The foremost mission of the Department of Health and Human Services (DHHS) is to address the health and welfare of the human species. That mission cannot be accomplished without relevant research using nonhuman primates as models. The mission of the U.S. Department of Agriculture (USDA) is to ensure animal welfare by governing their care and treatment. Nonhuman primates are commonly maintained in biomedical institutions to support high-priority research into issues of human well-being, and those institutions are obliged to address the welfare of the animals. However, the ability to generalize findings from our close primate relatives depends on maintaining animals in a state that is representative of normal functioning. Thus, research into the well-being of nonhuman primates is essential to ensure that the missions of DHHS, the Department of Veterans Affairs, the Department of Defense, and USDA are carried out effectively. Those and other agencies depend on the collection of data on nonhuman primates.
Assessment of the well-being of nonhuman primates depends on the development of an adequate theory of well-being and on an adequate understanding of the animals' psychological, social, ecological, and physical characteristics. Although much is known about laboratory primates, we still have much to learn about what is necessary to maintain them in a condition of well-being.
A working definition of psychological well-being has been presented. However, research is needed both to evaluate potential measures of psychological well-being and to develop techniques that promote it. Such techniques should be
appropriate, practical, cost-effective, and, above all, compatible with colony management and the conduct of research. Although professional judgment and lay judgment will continue to have their roles in programs to monitor and enhance the well-being of laboratory primates, intuition and personal judgment should be viewed as adjuncts or as the first step of hypothesis formation, not as a substitute for scientific investigation. Enrichment methods that have not been subjected to empirical testing should be viewed simply as invalidated ideas, regardless of how well intended they might be. Without appropriate measurement and verification, we might do more harm than good in our efforts to improve animal conditions.
With an eye to achieving a better and, insofar as possible, a scientific perspective of psychological well-being, we propose the following topics for research. We recognize that the list is not all-inclusive.
Theory of Psychological Well-Being
The phrase psychological well-being arose in a regulatory, rather than scientific, context. Psychological well-being is a hypothetical construct, and the validity of a hypothetical construct can be determined only in relation to a theory that defines its properties and in relation to empirical data that address the fit between predicted and observed phenomena. We do not have a theory to guide thinking and research on well-being; the development of a coherent theory of psychological well-being is an obvious research need. Such a theory ought to incorporate cognitive, behavioral, and physiological characteristics of the organism in an integrated view of well-being. It also ought to encompass the possibility of species, age, sex, and individual differences in responsiveness to the same immediate environmental situation. Some of the work that has been done so far in this domain is based on notions of how humans behave and how they react to environmental situations. That might be a suitable place to begin the development of hypotheses, but it clearly is not sufficient. We must develop a theory so that we can agree on empirical measures of well-being, rather than relying on conflicting subjective judgments about internal states in other species.
Indexes of Well-Being
The acceptance of particular behavioral or physiological measures as operational indexes or correlates of psychological well-being can follow only from a theory that defines well-being in compatible terms. The use of multiple measures of psychological well-being until such a theory has been developed has been advocated. However, the relationships among several commonly used measures are not well understood. For example, chronic stress has both behavioral and physiological components, but their interrelationships have not been characterized fully. Furthermore, if stress is considered a manifestation of ill-being under
some conditions, is the absence of chronic stress equivalent to well-being? The answer to the latter question should not be presumed to be yes.
In addition to identifying important research, this chapter addresses a fundamental problem in our approach to these concerns. Who would deny that all animals are equipped with capabilities for avoiding particular situations—the kinds of activities often referred to as flight-or-fight responses? The animals of concern are in fact wild animals and have little familiarity with captive-specific stimuli. In addition, these responses have a well-known physiological basis that enables the animal to cope with perceived conflict or danger. When precipitating stimuli are more or less continuously present, we conclude that the animals in question are chronically mobilized for responding, and so can be described as being "stressed." Psychological well-being would seemingly be enhanced, therefore, by also reducing the level of stimulation.
Conversely, captive animals display activities that have nothing to do with flight-or-fight, have a different physiological basis, and typically occur in relaxed contexts, e.g., grooming, play, foraging, or exploring. A paucity of these species-typical acts leads to the conclusion that animals are "bored," living in a chronic state of under stimulation. Not infrequently, animals under these circumstances invent their own ways of self-stimulation, often in the form of behaviors that we regard as undesirable (throwing feces, coprophagy, spitting, regurgitation, pacing, flipping, etc.). The enhancement of psychological well-being in these instances is realized through an increase in the level of stimulation, particularly with stimuli that bear an appropriate level of novelty.
A complete elimination of environmental stressors might be undersirable (NRC 1992; Selye 1974). Animals and humans seem to seek out some level of stimulation or stress that they find optimal. That level might be different across individuals and certainly across species. Alternatively, adaptive mechanisms might habituate a primate to the point where it is no longer physiologically stressed (e.g., animals adjusting to intermittent but continual pestering by cagemates). Research is needed to evaluate those ideas more fully.
In discussing atypical behavior, or stereotypies that either are detrimental to well-being or serve no adaptive function, it is critical to differentiate repetitive movements, stereotyped patterns, and potentially self-abusive behavior that indicate a lack of well-being from patterns that constitute potentially harmless idiosyncrasies (Berkson 1967; Mason and Berkson 1975). Often, it is not the occurrence but the frequency or the situation in which they occur that signifies a lack of well-being. Patterns of unusual behavior seen in play contexts are rarely viewed with concern, whereas the same patterns seen in other contexts might indicate ineffective adaptation. Many forms of behavior considered abnormal in adults are not viewed with alarm when seen in infants, e.g., digit-sucking and repetitive-movement patterns might be regarded as harmless in very young animals, but their appearance in adults, where they substitute for more effective typical coping responses, is cause for concern. Clearly, we need to conduct more research
aimed at identifying indexes of well-being and determining which manifestations of atypical behavior indicate a lack of psychological well-being.
Although considerable information is available on the natural history of some primate species, we do not yet know how to incorporate aspects of natural history into a practical, sensitive, and valid program of colony management that serves the dual interests of primate well-being and the research enterprise. We should avoid the conclusions derived from nature films that every animal's natural environment is either idyllic and peaceful or dominated only by bouts of hunting, wounding, and starvation. Rather, we should investigate animals' environments to identify characteristics relevant to well-being in captive animals. For example, how do the stressors encountered in captivity compare with those encountered in a given animal's natural environment in source, frequency, intensity, or duration?
A number of natural history variables seem to bear on the psychological well-being of captive primates, including arboreal and terrestrial activities, social organizational patterns and occasions for groups to form and disperse, dependence on long-term social affiliations or bonds, dietary needs and food-getting behaviors, preferred temperature norms and ranges, mating and infant-care patterns, natural communication modes, and modes of locomotion and movement. Further research is needed to determine the relative importance of those factors for different species of primates maintained in captivity. It should not be our goal to duplicate natural environments with all the hardships sometimes suffered by wild animals, but rather to identify favored activities, preferred patterns, and the general rhythms in life that organize behavior.
Research has already demonstrated that several species of monkeys and apes do not develop normally when reared under conditions of social deprivation (Fritz and Howell 1993a; Mason 1965, 1991; Mason and others 1962, 1963), and the adverse consequences of early social restriction are only somewhat ameliorated by later rehabilitation efforts (Davenport and Rogers 1970; Davenport and others 1973; Fritz 1986, 1989; Rogers and Davenport 1970). Efforts have been made to identify the exact types of social stimulation required for normal development in some species, and there is some debate concerning minimal amounts of stimulation that are required to produce normal social and reproductive skills. That is of practical importance not only for rearing orphaned animals or animals abandoned by their mothers, but also because the pressure to produce infants in a breeding colony can result in early weaning and separation to reduce lactational amenorrhea, when it is present, and advance the next conception. It is clear that
learning through social interactions is essential for many primates to be able to predict the social consequences of behavior directed toward live partners, to acquire appropriate communication skills, to care for infants properly, and to recognize predators. It is not clear, however, what types of experience will foster social competence or other aspects of species-normal behavior. Can cross-species fostering substitute for normal social environments, or does a cross-fostered animal lose some of its species-typical behavior (Seyfarth and Cheney 1997)? To what extent is housing with adult males, females other than the mother, and immature animals of both sexes necessary or sufficient for normal development? Can infants be successfully reared with aged adults or peers, thereby freeing young adults for rebreeding?
Although the importance of social stimulation during infancy is well established, considerably less is known about the influence of social contact during the juvenile and adolescent stages of development. The issue is important, inasmuch as primates are often moved from their familiar social groups and housed elsewhere. More information is also needed on the adaptation of wild-born animals to captivity and on how to prepare captive animals for release to free-ranging conditions or to the wild. We also need to know whether primates reared in free-ranging situations require strategies for promoting psychological well-being that are different from those required for animals born and reared in captivity.
Several characteristics of individual animals can influence their psychological well-being in captive settings. Age and sex play major roles. It is important to determine what factors are necessary for the development of social competence in males and females. This information, for various species, might have important implications for the long-term success of captive breeding groups. One could argue that animals that fail to develop social competence for life in a social group and successful reproduction and parenting are damaged animals. The extent to which such animals are psychologically damaged and the importance of these failures require further exploration. Likewise, we need to understand the effects of age on psychological well-being and to determine effective strategies for maintaining both very young and very old animals in a state of well-being.
Cage design can be important in fostering well-being in captive primates. To that end, research is needed to evaluate the effects of particular caging materials (plastic, wood, and metal) and cage designs (dimensions and cage configuration) in eliciting species-typical patterns of behavior in the different species of primates maintained in captivity (e.g., Crockett and Bowden 1994; Shimoji and
others 1993). The cage sizes for particular species are currently determined by the typical weight of species members. Such an approach fails to provide for performance standards based on typical postures or locomotor expression, such as stride length. The extent to which performance standards can be translated into appropriate cages and housing depends on thorough study of species, age, and sex differences. It is important to specify cage configurations, in addition to absolute size. Other kinds of information will also be needed. Perch height should be examined in the context of both body conformation and tail length. Information on species-typical resting patterns can be relevant, inasmuch as some animals sit and others sprawl with arms and legs extended. Various stride characteristics should be identified. For example, if an animal travels horizontally with a quadrupedal gait, how long is each normal stride? For arm swinging species, what is the arm span in travel (as opposed to maximal reach), and what height is necessary for the feet to clear the ground? How far apart should supports be for vertical clingers and leapers? Those basic measures should be characterized during development for both sexes of each species.
According to the experience of many primate-colony managers, the vertical dimensions of housing affect the well-being of some species. Under natural conditions, many primates spend much of their lives aboveground and escape upward to avoid terrestrial threats. Therefore, these animals might perceive the presence of humans above them as particularly threatening. In addition, such environmental variables as lighting, temperature, and airflow are likely to be affected by height, and these in turn could influence an animal's physiological state. Clarification of the contribution of such factors to well-being is needed. New cage designs should be developed to foster the humane capture of primates that are maintained singly or in groups. Cages should facilitate the regrouping of animals, their transfer between cages, and their access to different cages. Different kinds of materials should be examined that might provide for optimal use of limited intracage space by particular species and accommodate species-typical behaviors, such as marking, leaping, and chewing. Optimal use of available cage space might well depend more on the placement of perches, platforms, moving and stationary supports, and refuges than on cage size itself.
Primates are noted for their social behavior and proclivities. Although individual caging of primates might be required by protocols of approved studies or by reason of the social incompetence or health of a given animal, group caging is often more appropriate in light of the social needs of a species. Field data from various species have revealed long-term bonding relationships between parent and offspring, siblings, and others. Several factors should be studied to understand more fully the impact of social housing on psychological well-being. These include group formation, group composition, group size, and group stability. It is
essential that suitable procedures be developed for placing animals in groups. Although much is known about pair formation (Reinhardt 1989a, b) and group formation in some species (Bernstein 1969, 1971, 1991; Bernstein and Mason 1963; Bernstein and others 1974a, b; Williams and Bernstein 1983), relatively little is known about other species. Likewise, relatively little information is available on the importance of group composition (age and sex classes) in promoting psychological well-being. On a social level, do particular combinations of animals (species, sex, and age) provide more positive, rather than negative, social interactions and facilitate the development of relationships? With respect to reproduction, does the presence of mothers contribute to the reproductive success of daughters with their first infants (Fairbanks 1988)? Knowing the answer to the last question might be important in reducing infant losses by primiparous mothers.
Group size is typically dictated by a practical consideration—the size of the cage or enclosure. An additive principle is used today to determine cage size for groups: the amount of floor area needed is established by multiplying the cage size required for an individually housed monkey by the number of animals in the social group. Few data are available to indicate the soundness of this strategy for the various species of captive primates. Further research is needed to determine how cage floor space and volume should be modified as the number of animals increases.
Extensive evidence suggests that for most species, stable social groups are preferable to groups that are continually reorganized. Should captive maintenance ensure, whenever and wherever possible, long-term social pairings of primates? One implication is that timed mating programs might need to consider the normal social partners of individual animals that are ordinarily removed and paired briefly for reproductive purposes. Another implication is that social bonds might be necessary for normal physiological responses in some primate species. It has sometimes been stated that long-term social bonds might contribute more to the well-being of primates in biomedical experiments than any other factor. That assumption requires verification; the extent to which social companionship yields more-normal physiological responses should be established scientifically. Finally, if long-term social arrangements exist, what are the implications for translocation of animals between social groups, between laboratories, and so on?
The environment can provide a captive primate with many objects designed to stimulate the expression of cognitive abilities. The degree to which such activity improves psychological well-being is unknown. Measures are needed to show whether increased activity of various sorts is beneficial. A need also exists for research and development on environmental design for captive primates to determine what features enhance primates' interest in their surroundings; their
care, nurturance, and general well-being; their capture and transfer with minimal disruption and trauma; their physical comfort; the expression of species-specific behavior; and their social interactions with cagemates. Other than increasing activity itself, the beneficial effects of such devices as television, video tasks, mechanical puzzles, and manipulanda should be demonstrated. Research and development on these topics should be systematic, based on a comprehensive theory or concept of primate maintenance, and conducted and reported so as to contribute to the literature of animal science and animal behavior.
The committee believes that nonhuman primates benefit by having some control over their environment (see Markowitz and Aday 1998 for a discussion on providing captive animals control over their environment) and that, lacking such control, they generally benefit by being able to predict environmental events over which they have no control. Considerable research will be required to demonstrate that control and prediction enhance the psychological well-being of nonhuman primates. The benefits of predictability could vary in accordance wit whether the procedure is positive, negative, or innocuous to the animals. Extensive work will then be required to learn which features in the environment are most critical to this sense of well-being.
Primates in general have substantial cognitive capacities for complex learning and memory. We know that primates' cognitive abilities and the specifics of their attention and motivational processes vary, probably in relation to ecological factors that are important to their adaptation to their natural environment. In addition, each species is naturally curious about or interested in particular kinds of things and will readily learn and remember particular kinds of things. These special interests and abilities are often related to the natural lives of animals in their species-typical habitats.
The following questions, however, remain to be answered: Are the cultivation and use of cognitive capacity basic to the well-being of captive primates? How can primates' intellect (i.e., cognition) be challenged in cost-effective ways to help to sustain their well-being, particularly where social companions are few or absent? How can knowledge of primates' natural history contribute to the design of materials and tasks that will be of interest and will stimulate appropriate behavior and enhance well-being? What species-related criteria are relevant to the empirical assessment of well-being as influenced by the cognitive operations of learning and memory?
Husbandry practices are likely to have a substantial effect on psychological well-being, but little research has been done to underscore this point. For ex-
ample, to what extent should routine procedures be predictable, and to what extent should the primate be an active participant in the process? How can one measure the adverse effects of deviation from routine?
Special attention should be focused on such procedures as quarantine and shipment, which generally entail translocation, social separation, dismantling of social groups, and marked alterations in ambient temperature, diet, water, caging, and animal technicians. The effects of quarantine and shipment should be subject to analysis and assessment from the perspective of animal well-being. Such study might afford a unique opportunity to do research on stress and its attenuation.
Special attention and research efforts should be directed to understanding the few animals in a colony that show signs of distress or disturbance (e.g., a type of sentinel) despite the presence of measures that support the well-being of all other animals in a colony. Care for these individual animals is particularly troublesome, in that they represent isolated problems in a population that appears to be exhibiting overall well-being. The temptation is to invest disproportionate resources in the chronic care of one animal or to remove the problem by euthanizing the animal (see Kreger and others 1998). Systematic investigation of individual differences in psychological well-being that leads to an understanding of why particular animals ''slip through" despite the best possible plans will improve our understanding of the interaction of factors that contribute to the well-being of all animals.
Animal Technician's and Caregiver's Role in Well-Being
The animal technician's and caregiver's roles are pivotal to the social support of primates, particularly animals that are singly caged. Caregivers can serve as important points of social contact from which primates can garner positive interaction, instructions (where to go and what to do during cage-cleaning, transfer, etc.), and emotional security. They might also serve important roles in managing events when protracted fights, quarrels, and incessant hassling among primates break out. How might these relationships between animals and humans best be established, and how can existing relationships be improved? How can caregivers and technicians best be trained in key aspects of animal behavior? What aspects of training are most effective and deserve support? These ideas should be investigated further so that the caregiver's optimal role can be defined with reference to primate well-being. The effect of rotating caregivers in a colony and of changing caregivers on weekend, holiday, and other schedules merits investigation.
Some of the programs involving animals' well-being require time and effort (e.g., animal training) and commitments from the institution's senior management. Ways to increase the efficiency of human operations to help promote well-
being need to be documented. Important to research and zoological institutions alike is how federal and private inspectors can best be trained to make accurate appraisals of well-being.
Research procedures should not be sustained merely because they have been used in the past. Research procedures that entail a demonstrated negative effect on well-being should be subject to review and should be modified or supplanted with other methods that are less disruptive to well-being. That consideration is of particular importance where, by tradition, very young animals might be housed or chaired singly for long periods. Such procedures might solely compromise normal development because they occur during the formative, hence sensitive, years of early development.
How can biological samples be collected so as to minimize restraint, isolation, loss of social support, the stress of isolation, pain, and other factors that can disrupt well-being?
In sum, whereas a great deal is known about the natural history and behavior of nonhuman primates held in captivity, much more information is required. Even with substantially greater information, the development of prescriptive recipes for primate well-being would not be desirable. A variety of solutions might achieve the same general goal—animals that are maintained under conditions that promote their physical and psychological well-being. The aim of research in this regard should be to find means by which to assess psychological and physical well-being and to provide the knowledge necessary to develop programs to achieve the general goal—animals maintained for research, exhibition or education can all be maintained under conditions that are consistent with the goal and will provide for their well-being. It is the responsibility of all who keep nonhuman primates to ensure that personnel are appropriately trained to develop procedures consistent with the goals of the institution and with the psychological and physical well-being of the animals in their charge.