Equus caballus and Equus przewalskii

Mitochondrial deoxyribonucleic acid (mtDNA) studies suggest that a divergence occurred as long ago as a million years,^2,3 between the modern horse (E. caballus) and its more numerous and diverse predecessors. The link between them was thought to be the pony-shaped Mongolian Wild Horse, Przewalski’s Horse (E. przewalskii). Apart from Caspian ponies that appear to be polymorphic in the diploid number (some having 64 chromosomes while others have 66),⁴ E. caballus has 64 chromosomes. Equus przewalskii, on the other hand, has 66 chromosomes in the diploid state. However, the recent publication of the horse genome⁵ has shown that E. caballus is not only related to E. przewalskii, they are essentially the same species.

Table 1.1 Classification of equidae

This explains the similarities in their serum proteins and their blood groups and why cross-breeding between E. caballus and E. przewalskii produces fertile offspring. Geneticists explain that a single fusion mutation in the haploid state (32) could have brought about their differences. Similar intra-species chromosomal dimorphism occurs in the Asian ass (Equus hemionus).⁶

Subsequent isolation of the E. caballus and E. przewalskii gene pools could perhaps have arisen because of capricious selectivity on the part of Caballus stallions (see Ch. 11) and the retreat of Przewalski herds into regions of the steppe uninhabited by humans.⁷ While outnumbered by the similarities between the two species, the differences are fascinating. For example, the striking bistability of hock joints that allows a very rapid switch from one extreme position to the other is much more marked in domestic horses than in Przewalski horses and, for that matter, zebras.⁸

Przewalski horses became extinct in the wild in the 1950s.⁹ However, from a nucleus of 11 foundation animals, they survive today in captivity and in successfully re-introduced wild populations, e.g. in Mongolia.^10,11 The survival story of the Przewalski horse is an extraordinary one, and we are indeed fortunate to be able to study them in a variety of contexts. It is fascinating that we now have data to refute the notion that Przewalski horses are socially more aggressive than domestic horses.¹² However, studies of their behavior should be treated with some caution since they have emerged from a shallow gene pool that has been filtered through 20 generations of captivity. Therefore before assuming that present-day Przewalski horses behave as true wild horses, we should bear in mind that some would regard them as survivors of the first stages of domestication. In this text, I have selected examples of feral horse behavior rather than Przewalski horse behavior as benchmarks for what is normal. Discussions of such free-ranging behavior appear in each of the subsequent chapters.

Reasons for stabling horses

Horse owners throughout the world house their animals for a number of reasons. Among these, the most important is often the need to spare limited grazing areas from damage underfoot, especially in regions of and in times of heavy precipitation. The danger of disease associated with exposure, such as ‘rain-scald’ and ‘mud-fever’, in these regions can be reduced by providing field shelters and appropriate prophylaxis. Therefore, the economics of pasture management can prevail over horse welfare when owners consider stabling their horses for the winter.

By stabling their horses, owners eliminate food-stealing by dominant conspecifics and can have more control over food quality and intake. Food and water intakes are easier to monitor, with an eye toward early detection of disease, in the stable rather than the field. During the winter, time spent mucking-out and bedding a horse can be offset by the time saved by not having to groom a thick layer of mud off the horse prior to exercise. Clipping, made more feasible by the fact that the horse is sheltered, adds to this advantage since clipped horses sweat less in response to heavy work than their counterparts in a full winter coat and are therefore easier to clean.

Stabling may be favored by the owners of horses that are difficult to catch when at pasture. The limit on kinetic activity that is imposed by stabling is considered advantageous by trainers of performance horses since they can more easily control the amount of daily exercise taken by their charges. Post-inhibitory rebound in locomotory behavior after periods of confinement⁷⁴ may give trainers the impression that horses want to work.

Health advantages may accrue from stabling since injuries can be more effectively rested, worms can be better controlled and flies are less numerous than at pasture. Apparent advantages for the horses of being stabled include shelter (from sun, wind, rain and flies), freedom from bullying and a reduced physical requirement to work (e.g. forage) for food. This book should help readers decide whether these benefits outweigh the costs paid by the horse as it loses choice and control over social and ingestive activities.

Feeding

Horses are notoriously wasteful grazers (see Ch. 8). Many farmers remark that, in wet weather, when one considers the damage done by its unforgiving hooves, a horse is equivalent to the activity of five bovine mouths. The reluctance of owners to allow this damage means that time spent by horses at pasture in temperate climates is often limited during the winter months. While sparing the horses the pain of disorders such as mud fever and dermatophilosis, this intervention often has a deleterious impact on their behavior and nutrition.

Traditionally, horses are fed long fiber from a haynet or hayrack. This is intended to reduce wastage from contamination with urine or feces and possibly to decrease the likelihood of endoparasite transmission. However, there is growing evidence that this unnatural foraging position can have a deleterious effect on the efficacy of the mucociliatory escalator in clearing the upper airways of inhaled particles (Fig. 1.10), especially those inhaled from dried foodstuffs (which are to some extent unnatural in themselves).⁸² Similarly, haynets reduce the space available in the stable, may increase the risk of the horse becoming snared (e.g. by trapping its hoof), and elevate the forage so increasing its potential as a source of ocular foreign bodies. It is also argued that feeding from a net or rack may adversely affect muscles and nerves in the neck.⁸³ Perhaps this is why, when financial considerations are almost insignificant relative to the value and maintenance of performance (e.g. in the majority of racing yards⁸⁴) horses are fed roughage, whether it be hay, haylage or silage, from the stable floor.

Figure 1.10 Dust plumes from a haynet as a horse forages. The position of the food and its contamination with fungal spores are unnatural. The harmful effects on respiratory health are well recognized.

(Reproduced by permission of the University of Bristol, Department of Clinical Veterinary Science.)

There are marked seasonal variations in feeding practices according to the weather and the availability of pasture. We do well to remember that the incidence of colic follows a similar pattern, with changes from fresh grass to a dried diet being associated with rises in various enteropathies, especially impactions. The use of concentrated feeds and the periods of fasting with which they are associated have been linked to unwelcome consequences, including increases in gastric acidity that can result in rapid ulceration.⁸⁵ Compared with those who ride mainly for competitive reasons, those who ride mainly for pleasure are less likely to feed cereal-based foods according to the manufacturer’s recommendations during summer, when grass is available.⁸⁶

Bedding

For a variety of reasons, including availability, ease of disposal and tradition, straw remains the favored bedding for horses.⁸⁶ Although it raises concerns for some about the risk of impaction and chronic obstructive pulmonary disease (COPD), straw seems to have the added advantage of being the bedding preferred by horses themselves.⁸⁷ It may be worth noting that straw is regularly used as horse food in developing countries.

Depth of various bedding types that is needed to maximize their appeal to horses has not been studied. However, it has been noted that horses lie down on deep litter bedding rather more than do horses at grass or on daily mucked-out beds.⁸⁸ The common practice of providing limited bedding to save money can affect choice by reducing the extent to which the horse can comfortably lie down. Houpt⁸⁹ notes that if bedding is present but inadequate, horses tend to lie down as soon as more bedding is supplied.

Whereas horses have evolved to be very sociable animals, many owners feel that their charges should be given solitary quarters to make bullying less likely. At pasture, horses choose their affiliates and will spend time interacting with them, for example, while playing, mutually grooming,⁶³ or settling reasonably minor disputes. On a stable yard, managers tend to dictate the distribution of horses with efficiency of service in mind rather than the associations established between pairs of horses at pasture. It is a shame that there are so few data on the effects of separating horses from preferred companions, or exposing them to agonistic approaches from neighbors whom they would normally avoid.

Watering

While some military horses are still communally watered on a three-times-daily basis, modern textbooks of stable management rarely commend the limited availability of water. However, the maxim that food should not be given before work and that water should not be offered after food probably represents a considerable limitation on the choice that stabled horses can exercise in terms of drinking behavior.⁹⁰

Stable design

There are few publications on the design of stables, and these have been based largely on extrapolations of recommendations for shelter and ventilation of agricultural species. The Universities Federation for Animal Welfare (UFAW)⁹¹ states that for a horse of average size – 500–600 kg bodyweight and 1.5–1.6 m (15–16 hands) – the loosebox should be at least 4 × 4 m and preferably 4 × 5 m. Ensminger⁹² states that except for foaling mares and for stallions, there is no advantage in having box stalls larger than 3.6 m square. Evans et al⁹³ note that the popular size is 3.6 × 3.6 m and that 3 × 3 m is adequate for young horses but suggests that the more time the horse spends in the stall, the larger the stall should be. There are several differences between the housing of horses compared with that of other farm animals. Horses usually have individual living areas, and a much greater labor input per animal is present in stables than in farm-animal housing. A survey of racing stables in the Southwest of England found that floor space varied between 8.76 m² and 21.8 m² with a median of 12.1 m² per horse.⁹⁴ This compares favorably with the floor space provided for other agricultural species per unit of body size.⁹⁵

While box designs have been suggested by agricultural engineers and by horse-lore, there appear to be no recommendations in management texts about the amount of time that the occupants of these quarters should spend within them on a daily basis. There are a number of reasons for protracted confinement, including inclement weather, locomotory illness and isolation of contagious pathogens. Episodes of enforced confinement are popularly noted as being contemporaneous with the onset of stereotypies. Given that Evans⁹³ advocates the provision of more space for animals that are turned out less regularly, it could be argued that the design of all stables should meet the needs of the worst-case scenario that precipitates withdrawal of the daily turn-out period.

The traditional layout of a stable-yard includes the quadrangular enclosure of a central lawned area by boxes whose doors and windows face inwards. Walkways in many yards pass close to these portals and expose stabled horses to the unpredictable and arousing movement of humans, feed-buckets and conspecifics. Therefore, it has been suggested that regimented geometrical accommodation of this sort makes relaxation difficult for the occupants.⁶³

Despite similar problems with the distracting effects of activity in the passageways and others to do with air hygiene, barn-style housing (in which horses are housed individually, under one roof, in pens made largely from bars and grilles), which facilitates communication between neighboring horses, is increasing in popularity. Studies on the effect of increasing visual contact with conspecifics on weaving^96,97 (see Ch. 5) suggest that whether a stable, rather than a tie-stall, is acceptable depends on its walls more than its size in that isolation is more of a problem than confinement.⁸⁹ Group housing of young horses allows more normal social development⁹⁸ and reduces the risk of stereotypies^99,100 when compared with individual housing.

Modern stable management brings with it a number of disadvantages for both owners and horses. For the owner, the disadvantages of keeping their animals in stables rather than at grass revolve around time and money. Time commitments in stable management include bedding and feeding as well as having to exercise the confined horse on a more regular basis than its grazing equivalent in order to maintain a given level of fitness. The disadvantages of stabling from the horse’s perspective are considered later.

Feeding behavior

Because behavior is a response to an organism’s environment, the more restrictive an environment is, the more limited are the choices available to the organism. It is possible that where choice is limited or eliminated, welfare may be compromised.¹⁰⁴ Choice allows animals to perform the behaviors that are important to them,¹⁰⁵ although the choices they make are not exclusively in the direction of their own welfare.¹⁰⁶ While the debate about the importance of environmental choice in the welfare of animals continues, there appears to be a number of ways in which modern stable management limits choice.

Choice is certainly reduced in feeding behavior when horses are stabled. Feeding behavior in stables seems to show the most marked difference from that at pasture since concentrated rations may be consumed more rapidly than a pure forage diet (Fig. 1.11). While the feral or pastured horse may spend 70% of its day foraging, stabled horses on ‘complete diets’ may spend only 10% of their time feeding.¹⁰⁷ These diets for competition or maintenance can be eaten in less than 2 hours and have removed the feeding behavior of stabled horses even further from its evolutionary origins. In addition, modern diets for the stabled horse can be monotonous. There is compelling evidence that giving stabled horses variety in their forage ration and distributing various forages throughout the stable can help to bring their time budgets closer to normal.¹⁰⁸

Figure 1.11 Time budgets for horses in a variety of environments: (A) free-ranging Camargue horses, averaged over a year; (B) eight horses housed as a group with ad libitum hay and straw; (C) three horses in individual stables given ad libitum hay and straw and able to see and touch each other; (D) horses fed restricted fiber in stables where they cannot touch each other and only see each other over stable doors; (E) six crib-biters fed ad libitum hay and able to see and touch each other. NB: Crib-biting accounts for 20% of these horses’ ‘other’ behaviors.

(Charts (A)–(D) reproduced with permission of M Kiley-Worthington from The Behaviour of Horses in Relation to Management and Training.)

The time at which food is made available often adheres to a strict regime, which may have variable implications for gastrointestinal function. The designation of such feeding times may address the convenience of the operator rather than the needs of the horse. Horses have not been observed to fast voluntarily for more than 3–4 hours¹⁰⁹ but spells of this sort are imposed on many stabled animals.

It has been shown that after 2–3 weeks of continuous access to a single feed, ponies stabilize their bodyweights, consuming 2–3% of their bodyweight in dry matter in a 24-hour period.¹¹⁰ Despite this finding in support of ad libitum feeding of housed equids, stable managers prefer to control intake, since it is known that to do otherwise is to risk temperamental volatility and metabolic disorders, e.g. laminitis.

There is considerable anecdotal evidence for the effects of diet on behavior in the horse,^1,111 but little has been done to determine the relative importance of factors such as energy, protein and fiber levels. (The relationship between diet and behavior is discussed in more detail in Chapter 8.)

It has been argued that the sleep pattern of a species can be used as an index of adaptation.¹¹² It is therefore interesting that, while drowsing accounts for 8% of the resting time of stabled horses, this figure has been shown to rise to 14% at pasture.¹¹³ The concomitant dietary differences between intake at grass and in the stable may play a role in this phenomenon since a preliminary study in ponies has shown that, when oats replaced hay in the diet, total rest time increased.¹¹⁴ This merits further scrutiny (see Ch. 10). Although the number of feeds per day can affect behavior via factors such as disturbances and arousal,¹¹⁵ it could be that restlessness in horses peaks when stabled animals are fed small amounts of forage and after consuming them are left with little to do in so-called vacuum periods.

Social behavior

The importance of social behavior to stabled horses has yet to be fully quantified with consumer demand studies. However, anecdotes that describe horses performing operant responses, such as undoing bolts with their lips to escape from the stable (Fig. 1.12) may indicate that housed individuals will perform work in order to return to their conspecifics.

Figure 1.12 (A–D) Irish Draft Horse undoing the bolt of his stable door. (E) After a chain was secured across the threshold to prevent his escape, he moved his hay so that he could watch activities on the yard while eating.

It is known that frightened riding horses may bolt in a bid to return to their fieldmates, apparently because of an innate preference for the presence of equine company.¹¹⁶ There are considerable data to suggest that isolation can be aversive.^117–120 When the choice of equine neighbors in a stable yard is dictated by the manager, bonded affiliates may be separated while individuals with mutually low tolerance may be housed next to one another. Disruption of an established social structure in this way may be associated with heightened aggression, especially at times of concentrated food delivery.¹²¹

Generally, the opportunities for social interaction with conspecifics, favored or otherwise, are minimized by the individual housing of horses. While olfactory, auditory and visual communication can often occur between stabled neighbors, tactile communication, which is of considerable importance in groups of horses,^122,123 is rarely possible in most stable designs.^94,115 Similarly, interactions such as mutual fly-swatting cannot be performed by isolated horses. These concerns are of particular importance when tie-stalls are considered, with evidence to suggest that they do not sufficiently cater for horses’ needs to perform social, recumbent resting and allogrooming behaviors.¹²⁴

Stables of all sizes when designed for individual horses have the potential to conflict with many of their occupants’ survival instincts.¹³ While imposing the vulnerability of isolation and an unhelpful concentration of excretory products that may attract predation, they prevent detection of predators and escape. Until we cost out the value of resources or the absence of threats to safety and comfort, it behooves us to consider these impositions from the horses’ perspective. For example, while stallions and occasionally geldings create stud-piles of feces in the stable, eliminative behavior in the stable differs markedly from that in the paddock, since the occupants cannot easily avoid contact with their urinary and fecal waste. Because fresh layers of bedding regularly prompt urination in male equids, some consider this a form of marking.⁶³ So, the extent to which bedding is managed to meet human rather than equine needs bears consideration but it is only one feature of the stabled horse’s world that represents a removal of choice. In many ways horses attempt to meet their behavioral needs despite the shortcomings of intensive environments but may fail to do so because of management routines.

Kinetic behavior

Among the prevailing features that arise in intensively managed horses is restricted space. The physical limitations imposed on a horse by stabling mean that kinetic behavior is more difficult to perform than it is at pasture. For example, it has been estimated that an average sized horse needs a 6-m span to roll from one side to another, so many stabled horses are unable to perform this most basic of maintenance behaviors. The relative lack of space either may prevent the stabled horse from choosing to roll or may precipitate casting. Confinement within looseboxes is common and is an important limitation on the amount of exercise that a horse can take through choice. After periods of confinement, horses show post-inhibitory rebound^74,125 that may explain why unwanted behaviors during training occur more frequently among stabled horses than those at pasture.¹¹⁶

It has been suggested⁵³ that voluntary kinesis indicates that, rather than being simply a substratum for most behavior, locomotory activity has a motivation of its own.⁷⁴ However, the possibility of motivation for spontaneous locomotion is a contentious area because there is often difficulty in eliminating motivation to perform other behaviors for which locomotion is a prerequisite, e.g. intrinsic exploration.¹²⁷

Although unacquainted horses from different yards may be mixed on a national and international scale for equestrian competition, isolation is often practiced when new horses arrive on a yard, in a bid to control the spread of pathogens. Should such animals perform an unwelcome behavior pattern, they will often remain in isolation, since stereotypies are traditionally regarded as contagious.

Space is certainly restricted when horses are housed, particularly in the case of tie-stalls. The extent to which stabled horses can see conspecifics is restricted. In a traditional loosebox, this can often be achieved only by looking over the stable door, while many tie-stalls restrict visibility further by having tall dividers between neighboring pens. This view is a contentious one. Marsden¹²⁸ maintains that the behavior of horses kept in tie-stalls is similar to that when they are kept at pasture. This was based on the observation that, in tie-stalls, horses showed no significant increase in the time they spent performing abnormal behavior. The proximity of neighboring horses was thought to facilitate social behavior more than is the case in individual looseboxes. It may be that the small sample size (n = 4) could have resulted in the inadvertent selection of horses that were especially unreactive and not predisposed to perform stereotypies. This work did, however, indicate that stalled horses spent significantly less time lying down and moving, with significantly more time standing in stalls than at pasture. Subsequent studies of pregnant mares in tie-stalls have shown their time budgets to be similar to those of free-range horses.¹²⁹ It would be interesting to see whether the mares’ gravid state influenced this outcome. There is a danger in assuming that, even though horses do not choose to be so confined, tie-stalls somehow meet horses’ behavioral needs. Jones & McGreevy¹³⁰ have described a framework for judging the impact of and ethical justification for various interventions imposed on horses in the domestic context.

Health

The continual proximity to feces and urine may be aversive to a stable’s occupant and may also have an impact on health since humidity and airborne pathogen viability may rise, especially if air-changes per hour are insufficient.⁹⁴ An unnatural foraging posture, e.g. eating from a haynet for extended periods rather than from the ground, may compromise the function of the ciliary escalator in the respiratory tract¹³¹ and precipitate pulmonary disease. Rhabdomyolysis¹³² and chronic degenerative joint disease¹³³ are often exacerbated by the physical restriction that accompanies extended periods of stabling. A stable with insufficient or uncomfortable bedding may limit the willingness of its occupant to lie down.¹³⁴ This is thought to be associated with the horse’s working life being cut short.¹³⁵ Stabling has also been implicated as a predisposing cause of one of the most dreaded and frequently fatal equine disorders, colic.¹³⁶

Conclusion

The five freedoms (Box 1.1) have been established as a means of judging the extent to which the needs of domestic and captive animals are met by human carers.^137,138 If we apply this model of good animal welfare to horses, we can highlight areas in which wellbeing may be compromised. For example, we see many ways in which they are denied the freedom to express their normal behavior in a stable. When one considers the social nature of equids, one quickly appreciates the impediment that the isolation of single-horse housing represents to normal equine behavior. Therefore, while it has many advantages for the horse owner, the stable should be viewed not simply as a source of physical confinement but also as a limitation on behavioral choice.

Displacement and redirected behaviors

Displacement behaviors are regarded as responses that are inappropriate for the current situation. Of course, this rather arrogantly fails to acknowledge that observers may not always know what is appropriate and inappropriate from the horse’s perspective. Displacement behaviors are recognized in situations that involve behavioral conflict (see Ch. 13). For example, when a ridden horse is prevented from moving forward for longer than it can readily tolerate, it may start bending its neck laterally in an apparent attempt to groom its flank even though self-grooming is not the most appropriate response to restraint mediated through the bit. In the stabled horse, a similar self-grooming response as a result of frustration may be the kernel of stereotypic self-mutilation.

When a behavior (e.g. an act of aggression) is directed away from the primary target and toward another, less appropriate object, it is said to be redirected.¹⁴¹ The term must be used with caution since it requires that the observer has correctly identified the primary target. For example, horses that eat their bedding material are thought by some to be performing a redirected behavior that meets their physiological needs for dietary fibre.¹¹⁰ Others regard this as a form of coprophagia if the bedding is soiled, but since the remedies advocated involve increasing the provision of roughage (which redirects the behavior back to its evolutionary target), the distinction is descriptive rather than functional. Coprophagia is regarded as an adaptive behavior in foals^142–144 but as being an ethopathy in older individuals.⁵³

Another common example of redirected behavior in equine husbandry systems is wood-chewing, which is regarded as an ingestive behavior that would be more readily directed toward grass or other palatable fiber were it available.¹¹⁰ That said, it is known that horses at pasture will chew wood,^145–147 which implies that our understanding of the causes of this behavior is incomplete. It is possible that the ingestion of small quantities of bark may be adaptive as a means of acquiring micronutrients.¹¹⁰

Both redirected and displacement behaviors are subject to the rules of operant conditioning (see Ch. 4) and may be reinforced to become established unwelcome responses. For example, many horses paw at the ground or at the stable door prior to being fed. Although this may be a response that would be appropriate for a free-ranging horse that is being kept from its forage by a temporary barrier, such as snow, some would see the targeting of the door as a redirection of the behavior. If the horse is fed during or shortly after pawing the door, and is therefore rewarded, it will be more likely to paw the door in future.

Over-stimulation or under-stimulation of behavioral systems

While lack of stimulation is often cited as a cause of anomalous behavior,^50,148 including apathy, the opposite also merits consideration in the ontogeny of unusual behavioral strategies linked to housing. Kiley-Worthington⁸⁸ indicates that too much noise, excitement or exercise can cause over-stimulation, as can the presence of too many horses and humans to mix with socially. Raised levels of arousal in these horses may make them generally more reactive when being handled in the stable.⁶³ Similarly, inappropriate levels of stimulation are thought to cause psychopathologies such as hyperphagia nervosa⁵³ and polydipsia nervosa.¹⁴⁹

Learned behavior

Learning in companion animals is often not recognized and is therefore regularly misinterpreted. This often involves so-called superstitious learning whereby a response is acquired as a result of its accidental link with a reinforcer.¹⁵⁰ Misinterpretation arises because the handlers are unaware of the cues that their presence or actions represent to the animal. Chapter 4 is designed to allow readers to become conversant with learning theory and to understand why the term superstitious learning is virtually redundant, given that animals are learning all the time. Recognizing ways in which animals learn associations unanticipated by their handlers is a cornerstone of behavior therapy. A bizarre example is a case of psychogenic colic that was associated with a demand for human company.¹⁵¹

Another example is learned aggression to humans that can evolve as a result of inappropriate associations between agonistic posturing and the unwitting delivery of a food reward. Thus, arrival of the handler prompts hunger-motivated head-threatening behavior that is reinforced by the consequent acquisition of food as the human retreats. This agonistic behavior escalates when punished directly by the handler since his eventual retreat becomes a more valued goal.¹⁵² Other learned responses, including conflict behaviors that arise during horse–human interactions, are considered in Chapter 13.

Physical problems

Problems that arise as a direct consequence of the dimensions of the stable, such as ‘refusal to lie down’, have been blamed on some deficiency on the part of the occupants rather than the designers of the accommodation. Certainly, the area available influences resting behavior.¹⁵³ Other examples of physical problems include habitually getting cast, catching hips on the doorway (often exacerbated by learned fear that prompts faster, though not necessarily better judged, exits through doorways) and getting feet caught in haynets.

Historical accounts of stereotypic behavior

Stereotypic behavior is not described in the earliest text on horsemanship and stable management¹⁵⁵ but archaeologists have used erosion on the incisors of equine skulls as an indicator of crib-biting and therefore domestication. Donkey teeth gathered from what are thought to be ancient sites of worship in Syria have shown signs of wear that have been likened to the erosion one finds in crib-biters.¹⁵⁶ The use of incisor wear to identify Paleolithic horses as crib-biters is not without its critics.¹⁵⁷ The current prevalence of crib-biting in donkeys is negligible and, while it is possible that crib-biting was more common in the past and has been selected out of the population to some extent, it seems strange that three of the five donkeys in this site would have this stereotypy. The use of incisor erosion as a means of confirming that the bearer is or was a crib-biter remains controversial because pre-purchase evidence of erosion is cited in instances of litigation when the vendor fails to declare the behavior.

This link between crib-biting and domestication is made in the belief that stereotyped behavior occurs only in animals that have experienced captivity. Furthermore, the assumption that this wear is invariably the result of an oral stereotypy is not without hazards since sandy soils and particular species of grass may produce similar erosion on the labial surface and the tables of the incisors.

As well as crib-biting and a number of the other so-called stable vices, weaving is described in a number of antique texts on equine husbandry.^158–160

Characteristics of stereotypic behaviors

Stereotypic behaviors are not recognized in free-living feral horses and are not purely a product of domestication since they are also reported in captive examples of wild equids such as the onager mountain zebra¹⁶¹ and Przewalski horses.¹⁶² In the horse, these behaviors have therefore been linked to a number of management practices. A number of equine stereotypies have been identified,⁸⁸ including:

This is an exhaustive list and includes a number of behaviors that are not morphologically invariant and therefore might be regarded as redirected behaviors by other authors. Other behaviors in the above list are difficult to define because changes in form may accompany aging of the stereotypy, e.g. weaving may change from a side-to-side pacing to a stationary head-swing.¹⁶³ More than one stereotypy may be performed by an individual horse, e.g. a box-walker may well be seen weaving on occasions.¹⁶⁴ Indeed if a horse has one stereotypy, it has a greater chance of having a second than do normal horses.¹⁶⁵ The results of 13 studies into the prevalence of stereotypies and 6 into that of wood-chewing appear in Table 1.3.¹⁶⁶

Table 1.3 Reported prevalence (%) of stereotypies and wood-chewing from studies published between 1993 and 1998¹⁶⁶

Since many stereotypies are popularly regarded as being transmissible by mimicry¹³⁴ and some are associated with health and performance problems, horses exhibiting them are often isolated. Some (such as box-walking, weaving and wind-sucking) must be declared at auction and tend to lower the value of affected animals. These are dealt with more thoroughly in Chapters 5 and 8.

The functional significance of stereotypies

There has been much recent debate about the functional significance of stereotypies performed by captive domestic animals.^167–169 One influential theory is that stereotypies enable animals to cope with stress.^170–173 However, experimental studies to examine the effects of preventing animals from performing stereotypies in order to assess the validity of the stress-coping hypothesis have produced equivocal results.^174–178 Nevertheless, transient decreases in heart rates have been demonstrated in association with bouts of crib-biting (Fig. 1.13).^179,180 Furthermore, efforts to prevent the performance of equine stereotypies are linked to some increases in physiological stress parameters.^177,178 It has also been suggested that crib-biters may have higher basal sympathetic activity because they have been found to have higher overall mean heart rate.¹⁸⁰

Figure 1.13 Horse crib-biting on metal substrate. Although horses prefer to grasp wood rather than metal (perhaps because it cushions their teeth), they will crib-bite on metal if no other substrate is available. This helps to demonstrate a considerable behavioral need to perform the stereotypy.

(Reproduced by permission of the University of Bristol, Department of Clinical Veterinary Science.)

Stereotypies could help a horse cope with suboptimal environments or bring direct and immediate rewards that make the behaviors intrinsically gratifying. Endorphins have been implicated as a possible source of reinforcement for crib-biting¹⁸¹ because opioid antagonists can reduce crib-biting by 84%, suggesting that at least one of the perceived benefits of crib-biting (from the horse’s perspective) is mediated by opioid receptors at some point. However, because resting behavior in crib-biters was also significantly increased by opioid antagonists, it may be that the reduction in crib-biting was linked to a generalized sedative effect. The effects of opioid antagonists on weaving are extremely variable, with reports of both decreases^182,183 and increases.¹⁷⁸ While it is possible that weaving may also be opioid mediated, further work with larger numbers of weavers would be required before there can be a clearer understanding of the mechanisms involved.¹⁷⁸ Recent studies on the brain of stereotypic horses indicate altered dopamine activity.¹⁸⁴

A further suggestion is that a given stereotypy may retain a function within the motivational system from which it is derived.⁸⁸ Thus, an oral stereotypy such as crib-biting may provide a route to normal feeding and digestive activity within an environment that severely limits normal forage intake (e.g. an intensive training program characterized by the provision of high concentrate: minimal roughage diets).

Persevering with non-functional behaviors or previously trained but currently unrewarded responses is a characteristic of stereotypic animals.^185,186 There is evidence that stereotypic horses may have developed a general inability to suppress non-functional behavior.¹⁸⁷ While this may have important repercussions for the trainability of stereotypic horses, it should also be borne in mind when equine scientists select subjects for studies of learning.

Management factors and abnormal behaviors in stabled horses

It is clear that the prevalence of crib-biting and weaving is greater in Thoroughbreds than in other breeds.^188,189 The prevalence of box-walking, wind-sucking/crib-biting and weaving in UK Thoroughbred populations has been estimated at 1.1%, 4.2% and 2.8%, respectively (n = 1033)¹⁹⁰ and in Italy at 2.5%, 2.4% and 2.5% (n = 1035).¹⁹¹ Data from 4468 UK Thoroughbred horses in training showed the total prevalence of all three of these stereotypies to be 10.8%, similar to the prevalence of lameness.¹⁹² Estimates of the combined prevalence of all stereotypies are more difficult; for example, when lip-licking, pawing, tail-swishing, head-tossing, head-nodding and box-kicking were included, the prevalence on some yards reached 26%.⁸⁸ However, it remains difficult to dissect the effects of management and breed because Thoroughbreds are generally raced and therefore managed intensively.

In other species, cage-design,¹⁹³ isolation-rearing¹⁹⁴ and food-deprivation¹⁹⁵ have been implicated as proximate causes of stereotypic behavior. Arousal, generated by frustrated motivation, is a possible shared underlying cause,¹⁹⁶ although others emphasize the possible heterogeneity of the cause and effect of different stereotypies.¹⁵⁴ Despite much work on farm and laboratory species, the proximate causes of stereotypic and redirected behavior in the horse have yet to be verified. However, the possible causal factors associated with oral and locomotory stereotypies are discussed in Chapters 8 and 5, respectively.

The role of genes and environment

Although they have failed to control for variability in management factors, studies into the role that heritability plays in stereotypy frequency have consolidated the view that certain family groups are more likely than others to demonstrate stereotypic behavior.^107,191,197 More recently, it has been demonstrated that crib-biting at least is a canine oral stereotypy associated with certain gene arrays.¹⁹⁸ Meanwhile a growing branch of the literature identifies the importance of management factors that might frustrate motivation in the horse.^115,166 In horses bred for Flat racing, these include the amount and type of forage, the bedding type, the number of horses on the yard and the amount of communication that is possible between neighboring horses.¹¹⁵ A parallel study in dressage and eventing horses demonstrated that the amount of time spent in the stable correlated with the likelihood of stereotypies being reported.¹⁹⁹

Although feeding practices tend to have a greater effect than housing practices on the incidence of abnormal behavior,¹²⁸ a single causative factor is rarely to blame. For example, wood-chewing increased when high-protein rations were fed, because of a concomitant reduction in the total fiber content,¹⁰⁹ and when exercise was withdrawn.¹⁴⁶ Many authors have suggested other possible causes of anomalous behaviors, including factors associated with weaning, social contact, crowding, feeding, housing and/or training practices.^88,200

Some authors^135,164 emphasize the importance of mimicry rather than environmental deficits, believing that exposure to a stereotypic neighbor may increase the likelihood of stereotypy development or performance. Such social influences, known in voles,²⁰¹ may also affect stereotypy levels in horses,²⁰² but the evidence that horses can learn by observation is so far rather limited^203–206 Even if observational learning is not involved, we should not rule out the possibility that having a stereotypic neighbor may simply increase the arousal of observing horses and therefore predispose them to developing stereotypic behavior.

An exhaustive study of abnormal behavior on Thoroughbred studs gathered data on more than 11 000 horses, which suggested that, as a result of emancipation (where abnormal behaviors become detached from their initiating causes and thus less responsive to environmental enrichment), the prevalence of stereotypies tends to rise with age (Fig. 1.14).¹⁹² Therefore, it seems likely that few horses are ever ‘cured’ of these behaviors. This implies that once these behavioral anomalies are established, they persist to an extent, despite attempts to improve any potentially causative deficiencies in the horse’s management (Fig. 1.15). The continual reinforcement of stereotypic behaviors contributes to their resistance to therapy. That said, in one study effective treatment of crib-biting was offered as a possible reason (along with euthanasia) for the largest percentage prevalence in crib-biting rates being found in foals.²⁰⁷

Figure 1.14 The percentage of Thoroughbreds reported as showing stereotypic and redirected behaviors from a sample population of 1023 foals, 746 yearlings, 1001 2-year-olds, 711 3-year-olds and 6250 adults.

Figure 1.15 Two horses crib-biting at pasture. Since they have one another’s company and optimal foraging opportunities they provide a useful example of emancipation.

(Reproduced by permission of the University of Bristol, Department of Clinical Veterinary Science.)

Fundamentally the possibility of emancipation means that it is better to implement environmental enrichment prophylactically rather than therapeutically. Notwithstanding its effects in maintaining normal behavior in youngstock,²⁰⁸ it is also likely to increase learning ability²⁰⁹ while at the same time promoting musculoskeletal health.²¹⁰

The process of emancipation also means that stereotypic foals are better research subjects than mature stereotypers. Studies on cohorts of foals from neonates to 4-year-olds have yielded fascinating results. For example, time-related patterns for the development of oral behaviors differ from those of locomotory behaviors.²⁰⁷ This seems to emphasize that a single process such as ‘boredom’ could never fully explain the full spectrum of equine stereotypies. The results of these and other recent studies on oral and locomotory stereotypies are discussed in more detail in Chapters 8 and 5, respectively.

The future holds tremendous promise for sympathetic and productive understanding of equine stereotypies and the responsibility we must take for their emergence and effective management.

Management of stereotypies

Detailed therapeutic approaches to individual stereotypic behaviors will be considered in later parts of this book, but it is worth noting some overarching principles that apply to many of these ethopathies. Of all of the anomalous behavior patterns in horses, stereotypies seem to have proved most insoluble to horse owners. Work on other species by different groups^{169,174,211–213} often appears contradictory and has highlighted heterogeneity in stereotypies.¹⁶⁷ Since the study of equine stereotypies is still in its infancy and because information from other species can be applied only with caution, it is perhaps not surprising that hippologists have found vices so mysterious and frustrating and have met with so little success when attempting to modify these behaviors.¹³⁵

As the data demonstrating increasing prevalence with age suggest, stereotypic behaviors can become emancipated from their initiating causes.²¹⁴ Therefore, while stereotypies may arise in response to adverse management, they may persist in more enriched environments. Despite this, owners of stereotypic horses often feel responsible for their charges’ behavioral anomalies. The behaviors are therefore unwelcome not least because they are a potential source of embarrassment to owners.

Stereotypic horses are considered undesirable for many reasons. Generally diminished performance in stereotypers has been identified by some authors,^88,215 while health problems associated with these behaviors have been highlighted by others.^{116,134,216–219} Furthermore, it has been suggested that stereotypic horses in tie-stalls spend less time interacting with neighboring horses.¹²⁹ Prevention of crib-biting, involving physical or surgical approaches,^{1,134,220–224} is less commonly attempted presumably because of considerable variation in rates of success. Certainly surgery is falling out of favor as the causes of stereotypies become better understood.

There is recent empirical evidence that observational learning can occur in horses, but this is so far confined to a simple, following response, quite unrelated to the complex and covert oral maneuvers required in, say, crib-biting. Nevertheless, the belief that behaviors can be learnt or copied continues to affect the management of stereotypic horses. Physical prevention of the behaviors is the prevalent response, while isolation of affected horses from other horses is also common (Table 1.4). If these behaviors are not as functionless as assumed, e.g. if they constitute a coping response to a suboptimal environment, the common practice of preventing stereotypic behavior may be of welfare significance. Conversely, managing stereotypic animals so that they are subjected to minimal frustration, e.g. by feeding and exercising before other horses on the yard, is very helpful.

Table 1.4 Attitudes among professional horse owners to stereotypic behaviors^a

The need to prevent stereotypies for aesthetic and occasional health reasons has prompted searches for permanent cures. A remedy that is effective for every crib-biter remains elusive (but readers are directed to Chapter 8 for a discussion of current humane approaches). The resourcefulness of horses in satisfying their motivation to perform this behavior seems to overwhelm humans’ ability to physically prevent established forms of the behavior. In the light of evidence of brain chemistry differences in affected horses,¹⁸⁴ physical and surgical prevention attempts seem especially crude.

Environmental enrichment, often mentioned in the context of captive exotic species, is no less important in the horse. Turning stabled horses out to pasture offers a dramatic illustration of the effects of enrichment. However, for some owners this is not an option. While many owners reach for the blanket term ‘boredom’ to describe the problems faced by the stabled horse, some are beginning to explore ways to increase complexity of stables.²²⁶ For example, using a chain in place of the stable door is a surprisingly effective means of increasing the visual stimuli that enter a stable (see Fig. 1.12E). The key with any environmental enrichment is to ensure that the intervention is relevant and unlikely to provide a focus for simple redirection of abnormal behaviors, an outcome that is quite unsatisfactory. By studying this book, and most notably Chapters 5 and 8, readers will be well placed to design enrichment strategies that are likely to meet the behavioral needs of all horses, especially those that exhibit under-stimulation and frustrated motivation.

The web site http://research.vet.upenn.edu/HavemeyerEquineBehaviorLabHomePage/ReferenceLibraryHavemeyerEquineBehaviorLab/LabPublications contains a tremendous amount of information and advice on the behavior and management of stallions. Before attempting to modify the sexual responses of stallions, e.g. for semen collection purposes, readers are advised to visit this site and study the research papers it offers.

How to take a case history

Behavioral consultations can be considered under two main headings; the background and management details, which constitute a behavioral profile, and the unwanted behavior. Although entirely a matter for personal preference, it seems good practice to establish details of the horse’s management before exploring details of the unwanted behavior. This is because, having acquired a picture of the way the horse sees its world, one can better understand the motivation it may have to deploy unwelcome counter strategies. Box 1.2 gives an example of factors constituting a typical behavioral profile.

Because such potentially dangerous behaviors as bucking and rearing can be both expressions of joie de vivre and evasions from pain, veterinarians undertaking behavioral consultations should begin by performing physical examinations to eliminate somatic causes before embarking on a program of behavior therapy. This book explores the normal behavior of horses in considerable depth so that practitioners are equipped to consider the adaptive significance of many signs that present as behavioral problems. Horses that shy provide a useful example in that they should undergo ophthalmological examinations that are complemented by information regarding the frequency and variability of shying, and the spatial relationships of objects at which they tend to shy.

Problem behaviors, especially those under-saddle, can be acquired as a result of unfortunate experiences but may also reflect ongoing pain. To some extent it is a shame that the veterinary profession is called upon only as a last resort when a horse starts refusing jumps it is capable of clearing. As a result, lower limb pain is probably under-diagnosed as a cause of such problems.²³³ The application of various gadgets and escalated coercion often come before a full clinical examination.

Recognition of pain ranks as an extremely important feature of the examination in a clinical case with behavioral manifestations. Houpt²³⁴ suggests that a visual appraisal may help to indicate chronic pain, with the affected horse typically having a loose lip but clenched masseter muscles. Meanwhile, there may be a facial expression suggestive of ‘concern’, with decreased mobility of the eyes, puckering of the eyelids, backward inclination of the ears²³³ and dilation of the nostrils.⁵³ As social herbivores, horses might not be expected to demonstrate signs of pain more overt than teeth-grinding, since to do so may single them out for predation. However, it is proposed that when vocalization accompanies pain, it generally takes the form of a squeal if the pain is of somatic origin and a grunt if visceral.²³³ While reduced locomotion and postural attempts to shift weight away from the seat of pain can be spotted from a distance, guarding of muscle groups may be appreciated during palpation.

Although the role of neuropathic pain is now recognized in the etiology of headshaking (see Ch. 3),²³⁵ it remains unclear why some horses headshake most when being ridden in certain ways, e.g. with a flexed poll. Sometimes the role of pain in emergent unwelcome behavior is not revealed without analgesics. While one horse that is slow to leave from and quick to return to its yard may have become overly bonded to a companion, another may refuse to leave the yard until placed on a course of anti-inflammatory drugs. The latter would be one of those horses that have learned that being ridden, especially on resilient surfaces, is reliably painful and therefore should be avoided.

In behavioral terms, donkeys are not small horses. Behavioral peculiarities in the donkey are sometimes of clinical significance. Compared with horses, donkeys give fewer overt signs of pain, tending instead to lie down or stand still with their heads lowered. Sweating and hyperpnoea are more common responses to abdominal pain than rolling and flank-kicking (Jane French, personal communication). Even during intestinal strangulation or rupture, donkeys rarely lie down and roll but instead show little more than inappetance, a mildly elevated heart rate and a vague reluctance to move.²³⁶

Because behavior modification relies on being able to redirect motivation that has prompted unwanted behaviors, it pays to clarify the specifics of any behavior that seems to have departed from the norm. For example, if the animal is described as aggressive, considerably more information is necessary before the motivation to demonstrate agonistic behaviors can be understood. One should establish whether the horse is aggressive to other horses, other domestic species, humans or all of these. Aggression toward specific targets merits further questioning. For example, aggression toward humans can be investigated further with questions such as:

The unwanted behavior and the way it has evolved to its current form need to be clearly defined (Box 1.3). Given the apparent inconsistencies in the terminology and interpretation of equine behavior,²³⁷ practitioners are commended to equine ethograms designed to establish standard nomenclature for equid behavior.²³⁸ That said, there remains a need for a working horse ethogram that offers robust definitions on which the majority of experienced observers can agree.²²⁹ Because of the possibility of inconsistent labeling of behaviors,^192,235 direct observation or video material is certainly preferable to an oral account of the behavior. Problems in the ridden horse can be fully appreciated only if the patient is ridden saddled and unsaddled and, if feasible and safe, with and without a bit, preferably on more than one occasion by different riders.

With a clear picture of the ontogeny and motivation of the unwelcome behavior, the development of an effective, tailored strategy to remove any reinforcement of the behavior and redirect the horse’s motivation is likely to emerge.^239–242 This is behavior modification. Although a number of case studies are presented in this book, readers may find more exhaustive accounts of behavior modification elsewhere (for a series of 60 cases, see Why does my horse…?⁵⁴). Even when consistently applied, the results of behavior modification programs are rarely immediate in the horse, because of its innate resistance to extinction,⁸⁸ but they are usually impressive. There have often been years of inconsistent and inappropriate handling which leave habits that must be surmounted before the slate is clean enough for fresh lessons to be learned. It is therefore worth reminding owners that, because horses have excellent memories, ‘quick fixes’ are generally a rarity in equine behavior therapy. Behaviors with their origins in fear are especially prone to re-emerge in response to historic triggers. Some apparently expeditious solutions may mask pain, so they should be used with tremendous caution and only after the role of discomfort has been assessed.

In the following chapters, we will explore the variety of innate and learned behaviors seen in horses. Given this information, the success of behavior therapy depends largely on the versatility of the practitioner and the compliance of the owner.