What Makes a Training Method Good? The ISES Ten First Principles of Horse Training

Why does horse training need a method-agnostic framework?

Most horse training comes packaged as a method or a brand. Natural horsemanship, classical dressage, western pleasure, eventing, in-hand work, liberty work. Each has its own vocabulary, its own characters, its own equipment, its own loyal followers. Underneath all of them, however, the same set of biological, psychological and behavioural realities is at work, because the horse does not change its species depending on whose paddock it is in.

The Ten First Principles describe those underlying realities. They were developed and continue to be refined through international scientific consensus, drawing on peer-reviewed research in learning theory, ethology, and equine behavioural science^{1, 2, 5}. They are not a method themselves. They are the framework against which any method can be evaluated.

The first four principles are about the horse you are working with. The next four are about the science of how horses learn. The last two are about how the two come together in the everyday work between you. I’ll walk through them in groups.

In our own minds, we are not punishing the horse. We are “getting on with it.” But functionally, we are punishing a horse who has clearly communicated that the question we are asking is genuinely new to them.

Knowing the science

5. Correct use of habituation, desensitisation, and calming methods

Habituation is a form of non-associative learning. This is when a horse learns to stop reacting to stimuli that have no consequence for them. The horse is not learning that one thing predicts another, and they are not learning that their behaviour produces a result. They are simply learning that this particular thing, on its own, is not worth reacting to. It’s not a threat to them. The horse who flinched at the neighbour’s tractor on the first day stops flinching after enough days have passed with the tractor doing nothing bad.

Systematic desensitisation is the process we use to produce habituation deliberately. You expose the horse to a feared stimulus at a low intensity (the clippers across the yard, off, then in your hand) and raise the intensity in small steps. The horse stays calm at each level long enough to habituate to it before moving to the next. Empirical work has shown that this structured, stepwise approach produces calmer responses to novel stimuli than ad hoc or forced exposure⁶.

Calming methods sit alongside desensitisation and include counterconditioning (pairing the scary thing with something good), approach conditioning (letting the horse approach the object on their own terms), and overshadowing (giving the horse a stronger known cue that takes precedence over the fear response)⁵.

Done well, these methods produce a calm, sensible horse who can cope with the variability of the human world. Done badly, they tip into flooding, which is prolonged forced exposure to a stimulus the horse cannot escape from.

What this looks like in practice is working at a level of arousal the horse can manage. The horse who is alert but soft is learning. The horse who is rigid, wide-eyed, and holding their breath has gone past what they can process, and learning has stopped. We work below that line, building tolerance one small step at a time, using food or a wither scratch where appropriate, and giving the horse a way out of the situation if they need one.

What it looks like when the principle is being ignored is tying a horse to the post until they stop trying to leave, pushing them on and on and facing them up to a frightening object, spraying the bottle on a horse pulling against the rope, or running the clippers next to a sweating, planted horse who has clearly said no. The phrases sound familiar: “they have to learn”, “don’t give in”, “it’s just a piece of paper”, “don’t let them win”. Flooding can produce horses who appear calm but remain physiologically stressed, with the potential for sudden and explosive responses later, often in unrelated situations.

6. Correct use of operant conditioning

Operant conditioning is a form of associative learning. The horse does something, something happens as a result, and the horse is more or less likely to do that thing again depending on whether the consequence was rewarding or aversive. This is happening every minute the horse is awake, not just when we are actively training.

The horse leans through a fence because someone forgot to turn the energiser on, and they get the sweet green grass. Guess what, they will do that again. The tasty grass reinforced the behaviour. This is positive reinforcement happening naturally. Until we remember to turn the energiser on and they try again. This would be positive punishment in action. Touching the now hot fence added a very aversive consequence, lessening the chance they will try that again.

Positive just means something was added, reward or aversive. It also works the opposite way. We can take something away to change the behaviour.

Pressure-release training is negative reinforcement. We put our leg on, the horse moves off, and we take our leg off, releasing the pressure. Pressure is an aversive stimulus that we apply to get the response, and the right response means it disappears. Negative just means we are taking something away.

And the opposite of negative reinforcement is negative punishment. The horse becomes pushy at the gate, so we walk away. We are removing something rewarding (the human), which in turn punishes the pushy behaviour.

The principle is asking us to use all this correctly: the right cue, the right timing, and consistency every time, regardless of which quadrant of operant conditioning we are using.

What this looks like in practice is using the lightest cue that produces the response, then releasing pressure the moment the horse responds correctly. A fraction of a second too late and the horse is being reinforced for something other than what we intended. Pairing the release with a wither scratch or a piece of food when introducing something new, because combined reinforcement is more powerful than pressure-release alone, with peer-reviewed evidence showing positive reinforcement produces measurably less stress and faster learning in both novel training contexts and the retraining of established problem behaviours⁷. The cue you needed last week should be the cue you need this week, or lighter. If you find yourself needing more pressure than before to get the same response, something has slipped.

What it looks like when the principle is being ignored is pressure that stays on after the horse has responded, so the horse stops trying. Cues that change from day to day, with reinforcement that arrives late or not at all. The horse who paws at the gate before dinner, day after day, until pawing has become the reliable way to summon dinner. The horse who spooks at the corner, the rider gets off and walks home, and the horse has just learned that spooking ends the work. The horse is not doing it on purpose. The horse is doing what they have been taught, by us, because every consequence we provide is teaching them something whether we mean it to or not.

7. Correct use of classical conditioning

Classical conditioning is the other form of associative learning. While operant conditioning is the horse learning about their own actions (if I do this, that happens), classical conditioning is the horse learning about the world (when this appears, that follows). The horse learns that a previously neutral thing, a sound, a smell, a piece of equipment, reliably predicts something meaningful, and they begin to respond to the neutral thing as if it were the meaningful thing itself.

The rattle of the feed bin is classical conditioning at work. The bin does not feed the horse, but it predicts dinner so reliably that the horse is at the gate before we have arrived, head up, alert, mouth ready. The horse who panics at the sight of the float without being asked to load is the same mechanism in the other direction. The float does not hurt them, but it predicts a journey they have come to dread.

The principle is asking us to be deliberate about the associations we are building, because the horse forms them whether we mean to or not. Every time we pair two stimuli together, the horse is taking notes.

What this looks like in practice is pairing a light cue (a voice command, a hand signal) just before an established cue (rein contact, leg pressure), so the light cue eventually predicts the pressure-release sequence and produces the response on its own. Building positive associations with the float, the wash bay, the vet, the farrier, by pairing them with food, scratches, and quiet. And noticing when a neutral cue is starting to predict something aversive, and intervening before the association sets.

What it looks like when the principle is being ignored is a horse who panics at the sight of the float because every previous trip was stressful. A horse who becomes anxious at the smell of the clipping kit. A horse who tenses the moment the saddle comes off the rack, because the saddle has come to predict back pain. The horse is not being “difficult”, “naughty”, “stubborn”, “silly”, or “just trying it on”. The horse has accurately predicted, based on previous evidence, what is about to happen, and is responding to the prediction.

8. Correct use of shaping

Complex behaviours are built from simple ones, one small step at a time. The principle says to break tasks into the smallest achievable steps and reinforce each step toward the desired behaviour. Plan the training to make the correct response as obvious and easy as possible. Maintain a consistent environment to train a new task and only change one contextual aspect at a time, whether that is the location, the cue, the trainer, or the level of distraction.

What this looks like in practice is introducing a new behaviour in the quietest, most controlled environment possible. Reinforcing the smallest piece of the desired behaviour first. Adding difficulty gradually. Changing only one variable at a time once the behaviour is solid in the easy version, where “solid” means at least three to five clean repetitions in a row². The horse who walks calmly over the bridge obstacle from both directions, five times in a row, is ready for one variable to change, perhaps the bridge moved to a slightly different position, perhaps a different person leading, perhaps a different time of day. The horse who has done it once or twice is not.

What it looks like when the principle is being ignored is introducing a new behaviour in a busy or distracting environment and getting frustrated when the horse cannot focus, asking for the full finished version of a behaviour from the start, or changing the cue, the location, and the criterion all in the same session and then concluding the horse is “being difficult” when they cannot keep up.

The horse is not doing it on purpose. The horse is doing what they have been taught, by us, because every consequence we provide is teaching them something whether we mean it to or not.

What three things does every good training method depend on?

Across all ten principles, the operational expression comes down to three things any good training depends on: clarity of the cue, timing, and consistency. Principle 9 is the clarity. Principle 6 is the timing. Consistency runs through all of them. The principles are the framework; these three pillars of clarity of cue, timing and consistency are how they show up in your everyday work, in any method, with any horse.

The horses you work with cannot tell you in words which principles are being upheld and which are being violated. They can only tell you in behaviour. Once you have the principles, you can read that behaviour for what it is.