Why Measuring and Building Resilience Is Applicable to Zoo and Aquarium Animal Welfare

Abstract

In recent years, animal welfare scientists working in professionally managed settings have increasingly focused on promoting resilience to enhance the quality of life of individual animals. Resilience—defined as an animal’s capacity to be minimally affected by a disturbance or to rapidly return to the physiological, behavioral, cognitive, health, affective, and production states that pertained before exposure to a disturbance—involves various systems and dynamic processes. There is evidence that resilience can be measured using a suite of species-specific indicators, including both behavioral measures and physiological biomarkers. These indicators should be tracked for individuals of the same species over time and across various conditions, events, and experiences. Large-scale, multi-institutional studies allow welfare scientists to collect cross-sectional data to identify “resilient phenotypes” for the species of interest. Ultimately, the focus should be on improving outcomes for individual animals as they face particular stressors, challenges, and environmental disturbances over their lifetime. Animal care specialists play a crucial role in helping animals build resilience by providing opportunities to engage in cognitive challenges, stimulating environments, and species-appropriate social interactions. This review defines resilience for animal welfare scientists, as well as discusses how to measure and promote resilience in animals residing in zoos and aquariums.

1. Introduction

1.1. What Is Resilience?

Welfare scientists working in professionally managed settings are continually searching for ways to apply novel concepts, tools, and measures to improve husbandry practices, management routines, and ultimately the quality of life of individual animals. Recently, there has been increased focus on promoting resilience to enhance animal welfare [1,2,3]. Although we regularly and colloquially use the term “resilience,” it is crucial to define the term as it is should be employed by welfare researchers. In 2025, the American Psychological Association defined resilience as: “the process and outcome of successfully adapting to difficult or challenging life experiences, especially through mental, emotional, and behavioral flexibility and adjustment to external and internal demands” [4]. As applied to non-humans, resilience has been defined as, “the capacity of the animal to be minimally affected by a disturbance or to rapidly return to the physiological, behavioral, cognitive, health, affective and production states that pertained before exposure to a disturbance” [1] (p. 1962). Resilient animals may be capable of exhibiting their typical physiological and behavioral responses even when faced with novel challenges [3]. Indeed, in comparison to more vulnerable conspecifics, highly resilient animals may be less sensitive to disturbances, as well as more likely to recover quickly from adverse situations [1]. In general, higher systemic resilience is associated with lower morbidity and mortality, as well as a better quality of life [2].

Animal welfare scientists studying resilience are interested in an individual’s ability to regularly adapt to various challenges, including physical, social, and environmental stressors [1,2,3]. Resilience—which is not simply inherited but can be developmentally acquired across the lifespan—involves dynamic, complex processes and subsystems [2]. Ultimately, an individual’s resilience is determined by the resilience of these various subsystems. Fortunately, as we discuss in detail later in the review, it is possible to improve and build resilience [1,3]. Enhancing resilience not only minimizes harm to the individual but also promotes, “…functional competence to thrive that can provide eudaimonic wellbeing, positive health and positive welfare outcomes to the animal” [3] (p. 1440). There is evidence that positive, stimulating experiences and enriched environments can confer resilience [1,5,6,7]. It is also possible to build resilience by exposing animals to events and short-lived stressors that are congruent with their natural history, life history, and daily husbandry routine/conditions [1]. Ultimately, individuals can become more resilient by gaining agency, which can be achieved by overcoming challenges, problem-solving, and learning to predict and control the environment/stressors [1].

To examine resilience, researchers must consider the disturbance (typically episodic and context-specific), response to this disturbance, and the outcome [1,2]. The disturbances that animals face can be characterized by their frequency, severity, and type [2]. In terms of type, disturbances may involve macro-environmental factors (e.g., disease outbreaks, ambient temperature, nearby construction, exposure to visitors in zoos) or micro-environmental factors (e.g., social interactions, social introductions/separations, special diets) [2]. It is important to note that while environmental change is not inherently negative, animals can be challenged by environmental variation that is difficult to control and/or predict [3,8,9]. Furthermore, individual animals differ in their abilities to cope with variation and overcome challenges [10]. These differences impact how the individual responds or reacts to disturbances or events. Individual variation in resilience may stem from various genetic and non-genetic factors, the latter of which include both protective factors (e.g., social support/interactions, appropriate maternal care) and risk factors (e.g., genetic factors, early maternal separation for extended periods, poor maternal care, illness, traumatic events) [2,11,12]. For example, there is evidence that social interactions can offset risk factors and increase resilience in both rodents and primates, including humans, reviewed by [12]. In other words, while the outcome for one individual may be to tolerate or adapt to a disturbance, another individual may be more vulnerable or susceptible to this stressor. Ultimately, “the perception of animals to environmental stimuli is influenced by historical events, the current state of well-being, and cognitive or behavioral capacity,” with the outcome being resilience or vulnerability [2] (p. 19; see Figure 1).

As mentioned above, resilience involves various systems and dynamic processes [2,11,12,13]. For instance, brain derived neurotrophic factor (BDNF) is considered a biomarker of resilience, with low levels being associated with depression in humans and animal studies reporting a relationship between chronic stress and decreased expression of BDNF in the hippocampus [11]. Furthermore, there is evidence that rats (Rattus norvegicus) considered resilient to chronic mild stress exhibit increased levels of BDNF mRNA and that higher levels are negatively correlated with measures of anhedonia, i.e., the inability to experience pleasure and joy [11,14,15]. When considering the glutamatergic system, the expression of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors in the dorsal hippocampus differs significantly between vulnerable and resilient animals, with GluR1 and GluR2 being differentially regulated [2,11,12,16]. In addition, there is evidence that the peptide neurotransmitter, neuropeptide Y (NPY), may serve as a protective factor by modulating stress responses [17,18]. For humans participating in military survival training, higher blood NPY levels were associated with better performance, less psychological distress, and fewer symptoms of dissociation [19,20]. Research on humans also shows that the hypothalamic–pituitary–adrenocortical (HPA) axis plays a role in resilience, with recent studies suggesting that the ratio of dehydroepiandrosterone (DHEA) to cortisol (or DHEA sulfate to cortisol) may serve as a pro-resilience factor, reviewed by [17,18]. Finally, it should be noted that the medial prefrontal cortex (mPFC) plays a key role in developing resilience, or experience-driven resilience, by receiving/integrating information from the amygdala, hippocampus, and other structures [2,11]. Fortunately, the plasticity of the mPFC and other systems allows for resilience to be modified, which both reduces harm and improves functional competence, ultimately promoting better health and positive welfare [3]. A more thorough discussion of the processes and subsystems associated with resilience, including genetic, epigenetic, and developmental factors, can be found elsewhere [2,17,18].

The majority of research on resilience has been conducted on livestock, with a focus on both animal welfare and productivity [1]. Some of the earliest studies focused on resistance to parasitic infestation, as well as resistance to the negative effects of infection, e.g., [21,22]. More recently, the concept of resilience has been expanded to examine how individuals respond to environmental and social stressors [1,3,12,17,18]. For example, researchers have investigated resilience in the context of exposure to both low and high temperatures [23,24,25], weaning [26,27], management practices [28,29,30,31,32,33,34], food shortage [35], and diet [36].

Finally, it is important to note that resilience should not be confused with related terms such as adaptation, allostasis, coping, and homeostasis (

Table 1. Related terms used in the animal resilience literature. Definitions adapted from Colditz & Hine [1], Basile et al. [2], and the Animal Behavior Management Alliance’s glossary [39].

Term	Definition
Adaptation	Limited to a specific type of stressor. The acclimatization of the individual to certain environmental conditions. Does not indicate flexibility in successful adaptation to all new challenges over a lifetime.
Allostasis	The ability for systems to maintain physiological stability as changes occur. A dynamic, proactive process of adjustments to the internal environment so the body can adapt to changing conditions. Internal variables are modified to prevent errors, with stability being achieved through change.
Coping	The ongoing strategy or processes an animal employs when reacting to stressors. Behavioral, cognitive, and physiological responses to stressors. These stressors may reflect internal or external demands. Coping strategies help minimize the impact of stress and determine the degree of resilience or susceptibility.
Desensitization	The process of decreasing an organism’s reactivity to a stimulus.
Habituation	A progressive decrease in the intensity or probability of an elicited response that occurs as a result of repeated exposure to the eliciting stimulus.
Homeostasis	An individual’s self-regulatory capacity that allows the internal environment to remain constant despite variations in the external environment. Feedback mechanisms allow for individuals to maintain the status quo in terms of physiological regulation. A reactive process aimed at restoring balance and maintaining a stable internal environment.
Resilience	The capacity of the animal to be minimally affected by a disturbance or to rapidly return to the physiological, behavioral, cognitive, health, affective and production states that pertained before exposure to a disturbance. Involves dynamic and complex processes that allow for active, positive adaptation to new conditions and adverse events. Not limited to a specific type of stressor and relies on the reaction of the animal to stressors. Tends to be expressed in response to environmental challenges lasting a matter of days.
Systematic Desensitization	The process of exposing an individual to an anxiety-provoking stimulus in a stepwise manner while a state of relaxation is maintained. The goal is to gradually increase the intensity of the stimulus until the maximum-intensity stimulus does not elicit anxiety. Throughout the process it is important to always keep the stimulus below the threshold that causes anxiety.

). For instance, while animals may adapt or become acclimated to a particular stressor or disturbance, “adaptation” does not capture the individual’s ability to flexibly adapt to novel challenges over time and across diverse situations. An individual’s resilience is a trait that is consistently demonstrated across various contexts and challenges over the lifetime [1]. In fact, resilience has been specifically contrasted against “adaptation”, with some researchers suggesting that the biological costs of resilience may be lower than those of adaptation [37,38]. The term “coping” is also sometimes used interchangeably with resilience, though the former is defined as the ongoing strategy or processes—i.e., behavioral, cognitive, and physiological responses to internal and external demands—that ultimately determine the degree of resilience or susceptibility [1,2]. Finally, the terms “allostasis” and “homeostasis” are sometimes confused with one another, as well as resilience. However, the former reflects a dynamic, proactive process of adjustments to the internal environment to maintain physiological stability as conditions change, while the latter is more of a reactive process aimed at restoring balance and maintaining a stable internal environment despite changes in the external environment [1,2]. Finally, it should be noted that resilience cannot be trained and should not be confused with terms commonly used by animal care specialists such as desensitization, systematic desensitization, and habituation. In zoos, animals may habituate to the presence of visitors or the sound of food buckets, while trainers may design training programs/processes to decrease an animal’s reactivity to anxiety-provoking stimuli (desensitization). These terms are further defined in Table 1.

The current review article examines how to apply the concept of resilience in zoos and aquariums. Specifically, we discuss how to measure and build resilience in animals living under professional care, with the ultimate goal of enhancing the welfare of these individuals. We should note that this is the first article that focuses on defining resilience for the zoo community, as well as providing specific recommendations for promoting resilience in zoo-housed animals.

1.2. How to Measure Resilience in Professionally Managed Animals

Basile and colleagues [2] explain that because resilience is a process that involves, “a set of capacities of a system,” it can be challenging to measure directly. As a result, researchers generally incorporate animal-based measures that reflect resilience to environmental disturbance, such as behavioral measures, physiological biomarkers, and performance variables or indices of productivity (e.g., milk yield). The goal is to gain insight into the biological underpinnings of resilience by examining the individual’s perception of their environment/disturbances, their immediate stress response, and any adaptive responses [1]. Colditz [3] emphasizes that researchers should identify “parameters of individual normalcy” that capture an animal’s experience of their micro-environment, as individual differences exist even when there has not been an obvious disturbance to the shared macro-environment [40]. If possible, these measures should be tracked on a regular, if not continuous, basis and should capture stress-inducing events (e.g., changes in diet, social introductions/separations, weaning). Furthermore, when an animal is exposed to such challenges and stressors, the researchers should not only consider the magnitude of the response but also the “rate of recovery” to the individual’s baseline state for the variable of interest—whether it be behavioral, physiological, immunological, cognitive, etc. [1]. For example, it is important to examine both how much cortisol levels increase following a stressful event (e.g., veterinary exam), as well as how long it takes for levels to return to baseline. It is important to note that while longitudinal data allows the researcher to estimate deviation from an individual’s typical pattern, cross-sectional data estimates deviation from what is typical at the population-level [3,41]. After all, the concept of resilience is ultimately comparative, measuring the differences in outcomes following individuals’ exposure to a particular stressor or situation [1]. Ultimately, measures of resilience can provide insight into the individual’s welfare status, physical health, and biological functioning as the animal faces challenges.

Colditz [3] outlines various approaches to measuring resilience. Most commonly, resilience has been quantified by considering the performance trajectory of an individual animal for a particular trait related to production. This involves assessing production and measuring this against the performance of other animals. However, studies have shown that resilience and production are not equivalent [3]. In fact, a study of milk yield in dairy cows (Bos taurus) reported that resilience and productivity (i.e., milk yield) were negatively correlated [31].

It is also possible to assess adrenal activity and the reactivity of the HPA axis [12,17,18]. In humans, childhood trauma is associated with a hyperactive HPA axis later in life, and elevated cortisol levels have been reported for individuals diagnosed with major depressive disorder [17,42]. As noted above, there is evidence that DHEA and its sulfate ester (DHEA-S)—collectively referred to as DHEA(S)—as well as the ratio of DHEA(S) to cortisol, can provide insight into resilience and stress vulnerability [17,18,43,44]. In humans, research on post-traumatic stress disorder (PTSD) patients suggests that DHEA or the ratio of DHEA:cortisol may be a resilience factor, reviewed by [17]. For instance, while PTSD patients have elevated DHEA levels overall, higher levels are correlated with better coping and symptom improvement, and a higher DHEA:cortisol ratio is associated with less severe symptoms [45]. Some animal welfare researchers have also referred to the ratio of glucocorticoids:DHEA(S) as a potential resilience biomarker [46,47,48]. Trevasin and colleagues [47] found that the cortisol:DHEA ratio was elevated in pigs adjusting to a novel environment following transportation to a new facility. Similarly, an experimental study in dairy cows reported an increase in the cortisol: DHEA ratio for dairy cows who experienced deteriorating conditions [46].

Another approach to assessing resilience involves measuring an individual’s immune response when exposed to stressors [26,49,50]. Indeed, researchers can examine immune competence (i.e., the immune system’s absolute level of performance) in the context of stress-inducing situations and husbandry practices [3,26,50]. For example, Aleri and colleagues [49] discovered that in the context of routine husbandry (i.e., handling), immune competence and cortisol responses were inversely correlated in dairy cows. Specifically, as compared to their counterparts, subjects with below-average antibody and cellular immune responses following the administration of a commercial vaccine had higher serum cortisol concentrations after handling. Furthermore, antibody response was negatively correlated with internal parasite burden [49]. Previous studies have also reported a relationship between immune competence and disease outcomes [34,51,52].

It is also possible to measure behavioral responses to challenges and stressors. For example, individuals demonstrating heat resilience might perform specific behaviors, such as increasing water intake, reducing food intake, and resting in a shady area [53,54]. In general, reduced behavioral complexity may be indicative of reduced resilience to environmental stressors [3,55,56,57]. For instance, when exposed to a highly stressful situation (i.e., food limitation) chickens (Gallus gallus domesticus) displayed reduced complexity of locomotor behavior sequences [35]. Colditz and Hine [1] also suggest integrating vocalizations, and specifically the normality of vocalizations, into studies of resilience. Indeed, recent research has highlighted the potential of using vocalizations to evaluate responses to changes in the environment and/or routine [58,59]. Finally, researchers can attempt to gain insight into an animal’s affective state by measuring the “normality of demeanor” in studies of resilience [1]. Certain behaviors (e.g., exploratory behaviors, play behaviors, positive vocalizations, affiliative behaviors), as well as behavioral diversity, have specifically been used as indicators of positive affective states [6,60,61,62,63].

Colditz and Hine [1] argue that researchers should not hone in on one biomarker or behavior when measuring the “resilience syndrome.” After all, a given variable may change in response to one stressor but not another and may change only for individuals with certain coping styles. In fact, these authors suggest “that a diagnostic signature for resilience will be more easily identified by measuring summary characteristics of response variables rather than by attempting to identify correlative relationships between the numerous labile variables that change in diverse ways in differing challenge scenarios” [1] (p. 1972). Indeed, some researchers advocate for developing cumulative scores of resilience to examine an individual’s overall welfare status [29,64]. It is important to note that the suite of variables used to measure resilience may differ across species, especially because it may not be feasible to collect certain types of data non-invasively. Aside from the measures discussed above, Colditz and Hine [1] provide an overview of variables that may be useful for identifying “resilient phenotypes” including heart rate, heart rate variability, and core body temperature, which can be tracked using non-invasive or minimally invasive methods [65]. Other options include feed intake and growth rate [1].

To identify species-specific indicators of resilience, Colditz and Hine [1] recommend exposing individuals to an adverse event that: (1) the animals already encounter in their daily routine or environment, and (2) has both a social component and a novelty component. For example, zoos and aquariums could take advantage of management decisions such as changes in group composition and/or housing to identify appropriate indicators for measuring resilience (e.g., food consumption, vocal behavior, behavioral diversity). Such studies allow researchers to identify and characterize resilient phenotypes for the species of interest [1].

As welfare scientists conduct studies on resilience, they should be prepared for various challenges that may arise. Some limitations associated with measuring resilience include: (1) deciding how to incorporate a cumulative scoring system to avoid honing in on a single measure, (2) determining which events are stress-inducing (but will not result in distress) for particular individuals, and of course, (3) consistently monitoring individuals over the course of their lives. The latter is particularly challenging, as zoos and aquariums must invest both staff time and resources to regularly collect, analyze, and share data with other facilities.

1.3. How to Build Resilience in Professionally Managed Animals

If welfare scientists have the tools and resources to measure resilience, the next step is to increase efforts to build resilience in professionally managed animals. It is important to recognize that there are a variety of ways to promote resilience and that resilience can be achieved by different mechanisms for different individuals [1]. Please note that while this article does not consider the role of genetic selection, research on livestock has increasingly focused on identifying and selecting for resilience traits. For instance, there is evidence that heifers with greater resilience to environmental challenges have higher reproductive success (i.e., more likely to become pregnant, higher pre-weaning survival of offspring) [66]. Further information regarding genetic selection, including how to epigenetically reprogram resilience, can be reviewed elsewhere, e.g., [11,18].

There is evidence that resilience can be enhanced by not only providing a comfortable environment but also by offering enrichment—both cognitive and emotional [1,3,5,6,7,11]. In fact, exposure to enriched, stimulating environments early in life can confer resilience in the future. For instance, Chapillon and colleagues [67] reported that as compared to mice (Mus musculus) reared in standard environments, those reared in enriched environments (with tunnels, toys, running wheel, larger cage size, various height levels, a hut), were less fearful of new environments when tested post-weaning. Specifically, the latter took less time to enter novel environments, spent more time exploring, and demonstrated fewer anxiety-like behaviors (e.g., a stretched posture) [67]. Various studies have reported that offering enrichment is associated with greater behavioral diversity, a reduction in stereotypic behaviors, an increase in activity levels, lower glucocorticoids, and even an increase in behavioral indicators of positive affective states (e.g., play, exploration) [68,69,70]. Having the ability to explore and master one’s environment helps promote agency, build eudaimonic well-being, and foster resilience [5,6].

Indeed, animals should be given opportunities to overcome challenges, make choices, and learn to control the environment [6,7,71,72,73,74]. It is possible to confer “stress inoculation” by offering controlled, graded exposure to challenges and adverse situations that are part of the animal’s life history, natural history, and husbandry environment [1,5,6,75]. In other words, resilience can be improved as individuals learn how to predict, control, and reduce their exposure to stressors. Even though challenges in the environment may elicit short-term frustration or stress, an “appropriate challenge” [73] allows the individual to build competencies (e.g., strategies, skills) to address future challenges [6]. Indeed, there is evidence that presenting zoo animals with opportunities to make choices and exert control can be protective, serving as a buffer against stressors. In fact, there is evidence that gaining agency is associated with various positive outcomes, including: (1) a decrease in abnormal behaviors, inactivity, aggression, and HPA activity, as well as (2) an increase in exploratory behaviors and behavioral diversity, reviewed by [72]. Moving forward, one suggestion is that animal care specialists give individuals the ability to control how they attain the food/items that they like/want [7].

It may also be possible to promote resilience by ensuring that animals, and particularly social species, have the opportunity to engage in species-appropriate social interactions [12]. There is evidence from primates (both human and non-human) and rodents that social interactions can help an animal build resilience, serving as a protective factor against genetic and environmental risk factors [12]. In male rats, subjects exposed to a stressor (i.e., a loud white noise) were more likely to seek social contact and engage in social interactions with conspecifics than rats who did not experience this stressor [76]. Furthermore, subjects who were observed in groups displayed less immobility, a behavioral sign of fear, when exposed to this aversive stimulus than those who were alone [76]. These findings highlight the importance of cultivating environments and identifying management practices that promote social buffering to counter stressors/challenges. There is evidence that social buffering can positively influence endocrine function, immune function, cardiovascular health, and longevity [12,77,78,79,80,81,82]. For instance, in female prairie voles (Microtus ochrogaster), females who recovered from immobilization with their male partner exhibited no elevations in corticosterone or anxiety-like behaviors, while subjects who recovered alone demonstrated an increase in both [82]. Although the loss of a close relative was associated with increased fecal glucocorticoid levels in female Chacma baboons (Papio hamadryas ursinus), subjects also displayed an increase in grooming rates and the number of grooming partners over the following three months [79]. The authors argued that this “broadening and strengthening” of grooming networks may help explain why glucocorticoids returned to baseline soon after the loss [79] (p. 710). Finally, the unpredictability of social play can foster the development of long-term skills such as conflict-solving, emotional flexibility, and general stress resilience [60,83]. As Pellis and Pellis [83] point out, play helps animals prepare for the unexpected, with different actions and movements allowing individuals to experience the loss of control and unpredictable outcomes. As a result, animals may be more emotionally resilient when dealing with an unpredictable world [60,83].

2. Discussion

How can we apply what has been discussed above regarding measuring and building resilience to: (1) the research conducted in zoos and aquariums, and (2) refining best practices related to management routines and housing?

Let us consider the first question: how can zoos and aquariums measure resilience? First, facilities should attempt to create baseline profiles for individual animals by collecting the following on a regular basis: (1) behavioral data (e.g., 10-min of instantaneous sampling, 2–3 times per day, 2–3 times per week), and (2) physiological data (e.g., collecting fecal samples three times per week to calculate weekly averages for glucocorticoids, DHEA(S), etc.). Data should be collected for several weeks per year, across different seasons. This will allow animal care specialists to accumulate longitudinal data for the individuals in the collection. Furthermore, these behavioral and physiological profiles can be compiled across facilities to create reference intervals for individuals of particular age-sex classes. For instance, by analyzing data from 40 adult chimpanzees residing at 16 zoological facilities, Whitham and colleagues [84] were able to: (1) generate reference intervals for various behaviors and physiological measures, as well as (2) identify significant differences related to both sex and age for a handful of behaviors. Moreover, multi-institutional studies can help welfare scientists gain insight into how members of a particular species typically respond to events and conditions that commonly occur in zoo environments (e.g., concerts, construction, late-night events). Indeed, these large-scale studies can provide the data necessary for identifying species-specific resilient phenotypes [1]. For example, following a stressor (e.g., introduction to a new habitat), some individuals’ behavior may not deviate significantly from baseline, or perhaps rates/proportions of time spent performing certain behaviors (e.g., self-plucking) return to baseline within days rather than weeks. Moving forward, when another individual of this species faces similar events or conditions, data can be compared to the resilient phenotype. Additionally, if an individual were to encounter similar stressors/challenges over their lifetime, welfare scientists can determine whether the individual becomes less reactive, less likely to display indicators of negative welfare, or more likely to display indicators of positive welfare, etc. Tracking responses over time would be especially informative if the facility was actively working to build resilience. In fact, this ongoing monitoring would allow staff to develop an awareness of individual variation in resilience that could benefit the management strategies guiding conservation, breeding, and reintroduction programs.

That leads to the next question: how can zoos and aquariums promote resilience? First, facilities should continue to construct, modify, and enrich habitats to stimulate their residents. Animal care specialists should be involved in the planning process when designs and blueprints are drafted for new habitats and renovations. The goal is to provide opportunities for individuals to make choices, overcome challenges, and control the environment [6,7,70,71,72,73,74]. Fortunately, many zoos and aquariums now design animal habitats with built-in enrichment features and automated experiences that allow animals to control their environment (e.g., motion detectors, feeders) [85,86]. Animals residing in zoos and aquariums can be given access to carcasses, puzzle feeders, feeding logs/trees, and other devices/structures that stimulate individuals by encouraging them to “hunt” or forage for their food, e.g., [87]. Indeed, there is evidence that offering foraging challenges (e.g., puzzle feeders) at variable intervals can have positive behavioral impacts and be cognitively enriching [88].

As discussed above, for social species, resilience can also be fostered by increasing opportunities for social buffering [12,76]. In zoos and aquariums, this can be accomplished by ensuring that both the physical environment and management practices related to housing promote species-appropriate social interactions. Fortunately, facilities accredited by the Association of Zoos and Aquariums (AZA) must meet “social grouping” standards for particular taxa [89]. Furthermore, animal care specialists can determine whether certain individuals benefit from the presence of conspecifics when facing particular challenges. For example, an individual may be less reactive to large crowds, or perform fewer negative behaviors, when they share a habitat with conspecifics than when in the habitat alone. Ultimately, the question of whether social interactions and social buffering will help promote resilience will be influenced by several factors, including familiarity of the conspecific, hierarchical relationships, the responses/demeanor of the conspecific, and the nature of the relationship (e.g., partners, siblings, parent-offspring, etc.) [12,90]. Indeed, while some conspecific pairs may simply tolerate one another and rarely interact, other pairs may develop complex, mutually beneficial, and enriching relationships.

One zoo-based study that demonstrates how to begin measuring and building resilience was conducted by Glaeser and colleagues [91]. Over the course of this 4-yr project, Asian elephants (Elephas maximus) were monitored before and after being transferred to a new habitat using behavior assessments, hormone monitoring, and GPS data. The new habitat was designed to be more spacious, complex, and flexible, with the goal of offering more foraging opportunities, choices, and challenges. While fecal glucocorticoid metabolites (FGMs) peaked during construction, concentrations did return to baseline in the new habitat. In the new habitat, the elephants were more active, walked farther, foraged more, and spent more time exploring. Although the herd exhibited healthy social dynamics and species-typical behaviors in both habitats, individuals had more autonomy in terms of choosing how to utilize the habitat and who to interact with socially, given that they had regular access to indoor/outdoor areas and more space for creating distance. This study highlights how facilities can: (1) collect baseline data in preparation for major events that are presumed to be stressors (construction, transfer to a new habitat), (2) continue data collection post-stressor, and (3) design habitats that promote agency and offer challenges.

It should be emphasized that it is not necessary for zoos and aquariums to undertake large-scale renovation projects or construct new habitats to gain insight into the factors that may impact resilience. For instance, by monitoring FGMs in 26 fishing cats (Prionailurus viverrinus) living across 16 AZA-accredited facilities, Fazio and colleagues [92] identified potential stressors, as well as potential protective factors for this species. The researchers found that FGMs increased following events such as facility transfers, immobilizations, and aggression among breeding pairs. However, management factors such as providing access to off-exhibit holding areas, more formal training sessions, and social housing arrangements resulted in lower FGM concentrations. While it would be preferable to integrate other types of monitoring (e.g., behavioral observations) into studies like this, Fazio and colleagues [92] provide an example of how to conduct a multi-institutional study that captures baseline data and then tracks individuals’ responses to various challenges over time. Such studies can also help researchers identify resilient phenotypes.

3. Future Directions

What Next Steps Should Welfare Researchers and Animal Care Specialists Be Taking?

Facilities can start by creating behavioral and physiological profiles for individuals of the species of interest. This can be accomplished by developing a “toolkit” of species-specific indicators (e.g., vocalizations, behavioral diversity, DHEA(S):glucocorticoids). Over time, data can be compiled across facilities, with multi-institutional projects allowing for the development of behavioral and physiological reference intervals for particular age-sex classes. When facilities collaborate, they can also gain insight into how members of a species typically respond to specific challenges. For example, animals residing in zoos may repeatedly experience late-night events and concerts, as well as management-related changes such as rotation into different habitats, temporary social separations, and social introductions. Large-scale studies that consider how individuals respond to these challenges—both in terms of the magnitude of response and rate of recovery—can help us identify resilient phenotypes. Furthermore, future studies should consider the ethical implications of resilience research, including how individuals of the same species have different capacities or starting points for building resilience, and hence “appropriate challenges” must be carefully selected. After all, there is the possibility that attempts to elicit short-term frustration or stress could actually lead to fear or distress.

Perhaps most importantly, animal care specialists play a crucial role in helping individual animals build resilience on a daily basis by providing stimulating, enriching environments and appropriate social interactions. Ultimately, improving resilience will benefit individual animals as they face stressors and challenges across their lifetime. Researchers can track which outcomes (e.g., an increase in DHEA(S):glucocorticoids or in positive vocalizations) improve for individuals over time. Zoo-based resilience research will help us identify how resilience-building strategies vary across individuals of the same species, as well as across taxa.

4. Conclusions

Moving forward, welfare researchers working in professionally managed settings should increase efforts to both measure and build resilience in individual animals. To gain insight into an individual animal’s resilience to stressors and challenges, researchers should attempt to identify a suite of species-specific indicators that can be measured using non-invasive or minimally invasive techniques [1]. This “toolkit” for monitoring resilience should include both behavioral measures and physiological biomarkers. These indicators should be tracked for individuals of the same species, over time, and across a variety of conditions, events, and experiences. By conducting large-scale, multi-institutional studies across AZA-accredited facilities, researchers can collect cross-sectional data to begin constructing a “phenotype of resilience” for the population [1]. Ultimately, the goal is to regularly collect data on individual animals as they confront challenges and environmental disturbances that are common in professionally managed environments. Researchers should attempt to capture the individual’s perception of their environment and these disturbances, including their immediate stress response and any adaptive responses [1]. To do so, it can be informative to measure both the magnitude of response and the rate of recovery for the variable of interest. While comparisons can be made to the resilient phenotype, the focus should be on improving outcomes for individual animals.

Fortunately, animal care specialists can promote resilience, and hopefully make individuals less vulnerable to disturbances and stressors in various ways. For some individuals, increased opportunities to engage in cognitive challenges may help promote agency and foster resilience [5,6]. For others, being housed with certain conspecifics may be beneficial as they face everyday challenges. As Boissy & Lee [7] point out, alterations to an individual’s experiential environment or to the management routine can be, “…simple and inexpensive, with the potential to have strong impacts on animal welfare…” [p. 107]. In other words, animal care specialists and welfare scientists can work together to introduce practical and affordable changes, or even to maintain certain conditions (e.g., housing that promotes social buffering/support) in an attempt to foster resilience. Ultimately, improving resilience will benefit individual animals as they face stressors and challenges across their lifetime.