This was the second of three articles I wrote for Horsemanship Magazine....
Psychology of training - natural or not?
It’s a piece of marketing genius. First identify a population of horse owners who really love their horses, but need to learn more about how to do things with them whilst staying safe. Then convince them that the most “natural” way to develop a relationship with their horses is to emulate equine behaviour and engage in a number of techniques that are sufficiently aversive that those owners probably wouldn’t have been prepared to use them otherwise. Finally develop a progressive training method that ties clients to you in a state of dependency.
I’m not being entirely cynical; I believe that most “natural" trainers do so in the firm conviction that they are indeed helping horses this way. But when we start to assess the learning processes taking place - from the perspective of the horse, rather than the intentions of the trainer - I think a very different picture emerges.
Firstly, can we really emulate horses’ natural behaviour? Humans with their puny bodies, immovable ears and mixed messages, such as pretending to act confident when their body language betrays their underlying fear. In fact, it is hard to believe how anyone could believe that a horse could possibly view us as horses. To be fair, one could argue that I am being unreasonably pedantic here. It’s not about body shape, it’s about behaving in a manner in-keeping with equine ethology. And so even though horses don’t think that we are horses, they still find it easier to understand us if we behave the way they do.
So how about the assertion that we become the leader in our “herd of two”? Does this help? It is true that horses live in herds and that they have leaders of some description. But unlike a trainer assuming perpetual leadership, genuine horse-horse interactions have a mutual give and take and the leadership is earned rather than demanded. When aversive training techniques are implemented - using emotional and/or physical pressure - the horse may settle for the easy life, shut down and comply, sadly often confused with relaxation. Other horses, through fear, confusion and/or frustration, may resist and we are told they are “dominant”. The dominance paradigm has been around for a long time and, although widely debunked as a myth in canine behaviour, the horse version has remained somewhat persistent. The leader, or “dominant” horse in a feral herd is typically the mother of the offspring comprising the rest of the herd, not a bully attempting to take over the herd by force. She is simply the leader by virtue of knowing where desirable resources are and is worthy of following.
We also need to ask whether horses actually behave as portrayed in training methods. Horses certainly do use "advance and retreat” when they investigate something mildly frightening in their environment and so this is an approach that they indisputably understand. Being a flight animal they will flee, but even flight animals can’t spend their whole lives running. They need perspective, so will stop at a safe distance and investigate. Then maybe approach if it feels safe to do so. Maybe retreat again if they inadvertently over-face themselves. And over time they will learn what is unnecessary to flee from after all. But there is a key difference between this and the way in which “natural” methods encourage horses to face their fears. Choice. Horses choose their pace, they know exactly where their boundary lies, between feeling safe and feeling over-faced. When we accelerate this process we end up using a process known as flooding. We force the horse to confront his fears, albeit often apparently with success, but at significant cost to the psychological health of the horse. Wild horses do indeed occasionally send another horse into flight, but very rarely compared with the other, more affiliative interactions in which they all engage. Wild horses do indeed fight over resources but, again, very rarely in a truly natural setting. On the other hand, domestic horses are more likely to be kept in randomly assigned herds, rather than natural family groups, on too small a plot of land and with limited resources. This significantly increases the likelihood of agonistic interactions, wildly distorting our view of herd life and reinforcing our erroneous belief that there is a linear hierarchy with a “dominant” horse at the top.
So what do horses really think of our training? When we stand in the centre of a round pen and send a horse into flight, what message are we really giving that horse? Or when we thwack a rope on the ground next to a horse to encourage him to “raise his energy”? Or when we use pressure and release to encourage a non-loader into a horse-box? Are we tapping into “natural behaviour” or are we invoking some other form of learning?
Natural horsemanship techniques typically work via “pressure and release”. To apply the pressure, trainers send horses into flight, apply physical pressure via halters, restrict horses’ escape via round pens, expose horses to scary stimuli and all manner of combinations of these techniques - with varying degrees of aversiveness. The release of the pressure then indicates to the horse that he has performed the desired behaviour. This process is known as negative reinforcement, defined as the cessation of an aversive stimulus and meaning that something (i.e. pressure) is taken away in order to elicit a behaviour. The stimulus might not be very aversive, it could be very mild, but arguably it must be slightly aversive from the horse’s perspective for its release to have any effect.
Sometimes it will be the application of pressure that stops a behaviour from happening. This is the definition of positive punishment, i.e. adding something. Again, the pressure won’t necessarily appear to the human to be very big. But for it to have an effect on the horse’s behaviour, it must have been fairly meaningful to the horse. Punishment doesn’t need to apply physical pressure to a horse - a “thwack” on the ground with a rope to stop a horse coming too close can be just as punishing.
There’s no need to take my word for this. There is a growing body of scientific research investigating “natural methods”, in particular those methods using round-pen training (Henshall & McGreevy, 2014). It is widely agreed that the methods work but the reasons why
the methods work is more open to debate. The consensus emerging from research is that the methods appear successful, not because of the development of a harmonious, natural horse-human relationship, but because predictable patterns of behaviour are repeated according to learning theory (Krueger, 2006). One fascinating piece of research illustrated this point beautifully. The researchers employed a popular round-pen training technique but, instead of having a human send the horse into flight, a remote-controlled toy car was used to apply the pressure. The “relationship” developed between horse and car was remarkably similar to that of the horse and trainer - much less "nature" and much more mechanical stimulus-response conditioning in action (Henshall, Padalino & McGreevy, 2012).
Not everyone feels comfortable about using pressure-based methods of horse-training and will attempt to include positive reinforcement, arguably “natural” on account of the affiliative behaviours well-socialised horses express. For positive reinforcement to be genuine it is not enough to simply add a nose rub onto the end of an otherwise aversive training session. The reward needs to be truly motivating to the horse, sufficiently so that it is the reward that causes any change in behaviour. The inclusion of positive reinforcement may help to mitigate some of the aversive training preceding it, although is often fairly meaningless if the horse finds the session generally aversive. Paradoxically it might even contribute to the aversiveness if the horse feels under pressure to work out how to earn the reward. Like any other form of training, the use of positive reinforcement needs to be studied carefully and learnt with the help of a professional behaviourist.
The idea that our horsemanship is natural is undeniably appealing but there is no evidence to suggest that any methods are more natural than others. All methods appear to work via psychological processes of learning and, since horses learn via natural processes in the brain, we could argue that all horsemanship is as natural - or not - as any other. But that’s not to say the horses will necessarily be enjoying it. If we could only start looking at how to make our horsemanship truly beneficial and enjoyable to the horse then we will do a much better job of creating the horse-human relationship that we crave. This will be the subject of my third article for Horsemanship Magazine in the next issue...
Henshall C., McGreevy P.D., 2014, “The role of ethology in round pen horse-training - A review”, Applied Animal Behaviour Science, 155, pp. 1-14
Henshall C., Padalino B., McGreevy P., 2012, “The radio-controlled car as a herd leader? A preliminary study of escape and avoidance learning in the round-pen”, Proceedings of the 8th International Equitation Science Conference, p. 157
Krueger, K., 2006, “Behaviour of horses in the “'round pen technique’, Applied Animal Behaviour Science, 104, pp. 162-170
1. Most horse-horse interactions are harmonious and affiliative
2. Natural training? Or stressed, anxious horse performing under duress?
3. Natural relationship? Or tense, shut-down horse with tight tack?