Cynthia A. McCall, PhD
Professor and Extension Horse Specialist
Auburn University, Auburn, Alabama

Introduction
Many horse owners are content owning a pleasure horse which can carry them at a leisurely pace along a trail. Others want a horse that can perform successfully in competitive events. For a horse to be successful in either job it must be able to learn specific tasks. Most of these tasks are based on natural physiological responses of the horse, but they are performed with an intensity and duration that the horse normally would not use on its own. For example, successful polo ponies must discriminate and follow a fast-moving ball while running at top speed, and making hard turns and stops. All tasks have components of natural behaviors which have been developed to an extreme degree. Teaching horses specific tasks is a fairly simple procedure, but it is a procedure which is often mishandled in spite of good intentions.

 

How Horses Learn
Horses learn through stimulus-response-reinforcement chains which we commonly call trial and error. The stimulus-response-reinforcement theory states that a stimulus, or cue, perceived by a subject results in random responses (reactions) from the subject, and if the subject that is reinforced (rewarded) after a correct response it will begin to associate a specific cue with a specific response. A reinforcement can be either something that the horse likes and works to receive (positive) or something the horse does not like and works to avoid or eliminate (negative). Either way, the end result is the same: reinforcements increase the probability that a specific cue will produce a specific reaction in the horse. The trainer's job is to present understandable cues, know when the horse has made the correct response, and reinforce the response in the correct way at the right time.

Cue Presentation
Horses are adept at discriminating cues. Minute cues such as tightening of the handler’s facial muscles (Pfungst, 1907), large numbers of differing visual patterns (Dixon, 1970) and cues utilizing all the senses (Yeates, 1976, Mackenzie et al., 1987) are discriminated easily by the horse. Horses also are skilled at generalization (Dougherty and Lewis, 1991; 1993) which is detecting similarity between stimuli and giving the same response to very similar stimuli. And, there is some evidence that some horses can categorize stimuli into simple groups, for example, an outlined pattern with an open center versus a filled-in pattern (Hanggi, 1999), triangular shapes versus other shapes (Sappington and Goldman, 1994) or two identical visual stimuli versus one nonmatching stimulus (Flannery, 1997). However, horses may not be able to easily transfer a previously learned stimuli to a new task (Sappington et al., 1997: McCall et al., 2003).
When teaching your horse a new maneuver, begin with a simple, obvious cue that makes use of the horse's natural behavior and balance. For example, when teaching a young horse to turn to the right, a natural cue combination would be to put weight in your right stirrup and move your right hand out to the right (providing a leading rein which allows the horse's head and neck to turn right). The horse will instinctively step to the right to keep your weight balanced comfortably on its back and to follow its head and neck. While it is possible to teach the horse to turn right when you touch its left ear, this cue is not natural and will not lead the horse to the desired response easily. Using more natural cues will make training easier and more enjoyable for both the horse and the handler.

Cues used in animal training should be consistent and specific. Once you have decided that a specific cue will be used to ask for a certain action, the cue should not be varied until the horse learns the response. Suppose, in the above example, your horse is a real blockhead and fails to make the right turn consistently. This does not mean you abandon these cues and try a new combination of cues, or that you give up and try a left turn. It means that you need to make sure the same cues are presented each time you ask for a right turn. You can increase the intensity of the cues (really weight your right stirrup and move your right hand far out to the side), but do not confuse the horse by suddenly deciding to use a different cue combination.

Once your horse has mastered a specific cue combination, you can refine or completely change the cues used to get the response. This is done by repeatedly presenting your new cue to the horse and following it with the old, previously learned, cue. Again, both your new cue and old cue must be specific and consistent. For example, once your horse has learned the simple cues for a right turn, you might want to teach it to respond to a neck rein. You give the horse the new, less natural cue (the left rein placed against the neck) and follow this with the old cue (weight the right stirrup and the right leading rein). Repeated pairing of the cues in the specific order of the new cue first followed by the old cue is the most effective way to teach a horse (or any animal) to respond to a new, more subtle cue (Chance, 1994). Backwards pairing of cues, old cue followed by new cue, and simultaneous pairing of cues, new cue and old cue presented at the same time, are very ineffective in teaching new cues (Chance, 1994).

Small Responses Lead to Major Maneuvers
Once the horse has responded to a cue, the trainer must be able to recognize that the horse has made the correct response. It is extremely important to realize that the horse should be rewarded for pieces of the desired response during the initial training process. The process in which the trainer rewards successive approximations to the desired response is termed shaping. Shaping is essential in horse training to teach and refine tasks and to increase the intensity the horse puts into completing the task. For example, you want your horse to learn to face you when you open the stall. To do this you first reinforce the horse when it cocks its ear toward you when you open the stall. After the horse has grasped this task, you wait until the horse cocks its ear and turns its head toward you before you reward it. The next step is to begin rewarding the horse when it also begins to take a step toward you. You keep requiring more steps prior to the reward until your horse walks to the stall door when you open it. To use shaping to increase the intensity of a task, simply withhold the reward until the horse gives you a little more of the speed, strength, effort, etc. that you are looking for in the performance. Gradually require more and more of the desired intensity for the horse to receive the reward.

Chaining behaviors is a concept similar to shaping in which several finished responses are performed in particular sequence called a “response chain”. All major maneuvers performed by the horse are simply a series of responses connected together in a response chain. Horses do not learn major maneuvers such as calf roping all at one time. They learn many small responses (stand in the box, track the calf, stop, work the rope, etc.) that result in a finished roping performance when the responses are connected together. Chaining will result in a horse that can perform finished maneuvers confidently and effortlessly. The key is to recognize and reward the small responses that make up a finished maneuver and to successively add more of the responses which make up that maneuver. By shaping individual behaviors then adding each behavior into a chain, you can have a horse performing maneuvers correctly and quickly in relatively few training sessions.

During training sessions, it is important to review responses learned in the previous lessons. If the horse is having trouble with an intermediate response in the chain that makes a finished maneuver, you must make sure the horse can perform this response before you can expect it to perform the finished maneuver. Do not ask the impossible. You cannot expect a young horse to do flying lead changes before it can perform a balanced, collected canter with your weight on its back. This could be compared to asking a toddler to water ski. The strength, balance and coordination needed for the movement have not been developed yet. When asking for specific responses from the horse, be aware of your horsemanship. Sometimes horses cannot make the correct response because the rider lacks the balance and coordination needed to help the horse through the maneuver. If the horse is not performing a response correctly, always check your horsemanship to insure that you are not accidentally causing the problem.

Reinforcements - Making Learning Happen
Reinforcements are simply additional stimuli which come after a stimulus-response episode which let the horse know it has made the correct response. Reinforcers can be stimuli the horse innately understands, for example, food or electric shock, or can be stimuli which the horse must learn to associate with an innate reinforcer, for example, a pat on the neck or a vocal reprimand. Natural reinforcements, or primary reinforcements, are generally powerful incentives for the horse to perform specific behaviors but often are cumbersome to apply. So, most trainers utilize many learned reinforcers, or secondary reinforcers, in their training as opposed to natural reinforcers. Teaching a horse learned reinforcers often is accomplished through normal human interactions with the horse. For instance, when a trainer catches and holds a young foal, he waits until the foal quits struggling then praises and pets it (the learned reinforcers) and releases it to rejoin its dam (the natural reinforcer). Repeatedly applying of the learned reinforcer followed by the natural reinforcer (again, new stimulus followed by old stimulus) teaches the horse to associate the learned reinforcer with a “good” or “bad” state. Eventually the horse responds to the learned reinforcer similarly to the innate reinforced state. In the popular training methods of “clicker “or “target” training, the horse learns that the clicker or target is a reinforcer and then the trainer utilizes these during training. Research indicates that horses readily learn secondary reinforcers and utilize them in learning new tasks, but that the learned reinforcer should be paired periodically with the natural reinforcer to maintain its reinforcing properties (McCall and Burgin, 2002).

Reinforcements also can be categorized broadly as positive or negative (aversive) stimuli. Positive reinforcements are simply stimuli that are rewarding to the horse. Horses work to obtain positive reinforcements, such as food treats (natural) or pats on the neck (learned). Negative reinforcements are aversive stimuli the horse works to remove. Examples of common negative reinforcers are spurs (natural) and vocal reprimands (learned). Negative reinforcers can be delivered in two main ways. Escape is when the negative reinforcer is applied until the correct response is given. For example, applying a spur until the horse makes the correct response of moving away from the spur is escape conditioning. Round pen training in which the horse is moved around the pen until it responds correctly to the trainer is a popular training technique utilizing escape conditioning. The second delivery technique, avoidance, gives the horse a chance to prevent the negative reinforcer altogether by giving the correct response. For example, the horse is given a warning cue, such as leg pressure, and when the horse delivers the correct response of moving away it avoids application of the spur.

Giving the horse a chance to respond correctly (avoidance) is the more common delivery method of negative reinforcement in horse training. Avoidance conditioning usually results in a horse which responds to training with less anxiety and resentment than one which is simply trained with escape conditioning. Negative reinforcement probably is used to a greater extent than positive reinforcement in most riding situations simply because it is applied easily. For example, release of bit pressure and avoiding a spur or the tap of a whip are negative reinforcers trainers can apply easily and immediately. As long as they are applied correctly, both positive and negative reinforcers should result in learning in the horse (Haag et al., 1980). However, some research (Wolff and Hausberger, 1996; Visser et al., 2003) indicates that individual horses may respond better to either positive or negative reinforcers. So, good trainers make use of both positive and negative reinforcers and often may use both to reward a correct response, for example, releasing bit pressure while verbally praising the horse. Successful trainers also tailor reinforcements to individual horses, realizing that some horses work better with positive reinforcement and others with negative.

Reinforcements are the glue that connect a specific stimulus to a certain response. Correct timing of reinforcements is essential to successful training. To make a good association, the stimulus, response, and reinforcement must follow each other closely in time. All reinforcements should be given immediately after the horse has made the desired response. The horse cannot associate a response it made five minutes previously with the pat on the neck that it is getting now. Likewise, if you are using negative reinforcement, you must remove the aversive situation, such as your spur in the horse's side, immediately after it gives you the desired response. It is important that reinforcement is dependent on the horse making the correct response. Presenting reinforcements in the absence of any appropriate response from the horse reduces their effectiveness. Decide what the correct response is, and reinforce the horse only when you get that response. However, remember that a response does not have to be a finished major maneuver before it can be reinforced. You should set standards about what is the correct response for the horse in its particular stage of training.

When you are first teaching a horse a response, the horse will learn more effectively if it is reinforced every time it performs the response. However, once the horse has learned the response, successful trainers shift to an intermittent schedule of delivering reinforcements. All animals will work harder to obtain reinforcement if they do not know when the reinforcement will be delivered. A good comparison in humans is gambling. Since you do not know when you are going to win big money on a slot machine, you keep putting money into it. The slot machine gives you small intermittent rewards which keep your interest focused on winning the jackpot. To use intermittent rewards in horse training, you must be sure that the delivery of reinforcements does not accidentally fall into any pattern. If a pattern develops in number or timing or responses prior to reinforcer delivery, the horse quickly will figure it out and work only when reinforcement is imminent (Myers and Mesker, 1960). Be careful that you do not quit reinforcing the horse altogether when using intermittent reinforcement. Responses which are never reinforced weaken. This concept is known as extinction and can be very useful in horse training. Irritating responses, such as pawing when tied, usually weaken over time if they are not reinforced. This does not mean that the horse has forgotten how to perform the response. It simply quits exhibiting the response because it does not receive reinforcement for it.

Many handlers may inadvertently reinforce the horse for undesirable behaviors, so it is important to recognize undesirable behaviors and potential rewards. For example, a horse which kicks its stall at feeding time often is rewarded by being fed immediately so that the annoying behavior is stopped. A horse that is pawing while tied often is rewarded by attention. Withholding the reward and changing the environment so that the horse can not inadvertently reward itself (the noise from kicking the stall may serve as a learned reinforcer) can lead to extinction of undesirable behaviors.

Other Considerations in Training
When you train a horse for a specific performance event, you must consider the horse's athletic ability and suitability for that event. It does not matter how well you utilize the stimulus-response-reinforcement chain if the horse cannot physically perform the task. Keeping your horse well-fed, physically fit, and well-shod will increase training effectiveness. But all the good feed and good training in the world cannot make a world champion out of a horse that is unsuitable for that event. All horses can be improved by training, but do not expect the horse to perform at a level that is beyond its physical abilities.

Because training involves some type of physical activity from the horse, avoid drilling the horse until it is exhausted. Too much physical work will quickly teach the horse to dread training sessions. You do have to repeat the stimulus-response-reinforcement sequence enough times for the horse to learn it (McCall et al., 1993), but you can weave the lesson into other activities so that the horse is not continuously drilled. This keeps the horse more alert mentally and reduces physical fatigue. And research has indicated that horses trained on a specific task learned more efficiently when training trials were conducted once weekly as opposed to daily (Rubin et al., 1980). Horses probably have very short attention spans. A two-year-old can concentrate for approximately 10 minutes; a mature horse may be able to last about 20 minutes. So give the horse frequent breaks in learning. Train for three or four minutes, then trot the horse around on a loose rein and let it relax. Exercise does not have to be paired with training. You can exercise a horse sufficiently without overwhelming it with training.

Successful trainers realize that the horse is not a small child or a large dog. They realize that it is a horse, and they treat it accordingly. A successful trainer is always the dominant member in the human-horse relationship. Because horses are herd animals and innately understand a dominant-subordinate relationship, they will respond willingly and happily to a dominant trainer. Establishing dominance over a horse does not have to be a painful experience for either the horse or the trainer. It is accomplished easily by recognizing dominance challenges the horse presents to the handler and dealing with these challenges immediately and consistently. Dominant horses can threaten and move subordinate horses around without any retaliation from the subordinate. Dominant horses do not have to turn and face a subordinate when it approaches and are free to rub on subordinates and initiate mutual grooming activities anytime they desire. So if you are the dominant member of the horse-human partnership, your horse should never run over or past you, threaten to bite or kick, rub on you uninvited, or fail to turn and face you when you enter the stall. If your horse does any of these actions it is challenging your dominance status, and you must correct the horse immediately and consistently. A dominant horse (or you as the dominant human) never misses an opportunity to remind other horses that it is the boss. Remember that once you establish a rule, it is forever. Do not let dominance challenges go unanswered or you will slide down the dominance relationship. Similarly as the dominant individual, you must take over certain responsibilities. If the horse is frightened, you must be the voice of calmness. If the horse is confused, you must show it the correct response.

Once the rules have been established, it occasionally is necessary to punish a horse that breaks a rule. Punishment is like negative reinforcement in that it utilizes an aversive stimulus; however it differs from negative reinforcement in that it is used to eliminate an undesirable behavior. It is applied after the undesired behavior has occurred, where negative reinforcement is applied until the desired behavior occurs. Using punishment follows many of the same rules as using reinforcement. Punishment must be applied immediately after the undesirable behavior for the horse to associate the punishment with its previous action, and the trainer must think about which response he is really punishing. For example, your horse bites you on the arm. If you wait for pain to lessen, then go find an appropriately big stick, then catch the horse, then whack it with the stick, the horse will think it is being punished for letting you catch it. There is no way it can associate the whack with biting you three minutes previously. Similarly, punishment of a response that is partially correct and partially incorrect works to eliminate both the correct and incorrect parts of the response. Both punishment and negative reinforcement must be used cautiously because aversive stimuli have the undesirable side effects of escape, aggression and apathy (Chance, 1994). If the aversive stimuli is applied to the point that the horse becomes uncontrollable or is applied to the point that all behavior is suppressed then the horse definitely has not learned the intended lesson. However, research has indicated that beginning with a weak aversive stimulus and gradually increasing to a much stronger stimulus results in the animal enduring a more severe stimulus than it would have needed to suppress the behavior initially (Miller, 1960). So, when using aversive stimuli it is important to initially make the punishment strong enough to suppress the behavior. However, applying a strong enough aversive stimuli to influence a behavior and losing control of your emotions are closely linked in many people. Losing your temper with a horse almost never results in a good training experience.

It also is important to make sure the horse has an alternative behavior available when using aversive stimuli. In negative reinforcement the alternative behavior should be the desired response. When using punishment, alternative behaviors may be more difficult to identify. If the horse is punished for running over the handler, the obvious alternative to the human (do not run over me again) may not be as obvious to the horse. Trainers can help horses, and reduce the amount of punishment used, by correcting the horse and providing a task which is incompatible with the bad behavior. For example, your young horse gets excited at a horse show and runs over you. This horse would benefit from a mild correction, a yank on the lead chain, then giving it a well-understood task which is incompatible with running over you (back up a few steps). When the horse backs correctly it should be rewarded. This replaces the unwanted behavior with a more desirable alternative.

Another type of punishment used with humans is “time out” or “response cost” punishment. In this type of punishment, privileges are removed for misbehavior. A misbehaving child is given a few minutes to sit alone removing the fun it was having. A misbehaving adult is fined for speeding. These punishments work well for humans, but may not have much use in horse training because the horse is limited in its reasoning ability. It is hard to explain to the horse that the reason it is not getting turned out is because it bucked you off earlier in the day.

Conclusion
Whenever you work around a horse you are giving it stimuli, intentionally or unintentionally, and your horse is responding to these stimuli. You are training a horse continuously, and the horse is learning continuously. So, you should think about your actions and your horse's reactions at all times, and be consistent in what you expect from the horse. For example, if your horse is allowed grab the sleeve of your heavy winter coat in his teeth without any negative reaction from you, then the horse rightfully can assume that it is acceptable behavior to grab your bare arm in the summer. Using stimuli correctly and consistently, recognizing segments of behavior that constitute the desired response and reinforcing the response in a timely manner are the main components of a successful training program. Combining these correct training procedures with a basic understanding of horse behavior and empathy for horses will make training easy, enjoyable and fun for you and your horse.

Training Tips
1) Plan your training sessions carefully. Have a goal and a method of reaching that goal before you begin each session. Because horses can have “off-days”, you should have an alternate training plan available for days where the original plan is not effective in reaching your goal.
2) Warm up the horse both mentally and physically before training.
3) Horses cannot reason or think abstractly. Simple, consistent cues and immediate rewards result in the most effective learning.
4) Do not ask the horse to perform maneuvers that it is physically or mentally incapable of handling. However, remember that both you and the horse have to get out of your respective comfort zones to progress.
5) Repetition is needed for horses to learn. However, too much repetition during a single training session or during numerous consecutive training sessions is detrimental to learning.
6) Use both positive and negative reinforcements in training.
7) After a horse has mastered a maneuver, reinforcements delivered intermittently maintain high rates of responses.
8) Recognize when the horse is trying to make a correct response and reward small portions of the desired response.
9) Stop training sessions while the horse is performing correctly. You may have to return to a previously learned maneuver to end the training session with a positive performance.
10) Clearly know the lesson you are trying to teach and be persistent with that lesson. A few
correct steps of a particular maneuver are worth more than a multitude of partially correct steps.
11) Horses give us gifts. If your horse accidentally gives you more than you asked for, for example, a flying lead change rather than a simple lead change, take the gift and reward the horse for it.
12) Release of pressure is a very potent reward for a horse.
13) Be the dominant member of the horse-human partnership. Learn to recognize and immediately correct challenges to your leadership role. Be the leader when the horse is confused or frightened.
14) Find ways to keep the horse’s attention on the rider/handler. Changes in movement, direction, gait or tempo are easy ways to keep the horse entertained and focused on you. Convince the horse that training is a game which the horse and handler are playing.
15) Good horse management is essential to good training. Healthy, happy horses with correctly fitted tack will give you their best performances.
16) Horses will work harder if they respect and like the trainer.
17) Horses live in the present. They do not worry about yesterday or the future.

References

Chance, P. 1994. Learning and Behavior, 3rd Ed. Brooks/Cole Publishing Co., Pacific Grove, CA.

Dixon, J. 1970. The horse: a dumb animal?... The Thoroughbred Rec. 192(19):1654-1656.

Dougherty, D.M. and Lewis, P. 1991. Stimulus generalization, discrimination learning and peak shift in horses. J. Exp. Anal. Behav. 56:97-104.

Dougherty, D.M. and Lewis, P. 1993. Generalization of a tactile stimulus in horses. J. Exp. Anal. Behav. 59:521-528.

Haag, E.L., Rudman,R. and Houpt, K.A. 1980. Avoidance, maze learning and social dominance in ponies. J. Anim. Sci. 50:329-335.

Hanggi, E.B. 1999. Categorization learning in horses (Equus caballus). J. Comp. Psychol. 113:243-252.

Flannery, B. 1997. Relational discrimination learning in horses. Appl. Anim. Behav. Sci. 54:267-280.

Mackenzie, S.A., Giodano, A.P. and Monahan, E.A. 1987. Use of a conditioned stimulus to improve equine behavior in response to clipping. Equine Prac. 9(3):15-17.

McCall, C.S. and Burgin, S.E. 2002. Equine utilization of secondary reinforcement during response extinction and acquisition. Appl. Anim. Behav. Sci. 78:253-262.

McCall, C.A., Salters, M.A., Johnson, K.B., Silverman, S.J., McElhenney, W.H. and Lishak, R.S. 2003. Equine utilization of a previously learned visual stimulus to solve a novel task. Appl. Anim. Behav. Sci. 82:163-172.

McCall, C.A., Salters, M.A. and Simpson, S.M. 1993. Relationship between number of conditioning trials per training session and avoidance learning in horses. Appl. Anim. Behav. Sci. 36:291-299.

Miller, N.E. 1960. Learning resistance to pain and fear: Effects of overlearning, exposure and rewarded exposure in context. J. Exp. Psychol. 60:137-145.

Myers, R.D. and Mesker D.C. 1960. Operant responding of a horse under several schedules of reinforcement. J. Exp. Anal. Behav. 3:161-164.

Pfungst, O. 1907. Das Pferd des Herrn von Osten (der AKluge@ Hans). Leipzig: Barth. in Hafez, E.S.E. (Ed.) The Behavior of Domestic Animals. Williams and Wilkins Co., Baltimore, MD. p. 332.

Rubin, L., Oppegard, C. and Hintz, H.F. 1980. The effect of varying temporal distribution of conditioning trials on equine learning behavior. J. Anim. Sci. 50:1184-1187.

Sappington, B.F. and Goldman L. 1994. Discrimination learning and concept formation in the Arabian horse. J. Anim. Sci. 72:3080-3087.

Sappington, B.F. K., McCall, C.A., Coleman, D.A. Kuhlers, D.L. and Lishak, R.S. 1997. A preliminary study of the relationship between discrimination reversal learning and performance tasks in yearling and 2-year-old horses. Appl. Anim. Behav. Sci. 53:157-166.

Visser, E.K., van Reenen, C.G., Schilder, M.B.H., Barneveld, A. and Blokhuis, H.J. 2003. Learning performances in young horses using two different learning tests. Appl. Anim. Behav. Sci. 80:311-326.

Wolff, A. and Hausberger, M. 1996. Learning and memorization of two different learning tasks in horses; the effects of age, sex and sire. Appl. Anim. Behav. Sci. 46:137-143.

Yeates, B.F. 1976. Discrimination learning in horses. M.S. Thesis, Texas A&M Univ., College Station, TX.

Gallery Preview

2017

Upcoming Events

 Upcoming Events
Spruce Meadows 'National'

June 7-11, 2017

 

 

 

 

Industry Directory

HIAA E-News

Please Click Here to view our Industry Directory with links to equine sport groups, breed groups, facilities, calendars and publications in Alberta.

Contact

Contact Us
Office Address:

97 East Lake Ramp NE
Airdrie, AB
T4A 0C3
Phone: 403-420-5949
Fax: 403-948-2069