses to the loss of the mother range across a wide spectrum of behavior. Indeed, a special relevance of the work on nonhuman primates in this domain derives from the fact that the higher primates show a diversity of response, as humans do. By pursuing the sources of this variation, and the forces that intensify or ameliorate the potentially debilitating effects of loss, understanding of the human condition can be advanced.

In monkeys, individual variation in response to the loss of the mother early in life can range from merely several hours of intermittent crying and restlessness to a pattern in which the infant cries and moves incessantly immediately after experiencing the loss and then virtually collapses into an unresponsive heap the next day. Some infants show very strong reactions in the first two to four days but appear to begin recovery quickly; others may still be quite disturbed weeks after the separation. It should be noted, however, that the absence or lessening of behavioral signs of disturbance does not necessarily imply that stressful reactions may not be emerging within several physiologic domains. 4

Over the last 20 years, the responses of more than a half dozen species of primates to the sudden loss of the mother or other close companions have been studied. The bulk of this work has focused on several Asian macaque species and on the South American squirrel monkey. This range of species has been particularly important because the combination of closely related species and a phylogenetically quite distant one allows an assessment of both the effects of a number of social and environmental conditions and the role of clearly primate, but nonetheless quite diverse, genetic predispositions.

These studies make it clear that although each of these species shows emotional disturbance following a loss, the intensity and duration of the reaction as well as its qualitative form may vary from species to species. The rhesus and pigtail macaques, for example, appear most susceptible to the severest forms of depressive reaction in response to loss of the mother.

One final element in consideration of the potential genetic source of variations in some aspects of the response to separation or loss is the role of individual genetic, or at least prenatal, influences. Recent work by Suomi 43,44 has shown that an infant rhesus' emotional reactivity to an auditory conditioning procedure at three weeks of age significantly predicts subsequent intensity of response to maternal loss. Also, half-siblings (infants with the same father) seem to resemble each other in the intensity of their reactions.

It seems reasonable to conclude from these data that although all primates are emotionally responsive to the experience of loss, there are important congenital contributions (genetic, prenatal, or both) to the form and intensity of that response. This ''background" source of variation should be kept in mind when considering the developmental, social, and environmental factors being considered in research results.


In general, studies of the response to sudden loss in primates have identified several successive phases through which the infants pass. Before the factors that have been identified as contributing to variation in response are discussed, the behavior that has been observed and the time dimensions along which it proceeds must be considered.

The "Protest" Phase

In almost every study reported in the literature, the initial reaction of an infant to the sudden disappearance of its mother includes frequent, loud, repetitive, rather plaintive calls. This high-pitched, relatively pure wail takes the form of the "coo" vocalization of the separated infant macaque or the ear-piercing, repetitive, relatively pure tones of the lost squirrel monkey baby. Accompanying the vocalizations is rapid activity, often distracted rather than focused. 3,37,38 High outbursts of energy (a factor of no little significance if the youngster is expected to survive for very long on its own) are expended in the immediate aftermath of the loss.

During this "protest" period the infant may briefly interact with either its social or physical environment. Even brief play bouts may be seen. But the form of infant activity is clearly altered. This ubiquitous initial phase of protest appears likely to represent a relatively closed genetic system 18 that leaves only limited room for individual variation, regardless of other environmental or experiential factors.

Most authors agree that the repeated vocalizations and high levels of locomotor activity are designed to effect rapid reunion of the separated infant and its mother or some other member of its group who might offer the protection and care needed for survival. This investment of limited and thus quite precious energy resources by the infant in an effort to regain contact as rapidly as possible may well reflect the fact that, under natural conditions, infants who become separated during the first year or so of life rarely survive (e.g., Rhine et al. 30). Numerous laboratory observations of animals requiring support during separations attest to the potentially devastating combination of limited physical coping capacity and the trauma of the sudden loss of an important attachment figure. 6

Despair/Depression Phase

Infants begin to vary considerably in their pattern of response within 24 to 36 hours following the loss of the mother or rearing partner. The most extreme responses have been observed in several laboratory settings (as well as in the field; e.g., van Lawick-Goodall 14) and in a number of species. These vulnerable subjects, following protests of varying intensity, cease spontaneous locomotion and lose virtually all interest in the world around them. They will neither initiate play nor respond to the efforts of others. Even the presentation of a novel object, normally the stimulus for great excitement and interest in young monkeys, fails to arouse them. While immobile, the infant is unable to maintain the normal alert posture, with head up and eyes open, and is changed into an infant seemingly rolled into a ball, its head lowered to the floor and pressed against its body. Its eyes are closed much of the time, even dur ing the day, when infants never sleep if out of contact with the mother. In addition to the postural collapse, these subjects often begin sustained oral contacts with their fur, limbs, digits, or genitalia.

In most instances, even the most severe cases of this type begin to improve five to ten days after the separation. The infant gradually begins to sit more normally, and the hunched-over posture is increasingly reserved for periods in which the infant is frightened. Its eyes are generally open now and oral contacts with its own body diminish. Finally, following a gradual return of interest in the physical environment, social play responsiveness and then social initiations reappear at increasing levels. By the end of two to four weeks, many infants look overtly like normal infants of their age, although more precise behavioral records indicate that recovery is by no means complete even a month after the loss.

Several important issues must be addressed to appreciate more fully the significance of these depressed patterns as one form of response to loss. First, are these patterns as common in primates as the protest behaviors described earlier? A review of the literature 38 reveals that many studies fail to record any instances of the despair/depression pattern. In general, even in studies in which these extreme reactions do appear, only a portion of the subjects show it at all. Thus, as in humans, the most severe forms of behavioral and emotional debilitation occur in a limited number of subjects. Study of these extremely disturbed individuals could reveal much about the potency of certain factors, perhaps including specific phylogenetic or ontogenetic (developmental) ones, in producing at least certain types of depression. It is at this form of response that efforts at intervention may best be directed.

Before considering the behaviors of these animals further, one should ask if the severe reaction just described could be merely the product of a physiologically disturbed youngster. Perhaps the sudden change in diet is important; infant monkeys generally continue to receive some mother's milk until their next sibling is born, after about a year. Or perhaps the reaction merely reflects the fact that sleep patterns have been abruptly disturbed because its mother, on whose ventrum it has always slept, is no longer available. The pattern described above, this line of cautious concern suggests, is merely the response of a tired young infant whose stomach is upset. In a week or 10 days it will start to feel better and its behavior will improve. This is a significant alternate hypothesis to the one proposed earlier—that the loss of the mother is extremely disturbing psychologically and emotionally for nonhuman primates.

Observations of the presumed depression after loss in primates make clear that the separation pattern is indeed the result of severe emotional response to the loss, and not merely the product of digestive or sleep disturbances. Data presented below indicate that nonhuman primates may well experience a period of detachment from the lost object as a later phase of response to loss.

Detachment Phase

Several studies have been done to assess the actual nature of the most severe forms of depressive response to the loss of mother, as well as to test the applicability of a "detachment phase" to nonhuman primates. In a variety of circumstances during reunions following a prolonged separation from their parents, human infants have been observed either to avoid the parents or to behave in an emotionally detached manner in their presence 7,31 Such behavior seems counterintuitive if not paradoxical. There is a clear suggestion that some alteration in the infant's emotional response to the lost care-giver has emerged during the separation. At the very least, such infants may be seen as having conflicting approach/withdrawal motivations in place of the initially unambiguous drive to move to mother at all costs. Main 16 has suggested, for human infants, that avoidance of the mother under these circumstances may be a part of the particular pattern of sustained mother-infant interaction. She suggests that an infant who is frequently rejected in its attempts to achieve contact with its mother may develop a pattern of avoidance when in proximity to her. Nonetheless, both Main 16 and Bowlby 2 have suggested that the apparent detachment behavior at reunion reflects an effort to cope with the conflict between attachment to and anger at the lost parent.

In laboratory studies of nonhuman primates, at a reunion following the enforced separation of the mother-infant pair, the mother almost always retrieves her infant immediately. Thus the infant has little opportunity to express hesitancy or avoidance if it were so inclined. Nonetheless, in an investigation of separated rhesus infants (8-20 weeks of age), Abrams1 observed unusual behavior on the part of the infants in 25 percent of reunions: " .. . the mother entered the cage and retrieved the infant as before, but the infant usually screeched as she did so. Ventral contact was established; but after some period of time, from one to five minutes, the infant broke contact and withdrew from the mother. .." Other instances in which the returning infant actually avoided the mother's initial efforts at retrieval are described as well. These data suggest at least ambivalence on the part of these reunited infants, indicating perhaps an altered emotional response to the previously lost parent.

Studies by Rosenblum 32,33 provide strong evidence in support of the emotional disturbance hypothesis and indications of the detachment phase as well. In one portion of this work, five pigtail infants, 7-8 months of age, were separated from their mothers for a period of 8-10 weeks. The infants, separated two at a time, remained with the rest of the social group in the pens where they had been raised. During the separation the infants showed the normal range of response, from severe depression in two infants and a moderate depressive reaction in a third, to two infants who showed rather minimal reactions after the protest phase. Recovery progressed normally and the depressive phase began to disappear after about 10 days following the loss. Two to three times a week during the separation, the mother of one infant at a time was returned to the home pen for half an hour. For control purposes, both the infant whose mother was returned and the other separated infants were observed in each return trial, thus allowing determination of the specificity of any observed reactions.

Most striking were the reactions of the several infants who had shown strong "depressive" patterns in the days following the loss. Well after these infants had shown relative recovery in the absence of the mother, they showed a virtually complete return of the depressive pattern each time the mother was brought back to the pen. Even nine weeks following loss, one infant that generally looked quite normal, playing socially and actively exploring and playing with its environment, had dramatic reactions to the returned mother.

During these reinvoked depressive episodes, infants would not move toward the cage in which their mother was restrained (for control purposes), although they readily approached the cage when someone else's mother was present. Upon removal of their mother, these infants would almost immediately return to very high levels of activity and play. As a follow-up to this work, the reactions of young infants to color videotapes of their mothers and other social stimuli have recently been studied. In one instance, a 10-month-old pigtail infant that had been terminally separated from its mother three months earlier had an opportunity to produce a taped image of its mother in an operant situation. After several minutes of alternating between the mother's image and that of another female, the infant turned on the image of mother, began to "coo" softly, then closed its eyes and gradually dropped into the collapsed posture typical of the depressive pattern. Holding the lever to maintain the mother's image for the remaining 13 minutes of the trial, the infant stayed in the depressed posture, only occasionally looking up at the image of its mother, cooing briefly, closing its eyes, and dropping its posture once again these observations indicate that it is the emotional disturbance in reaction to the loss of mother and the adaptive demands of the separation environment that produce the second-phase effects that have been labeled despair/depression. At the same time, the data suggest that some aspects of the pattern that have been described as reflecting "detachment" in humans may have their counterpart in the reaction of at least some monkey infants to a severe loss experience.


The Nature of the Attachment Bond

For a strong emotional relationship to be established between an infant and its mother, that relationship must be specific and unambiguous. Indeed, it is a precondition for any consideration of attachment that a particular set of responses be directed selectively toward a specific partner. Despite early anecdotal reports that infant monkeys come to recognize their mother in the first days or week of life, experimental evidence indicates that, as in humans, such recognition matures more slowly and may be affected by a number of factors.

Consider the case of bonnet macaque infants, raised in a complex social group that contains a number of mothers and infants. If offered a controlled choice of responding to their mother or to a complete stranger of the same species, it is not until about 12 weeks of life that significant preferences for the mother are shown. 35 Males, incidentally, appear to be several weeks slower in achieving this capacity than females. If, however, an infant bonnet monkey is reared alone with its mother, without other adults or peers, even as late as six to eight months of age it moves to a complete stranger as readily as it does to its own mother. Some infants reared in this condition failed to select their mothers consistently even as late as a year of age. 39

In a final study in this series, it was shown that if an infant is reared by its mother in the company of another female-infant pair, the infant at about three months responds selectively to its mother rather than to the familiar female, but may not select her in preference to a stranger until two to three months later. Thus the specificity of an infant's response to its primary care-giver may vary as a function of circumstances. As Spitz 42 initially suggested for human children, individuals will, at the very least, be expected to differ markedly in response to loss in the periods before and after the acquisition of selective, individually focused attachment behavior.

According to Bowlby, 2 the reliability with which the attachment figure is able to respond appropriately to the infant's needs (the degree to which the mother is "available and accessible" 22) influences the security of the attachment. Thus one speaks of relatively "anxious" or "secure" attachments, with those subjects at the anxious end of the continuum more likely to show severe reactions to loss.

Although some evidence regarding this issue has been provided by efforts to account for variations in response among members of single treatment groups, 10,41 most of the relevant information derives from comparative studies. One striking illustration may be seen in detailed comparisons of the bonnet and pigtail macaques. 34,37

The pigtails, studied in the laboratory in social units of unrelated individuals, form rather hostile groups in which animals rarely sit or sleep close together. The infants are zealously guarded by their mothers and in the early months rarely have contact with other adults. The mothers, in the early weeks, not only prevent others from contacting their baby, but initially restrain the efforts of their infants to leave. Even as the infant grows more competent, the pigtail mother retrieves it frequently as it attempts to run and play about the pen. As the time for rebreeding approaches (when the infant is four to five months old), the pigtail mother begins active, often punitive weaning and removal of the infant from her body. Such rejections last for varying periods of time. It is not until the infant reaches eight or nine months of age that this "punitive deterrence" abates.

In comparison, bonnet macaques rapidly form close, gregarious groups in which members often intertwine while resting and sleeping in close contact. When infants are born this adult pattern remains undisturbed, and from the first day of life onward the infants are the frequent object of communication and contact by many others in the group. As the bonnet infant matures, the mother, although protective when necessary, is typically passive to its comings and goings, neither preventing nor actively encouraging its departure.

In terms of the concepts described above, as a result of the inconsistency of maternal response and the limitations on their range of social and environmental experience early in life, the pigtails might be expected to develop more anxious "attachments" than their bonnet counterparts, and should therefore show the more severe responses to loss. This has indeed been the case. In numerous studies 12,29,32 pigtails have often shown very severe depression following loss, while bonnets rarely show the most extreme forms of negative response. 13,24,37 Bonnet infants often pass through the loss period relatively unscathed, whether they are "adopted" by others or not. 32

Further support for the influence of the security provided by the attachment bond has been obtained in other recent research on bonnet macaques. 38 In an effort to determine the effects of maternal "employment" on mother-infant relations and infant development, "working" bonnet mothers were required to spend a portion of their day searching for food, which was hidden within specially constructed "foraging devices." Control mothers, living in identical pens, had their food provided for them without any work required. During the course of early development, the infants of the foraging mothers appeared to be functioning more independently than the infants in the no-work group. They were apart from their mothers more and for longer periods. Although weaned somewhat more than the infants in the other group, the forager infants were not dramatically or consistently rejected by their mothers, at least during laboratory observations. However, the normal equanimity of bonnets was quite disturbed in the work group. More overt hostility and less gregarious behavior were observed. The group situation was clearly more tense.

Nonetheless, when pairs were separated it was the infants of the foraging mothers that showed the most severe reactions (including depression) and were the slowest to recover even nominally normal functioning. It seems reasonable to suggest that the mothers required to spend a portion of their day preoccupied with their search for food were often not as available to their infants as the control mothers were. Indeed the records showed that the foraging mothers generally were engaged in active foraging when their infants were out of contact. The requirement that the mother share time between the infant's care and other survival needs, coupled with the decreased friendliness of the social group, may well have left the infants less secure in their attachment to their mothers and less able to learn the skill and approaches necessary to cope with the requirements of the surrounding environment. 26

In recent research in which bonnet mothers and infants lived in environments where the work requirements (of foraging) changed repeatedly over time, the social group became increasingly aggressive. Mother bonnets became unusually rejecting of their infants as they became more distressed themselves. The infants, forced to spend periods of time apart from the mothers, showed increasing disturbance, in some cases culminating in the repeated expression of the full depressive pattern. In some ways this debilitating response duplicated the reactions of the pigtails to the repeated returns of the "lost" and rejecting mother. As in the case of their reinvoked depressions, these apparent "loss patterns" in the presence of a psychologically unavailable figure alert re searchers to the continuity of emotional responses between a pre-loss and loss period.

It appears from these comparative data that in nonhuman primates, and presumably in humans beings, many of the types of behavior that are clearly recognized as emotional responses to loss may appear in various degrees and configurations as part of the ongoing pattern of interaction prior to the final loss experience. 47

Artificial Mother Surrogates

Although it has been clear since the early work of Harlow that monkeys can form very strong emotional attachments to artificial mother surrogates that embody certain specific stimulus characteristics, new evidence makes clear that these attachments differ from those formed toward biological mothers. Although a surrogate may offer emotional support during rearing, the loss of this object does not impair the functioning of a young infant to the same degree that loss of a biological mother does. This important difference in the response to loss has now been demonstrated in rhesus, 20 pigtail, 28 and squirrel 19 monkeys.

What is there about the relationship of infant and surrogate, as compared to infant and mother, that results in less severe response to loss? Two factors seem relevant. First, the security of the relationship depends in part on the reliability and consistency of the responses of the caretaker. Second, the nature of the infant's experience with the nonmaternal environment depends, at least in part, on the affective state of the infant during encounters with the outside environment.

In the case of the surrogate mother, there can be no questions of consistency or availability. Thus, there is the possibility of a rather secure attachment being formed. Moreover, within the confines of the relatively simple environments in which surrogate-rearing usually occurs, the infant is free to explore and contact its environment whenever it feels comfortable doing so. It may return to the consistently available surrogate when its level of arousal or fear becomes too great and may return once again when it feels comfortable. From a variety of perspectives, an individual's opportunities to learn how to deal effectively with the requirements of its environment will be affected by the individual's emotional status at the time of potential learning. The more complex the situation, the more significant the role of the emotional state. Thus it seems reasonable to suggest that the relatively securely attached surrogate infants, having had the most opportunity to learn the nature of their cage environment to best advantage, may be expected to fare better than infants raised by biological caretakers, at least when tested "at home."

Response of Older Subjects to Loss

Nonhuman primate research on the effects of a disruption of social bonds in subjects older than about one year of age has not progressed as far as the work on early infant-mother separation. It does appear clear, however, that nonhuman primates show some evidence of emotional disturbance after the loss of a close partner at virtually all ages tested. For example, monkeys raised in so-called "nuclear families" have been shown to be markedly disturbed following separation from their family units at three, four, and even five years of age, which is well after puberty. 21 Similarly, three-year-old, peer-reared rhesus monkeys have shown very marked depressive behavior after repeated separations from their lifetime partners. 23 Finally, there are suggestions of depression in adult female pigtails after the loss of a close associate, 27 and transient emotional changes do occur in mothers after loss of their infants, although in general these responses are often rather limited.


The primary focus of most of the early work on separation has been the effects of the "loss" per se. In certain respects, too little attention was paid to how the nature of the separation environment might have influenced the individual's capacity to deal effectively with the loss. Consequently, at times there has been a confounding of the loss experience with simultaneous alterations in the social or physical environment. It has now been clearly established that the circumstances in which the grieving individual must function markedly influence the response to loss.

Social Environment

A number of studies of bonnet macaques, 37,38 squirrel monkeys, 4,36 pigtails, 32 and langurs 5,6 have indicated that when an infant is able to receive substitute caretaking from some other member of the group, the overt disturbance of behavior following loss of the mother is markedly reduced or even eliminated almost completely. This ameliorative effect of substitute care-giving in nonhuman primates fits well with the general experience with human infants under comparable circumstances. 2

In general, there is an indication that the impetus to ''adoption" often lies with the infant, 6 and that this capacity to move to others and to transfer filial responses to a foster care-giver may be a crucial coping skill for dealing effectively with the loss experience. The target of the transfer of attachment behavior does not seem to be rigidly fixed, as infants have been shown to shift to other females, adult males, juvenile siblings, other peers, other species, 17 and artificial surrogates after the loss of mother.

If they occur immediately after the separation, these transfers of attachment may preclude both the initial protest behaviors as well as the subsequent despair patterns. Initial vocalizations in squirrel monkeys are virtually eliminated when separation occurs in the presence of an "aunt" (already a partial caretaker of the infant) who immediately accepts the infant. Even the sensitive indicator of infant affect—social play—may show insignificant changes during separation if the infant has been adopted. As these foster relationships continue past a week or two, the new attachments may become so strong that the infant will fail to return to the biological mother when she is returned to the group.

The nature of the infant's familiarity with the other animals appears to be important even if an actual "adoption" does not occur. In one case, for example, a bonnet infant separated from its mother and all other members of its own species but left with several pigtails with whom it had been living showed a marked depressive response to separation. As noted earlier, bonnets left in more supportive groups of their own species generally show minimal disruption of behavior following loss of the mother. Similarly, there is an indication in rhesus monkeys that infants left with familiar peers are less emotionally disturbed following separation than those left with unfamiliar peers. 45

Physical Environment

Current data make it clear that when separation is confounded with the transfer to a novel environment, the pattern of response is altered. Unfamiliar environments appear to increase and prolong initial protest reactions of the infant and either delay or ameliorate subsequent despair behavior. 8,9 Similarly, when surrogate-reared squirrel monkeys are separated from the surrogate, only when they are simultaneously placed into a novel environment do they show marked emotional distress (reflected particularly in their hormonal stress response 46).

There is now reason to believe that the requirements of the environment, if appropriately tailored to the capacities of the infant, may actu ally facilitate recovery following loss of the mother. In a recent study of pigtail macaque infants, 25 subjects were reared by their mothers in a no-work environment. In a series of separations from their mothers, however, these infants either were required to spend several hours a day digging for their food in a layer of sawdust on the floor or were provided their food freely. During the separations in which work was required, infants showed diminished symptoms of emotional distress and depression. Most important, this easing of disturbance occurred not only during the actual foraging activity, but also carried over to the remaining part of the day as well.

It is not simply a matter of "work" per se, or work effort exclusively, that might prove beneficial under these circumstances. It may be that any focused activity that the infant is motivated to engage in and that it can perform successfully would serve as well or better than the foraging task.

Suggestions from the Environmental Data

Available information suggests ways to lessen the intensity of the inevitable emotional response to loss and to increase an individual's capacity to overcome problems and return to effective functioning. It is clear from the data on nonhuman primates that the more familiar a subject is with the setting confronted following a loss and the more supportive the individuals within it are, the less severe the sustained emotional response will be and the more rapid the recovery. Furthermore, individuals confronted with a readily accomplishable and rewarding task may show fewer behavioral difficulties after a loss. In a broad sense, both the "coping" skills brought to the loss experience that are specific to the characteristics of the surrounding environment, and the individual's past successful experiences coping with a variety of situations prior to separation, markedly affect the severity of a bereavement. 15


Much of the current evidence argues against earlier views 11,40 that the nominal "phases" of protest, despair, and detachment are "phases of a single process" 2 inevitably tied to one another as a result of evolution. The patterns of behavior, characterized more by an intersubject and intersituational variability than by consistency, are perhaps best studied from the "coping" perspective on separation phenomena that has been developed by Levine and his associates. 4 In this view, it is the infant's active efforts to cope with the imposed loss, both in terms of behavioral and physiologic responses, that are seen in each phase of the loss pattern. These efforts are shaped largely by the social and physical circumstances during the loss period, not by a prior evolutionary history of adaptation to loss per se. This view also stresses the importance of prior experience with the attachment figure and the prior environment, which are seen as affecting the capacity of the subject to cope with or strategically adapt to suddenly altered conditions. These environmental features, as well as the contingent, interactive qualities of past experience, may reveal the factors controlling the severity and extent of a bereavement and the types of intervention that can ameliorate the potentially disastrous effects of the loss.

Although research on the response to loss of significant partners has shown considerable progress during the past two decades, in many areas researchers have just begun to examine a number of important issues. Pursuit of these topics in further animal studies is sure to enhance researchers' growing understanding of the bereavement process in human beings and help them to distinguish among the numerous interacting factors that are often confusingly interwoven at the human level.

The data base on response to loss throughout life and on continuities or changes in the patterns of response in different groups should be expanded. Moreover, data on the effects of early loss upon emotional response to subsequent social disruptions are extremely limited and must be increased through longitudinal studies. Again, the advantages of the nonhuman primate models here are obvious, as the pace of maturation allows for examination of these continuities and discontinuities across a life span of manageable length.

A wider variety of social partnerships needs to be studied in the context of the loss experience. How do sibling, peer-peer, peer-adult, and adult-adult relationships parallel or differ from the infant-mother bonds that have been the subject of extensive study? How are these differences reflected in patterns of behavior after loss of the specific partner?

Given current thinking about the effect of the availability and contingent responsivity of a partner on the initial bonding and subsequent response to loss, more detailed observations prior to loss are needed. This might help account for at least some of the notable variation in the form and extent of the capacities needed to handle the trauma of loss.

It is now clear that in a number of ways the social and physical environment in which adaptation to loss must occur significantly influences the bereavement reaction. Further studies can investigate the factors that promote or retard the support of other individuals for the grieving subject, as well as the environmental factors that may facilitate recovery or exacerbate the negative responses of various individuals. What features engage the subject productively? What features overwhelm the already disturbed individual?

That these same questions arise in discussions of the human experience points to the importance of following current leads under the controlled conditions obtainable in animal research.


Abrams, P. Age and the effects of separation on infant rhesus monkeys. Paper presented at APPA meeting, Mexico City, 1969.
Bowlby, J. Separation: Anxiety and Anger . New York: Basic Books, 1973.
Coe, C.L., and Levine, S. Normal responses to mother-infant separation in nonhuman primates. In: Anxiety: New Research and Changing Concepts (Klein, D.F., editor; , and Rabkin, J., editor. , eds.). New York: Raven Press, 1981.
Coe, C.L., Rosenberg, L.T., and Levine, S. Endocrine and immune responses to maternal loss in nonhuman primates. Paper presented at American Psychological Association meeting Anaheim, California, 1983.
Dolhinow, P. An experimental study of mother loss in the Indian Langur monkey (Presbytis entellus). Folia Primatologica 33: 77-128, 1980. [PubMed: 7191391]
Dolhinow, P. Langur monkey mother loss: profile analysis with multivariate analysis of variance for separation subjects and controls. Folia Primatologica 40: 181-196, 1983. [PubMed: 6884906]
Heinicke, C., and Westheimer, I. Brief Separations . New York: International Universities Press, 1965.
Hinde, R., and Davies, L. Removing infant rhesus from mother for 13 days as compared with removing mother from infant. Journal of Child Psychology and Psychiatry 13: 227-237, 1972. [PubMed: 4632823]
Hinde, R., and McGinnis, L. Some factors influencing the effects of temporary mother-infant separation: some experiments with rhesus monkeys. Psychological Medicine 7: 197-212, 1977. [PubMed: 406620]
Hinde, R., and Spenser-Booth, Y. Effects of brief separations from mothers on rhesus monkeys. Science 173: 111-118, 1971. [PubMed: 4996549]
Kaufman, I.C. Developmental considerations of anxiety and depression: psychobiological studies in monkeys. In: Psychoanalysis and Contemporary Science (Shapiro, T., editor. , ed.). New York: International Universities Press, 1977.
Kaufman, I.C., and Rosenblum, L.A. The reaction to separation in infant monkeys: anaclitic depression and conservation-withdrawal. Psychosomatic Medicine 29: 648-675, 1967. [PubMed: 4966515]
Kaufman, I.C., and Stynes, A.J. Depression can be induced in a bonnet macaque infant. Psychosomatic Medicine 40: 71-75, 1979. [PubMed: 414290]
van Lawick-Goodall, J. In the Shadow of Man . New York: Houghton-Mifflin, 1971.
Levine, S. A psychobiological approach to the ontogeny of coping. In: Stress, Coping and Development in Children (Garmezy, N., editor; , and Rutter, M., editor. , eds.). New York: McGraw-Hill, 1982.
Main, M. Analysis of a peculiar form of reunion behavior in some day care children: its history and sequelae in children who are home-reared. In: Social Development in Children: Day Care Programs and Research (Webb, R., editor. , ed.). Baltimore: Johns Hopkins Press, 1977.
Mason, W., and Kenney, M. Redirection of filial attachments in rhesus monkeys: dogs as mother surrogates. Science 183: 1209-1211, 1974. [PubMed: 17789224]
Mayr, E. Behavior programs and evolutionary strategies. American Scientist 62: 650-659, 1974. [PubMed: 4440939]
Mendoza, S., Smotherman, W., Miner, M., Kaplan, J., and Levine, S. Pituitary adrenal response to separation in mother and infant monkeys. Developmental Psychobiology 11: 169-175, 1978. [PubMed: 416983]
Meyer, J., Novak, M., Bowman, R., and Harlow, H. Behavioral and hormonal effects of attachment object separation in surrogate-peer reared and mother-reared infant rhesus monkeys. Developmental Psychobiology 8: 425-435, 1975. [PubMed: 817950]
Mineka, S. Depression and helplessness in primates. In: Child Nurturance: Studies of Development in Nonhuman Primates (Fitzgerald, H.E., editor; , Mullins, F.A., editor; , and Gage, P., editor. , eds.). New York: Plenum, 1982.
Mineka, S., and Suomi, S.J. Social separation in monkeys. Psychological Bulletin 85: 1376-1400, 1978. [PubMed: 104324]
Mineka, S., Suomi, S.J., and DeLizio, R. Multiple peer separation in adolescent monkeys: an opponent process interpretation. Journal of Experimental Psychology 110: 56-85, 1981. [PubMed: 6453185]
Paully, G., and Rosenblum, L.A. Variability of environmental characteristics and mother-infant relations. Child Development 55: 305-314, 1984. [PubMed: 6705632]
Plimpton, E. Environmental variables and the response to maternal loss. Unpublished doctoral dissertation: State University of New York—Downstate Medical Center, 1981.
Plimpton, E., and Rosenblum, L.A. Ecological and temporal dimensions of maternal loss in nonhuman primates. In: The Psychology of Separation Through the LifeSpan (Bloom-Feshbach, J., editor; , and Bloom-Feshbach, S., editor. , eds.). San Francisco: JosseyBass; (in press).
Rasmussen, K., and Reite, M. Loss-induced depression in an adult macaque monkey. American Journal of Psychiatry 139: 679-681, 1982. [PubMed: 7200331]
Reite, M., Short, R., and Seiler, C. Psychological correlates of maternal separation in surrogate-reared infants: a study in altered attachment bonds. Developmental Psychobiology 11: 427-435, 1978. [PubMed: 99340]
Reite, M., Short, R., Seiler, C., and Pauley, J.P. Attachment, loss and depression. Journal of Child Psychology and Psychiatry 22: 141-169, 1981. [PubMed: 7196919]
Rhine, R., Norton, G., Roetgen, W., and Klein, H. The brief survival of free-ranging baboon infants (Papio cynocephalus) after separation from their mothers. International Journal of Primatology 1: 401-409, 1981.
Robertson, J. Young children in brief separation: a fresh look. Psychoanalytic Study of the Child 26: 264-315, 1971. [PubMed: 5163230]
Rosenblum, L.A. Infant attachment in monkeys. In: The Origins of Human Social Relations (Schaeffer, R., editor. , ed). New York: Academic Press, 1971.
Rosenblum, L.A. Affective maturation and the mother-infant relationship. In: The Development of Affect (Lewis, M., editor; , and Rosenblum, L.A., editor. , eds.). New York: Plenum, 1978.
Rosenblum, L.A. The influence of the social and physical environment on motherinfant relations. Annali dell Istituto Superiore di Sanita 18: 215-222, 1982. [PubMed: 7186269]
Rosenblum, L.A., and Alpert, S. Fear of strangers and the specificity of attachment in monkeys. In: The Origins of Fear (Lewis, M., editor; , and Rosenblum, L.A., editor. , eds.). New York: Wiley, 1974.
Rosenblum, L.A., editor; , and Cooper, R.W., editor. (Eds.). The Squirrel Monkey . New York: Academic Press, 1968.
Rosenblum, L.A., and Kaufman, I.C. Variations in infant development and response to maternal loss in monkeys. American Journal of Orthopsychiatry 38: 418-426, 1968. [PubMed: 4967093]
Rosenblum, L.A., and Plimpton, E. Adaption to separation: the infant's efforts to cope with an altered environment. In: The Uncommon Child (Lewis, M., editor; , and Rosenblum, L.A., editor. , eds.). New York: Plenum, 1981.
Rosenblum, L.A., and Smiley, J. Therapeutic effects of an imposed foraging task in disturbed monkeys. British Journal of Child Psychology and Psychiatry (in press), 1984. [PubMed: 6540275]
Solomon, R., and Corbitt, J. An opponent process theory of motivation: I. temporal dynamics of affect. Psychological Review 81: 119-145, 1974. [PubMed: 4817611]
Spenser-Booth, Y., and Hinde, R. Effects of six-days separation from mother on 18 to 32-week-old rhesus monkeys. Animal Behaviour 19: 174-191, 1971.
Spitz, R.A., Anaclitic depression. Psychoanalytic Study of the Child 2: 313-347, 1946. [PubMed: 20293638]
Suomi, S.J. Genetic, maternal and environmental influences on social development in rhesus monkeys. In: Primate Behavior and Social Biology (Chiarelli, A.B., editor; , and Correuccino, R.S., editor. , eds.). Berlin: Springer, 1981.
Suomi, S.J. Social development in rhesus monkeys: consideration of individual differences. In: The Behavior of Human Infants (Oliverio, A., editor; , and Zappella, M., editor. , eds.). New York: Plenum, 1983.
Suomi, S.J., Collins, M.L., and Harlow, H.F. Effects of permanent separation from mother on infant monkeys. Developmental Psychology 9: 376-384, 1973.
Vogt, J.L., and Hennessy, M.B. Infant separation in monkeys: studies on social figures other than the mother. In: Child Nurturance: Studies of Development in Nonhuman Primates (Fitzgerald, H.E., editor; , Mullins, J.A., editor; , and Gage, P., editor. , eds.). New York: Plenum, 1982.
Weintraub, M., and Lewis, M. The determinants of children's responses to separation. Monographs of the Society for Research in Child Development 42: 172, 1977. [PubMed: 611417]


This chapter was prepared by Leonard A. Rosenblum, Ph.D., consultant to the study, Professor, Department of Psychiatry, and Director, Primate Behavior Laboratory, State University of New York, Brooklyn, New York.

Copyright © 1984 by the National Academy of Sciences.
Bookshelf ID: NBK217852


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (2.5M)

Related information

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...