DAVID SHEPHERDSON AND KATHY CARLSTEAD

Center for Species Survival, Smithsonian Conservation Biology Institute, Front Royal, Virginia, USA

When it comes to maintaining mental health, or psychological well-being, zoo animal populations are unique in a number of important ways. In contrast to most other animals in human care, zoo animals are ‘wild’ rather than domesticated, meaning that they have not been intentionally bred to favor traits compatible with captivity and human management. Certainly, genetic selection must exist in the zoo environment and it may result in small changes that favor survival and reproduction in zoos, but most zoo-held species have not been separated from their natural habitat for many generations and breeding programs exist to maintain genetic diversity and minimize directional selection (reviewed by Hutchins and Wiese, 1991; Maple, 1992).

The number of different ‘wild’ species that are cared for in zoos results in tremendous variation in ability to adapt to and cope with zoo environments and represents a huge challenge for zoo animal managers. Some species may come from similar environments and climates to the host zoo, whereas others come from radically different ones. Some species have small home ranges and some have large ones. Some species are shy by nature and others are bold. Some species are highly social and others more solitary. Indeed, Mason (2015) has pointed out that this variation could be a powerful comparative tool in identifying the factors that predispose a species to adapt well to captivity and has shown, for example, that species with larger home ranges in the wild tend to spend more time engaged in stereotypic behavior in zoos than those with small home ranges (Clubb and Mason, 2007).

Zoo populations of a species also tend to be relatively small and geographically dispersed, valuable to the zoos that hold them (due to the difficulty of replenishment and emotional attachment of staff and visitors), and, while kept in confined spaces and in view of the public, are also intensively managed with much higher ratios of husbandry and veterinary staff per individual than most other animals in human care. On the other hand, since these are mostly species that have not been intensively studied in the wild, we often lack basic biological knowledge about them.

In modern zoos, individuals of most species are now captive born and bred. Because one of the central missions of modern zoos is to conserve endangered species, zoos have made major commitments toward research designed to benefit captive populations in fields such as genetics, demographic analysis, reproductive physiology, and assisted reproductive techniques. We would also argue that concern for the welfare of individual animals is and has been a strongly held ethic by most ‘good’ zoos. The interpretation of good welfare, however, has changed over time. Traditionally, zoos have valued measures such as longevity, breeding success, and physiological health. As the science of zoo animal welfare advances, more measures of psychological well-being are being adopted, environmental enrichment techniques have become an everyday part of husbandry, and welfare assessment and real-time monitoring tools are being developed.

21.1 Abnormal Behaviors in Zoo Animals

The abnormal ways that many animals behave while confined in zoos was largely accepted by managers and the visiting public before the early to mid-twentieth century. One of the first of a new breed of zoo biologists was past director of the Zurich Zoological Garden, Heini Hediger (see obituary by Maple, 1992). He was one of the first to acknowledge, study, and write about abnormal behaviors and the importance of psychological or mental health in zoo animals, and suggest solutions based on their natural history. Hediger (1950, p.158), wrote that ‘Clearly one of the most urgent problems in the biology of zoological gardens arises from the lack of occupation of the captive animal.’ Hediger’s strategy was to apply his understanding of the biology, behavior, and motivation of a species in the wild to ‘see’ the zoo environment through their eyes. Hediger was particularly interested in studying stereotyped motor reactions in captive animals, work that was continued by Meyer-Holzapfel (1968) and published, among other places, in M.W. Fox’s (1968) seminal publication Abnormal Behavior in Animals. Meyer-Holzapfel spent many years studying abnormal behavior in zoo animals, observing the development and form of mostly locomotor stereotypies in a variety of species, and was one of the first to document the progression through time of some stereotypies from stimulus dependent to ‘fixated by habit’. The significance of stereotypic behavior as a potential indicator of stress has been confirmed many times since (Mason et al., 2007; Shepherdson et al., 2013; Kroshko et al., 2016). Other abnormal behaviors observed in zoos include self-mutilation (Pizzutto et al., 2015), regurgitation and re-ingestion (Gould and Bres, 1986), coprophagy (Akers and Schildkraut, 1985), and excessive aggression (Erwin et al., 1976).

21.2 Challenges Particular to Zoo Animals

Zoo environments have challenges to well-being that other contexts may lack. There are many sources of aversive stimulation (stressors) that are known to elicit protective behavioral and/or physiological responses in animals. Many are ultimately associated with space restriction and confinement that limit an animal’s ability to escape what is fear-inducing or prevent the ability to adapt to it. Some examples are noise, excessive heat or cold, overcrowding, social group instability, and the actions of caretakers (reviewed by Carlstead and Shepherdson, 2000; Morgan and Tromborg, 2007). Hediger (1964) describes situations in which captive animals injured themselves or failed to breed because of the inability to escape from caretakers or visitors. Persistent uncertainty about the actions of caretakers or veterinarians may contribute to the stressfulness of some captive situations if the animals have no reliable predictive cues (Bassett and Buchanan-Smith, 2007; Thines et al., 2007). Other stressors may be unique to zoos, specifically, olfactory and auditory cues from predators (Buchanan-Smith et al., 1993; Carlstead et al., 1993; Wielebnowski, 2003) and construction noise (Cociu et al., 1974; Powell et al., 2006).

There is one potential stressor that is common to zoos, aquariums, and wild animal parks, and very uncommon or rare in other animal contexts: the presence of zoo visitors. There is evidence that visitor presence, activity, size, and viewing position causes changes in the behavior of zoo-housed mammals (reviewed by Davey, 2007). The ‘visitor effect’ has been associated with a number of changes suggestive of stress, including increased vigilance, closer social spacing, reductions in affiliative behavior, and increases in aggression (Glatston et al., 1984; Hosey and Druck, 1987; Chamove et al., 1988; Thompson, 1989). Visitor presence can be stress-inducing for black rhinoceros (Diceros bicornis) as demonstrated by Carlstead and Brown (2005) in a study of 26 rhinos at ten zoos. They found positive correlations between the degree of public access around rhinos’ enclosures and the rhinos’ mean fecal corticoid metabolite (FGM) concentrations. An earlier study by Carlstead et al. (1999a) of black rhinos at 23 zoos suggested that high visitor access around the enclosure perimeter was stressful because it correlated with the zoo’s 20-year mortality rate for rhinos and with ratings for behavior trait fear. FGM concentrations of eight Indian blackbuck were significantly higher on days when zoo visitor density was extremely high compared to zero or extremely low (Rajagopal et al., 2011), and in spider monkeys (Davis et al., 2005) and Mexican wolves (Pifarré et al., 2012) an increase in the number of visitors was associated with an increase in cortisol. In clouded leopards, FGM concentrations were measured as higher in the individuals kept on public display compared to those maintained off exhibit (Wielebnowski et al., 2002). Obstructing the view of visitors has been reported to decrease fecal corticoids and aggression in black-capped capuchins (Sherwen et al., 2015). Sometimes, however, studies of visitor effects on zoo animals show that the visiting public has no effect (neutral) on animals, or is even enriching (Davey, 2007). The most negative effects of visitors have been found among animals with poor baseline welfare combined with high levels of stereotypic locomotion, intragroup aggression, and self-injurious behavior (Farrand, 2007).

Lack of mental or cognitive stimulation, analogous perhaps to boredom in humans, is often presumed to be a problem for captive exotic animals and has been well documented (Špinka and Wemelsfelder, 2011). Animals in the wild are constantly faced with cognitive challenges. They have to remember the location of important resources that may change through time, and they have to problem solve to find and gain access to the resources that they need. Predators have to outsmart their prey, and prey animals have to avoid predators. Individuals representing a rapidly growing list of species use tools to get what they want (Seed and Byrne, 2010). Information important to survival surrounds wild animals, and finding and interpreting this information correctly can determine the difference between life and death. Exploration is the outward expression of the desire to ‘know’. The importance of information seeking and strategies for providing opportunities to meet this need have long been recognized (Shepherdson et al., 1993) and, as we shall see, there are many ways in which zoo environments can be enriched to provide meaningful opportunities for information seeking. However, Meehan and Mench (2007) mount a spirited argument in favor of not just providing information rich environments, but of specifically providing animals with biologically appropriate challenges, the solving of which results in strong positive affect and associated well-being.

21.3 Positive Affective States and Well-being

Animal welfare scientists have become interested in positive emotions and psychological well-being, placing an emphasis on what increases thriving as opposed to reduces suffering. Boissy et al. (2007) and Yeates and Main (2008) have proposed a change in focus to a broader perspective of welfare measures that increases the emphasis on positive experiences. For example, they encourage assessment of behavioral indicators of contentment, such as play and exploration, self-grooming and maintenance behaviors, and affiliative relationships between animals as well as between animals and their caretakers (Boissy et al., 2007; Hemsworth, 2007).

Zoo and aquarium industry organizations in America, Europe, British Isles, and worldwide (e.g., Association of Zoos and Aquariums [AZA], European Association of Zoos and Aquaria [EAZA], British and Irish Association of Zoos and Aquariums [BIAZA], World Association of Zoos and Aquariums [WAZA]) have adopted frameworks for assessing and ensuring zoo animal welfare. All of these frameworks encompass physical and mental health components, including positive and negative experiences such that one can say, in principle, that there is a wide consensus on the importance of mental health for zoo animal welfare. The EAZA (2019) states that welfare should include ‘the provision of effective veterinary care, meeting dietary requirements, providing individuals with the opportunity to perform their species-specific behavioral repertoire and promoting positive emotional states’. There is, therefore, an emerging recognition in the zoo and aquarium industry of the importance of considering the experiences of individual animals and promoting subjective affective states; the absence of suffering due to pain, fear, or anxiety alone is not enough to ensure good welfare (Whitham and Wielebnowski, 2013). In fact, some scientists argue that positive experiences may offset the impact of negative experiences (Spruijt et al., 2001; Duncan, 2006; Van der Harst and Spruijt, 2007; Yeates, 2011).

Measures of positive affective state potentially can be determined from responses to signaled events (anticipation) (Watters, 2014), cognitive tests that assess optimism about future events (such as cognitive bias tests; Keen et al., 2013), and even pleasure-associated vocalizations (Panksepp and Watt, 2011) and facial expressions (for a review, see Whitham and Wielebnowski, 2013). Boissy and Lee (2014) discuss investigation of affective states in animals using cognitive approaches, such as tests of judgment bias, which have been developed recently for a variety of species (Boissy and Lee, 2014; Bethell, 2015; see also Chapter 23, this volume). Basically, these methods observe how an animal judges the presentation of an ambiguous stimulus cue after having been trained to distinguish presentations of the same stimulus under conditions in which it is unambiguously rewarding or aversive. The interpretation of reactions to ambiguous cues is considered to reflect whether the animal’s emotional state is optimistic or pessimistic. A review of 64 laboratory studies of judgment bias to date indicates that these measures are highly suitable for understudied taxa and can provide new insight into welfare in endangered species housed in zoos and aquariums (Bethell, 2015). The first application of a cognitive bias test to aquarium-housed mammals – bottlenose dolphins – linked the affiliative behavior of synchronous swimming to optimistic judgment of ambiguous cues in a spatial location bias test (Clegg et al., 2017). A study of tufted capuchins (Cebus apella) at a research animal center revealed a positive correlation between levels of stereotypic behavior and a pessimistic judgment bias (Pomerantz et al., 2012). Individuals that exhibited a negative bias while judging ambiguous stimuli had higher levels of stereotypic head twirls and had higher levels of fecal corticoids compared to subjects with lower levels of head twirls.

Studies on anticipatory behavior in expectation of predictable events have revealed links with animals’ affective states (Van der Harst and Spruijt, 2007). Anticipatory behavior is expressed as an increase in activity in response to a signaled reward such as food or enrichment, and demonstrates the perceived importance of a given reward to the animal. Watters (2014) suggests that zoo animal behaviorists should pay attention to anticipatory behavior, and this technique is beginning to be used to study motivation and related affective state of aquarium-housed cetaceans. Clegg and Delfour (2018) recently used anticipatory behavior of bottlenose dolphins, in this case increased frequencies of surface looking and spy hopping, to demonstrate that they perceive nonfood interactions with humans as more rewarding than toy provisioning. They also found that the frequency of anticipatory behavior is correlated to the level of participation in the following event.

21.4 The Importance of Keeper–Animal Relationships for Psychological Well-being

Optimization of mental health in zoo animals demands better understanding of their psychological needs under human management. We maintain that the daily caretakers of zoo animals are best positioned to meet the challenge of determining what animals feel, and that research on the relationships between keepers and animals is beginning to be a fruitful research approach to understanding animal emotions and psychological well-being. Whitham and Wielebnowski (2013, p.256) state ‘it is vital to recognize that animal keepers are a central element of each zoo animal’s environment, and that the quality of a given keeper–animal relationship may influence an individual’s well-being’. Furthermore, interactions between keepers and animals can be a major source of positive experiences and emotions, in effect becoming a form of environmental enrichment for animals (Claxton, 2011). On a regular basis, individual zookeepers feed, clean, groom, shift, exercise, monitor health, train, and provide enrichment to their animals. Developing a positive keeper–animal relationship (KAR) requires becoming familiar with the communication signals and patterns of the animal toward the keeper and of the keeper towards the animal – from the animal’s perspective – and responding appropriately. An important aspect of ‘true relationships’ is that interactions have reciprocal effects on both partners, which may include benefits to the well-being of each (Hemsworth, 2003; Clegg and Delfour, 2018).

In zoos, there has been limited research on the characteristics of positive KARs. Mellen’s (1991, p.99) study of 20 small felid species at eight US zoos was the first to report that ‘keepers who spent a great deal of time with each cat, soliciting contact and talking to the cat’ had significantly better reproductive success among their cats. Similarly, among 72 clouded leopards, stress, as assessed by fecal corticoids, was lower if keepers spent relatively more time around the animals and higher if there was a greater number of keepers working with the animals (Wielebnowski et al., 2002). A colony of captive common marmosets increased play and grooming activities after a program of daily 20-minute, positive interactions with caretakers was added to their husbandry routine (Manciocco et al., 2009). Keepers anecdotally report the emotional benefits of a caretaker just sitting quietly near an animal. Spending more time around animals could function as ‘safety signals’, i.e., a learned cue that predicts the nonoccurrence of an aversive event. Safety signals are potent inhibitors of fear and stress responses (Christianson et al., 2012) and are thought to be disrupted in people suffering from posttraumatic stress disorder (Lohr et al. 2007). Keepers’ activities are often signals to animals that something is imminent that will affect them, be it feeding, cleaning, shifting, separation from conspecifics, etc., so periods of inactive copresence of a keeper provides a direct safety signal that nothing will happen for a period of time. Hence, this provides the animal an opportunity to visually inspect the keeper, approach in their own time, and relax in keeper presence.

Animals may have varying needs for predictable behavior from keepers, and unpredictable actions may cause stress (Bassett and Buchanan-Smith, 2007). Carlstead (2009) explored KARs among 82 keepers working with four different species: black rhinoceros, cheetah, maned wolf, and great hornbill. She found that the animal’s affinity to the keeper (approaching spontaneously and allowing touching, etc.), was modulated by environmental factors that affect keeper predictability such as keeper visibility to the animal, the frequency and timing of feeding, and whether the keeper enters the enclosure space or not. Rimpley and Buchanan-Smith (2013) have shown that making keeper entries into indoor brown capuchin enclosures more predictable and less startling by providing a reliable entry signal (always knocking on the door before entry) reduces anxiety-related behaviors.

To avoid negative interactions and poor relationships with zoo animals, keepers should be aware of and avoid fear-evoking behaviors. For example, at 32 zoos, keepers were videotaped calling to maned wolves and cheetahs from the visitor area, an unusual location for the keeper to solicit interaction. Both species responded with aggressive behaviors more often to keepers that performed a high amount of locomotion when calling, and they approached keepers less often if the keeper made nonverbal noises (clapping, banging on fence) (Carlstead, 2009). When caretakers spent additional time engaging in positive interactions with chimpanzees, such as playing, grooming, feeding treats, and talking, the animals performed higher levels of allogrooming, exhibited fewer abnormal behaviors, and were less reactive (Baker, 2004). Unstructured affiliative interactions between 12 zookeepers and two resident groups of zoo-housed chimpanzees and two groups of gorillas were associated with the animals exhibiting fewer self-directed and abnormal behaviors (Chelluri et al., 2013). However, after interactions with keepers, agonism in the groups increased, leading the authors to suggest that managers should consider practices that minimize the degree to which zookeepers insert themselves in the social environment of great apes.

Finally, zoo animals and keepers may form mutual bonds that may reflect a psychological need for emotional rewards on both sides of the bond. In a survey of zoo staff working with various species, 92% of zookeepers reported having a bond with a zoo animal (Hosey and Melfi, 2012). Furthermore, the benefits to keepers of bonding were reported to be of two sorts: (i) operational benefits that enabled better, easier, or more efficient management of the animals; and (ii) affective benefits that included general enjoyment of the relationship and emotional rewards. An epidemiological analysis of keeper–elephant relationships (KERs) in North American zoos (Carlstead et al., 2019) revealed that bonds between keeper and elephant are associated with welfare benefits for both. Keepers’ attitudes about interacting with elephants were surveyed at 60 zoos and compared to mean serum cortisol responses of the elephants for which they cared; the more positive the keeper’s attitude, the lower the elephants’ mean serum cortisol concentrations, indicating that good KERs are associated with lower stress responsiveness of elephants. When queried about the strength of the bonds keepers have with particular elephants, 70% of keepers reported that they have a very strong bond, and the stronger the bond the less likely keepers were to report dissatisfaction with their job. In light of research in occupational medicine demonstrating a strong relationship between job satisfaction and both mental and physical health of workers (Faragher et al., 2005), keepers likely experience multiple benefits from having a bond with a zoo animal.

21.5 Tools for Enhancing Psychological Well-being: Positive Reinforcement Training

The application of positive reinforcement training (PRT) for zoo animals originated in the 1990s and has grown as an enrichment and management technique ever since (Laule, 2003). Described as ‘occupational therapy’ by Hediger (1950), the benefits of this technique to animals and to husbandry and veterinary management have proven to be numerous. With PRT, animals are reinforced with pleasurable rewards for performing a behavioral response asked for by a trainer. The technique relies on voluntary cooperation by the animal and the animal is not food deprived (Laule, 2003). PRT is welfare enhancing because it provides animals the opportunity to work for food rewards, achieve greater choice and control over daily events, and experience mental stimulation. Also, by using a properly applied PRT training technique called ‘cooperative feeding’ with socially housed animals, introductions can be enhanced, dominance-related problems mitigated, aggression reduced, and affiliative interactions increased (Laule, 2003).

Few studies have been undertaken to examine the long-lasting impact that PRT has on zoo animals’ daily lives. Shyne and Block (2010) examined the effects of PRT on African wild dogs in a zoo, and found reductions in stereotypic pacing behaviors following training sessions. PRT training of chimpanzees decreased abnormal and stress-related behaviors and increased prosocial affiliative behaviors (Pomerantz and Terkel, 2009). Among polar bears at 20 zoos, Shepherdson et al. (2013) found that bears receiving PRT had lower levels of stereotypy and lower fecal corticoid concentrations. All of these studies discuss these long-term effects as being the result of the training process, but evidence suggests that PRT training alters the animal’s relationship with the keeper and removes at least one important stressor associated with the keeper’s actions that might have been causing anxiety-related behaviors prior to training; examples of such stressors include unpredictability, uncertainty, confusion, need for a safety signal, or even lack of attention. Training provides a stimulating human–animal interface (Reinhardt, 1992; Claxton, 2011). Once training begins, communication with the keeper is improved and stress-related behavior declines.

Studies to establish the impact of PRT on KARs are increasing. Savastano et al. (2003) reported positive changes in keeper–animal rapport among callitrichid primates as a result of PRT, including a reduction in keeper-directed aggression and avoidance. Ward and Melfi (2013) reported shorter latencies to approach keepers and reductions in time to perform novel behaviors after PRT in black rhino, zebra, and Sulawesi macaques, concluding that PRT can decrease fear of humans and contribute to positive KARs. In a later paper, Ward and Melfi (2015) demonstrated that each keeper–animal dyad is unique and animals respond differentially to each keeper’s ‘stockmanship style’, which consists of two factors: attitudes towards animals and knowledge and experience of animals. Therefore, PRT is not only an enriching cognitive challenge for the animal, it is also a means of communicating directly with the keeper in a common language.

Recognition of the importance of the caretaker to the animal’s daily life could significantly increase the use of PRT, increase the amount of time the keepers are allowed to spend around animals, and improve positive and reduce negative communication between animal and keeper. We want to emphasize that one of the future directions of zoo animal welfare science is further study of KARs, as evidence is mounting that the benefit to psychological well-being of animals, as well as of keepers, is of major importance.

21.6 Tools for Enhancing Psychological Well-being: Environmental Enrichment

Environmental enrichment has over the last three decades become de facto the primary tool for addressing psychological well-being in zoo and aquarium animals. Compared to other animal contexts, enrichment in a zoo context is permitted to be much more boundless and creative to the extent that zoos perhaps lead the field of enrichment in terms of innovation. However, theory development, formal experimentation, and reporting in the literature may be more restricted than for animals in laboratories and on farms due to the diversity of species kept in zoos, the wide variety of enrichment ideas, and the limited number of subjects in a given zoo. Nevertheless, enrichment methods and successes are widely shared among zoos via scientific publications, newsletters, websites, and workshops.

Many different interpretations of the term ‘environmental enrichment’ exist today and other names have been used to describe this process. The following definition is commonly accepted by the zoo community and the American Zoo and Aquarium Association:

Stay updated, free articles. Join our Telegram channel

Join