Equitation Science

Your Horse is Always Learning. The Question is What.

WRITTEN BY: HARRIE PHILLIPS

PGCertClinEd, BAdVocEd (VocEd&Trng), DipVN (Surgical, ECC), DipBus, DipTAE (Development & Design), TAA

PUBLISHED: 15 May 2025

Harrie Phillips, Founder and Managing Director of the Australian College of Animal Care

A few years ago, when I was still relatively new to owning horses, I purchased an 18-year-old dressage schoolmaster mare named Rory, a kind and generous horse with a long and unblemished record of being reliable in just about every situation she had encountered, including, among many other settled habits, the simple matter of being loaded onto a float. Within months of being in my care, she would not load at all.

The timing of her arrival did not help. I bought her during the COVID lockdowns, which meant that for the first stretch of our partnership we had no real cause or opportunity to float anywhere, and so the question of whether she still loaded easily simply did not come up until restrictions lifted and I went to float her for the first time. When that day arrived, the wheels came well and truly off. She would not approach the ramp willingly, and when I persisted she would swing sideways, plant herself, or back up entirely, turning what should have been a short and uneventful task into sessions that stretched into hours. I started to avoid going anywhere, if I did need to go, gave myself extra time to load and just accepted that this was her. Help from experienced horse people along the way, kindly offered, only made the situation progressively worse, and eventually I had a horse who actively dreaded the float and a problem I had no idea how to solve.

What I worked out, eventually, was that Rory had not become a difficult loader; I had become an unclear handler. My pressure-release timing was inconsistent, my cues were not landing where I thought they were landing, and over the course of our quiet lockdown partnership those once-reliable responses had been slowly diluted by my own confused signalling. The fix, when I finally found it, involved positive reinforcement, counterconditioning, and a complete reset of how I was applying pressure, and within a few weeks Rory was self-loading.

Rory taught me more about training horses in a few months than several years of riding had before that. More importantly, she pushed me to start reading the equitation science literature properly, and to start thinking carefully about the way horses actually learn, drawing on my previous life experience of training dogs.

Every interaction we have with our horses is training, whether we intend it to be or not. Once you recognise this, you cannot unsee that fact, and you will probably start handling horses very differently.

What equitation science actually is

Equitation science is the application of established principles from learning theory, , and behavioural science to how we train and ride horses. As a field it has been around for more than two decades, supported by peer-reviewed research, dedicated journals, and an international society (the International Society for Equitation Science, ISES)1. Despite this, most everyday horse owners have never heard of it.

The gap between what equitation science research has established and what is taught in everyday horsemanship is wide1, 2. Most riders are taught one or two principles of learning theory, typically use pressure-release, and assume that is the whole picture. It isn’t, and the consequences of that gap show up in the behaviour and welfare of horses every day. They showed up plainly in my paddock with Rory, and once you have the framework, you start to see them everywhere.

There is enormous variation in training methods used across the horse world. Some methods work well, some don’t, and some work for the wrong reasons. The point of this article is not to dismantle methods, horsemanship leaders or name brands, but to give you the underlying knowledge to evaluate any method you encounter against what we know about how horses actually learn.

The four quadrants of operant conditioning

Operant conditioning describes how an animal learns from the consequences of its behaviour3. There are four quadrants, and most of us only know one of them well but would be using all 4 daily without realising it.

  • Negative reinforcement: removing something (such as pressure) to increase a behaviour. The horse moves forward in response to leg pressure, and the leg comes off.
  • Positive reinforcement: adding something pleasant to increase a behaviour. The horse stands quietly at the mounting block and receives a wither scratch or food reward.
  • Positive punishment: adding something unpleasant to decrease a behaviour. The horse nips and receives a sharp verbal correction.
  • Negative punishment: removing something pleasant to decrease a behaviour. The horse becomes pushy at the gate, and the handler walks away.

The terms positive and negative are commonly misunderstood. They do not mean good or bad; they mean addition or subtraction. Reinforcement increases the likelihood of a behaviour occurring again, and punishment decreases it3.

Negative reinforcement is the dominant learning mechanism used in ridden work. Pressure is applied through the legs, seat, reins, or whip, and removed when the horse offers the correct response. When done with good timing and clear cues, it is an effective and ethical method of training1. The word “negative” carries no judgement here, since it describes the subtraction of pressure as the reinforcing event rather than any quality of harshness or cruelty.

Positive reinforcement, particularly food-based, is well established in the literature as effective and welfare-positive4, and yet it remains under-used in everyday horsemanship, often dismissed as “treat training” or assumed to create pushy horses. When applied with appropriate protocols, neither concern is supported by the evidence4. Positive reinforcement is particularly useful when retraining a horse with negative associations, when working with fearful or shut-down horses, and when teaching a behaviour from scratch where pressure alone would be confusing or stressful for the horse.

This is the part most riders find clarifying. You are not choosing between these quadrants; you are already using all four of them every day. You scratch the wither when the horse stands quietly, you release lead rope pressure when they step back out of your space, you growl when they nip, and you walk away from the gate when they become pushy. The question is not whether you are using these tools, because you are, just maybe you have never named them. The question is – are you are using them with purpose?

The two learning processes most of us overlook

Operant conditioning is not the whole picture. There are two further learning mechanisms worth understanding, both of which are running in the background of every interaction you have with your horse.

Classical conditioning describes the formation of associations between stimuli. The horse learns that the rattle of the feed bin means dinner is coming. The horse learns that the float means a long, uncomfortable journey, or alternatively, that the float means hay and a comfortable arrival somewhere good. These associations form whether we intend them to or not, and they have powerful effects on a horse’s emotional response to everyday events1.

Habituation is the process by which a horse stops responding to a stimulus through repeated, harmless exposure. The pony who stops flinching at the tractor in the next paddock has habituated, and done well with intent to desensitise, this is the foundation of producing a calm, sensible horse who can cope with the variability of the human world5.

Done badly, habituation can tip into flooding, which is the prolonged forced exposure to a stimulus the horse cannot escape. Flooding can produce horses who appear calm but remain physiologically stressed, with the potential for sudden and explosive responses later5. The distinction between habituation and flooding is not always obvious from the outside, and getting it right is one of the most important skills in working with young or fearful horses.

What was actually going wrong with Rory

Rory had a long history of reliable cue responses; she knew what leg pressure meant, what a lead rope cue meant, and what being asked to walk forward looked like, and none of that knowledge was lost when she came to me. What I did, slowly and without realising it, was degrade those responses through inconsistent application of my cues and reinforcers.

This is the part most horse owners are never taught, and it matters: a trained cue is not a permanent installation but a learned response that has to be maintained through consistent reinforcement. Every time you apply a cue and release at the wrong moment, you are not teaching the horse nothing; you are teaching them something, and often what you are teaching them is that the cue no longer means what it used to. Equitation science researchers describe this as the dilution or extinction of trained responses through inconsistent timing6. Releasing pressure a fraction of a second too late teaches them that what they were doing at that moment was correct, so if they were bracing, you have reinforced bracing, and if they were stopping, you have reinforced stopping. Do this enough times, and the original cue stops producing the original response. The horse is not being naughty; they are responding accurately to the new pattern you have unintentionally trained.

Three things determine whether any training method works: clarity of the cue, timing and consistency. With Rory, my cues were not clear, my timing was off, and my reinforcement was inconsistent. All three had quietly slipped, and the original training had quietly slipped with them.

This is one of the most common problems I see, particularly with new horses in a new home. Riders assume that because the previous owner had the horse responding well, the horse is “trained” and will continue to respond regardless of how they are now being handled. The horse may indeed be trained, but training is a moving target, and within weeks of inconsistent handling those responses can be quietly diluted into something quite different. This might be a familiar scenario, new horse owners that within a month or so are complaining that the horse sold to them was “not as advertised”.

A trained cue is not a permanent installation. It is a learned response that has to be maintained through consistent reinforcement.

In Rory’s case, my inconsistent pressure-release at the float diluted her loading response over a series of attempts that I genuinely thought I was handling well. Worse, the repeated unpleasant sessions began to layer a classically conditioned negative association onto the float itself, and each helper who applied more pressure compounded both problems. We were not just dealing with a confused horse; we were dealing with a horse who had been actively retrained, by accident, to do the opposite of what we expected, and who was beginning to find the entire context aversive.

Fixing it

The fix required two changes. First, the negative association with the float needed to be broken down, so I stopped trying to push her on at all and instead broke loading into very small approximations, using positive reinforcement (food and calm relief) to reward any forward movement toward the float. This was straightforward counterconditioning. The float needed to mean something good again before any cue I applied could be expected to work properly.

Second, I had to clean up my own pressure-release. Once Rory was relaxed around the float, I reintroduced familiar cues with deliberate, consistent timing, and paired them with positive reinforcement when she responded correctly. The combination of clear pressure-release and positive reinforcement is well supported in the loading literature as both more effective and lower-stress than pressure-release alone4. She loaded reliably within a small number of sessions and was back to self-loading shortly afterwards.

Rory had not changed and the principles of learning theory had not changed; what changed was my own awareness of what was actually happening between us, and my ability to apply the right tool for the situation.

Learning and welfare are linked

Understanding how horses learn is not only a training issue, but also a welfare issue.

A horse who is frightened, flooded, or shut down is not learning effectively, because stress hormones interfere with memory consolidation and disrupt the cognitive processes required for new learning to occur8. A horse who has gone quiet because they are overwhelmed may look, to an untrained eye, like a horse who has accepted the lesson, but they actually have not. They are coping. The behavioural outcome can look identical from the outside, but the internal state, and the long-term consequences, are very different. You have a horse that may become suddenly unpredictable, and that is a safety issue.

A horse who stops resisting might have learned what we intended, but they may also have ‘given up’, and even experienced handlers could mistake one for the other.

Now, some of the well-meaning advice I was also given when Rory and I were struggling, was to be more assertive and show her who the boss is. This might sound familiar. However, research has substantially dismantled the idea that horses operate on a strict linear and that humans need to be the alpha or the leader of the herd. Studies of free-living and feral horse populations have consistently found that equine social structures are more nuanced and based on long-term affiliative relationships rather than rank7, 9. The “be the boss” framing common in some training traditions is not supported by current research on equine social behaviour, and the persistence of this framing in horsemanship has been raised repeatedly as a welfare concern7.

Where to from here

For now, the most useful thing you can do is start paying close attention to what you are actually reinforcing. Spend a single ride or grooming session watching yourself rather than your horse, noting every release of a cue, every scratch, and every walk-away, and ask yourself whether the horse had just done what you wanted at that moment or whether your timing was a beat off. If you have a new horse who used to do something well and is now doing it less well, ask yourself the same question, because the answer is almost always in the consistency of your own handling rather than in the horse.

Every interaction you have with your horse is training them, whether you intend it to be or not. Your horse is always learning. Understanding this is what allows you to be deliberate about what they are learning from you.

References

  1. McGreevy, P., & McLean, A. (2010). Equitation Science. Wiley-Blackwell.
  2. Randle, H., & Waran, N. (2019). Equitation science in practice: How collaboration, communication and change can improve equine welfare. Journal of Veterinary Behavior, 29, viii–x.
  3. Skinner, B.F. (1938). The Behavior of Organisms: An Experimental Analysis. Appleton-Century.
  4. Hendriksen, P., Elmgreen, K., & Ladewig, J. (2011). Trailer-loading of horses: Is there a difference between positive and negative reinforcement concerning effectiveness and stress-related signs? Journal of Veterinary Behavior, 6(5), 261–266.
  5. Christensen, J.W., Rundgren, M., & Olsson, K. (2006). Training methods for horses: habituation to a frightening stimulus. Equine Veterinary Journal, 38(5), 439–443.
  6. McLean, A.N., & Christensen, J.W. (2017). The application of learning theory in horse training. Applied Animal Behaviour Science, 190, 18–27.
  7. McGreevy, P. (2012). Equine Behavior: A Guide for Veterinarians and Equine Scientists (2nd ed.). Saunders Elsevier.
  8. Mendl, M. (1999). Performing under pressure: stress and cognitive function. Applied Animal Behaviour Science, 65(3), 221–244.
  9. VanDierendonck, M.C., de Vries, H., Schilder, M.B.H., Colenbrander, B., Þorhallsdóttir, A.G., & Sigurjónsdóttir, H. (2009). Interventions in social behaviour in a herd of mares and geldings. Applied Animal Behaviour Science, 116(1), 67–73.
  10. International Society for Equitation Science. www.equitationscience.com

Every due care has been taken to ensure the information herein is based on sources Veterinary Nurse Solutions believes to be reliable, but is not guaranteed by us and does not purport to be complete or error-free. As such, we do not warrant, endorse or guarantee the completeness, accuracy, and integrity of the information. You must evaluate, and bear all risks associated with, the use of any information provided hereunder, including any reliance on the accuracy, completeness, safety or usefulness of such information. As part of our quality control of information contained within this document, it has been peer-reviewed by qualified animal care professionals.

Veterinary Nurse Solutions acknowledges that there is more than one way to carry out many of the tasks described within this website, and techniques omitted are not necessarily incorrect.