VICTORIA CUSSEN AND PAMELA J. REID

Anti-Cruelty Behavior Team, American Society for the Prevention of Cruelty to Animals, New York, USA

The realities of animal sheltering have changed drastically over recent decades. Over six million animals are estimated to be held in shelters every year in the United States (American Society for the Prevention of Cruelty to Animals, 2017a, 2018). However, the percentage of the pet population that is free roaming has decreased dramatically, and intake rates have also decreased in some areas of the country (Scarlett, 2013).

A major focus of animal sheltering is reducing the number of animals entering shelters and increasing the proportion of shelter animals that leave alive every year. In this respect, efforts have been impressively successful, with live release rates as high as 96% in some states (Shelter Animals Count, 2016); however, this is highly variable across the country and difficult to quantify accurately (see Scarlett, 2013 for an in-depth consideration of shelter reporting). Transportation of animals from shelters with lower live release rates to other shelters with higher live release rates, or to rescue groups, has increased as differences in excess shelter capacity, driven by intake and release rates, have emerged (Scarlett, 2013).

At the same time, length of stay (LOS) – how long the animal spends in the facility from intake to release – appears to be increasing. Twenty years ago, a national survey indicated dogs spent an average of 9.5 days in the shelter (Wenstrup and Dowidchuk, 1999 as cited in Gunter et al., 2016). A more recent study reported a median LOS of 14 days, with a range from 1 to 71 days (Protopopova et al., 2014). An animal’s LOS is influenced by many factors (Marston and Bennett, 2003), from shelter location (Kay et al., 2018) and behavior (Protopopova et al., 2014), to visitor biases (Brown et al., 2013; Gunter et al., 2016; Protopopova and Wynne, 2016).

However, the overall shelter population picture is obscured because reporting is far from evenly distributed either within or between organizations (Scarlett, 2013). Furthermore, commonly reported metrics logically focus on intake and live release rates and do not include LOS (Shelter Animals Count, 2004). Adding overall duration in shelters or rescues for dogs that are transferred from one to the other, or for animals adopted and then returned, further complicates the picture.

Shelter policies vary and some hold animals for much longer than others. For example, a study of two shelters reported an average LOS of 42 days for adult dogs, with five individuals held for over a year (Brown et al., 2013). Italy introduced national legislation forbidding the euthanasia of medically and behaviorally sound stray dogs, which has created ‘chronic overpopulation’ and welfare concerns in shelters due to life-long holds for the dogs (Cafazzo et al., 2014). Animal sanctuaries are, by definition, intended to hold unplaceable animals for the remainder of their lives (Best Friends Animal Society, 2019).

Other long-term hold situations stem from legal proceedings, such as animal cruelty prosecutions. In the US, animals seized in cruelty cases are evidence in the criminal case and remain the property of the defendant unless there is a voluntarily relinquishment or a court-ordered forfeiture. Until those issues are resolved, the seized animals cannot be adopted and options for humane euthanasia are severely limited (Lockwood, 2006). Due to judicial backlog, laws that do not provide for prompt adjudication of ownership, and/or vigorous litigation caused by defendants, cases can sometimes take years.

The welfare of dogs in shelter has been studied since the 1990s (Beerda et al., 1997, 1998). LOS has important implications for physical health (University of Wisconsin Shelter Medicine, 2015) and also for mental well-being (McMillan, 2013). The effects of chronic stress are cumulative (Mills et al., 2014), meaning welfare impacts are compounded over the duration of an animal’s shelter stay. This can result in a vicious feedback loop where LOS erodes an animal’s welfare, leading to the development of behaviors that make them less adoptable, which increases their LOS, and on and on (McMillan, 2013). These cumulative welfare deficits can have long-lasting impacts on the animals. This chapter will focus on the mental well-being of dogs in shelters, especially with regard to long LOS.

18.2 Relinquishment Reasons with Mental Well-being Implications

The accepted wisdom is that behavior problems are major drivers for relinquishment of pets to shelters (Segurson et al., 2005). This area has received much attention from welfare organizations and applied scientists. In a recent scoping review, Coe and colleagues (2014) identified almost 200 primary articles, reviews, and commentaries on relinquishment – almost all aimed at clarifying why animals are relinquished to shelters.

From the perspective of mental well-being, understanding why animals end up in the shelter environment may tell us what to expect of them while there. Because of differences in population demographics, source, record-keeping in shelters (Scarlett, 2013), and study design and conclusions in research articles (Lambert et al., 2015), knowledge of psychological problems that cause the animal to be relinquished to the shelter is lacking.

Aggressive behavior was cited in 58% of reviewed publications, with moving house and owner’s personal issues cited as the two major nonbehavioral reasons (54% and 50% of publications, respectively; Coe et al., 2014). Caution should be used, however, before assuming dogs end up in the shelter because of pre-existing problematic behaviors that resulted in their relinquishment. In a later meta-analysis from the same institution (Lambert et al., 2015), a large degree of variability was reported across studies, with, for example, moving reported two times as frequently as behavior problems in some studies and the reverse (i.e., behavior two times as frequently as moving) in others.

Further complicating the picture is that the behavioral background of most of the shelter population is unknown at the time of intake. Reasons for this include owners’ reluctance to honestly report certain types of behavior problems (Segurson et al., 2005), and a large subset of animals entering the shelter as either strays (Gunter et al., 2018) or cruelty/neglect victims (Reid and Collins, 2015). Regardless of behavior on intake, animals in shelters do have behavior problems, and those problems can have direct impacts on mental well-being by increasing the animal’s LOS, rendering them less able to cope with the shelter environment during their stay, or both (McMillan, 2013).

Shelter staff report that aggression is the most pressing behavior concern in shelters (Reid and Collins, 2015). Aggressive behavior is undesirable from the perspectives of staff safety and ultimate disposition (the latter because aggressive animals are more likely to be euthanized), but it is not necessarily directly troublesome from a mental well-being perspective. Aggression can be a normal behavior that may be driven by multiple underlying motivational states (Lindsay, 2005). Offensive aggression may not be mediated by a negatively valanced affective state – but if the behavior causes the dog to be more isolated or restricted while in the shelter it could indirectly result in psychological suffering.

Defensive, or fear-based aggression, also has implications for mental well-being, because it is indicative of a negative internal state – i.e., the animal is afraid and attempting to defend itself from perceived threats. The brain circuitry necessary for fear and panic is found across all mammals and animals will work to avoid fear-invoking situations or stimuli (Panksepp, 2011). Fear, either in conjunction with aggression or on its own, is a normal emotion – but one with a negative valence. In dogs, chronic states of fear cause psychological suffering, increase risk for certain diseases, and decrease life span (Dreschel, 2010).

Fear of strangers is a commonly reported behavior cited by owners relinquishing their animals (Segurson et al., 2005). Fear across multiple contexts is prevalent in animals that did not receive appropriate opportunities for social interactions in early life (McMillan, 2013). Because of inappropriate housing and/or aversive early life experiences, fear can also be experienced to such a degree that the animal cannot function in the normal day-to-day life of a companion animal (Miller et al., 2018). Approximately one-third of 3,000 dogs seized from cruelty cases exhibited extreme, debilitating fear (Cussen and Reid, unpublished data). While extreme fear necessitates specialist behavior interventions such as is provided at the American Society for the Prevention of Cruelty to Animals’ (ASPCA) Behavioral Rehabilitation Center (Miller et al., 2018), even mild to moderate fear is also psychologically detrimental and, sadly, likely more widespread across shelter populations.

18.3 Shelter Environment Aspects with Mental Well-being Implications

While fear may not be the direct cause of relinquishment, it is likely experienced by a significant proportion of the shelter population. Reid and Collins (2015) called fear ‘the most insidious behavior concern among shelter dogs’. They explained that fearful dogs may be quiet and withdrawn (Reid and Collins, 2015) and may still be viewed as easily adoptable by shelter staff. This could result in a situation where fearful dogs essentially ‘suffer in silence’ during their shelter stay.

Classic fear behaviors include panting, pacing, and trembling (Overall et al., 2001). Freezing and hiding behaviors are also associated with fear and are shown in response to threatening stimuli such as predators (Fureix and Meagher, 2015). More subtle indicators include tongue flicks, looking away, and a slightly lowered head or body carriage (Horwitz and Mills, 2009). These behaviors are frequently displayed by dogs in the shelter environment, but not all volunteers or staff may pick up on them (Flint et al., 2017). It is important that all staff and volunteers are trained in body language for the species they are working with in the shelter. Even those staff not working directly with animals benefit from such training, because it allows them to adjust their own behavior based on the animal’s behavioral indicators of emotional discomfort before the situation becomes overwhelming to the animal. For example, a facility’s staff moving equipment past a row of kennels could notice stress indicators and slow down to reduce noise and minimize stress.

Shelter programs dedicated to the behavioral well-being of their animals are an important and necessary resource to address psychological suffering in fearful and other psychologically distressed animals (Reid et al., 2004; see also Chapter 24). But there are simple interventions any staff or volunteers can do, such as the ASPCA’s ‘Drive by Treats’ protocol where all people offer or drop treats into kennels as they walk past to create positive associations with activity outside the kennel (see Reid and Collins, 2015).

It is important for shelters to include psychological suffering and behavioral interventions in calculations of capacity for care. One recent example of time allocation at a facility indicated the required staff hours for minimum standards of care (based on a 15-minute-per-animal-per-day minimum standard) was between two to four times less than what the population size warranted and led to periods where staff time averaged 2.5 minutes per animal per day (Koret Shelter Medicine Program, 2019). This is clearly inadequate to provide for anything but the most basic physical needs of the animal and cannot provide for their psychological well-being. Fearful animals will be especially impacted by a lack of human attention, given the generally stressful nature of the shelter environment, which will be briefly summarized next.

18.3.1 Shelter environment and ‘overall’ welfare

While sheltering has advanced significantly in recent decades (Scarlett, 2013), it still holds true that dogs in shelters endure spatial and social restrictions because of practical realities of housing large numbers of animals. There is widespread concurrence across sheltering professionals and shelter-based researchers that the shelter environment is not designed to provide for the mental well-being of animals long term (Fig. 18.1).

Captivity is known to expose animals, including domesticated species, to a wide array of stressors, including loud noises, smells, restricted movement, forced proximity to humans and conspecifics coupled with the inability to interact directly, and others (Morgan and Tromborg, 2007). Most research has concerned the effect of the shelter environment or management on animals’ behavior and overall welfare. The general approach is to compare some type of management change – e.g., space allowance, single or group housing, human interaction – and see how it impacts animals’ behavior (Taylor and Mills, 2007a). Beerda and colleagues (1997, 1998, 1999a,b) did a series of seminal studies in the late 1990s that established references for physical and physiological manifestations of ‘stress’ in dogs. The measures are still commonly used in shelter-based research but typically not in a way that allows interpretation of results (see Hennessy, 2013 for a review of the difficulties). For example, the finding of a null effect on cortisol is ambiguous: was the animal unresponsive (and therefore not stressed) or was the hypothalamic–pituitary–adrenal axis hyporesponsive due to chronic stress (Hennessy, 2013)?

Results to date for behavioral, physiological, and immunological outcomes of shelter-based research are variable and, at times, contradictory (Part et al., 2014; Protopopova, 2016). Multiple reviews have been written on the subject, and readers are referred to them for more information (Taylor and Mills, 2007a). The equivocal results on shelter animal behavior likely result in part from looking for effects at the group, rather than the individual, level (Taylor and Mills, 2007a). Additionally, multicenter studies would be a welcome addition to the literature, as researchers using this approach found that the random variability between facilities explained more of the variation in the data than any of the wide range of management factors under study (Kiddie and Collins, 2014). This is not surprising, given the typically small sample sizes and/or restricted number of observation points in shelter research. From a practical perspective, shelter managers should be cautious about adopting widespread management changes based on one or two studies or successes at other shelter facilities.

That said, the preponderance of evidence supports some aspects of the environment as being important to an animal’s welfare. As noted in the introduction, the current leaders in shelter medicine agree the LOS is the largest risk factor for illness in shelter animals (Koret Shelter Medicine Program, 2019). Long LOSs are known to cause within-individual changes in behavior that may be indicative of negative affective states, such as spending more time at the back of the kennel (Wells et al., 2002a). In the authors’ experience, the same is true for psychological suffering – the longer animals spend in the shelter the more likely they will display behavior that is indicative of poor mental well-being.

Restricted movement of mammals is known to be a powerful stressor and is linked with stereotypical or repetitive behaviors across a wide range of species (Mason, 2010). Studies on exercise in kennel-housed dogs do not consistently find that exercise per se is enriching (Jongman et al., 2018), but conclusive evidence is limited because exercise is usually confounded with increased space, social enrichment, and other factors (Cafazzo et al., 2014). The effect of exercise on shelter dog behavior is usually framed within the paradigm of increasing adoptability (by promoting behavior desired by humans). Because of that perspective, research focuses on short-term changes in behaviors related to kennel presence – as opposed to determining if exercise can improve the mental well-being of the animal directly (Protopopova et al., 2018). The authors are not aware of any investigations where shelter animals were given access to treadmills or other similar devices that would allow for increased exercise while holding other variables constant. This could be a promising area for future shelter-based research, as exercise is known to reduce stress in dogs generally (Horwitz and Mills, 2009) and allowing individuals to self-regulate could promote mental and physical health (see Section 18.4).

Noise is another known stressor in the shelter environment. It interferes with dogs’ ability to rest and sleep (Fullagar et al., 2015), elicits a physiological stress response (Beerda et al., 1998, 1999b), and is related to poor postoperative recovery rates in human medical patients (Hekman et al., 2014; Fullagar et al., 2015). Dogs in shelters are frequently exposed to noise levels over 100 dB (Coppola et al., 2006a) and persistent exposure over 3 months led to hearing loss in shelter dogs (Scheifele et al., 2012). Design of shelters should incorporate noise dampening elements into the construction plans (Coppola et al., 2006a). Most shelters do not have the luxury of creating new facilities, and so pragmatic approaches are needed. Since much of the noise is created by the barking of the dogs themselves, decreasing arousal is a high priority in shelters. Limiting visitor access during the day can significantly lower noise levels and promote resting behavior in shelter dogs (Hewison et al., 2014), and many shelters now use ‘quiet times’ where all staff and volunteers have to be off the floor for a period during the day. These strategies could also be leveraged to decrease frustration and increase predictability, which would be beneficial for the mental well-being of shelter dogs (see Section 18.3.2). However, human interaction also has benefits, so balancing restful periods against social enrichment is important.

Providing social enrichment through human interaction consistently changes behavior and physiological indicators of stress in dogs in shelter (see Pullen et al., 2012 for an overview) (Fig. 18.2). Even opportunities for passive proximity to humans outside the home kennel decreased vocalizations and panting – compared to control dogs that were removed from their kennels but did not have human interaction – as much as petting or playing with the dog during the session (Shiverdecker et al., 2013). Relatively short sessions can be as effective as interactions twice as long (Willen et al., 2017). Dogs seem to prefer to interact with familiar people (Pullen et al., 2012) but also appear to benefit from contact with unfamiliar people (Coppola et al., 2006b).

There is some evidence that the beneficial effects of human interaction are short-lived (Willen et al., 2017) and influenced by the dog’s history (Pullen et al., 2012). Because length of interactions vary between studies there is no rule of thumb for how best to set up human interaction. Maximizing time spent with the dog may increase welfare and provides an opportunity to gather information about that animal’s temperament (Coppola et al., 2006b). Care should be taken to avoid one-size-fits-all mandates for interactions because individual differences in personality and life history can affect whether dogs actually enjoy, and therefore benefit from, interacting with humans. Much of the human-interaction work is framed in an adoptability paradigm (see Pullen et al., 2012), so the long-term effects are largely unknown. One study found that areas of the dog’s brain associated with reward were activated by smelling the scent of a familiar – but not an unfamiliar – human (Berns et al., 2015). Many shelters use olfactory enrichment already (Miller and Zawistowski, 2013). Incorporating scents specific to humans familiar to the dog may be a novel way to extend the benefit of social enrichment after the dog returns to the kennel, but this has yet to be studied.

Social enrichment through interaction with conspecifics is another powerful way to offset the stress of the shelter environment. Providing dogs opportunities for social interaction with conspecifics through pair housing is known to reduce stress (Grigg et al., 2017). Dogs can be separated for feeding or enrichment if there are concerns about aggression, but aggression occurs at very low levels and is typically agonistic as opposed to actual fighting (Taylor and Mills, 2007a). Precautions against disease transmission should be included in social enrichment protocols (Miller and Zawistowski, 2015). Another way to provide social enrichment is playgroups. Playgroups are used by shelters to teach dogs appropriate social skills and promote positive mental health (American Society for the Prevention of Cruelty to Animals, 2017b). Depending on resources and shelter population, playgroups vary in their complexity (e.g., dyads versus group play) and formality (e.g., volunteer versus behavior staff, information collected during playgroup). The unifying idea is that interacting with other dogs is a powerful method for reducing the effects of shelter stress (Sadler, 2018). Anecdotal reports support the benefits of using playgroups (Dogs Playing for Life, 2018), but the effects on in-kennel behavior outside the play yard are yet to be systematically studied. Play has been extensively studied in a range of animals as a potential indicator, or even cause, of both positive emotions and welfare (for a review, see Held and Spinka, 2011). In general, play behavior decreases as welfare decreases, but some forms of play are known to increase during periods of stress (Held and Spinka, 2011). For example, in both humans and animals, increases in solitary play (parallel play or object-directed), as opposed to interactive play, may indicate stress and negative mental well-being (Ahloy-Dallaire et al., 2018). Propensity for play in dogs is known to vary with breed (Svartberg, 2006; Asp et al., 2015) and temperament (Starling et al., 2013). Play in dogs is generally thought to increase social cohesion but it is also heterogeneous – i.e., forms include solitary, intraspecific, and interspecific – and types of play are mediated by different factors (reviewed by Sommerville et al., 2017). For example, intraspecific play may be more competitive than interspecific play (Bradshaw et al., 2015). These points have practical implications, inasmuch as absolute recommendations for amounts of play in shelter, or viewing all play as good play, are likely misleading. Here again, documenting baseline and within-individual changes in behavior is necessary to track how well animals are coping with the shelter environment. Collecting and pooling such information across shelters could be a powerful way to identify risk factors for a variety of states associated with negative mental well-being (e.g., hyperarousal or anhedonia, see Section 18.4).

Importantly, the subjective experience of the shelter environment is influenced by the animal’s previous experience. Hubrecht and colleagues (1992) found that dogs were less negatively impacted by individual housing in a kennel if they had experienced adverse social or thermal conditions in their housing prior to indoor kenneling. After entering a shelter, behavioral indicators of habituation increased over the first 10 days (increased self-care, decreased panting and paw lifting) but a physiological stress indicator (cortisol to creatinine ratio) of owner-surrendered dogs increased over time, whereas strays showed the reverse (Hiby et al., 2006).

The physiological results may be explained by differences in adversity prior to intake: strays may be exposed to more physical stressors, but the physiological indicators did not relate to ‘stress’ behaviors within individuals (Hiby et al., 2006). Similarly, a within-subjects design study found that prior habituation to the kennel in a home environment significantly lowered the magnitude of the dogs’ stress response when they entered a training kennel compared to the control group dogs. Further, those dogs with no prior experience failed to adapt to the kennel environment, even after 3 months (Rooney et al., 2007).

While at the time of intake owner-surrendered dogs did not differ significantly from strays when comparing hair cortisol levels (a measure of long-term cortisol release), they did not show a physiological response to human interaction, whereas stray dogs showed an acute decrease in cortisol (Willen et al., 2017). That may be counterintuitive, but dogs in shelters have been shown to form attachments more quickly than pet dogs (Thielke and Udell, 2018) and to seek proximity to humans, especially familiar humans, more than laboratory-reared dogs (Pullen et al., 2012). Owner-surrendered dogs may be more distressed due to the loss of a stable attachment figure (Tuber et al., 1999). So, perception of and response to various environmental aspects likely differ between dogs with varied histories and or personalities (see Section 18.4).

18.3.2 Ontogeny in a shelter environment

A special note is warranted on pregnant bitches and puppies whelped in animal shelters. Regardless of LOS, puppies born in a shelter environment are indirectly impacted by their mother’s prenatal and postnatal stress and are directly exposed to a stressful environment during an important developmental period. Therefore, the impacts are outsized in relation to the LOS and could potentially impact the puppies’ mental well-being for their entire lives (Fig. 18.3).

The importance of early postnatal development periods is well known (summarized by Scott and Fuller, 1965). There are breed-specific differences in ontogenic timing (e.g., of fear, see Morrow et al., 2015). Puppies need to experience some degree of novelty so that, as adults, they can cope with the everyday things experienced by pet dogs (Estep, 1991). Controlled and gradual exposure during early development is important for reducing fear and anxiety in dogs (Rooney et al., 2016). Environments that fail to meet puppies’ developmental needs can cause a range of behavior problems including fear, reactivity, impulse control deficits, and anxiety (Hammerle et al., 2015). The negative impacts are especially pronounced when dogs endure inappropriate conditions from a young age (McMillan et al., 2016).

Although direct comparisons cannot be made between shelters and commercial breeders or hoarding cases, puppies in shelters face many of the same challenges, such as prenatal maternal stress, restricted exposure to environmental and/or social stimuli, and maternal separation (reviewed by McMillan, 2017). Hetts and colleagues (2005) recommended developing proactive behavioral health programs with criteria that reflect mental health (see also Chapter 10). Howell and Bennett (2011) specifically addressed possible approaches to ensuring the behavioral health of puppies raised in shelters. Ensuring the appropriate behavioral development of those puppies could potentially stave off development of behaviors that would cause them to be returned to shelters later in their life. This represents an opportunity to protect mental well-being twice in an animal’s life by using interventions that occur at a single point of development.

18.4 Psychological Environment

While much shelter-based research has examined aspects of the animal’s physical and social environment there is little in the way of addressing what Taylor and Mills (2007a) call the ‘psychological environment’. This aspect is arguably of the most importance for the subjective experience of the animal, because it involves agency and choice (Meehan and Mench, 2007). Agency – or lack thereof – may be the most directly relevant aspect of the shelter environment as concerns mental well-being (see Chapter 6 for more on personal control in animals). A sustained absence of control, in conjunction with forced proximity to unpredictably available rewards or aversive stimuli, can lead to extreme frustration or apathy (Brando and Buchanan-Smith, 2018).

18.4.1 Hyperarousal and frustration

As discussed above, shelters are loud, with much of the noise coming from the dogs themselves. Sounds associated with food preparation and people walking past the kennel fronts tend to trigger high arousal behaviors such as barking, spinning, and wall bouncing in kenneled dogs (Wells and Hepper, 2000; Denham et al., 2014). Hyperarousal is a common complaint in the shelter and several studies have looked at training low-arousal behaviors such as sitting, eye contact, and not barking in kennel to increase adoptability (see Reid and Collins, 2015 for an overview of this research) (Fig. 18.4).

High arousal behavior is likely maintained, at least partially, by a conditioned emotional response to the appearance of people and high-value reinforcers such as time outside of the kennel, human interaction, or food delivery (Protopopova et al., 2018). To a point, increasing anticipatory behaviors can indicate a positive subjective state, because the animal has an expectancy of obtaining an impending reward and is not suffering from anhedonia (Spruijt et al., 2001; van der Harst and Spruijt, 2007). Above a certain point, however, excess anticipatory behavior may appear almost manic, and may indicate a sensitization to reward (Bassett and Buchanan-Smith, 2007). Reward sensitization can occur when the environment offers so few rewards that any reward becomes highly salient and the animal is neurologically primed to have a large emotional response to it (Spruijt et al., 2001; Cabib, 2006). The degree of the anticipatory response could be a potential metric of affective state in shelter dogs (van der Harst and Spruijt, 2007). This is an area that remains unexplored and is likely linked to individual personality or coping style.

Dogs in shelters repeatedly see and hear other dogs and humans, with no means to interact with them, which can lead to frustration (Taylor and Mills, 2007a). All mammals, and possibly all vertebrates, are thought to share several ‘core’ emotional systems, including seeking, rage, joy, and others (Panksepp, 2005). The ‘seeking’ system supports the expression of goal-directed behaviors such as foraging. Animals in shelters may be motivated to engage in appetitive behavior, such as seeking social interaction, but due to the physical restrictions of the kennel environment they may be unable to reach the consummatory phase of behavior, e.g., interaction with a person or a conspecific. The core emotional systems in the brain are interconnected by inhibitory and excitatory synapses. Repeated frustration of goal-directed behavior causes excitation of the ‘rage’ system, which is associated with aggressive behavior in animals (Panksepp, 1998). Whether or not this mechanism underpins the development of barrier aggression or reactivity in shelter dogs is not known. However, frustration and conflict are two known risk factors for repetitive behaviors in companion animals (Luescher, 2009).

Stereotypical behavior is a subset of repetitive behavior. Repetition is necessary, but not sufficient, to classify a behavior as stereotypical. A dog that spins two or three times in response to environmental triggers and/or anticipation (Denham et al., 2014) is different from an animal that paces invariantly in the absence of environmental triggers (see Mason, 2010). In practice, however, shelter-based research tends to conflate the two, making study comparisons and interpretation of results difficult (Protopopova, 2016). Physical or chemical prevention of repetitive/stereotypical behavior is not advised, because it is likely some of these behaviors serve as a coping mechanism to help the animal deal with a suboptimal environment (Mason and Latham, 2004). However, behavioral indicators of chronic stress in shelter animals are still unsettled (Hewson et al., 2007). This is in part because of the loose definitions mentioned above, but it is also likely because there are multiple mechanisms underpinning stereotypies (Mason and Latham, 2004). Altering known risk factors, including the animal’s physical, auditory, and/or social environment (see Mason and Rushen, 2008 for a complete overview), should be the first step in addressing these behaviors in the shelter.

18.4.2 Anhedonia and helplessness

The flipside to reward sensitization is the loss of neurological or behavioral responsiveness to environmental rewards. This phenomenon is called ‘anhedonia’, or a loss of enjoyment in previously enjoyed activities or rewards, and has been used in animal models of human depression (Fig. 18.5). Outside of the laboratory, however, anhedonia can be difficult to distinguish from relaxed resting if relying only on behavioral measures (Mason and Latham, 2004). For example, dogs suffering from a depressive-like state probably spend the majority of their time lying at the back of their kennel, much the same as would a dog that has adjusted to the shelter environment and is lying down to relax (Wells et al., 2002a; Mason and Latham, 2004).

Stay updated, free articles. Join our Telegram channel

Join