|
||||
|
Training and Behavioural Rehabilitation in the Horse Introduction Although the domestic horse is highly adaptable and has to a certain extent been selected to meet various human requirements, the highly intensive conditions in which most performance or sports horses are maintained is very different to those of the feral horse. Degrees of intensity vary between horses kept more for recreational purposes, where there is often a greater proportion of time spent grazing, and those kept for sports purposes, where animals are housed for longer periods. Sport horses also experience intensive training, controlled feeding regimes, and regular transportation. The value of understanding what motivates the horse and the principles of learning are obviously central to interpreting and resolving undesirable behaviours. A great deal has been written about the methods used for training horses, and it is clear from the history of these techniques that modern practices are in the main based upon traditional methods that have developed since the earliest days of horse domestication. Methods of training and rehabilitation, therefore, need to be evaluated in terms of modern day understanding of equine motivations, psychology and welfare. Features of the Domestic Environment that Can Cause Problem Behaviour The causes of problem behaviour relate to the methods used for training purposes, the intensive conditions the horse is maintained in, the unnatural situations it is exposed to and the feeding regimes it must deal with. In essence there exists an uneasy relationship between the adaptive ability of the domestic horse and the environmental conditions it is exposed to. For example horses are not only physiologically adapted to feed upon a herbivorous diet, their body structure and behaviour are also selected for this purpose. In the domestic situation, the sports horse will require more nutrients than are available through a grass based diet, and so it will receive a concentrated grain based diet, with limited access to fibre. This will not only cause digestive problems, but will also restrict normal foraging activity, often leading to the development of abnormal behaviour such as crib-biting and wood-chewing (see Winskill et al, 1995). Prevention of such abnormal behaviour relies on an understanding of the causal factors. Training The application of learning theory in modern training methods for horses has received surprisingly little scientific interest until recent years. It has long been normal practise to train horses through operant conditioning using mainly negative reinforcement, which is where an unpleasant stimulus is applied to the horse until it performs the desired response, or positive punishment. In fact if we examine many of the responses that are required of a horse during traditional training processes, we find that for the most part we make use of their natural tendency to avoid a painful and at times, frightening, stimulus. Housing Since most performance horses are frequently housed for the greater part of the day, this often means they are at least partially isolated from con-specifics. Lack of adequate social interaction can lead to the development of various problem behaviours such as what has been termed ‘isolation induced aggression’ (Houpt, 1981). Problem aggressive behaviour such as this can often be directed at other horses and/or the handler. The function of normal aggressive behaviour is to enable an animal to gain a competitive advantage, giving him or her better access to a limited resource important for survival. For the horse this may be a fitter mate or a better feeding or foraging opportunity. By understanding the importance of different resources to horses and in addition the importance of the resource to a specific individual, it is possible to determine if an animal is behaving abnormally or if it is behaving perfectly normally, if inappropriately from an owner’s perspective. In the case of isolation-induced aggression particularly in stallions, the cause is probably the result of mismanagement, particularly during early development. Any horse that is reared in relative isolation will have few opportunities to learn about normal social interaction due to their restricted access to other horses during early development. Situations in which horses are anxious or frustrated will also decrease the threshold at which aggressive behaviour is displayed (Luescher et al, 1991). Interestingly, excessive aggressive behaviour is often thought of as a natural behaviour in stallions. However, problem aggression in stallions may be due to in part to poor training, as well as inappropriate housing and management. Exposure to unnatural situations. The performance horse needs to be able to cope with a variety of unnatural and potentially fear-provoking situations. Although it is possible to habituate the horse to a large number of the treatments and situations it will experience in its life as a performance animal, it is not possible to cover everything. The performance horse has to be ‘flexible’ in its behaviour in order to deal with new situations as they arise. Such flexibility, relies on good and varied early experiences. For example, it is known that orphan foals reared by humans are less emotional when placed in a novel environment as compared with normally reared foals (Houpt and Hintz, 1983). However horses that have been extensively handled, learned which way to turn in a maze more slowly than minimally handled horses, but faster than unhandled horses. This finding suggests that there is an optimum amount of handling early in development that is most effective for horses when placed in a problem-solving situation without human help (Heird et al, 1981). By intensively handling foals Miller (1989), claims early bonding (imprinting) between human and horse takes place, which enhances the ability of the horse to cope with training and management later in life. There is no real scientific evidence as to the optimal period for handling or to the type or amount of handling that optimises foals’ response to humans later in life, and this is an important area for further research. Ensuring that foals experience the situations likely to be encountered as adult horses, such as entering an enclosed space and moving away from conspecifics, is also likely to decrease the chance that such activities will be fear-provoking later in life. Diagnosis and Treatment of Problem Behaviour Importance of good diagnosis. The first stage in the rehabilitation of horses with ‘behavioural disorders’ is to obtain an accurate diagnosis. This will take into account the origin of the problem behaviour, the horse’s motivation and specific contextual cues in which the problematic behaviour occurs. In order to do this an understanding of two areas are essential: · Equine ethology – the ‘natural’ behavioural repertoire of the horse, where motivations are determined by the evolutionary history of horses. · Learning theory – the types of learning experiences that each individual horse has been exposed to which have shaped its behavioural response to its environment from the first weeks of its life. Since learning opportunities will inevitably differ between each horse, individual ‘problem’ behaviours have to be approached as unique cases for which a detailed historical and observational analysis is conducted. There are, however, similarities in types of problem encountered because a high proportion of handling problems in horses derive directly from behaviours that are essentially ‘normal’ in the horse, such as behaviours designed to retain contact with the herd, or avoid entering a dark enclosed space (Cooper and Mason 1998). Equally commonly, behaviours derive from associative learning that occurs as a consequence of ‘normal’ equine motivations. These are due to the horse’s ability to learn, through operant conditioning, to avoid situations that are aversive and to move towards situations that are rewarding. Additionally the horse learns to predict important events through linking stimuli that occur contiguously to the events, through classical conditioning. Behaviours that are not within the normal equine behavioural repertoire, such as stereotypies, are considered to be abnormal behavioural responses. These behaviours occur in response to acute stress or chronic frustration of behavioural motivations, where the environment that a horse is in, does not meet its behavioural needs (Luescher et al, 1991). Despite the fact that scientific knowledge about both equine ethology and learning theory has been in the public domain for several decades, it is still frequently the case that those owning and caring for horses have a tendency to attribute behaviour problems to anthropomorphically derived causes. Comparative neuroanatomy suggests that horses do not have the same abilities as humans to form abstract ideas (Mayhew 1989), and hence equine behaviours can most easily be explained using more parsimonious explanations based on simple associative learning. Treatments that are traditionally described for dealing with problem behaviours are generally highly prescriptive. They often rely on subscribing to a particular philosophy or form of training, rather than encouraging owners and caretakers to develop a more fundamental understanding of the behaviour of their horses. Each equine behaviour problem case has a unique series of events that have led to the development of the problem. The temperament of the horse, the early environment it experienced, its previous learning experiences, and the current context in which the behaviour occurs, are all relevant to an accurate diagnosis. It is also often the case that the environmental factors that initiate or precipitate a behaviour are not the same as those, which maintain it. For example, the cause of the behaviour of a horse that turns to kick at humans trying to catch it could be considered as a learnt response to avoid being removed from conspecifics. However, further investigation of the history of such a behaviour, may reveal, that the behaviour first appeared when the horse was developing splints and was still being exercised regularly. Hence, the association between being caught and the pain of exercise would be a likely initiating factor for the problem behaviour in this case. Even once the pain is gone, problem behaviours such as this are often maintained by horses, as they do not have the chance to learn that the predictive stimuli are no longer associated with pain. In addition, once an avoidance response such as this is learnt by an individual, it is frequently then used to achieve other aims – in the example given it could be that the behaviour was maintained as a mechanism to avoid leaving the herd. Treatment for such a case should not, therefore, just involve a ‘prescriptive’ treatment for horses that refuse to be caught. It should also include a plan that aims to break the associative links that have been formed between the stimulus of a person entering the field and the consequences that the individual horse has learnt about this event. In the identification of factors that are associated with a change in behaviour, the recognition of events that are painful to the horse is extremely important. This is because the avoidance of painful stimuli is likely to be particularly pertinent to a prey species and horses are therefore adapted to respond rapidly to painful events. Pain avoidance is a problem in the handled horse – pain from touching or medicating an injured area, or over vigorous grooming of a sensitive horse, frequently results in avoidance responses, or, where these are unsuccessful at preventing pain, defensive aggression. Treatment Effective treatment of behaviour problems in the horse depends on accurate diagnosis, a realistic and individual treatment plan, and owner motivation. Treatment is most effectively carried out by the owner or caregiver of the horse in the horse’s normal environment. This is because learned behaviours are often context specific. Thus taking a horse away from its environment and owner for ‘rehabilitation’ may cause a short-term change in behaviour in the new environment, but the horse may well resume its previous behaviour on returning home. Since it is usually the owner who carries out the treatment, ensuring that these people understand the origin of the problem and the rationale for treatment is at the crux of effective behaviour modification. Dealing with behaviour problems therefore involves not only sufficient understanding of the underlying science and horse handling skills, but also good interpersonal communication skills. The diagnosis of the problem behaviour identifies the associative links that have been made, the main motivational factors for the horse, and the specific contextual and discriminative cues in which the behaviour occurs. This understanding can then be used to develop a plan that will result in changing both the consequences of the horse’s own action, and the underlying motivation that originally caused the behaviour to develop. For example, where the underlying cause of a behavioural response is fear, the aim of the treatment is to disassociate the fear from the events that cause it, and also to make the behavioural response of the horse unsuccessful in removing itself from the fearful stimulus. General Methods used in Behavioural Rehabilitation. Various methods are used in equine behavioural rehabilitation. Traditional methods of training and rehabilitation have relied predominantly on negative reinforcement and punishment, and many of those working with problem horses still use such methods. Although in some cases aversion therapy is necessary for a horse to learn the negative consequences of its behaviour, this can be carried out so that the horse is given the chance to learn the undesirable consequences of its behaviour in a controlled and systematic way. Using punishment without allowing the horse to learn an alternative appropriate behaviour only causes confusion in the horse, which generally results in the animal becoming uncooperative, aggressive or leads to the development of learned helplessness. Through the use of positive and negative reinforcement, the horse is offered a choice between performing an appropriate behaviour or an undesirable one. Negative reinforcement is better since the consequences of the unwanted behaviour are immediate and therefore more obvious to the horse. The use of positive reinforcement is preferable to negative reinforcement or punishment where it is feasible. Using positive reinforcement decreases the chance that the horse will learn to associate negative events with the handler or other unintended aspects of the environment, such as the stable. Whether punishment or reward is used for retraining, the effectiveness of the process depends to a great extent on timing. The reward or punishment must occur rapidly after the behaviour of the horse in order for the horse to associate the consequence with its own action. In most cases where modification of behaviour using any of these techniques does not appear to be effective it is because of poor timing on the part of the owner or trainer. As well as using direct reward and punishment in behavioural modification, conditioned reinforcers and punishers can also be effective methods for retraining. Essentially these involve associating a previously unimportant stimulus with a salient (rewarding or punishing) one. With repeated pairings of the unimportant stimulus with the salient one, the former becomes predictive of the latter and becomes reinforcing in its own right. Conditioned reinforcement has been used successfully in horses through the use of specific words or sounds for many years, but using novel stimuli makes the conditioned reinforcer more obvious and consistent for the horse – an example of this technique is clicker training (Kurland 1998). Desensitisation is a technique developed in the field of human psychology that involves the staged exposure of the animal to an increasing intensity of a fear-provoking stimulus. Counter conditioning is the process by which a behaviour that is incompatible to the undesired response is reinforced in the context of the fear-provoking stimulus (Voith, 1986). In practical terms, desensitisation and counter-conditioning are carried out concurrently (see Mackenzie et al, 1987). The fear-provoking stimulus is reduced to a low intensity and ‘relaxed’ or trained behaviours are reinforced, and then the level of the stimulus is gradually increased whilst the trained behaviour is reinforced. Conclusion Rehabilitation of horses with behaviours that are undesirable cannot be prescriptive. This is because the individual learning experiences of each horse needs to be matched with an understanding of the kinds of behavioural responses that are likely to occur in response to environmental change as well as the motivations, which underlie these responses. An accurate assessment of the factors contributing to each problem is necessary before an effective treatment plan can be developed. References Cooper JJ, Mason G (1998) The identification of abnormal behaviour and behavioural problems in stabled horses and their relationship to horse welfare a comparative review Eq Vet J Suppl 27: 5-9. Heird JC, Lennon AM, Bell RW (1981) Effects of early experience on the learning ability of horses J Anim Sci 53: 1204-1209. Houpt KA(1981) Equine behaviour problems in relation to humane management Intern J Stud Anim Prob 2: 329-336. Houpt KA, Hintz HF (1983) Some effects of maternal deprivation on maintenance behaviour spatial relationships and responses to environmental novelty in foals Appl Anim Ethol 9: 221-230. Kurland A (1998) Clicker Training for Your Horse. Luescher UA, McKeown DB, Halip J (1991) Reviewing the causes of obsessive compulsive disorders in horses. Vet Med 86: 527-530. Mackenzie SA, Giordano AP, Monahan EA (1987) Use of a conditioned stimulus to improve equine behaviour in response to clipping. Eq Pract 9: 15-17. Mayhew IG (1989) Large Animal Neurology: A Handbook for Veterinary Clinicians. Lea and Febiger, Philadelphia. Miller RM (1989) Imprint training the newborn foal. Large Anim Vet 44(4): 21. Voith VL (1986) Principles of learning. In Veterinary Clinics of North America, Equine Practice 2 (3): 485-506. Winskill L, Waran NK, Channing C, Young R (1995) Stereotypies in horses: Causes treatments and prevention. Cur Sci 69 (4): 310-316. |
||||
