Benefits and risks of enrichment

The primary goal of enrichment is to maintain the animal in good physical and psychological health. On entry to a shelter environment, the initial focus is to support the dog’s adaptation to the new environment, to manage stress levels, and to see the return of a normal range of behavior. When confined in a shelter for longer time periods, enrichment can provide the dog with appropriate stimuli to encourage learning and prevent boredom. Dogs in good physical and psychological health will be easier to care for and home and have better welfare.

A central concept in animal welfare (Appendix 3) is “The Five Freedoms,” which succinctly describe the basic needs of all animals (Farm Animal Welfare Council 1992). Young (2003), building on the work of Scott et al. (2000), suggests utilizing the freedoms to generate questions for the assessment of animal welfare. In this way, one can objectively determine what is being provided to satisfy the five freedoms (e.g., rather than “freedom from hunger and thirst,” Young asks if there is “provision of food and water”).

Enrichment strategies can contribute to ensuring that basic needs are met and animal welfare is protected by encouraging species-typical patterns of behavior and enhancing the behavioral repertoire, by preventing, reducing, or eliminating aberrant patterns of behaviors, such as stereotypical behaviors, and by increasing the ability to cope with challenges and reduce “stress” (Wells 2004a).

Concerns have been raised about additional time, resources, and costs that might result from applying enrichment techniques. An increased risk of abnormal or unwanted behavior (e.g., guarding) and threats to hygiene and health have also been raised (Bayne 2005). These risks are low provided the enrichment technique has been risk-assessed as appropriate (Young 2003) and must be balanced against the potential benefits of enrichment. If implemented correctly, benefits of enrichment, such as prevention of undesired behaviors and expression of desired behaviors, which may increase adoption success, should well outweigh the investment of time, resources, and costs (Wells 2004a).

Stress in a shelter environment

Upon entry to a shelter, the novelty of the environment will pose a challenge to most dogs, as their previous environment will, in all likelihood, have differed greatly. Challenges may include an unfamiliar and more restricted housing system; a change in the dog’s daily routine; unfamiliar food, sounds, sights, and smells; the presence of a large number of unfamiliar animals and humans; or the lack of previous attachment figures. All of these challenges are likely to cause “stress.” One definition by Moberg (2000) describes stress as “the biological response elicited when an individual perceives a threat or ‘stressor’ to its homeostasis.” As a consequence of the stressor, the animal will then undergo behavioral and physiological adjustments to avoid or adapt to the stressor and to return to homeostasis. How enrichment can facilitate some of these behavioral adjustments will be discussed later.

When an animal shows a behavioral or physiological adaptation in response to an environmental stressor and as a consequence the stressor disappears, the animal is said “to have control over this environmental stressor” (Mazur 2006). Control over the environment is a key concept to reduce stress and improve welfare. A perceived lack of control over aversive events may lead to learned helplessness and associated motivational, cognitive, and emotional impairment in a variety of species (for a review, see Mazur 2006). On the other hand, perceived control over some nonaversive aspects of the environment, such as water, food, and light, has been shown to reduce emotionality (the observable behavioral and physiological component of emotions) and fear responses in mice and rhesus macaques (Joffe et al. 1973; Mineka et al. 1986). Shelter dogs have only limited control over aversive (e.g., noise levels or aggression from other dogs in the pen) or nonaversive (e.g., food) aspects of their environment. One goal of enrichment strategies in a shelter therefore should be to provide an opportunity to increase the degree of control that a dog can exert. Providing an elevated place to retreat, the ability to move out of sight, and providing food-dispensing devices are some examples of offering control through environmental enrichment.

Short- and long-term stressors and the individual perception of stress

“Short-term stressors” affect a dog immediately upon arrival in the shelter environment, such as the sensory input of an unfamiliar kennel environment, but over time the dog can adapt. On the other hand, “long-term stressors,” for example, social or spatial restriction, may be less apparent upon arrival but become more challenging during long-term stays. If the animal is not able to adapt to the stressor, any short-term stressor can turn into a long-term stressor. In fact, the separation between short-term and long-term stressors is somewhat imprecise, as not every stressor may be clearly assignable to one of the two categories. Despite the concept’s limitations, it is nevertheless useful, as short- and long-term stressors pose different welfare concerns and require different enrichment strategies.

The perception of a threat or “stressor” is subjective and depends on a variety of factors, such as the dog’s genetic predisposition, previous experiences, or age (Stephen & Ledger 2005). For example, a stray dog that has not been socialized to humans may perceive the presence of human caretakers as a significant stressor. In contrast, a dog that has been strongly attached to his or her owners and surrendered may perceive the absence of human contact as the biggest challenge. Therefore, to design an effective stress-reduction program and identify a dog’s individual enrichment needs, consideration must be made of its perception of environmental challenges. To be able to evaluate a dog’s perception of its environment, a good understanding of species-specific behavioral and physiological indicators of stress is necessary, which are described in the next two sections.

Behavioral responses to stress

Behavioral adjustments are often a dog’s first response to a stressor. Depending on the stressor, behavioral adjustments may include avoidance or flight, aggression, behavioral inhibition, or other responses that may be specific to the stressor, such as shivering for temperature regulation. Behavioral responses can therefore be used as one indicator of how stressful an animal perceives its environment to be. The following canine behaviors have been discussed in the context of acute or chronic stress: behaviors associated with fear or conflict, such as paw-lifting, lowered posture, body-shaking, snout licking; increased restlessness; altered behavior patterns (e.g., alterations in exploratory behavior or sleep patterns); behavioral inhibition (e.g., reduced activity levels); repetitive behaviors, such as pacing, wall bouncing, and tail chasing; excessive grooming, self-mutilation, and increased vocalization (Hubrecht et al. 1992; Beerda et al. 1997, 1998, 1999b, 2000; Hiby et al. 2006; Hewson et al. 2007; Dalla Villa et al. 2013). Conversely, affiliative behavior, play behavior, drinking, and grooming have been linked with decreased stress and good welfare in the context of a shelter environment (Hiby et al. 2006; Dalla Villa et al. 2013).

Behavioral responses to stress can vary for short- and long-term stressors. Some behaviors associated with stress appear to be more common in the first few days of the stay in the shelter, therefore likely reflecting the adaptation to the novel shelter environment. For example, in one study by Hiby et al. (2006), panting and paw-lifting decreased over a 10-day period following entry to a re-homing kennel. They also found an increase in behaviors associated with decreased stress, such as drinking and grooming during the same time period. Similar findings are described by Wells and Hepper (1992) who observed that dogs (singly housed) in one shelter ate their food faster on the fifth than on the first day of their stay, possibly reflecting that dogs at that time point were more comfortable in their environment. Repetitive behaviors may be more relevant as an indicator for chronic stress (Beerda et al. 1997, 1999b). For example, when Stephen and Ledger (2005) observed a group of dogs over a period of 6 weeks from admittance to a shelter, they noted an increase in repetitive behaviors, such as pacing and wall bouncing.

Physiologic responses to stress and their measurement

In response to a stressor, dogs react not only with behavioral but also with physiologic adjustments; these may include responses of the neuroendocrine, autonomic nervous, or immune systems. A review of these mechanisms is beyond the scope of this chapter, but as the hypothalamic–pituitary–adrenal axis (HPA axis) has been the primary neuroendocrine axis monitored in the majority of stress studies in shelter dogs, we will discuss some limitations when measuring glucocorticoids to evaluate stress in shelter dogs. For a more in-depth discussion, please see the recent review by Hennessy (2013).

Measuring changes in glucocorticoid levels can be helpful in understanding the impact of short-term stressors, but there are several limitations to this measure. Glucocorticoids show a circadian rhythm with peak levels near the time of wakening in dogs and are released by the adrenal gland in a pulsatile fashion. Further, when measuring acute stress, it is important to consider that handling an animal for sampling itself could cause an elevation in glucocorticoid levels, especially in dogs that are not accustomed to regular handling and if preparation of the dog for taking the sample requires more than a few minutes (Hennessy 2013). Therefore, single glucocorticoid samples alone should never be used as a diagnostic device to assess the stress level of one individual. Glucocorticoid sampling can be useful if repeated samples are taken from an individual over time or if samples are averaged across a group of animals and compared with another group that differs in one independent variable of stress (Hennessy 2013).

Plasma cortisol levels have been used as a measure of stress in shelter dogs, for example, in two studies by Hennessy et al. (1997, 2001). The investigators found that dogs new to the shelter had higher levels of cortisol than dogs having stayed in the shelter for longer than 9 days and that plasma levels of cortisol in individual animals decreased over time. A more recent study by Shiverdecker et al. (2013) used blood cortisol levels of dogs recently admitted to a shelter to compare the effect of three different types of human interactions (petting, play, or exposure to a passive human) against a control group that entered the shelter with no interaction. The authors found that all three interaction conditions reduced cortisol levels as well as behavioral signs of excitation and fear compared with the control group.

The effect of long-term stress on the HPA axis is complex. Glucocorticoid concentrations may remain elevated during more prolonged challenges. Glucocorticoid and glucocorticoid metabolite concentrations have been measured not only in blood but also in feces, urine, saliva, and recently hair (in an attempt to evaluate the effect of chronic stress (Beerda et al. 1999a, 2000; De Palma et al. 2005; Siniscalchi et al. 2013)). However, in the presence of a chronic stressor, the HPA axis may become dysregulated over time, resulting in low glucocorticoid concentrations despite the presence of a chronic stressor. Tests that measure response of the HPA axis to stimulation or suppression, concentrations of hypothalamic and pituitary hormones, or the ratio between glucocorticoids and other hormones may therefore be helpful for the evaluation of chronic stress (Beerda et al. 1999a, 2000).