Professor Paul McGreevy, Faculty of Veterinary Science, University of Sydney, 14 November 2014
MARTHA SEAR: Hello everybody. It’s terrific to see you. Thank you for coming along today to the National Museum of Australia for the third in our Spirited: Australia’s Horse Story lecture series.
It’s a great pleasure for us today to welcome Professor Paul McGreevy from the Faculty of Veterinary Science at the University of Sydney to the Museum to speak with you. Paul was involved in the exhibition’s development as a scientific adviser. Many of you will know his work. He is a member and co-founder of the International Society for Equitation Science, and his research interests include equitation science, learning theory as it applies to animal training and behaviour modification. It’s a real privilege to have Paul speak to us today about how horses think and learn.
The format of today will be that Paul will give his presentation and then we have some questions that have been submitted by email. We will talk through those, and then there will be an opportunity for a general discussion. So if you have questions, we will look forward to getting those from you toward the end of the session today. I did just want to note that, because Paul is a vet, he can’t give advice on specific clinical or behaviour cases today as he has not himself been able to see the animal. If we can keep the questions in a more general vein, that would be fantastic.
It’s a great pleasure to welcome Paul and to invite him to tell us about what he has learnt about how horses think and learn. Thank you, Paul. [applause]
PAUL McGREEVY: Thank you, Martha, and thanks everybody for making me welcome. It feels as though I am among friends. I am sure that you are all as pony mad as I am. If you see something that you’d like to question me on during the presentation, that’s fine too. It’s not a formal lecture, and you won’t be examined at the end.
I would like to thank the Museum for all the opportunities it has given me over the last 12 months and also the faculty, because the Faculty of Veterinary Science is very supportive of my work even though at times it is quite contentious work that I am presenting. These are my research colleagues [image shown]:- these are my beautiful horses. That’s the front office [image shown]. They teach me all the time. I am not supposing for a moment that I remotely know everything. I am just fascinated on the journey that the horses are taking me along with them.
The way they communicate with each other is a study in itself. When you see a photograph like this [image shown], you realise there is communication going on all the time. They are communicating with each other in terms of the space they are going to occupy, who they are next to, who they want to avoid being next to - and all of this can happen at the gallop and in terrain they don’t actually necessarily know. That’s a reminder that they are learning and they are functioning in a completely different way to humans. They see the world through different eyes and they feel the world differently through their skin and through their feet.
A lot of what we do is to do with taking a humility pill and accepting that we actually can’t clone inside their brains and know what they are thinking, experiencing or suffering, but if we spend enough time with them I think we can learn some valuable lessons. I have been lucky enough to spend a total of three sabbatical periods, six months each, just studying horses in paddocks and in the feral state - so watching brumbies - and that’s where I get a lot of my research ideas from.
Just another gratuitous horse photo really [image shown] - these are Friesian crosses, and most of them have been born on the property. I have seen them develop and it’s been a fascinating journey. There are some really interesting familial relationships that I can observe.
My partner Rob took this photograph [image shown]. It speaks volumes about the dynamics of social interactions between horses. These are two mares with their foals, and they are just meeting a younger sister. She is on full observational duty to remain safe, and these foals have arranged themselves to remain safe as well. It reminds me about the importance of safety to horses. Horses are looking for safety signals all the time, and clearly it’s more valuable to horses than food on occasions. So the way we use safety in our training is, I think, going to become more and more clear and important. Yes, they use food. I have a really nice question that has been sent in about the use of food in training. I am really interested in how it can be used and the limitations of using food in training.
I mentioned that horses are social and they are tactile. [image shown] This is a lovely illustration of the areas on the horse’s body that are regularly or rarely allogroomed. Autogrooming is grooming yourself; and allogrooming is grooming another. This is a bonding behaviour, especially after horses have been through some sort of traumatic event. Say, for instance, they have been through a saleyard they will often find a horse to bond with. The fascinating thing is they lower their heart rates when they are doing this so they are in a very safe place. A horse with a lower heart rate is largely a happy horse because it’s not being preyed upon; it’s not in a dangerous situation. We can actually hijack that by scratching the horse in those hot spots. That red area is an area that we can very easily reach from the saddle and we can scratch horses and caress them there. That’s a way of rewarding horses.
We could also groom them in places where other horses can’t. Just between the front legs is a place that many of you will recognise is a really itchy spot for horses, especially young horses. Foals will just go off into seventh heaven when you groom them in that area. So we can actually do things for horses that other horses can’t do.
My use of grooming is how I first get on horses when I am doing foundation training. [image shown] I realise I should be wearing a helmet and not a stupid grin. The point is that these horses are just having a lie down in the paddock. Then the bipedal, the monkey comes up, and starts giving them some really important scratches and, if they stay put, the scratching carries on. So they get paid for being quiet and still, and that allows me to be the monkey on their back eventually. It’s all mediated through touch and scratching. They are not being given food rewards. There is absolutely no fear. They are actually being paid to stay still. That’s the first time they have seen a monkey on their back. The moment they look at though they are going to move, the scratching stops. If they stay still, they just get scratched and they seem to thrive on that.
The whole area of horse behaviour is a booming area for researchers. When I wrote this book in 2004 [Equine Behaviour: A Guide For Veterinarians and Equine Scientists] I was amazed that I could find 1,000 peer reviewed references for horse behaviour and they went into that book. There was a chapter in that book called ‘Equitation science’ about the training of horses. That’s something that has developed over the last 12 years largely thanks to my greatly valued colleague and friend Dr Andrew McLean. Andrew and I appreciate the different styles of horse training and the wonderful ways in which horses can be trained in different disciplines, but we have both had this scientific background so we are required to understand the mechanisms that underpin horse training.
It’s actually round pen training that forced us to demystify what’s going on in a round pen, and none of this should ever devalue the love we have for horses or that horses may even have for us. So please don’t think that, because I am a scientist, I am here to take all the fun away. I am still in awe of what we can do for horses and what horses do for us. But understanding the mechanisms is really important because it takes away the confusion. Confused coaches and confused riders create confused horses. Confused horses are not having a good time; confused horses are dangerous. That’s what motivates me knowing we can begin to clarify some of the mythologies that are doing horses no favours.
Equitation science has really taken off. There is a series of annual conferences. It started in Melbourne and the last conference was in Denmark. They are all over the place. If you are interested, the next one will be in Vancouver, Canada. We have a Canadian colleague here to fly the Canadian flag. The last conference that I came from in Denmark was absolutely exceptional. There was brilliant science being rolled out. What we have now is an ability to encourage young scientists in indulge their passion for horses and go on discoveries using the horse as a vehicle. It’s a really exciting journey. You are welcome to join the society; you don’t have to be a scientist. There are opportunities for lay people to join the society. http://www.equitationscience.com/
The purpose of equitation science is to basically explain what happens when horses learn, and that also explains when horse learn to behave in ways we don’t want, which that includes some of the dangerous conflict behaviours we know about. We have developed technology now that can pick up tension in the rein and can pick up pressure under the saddle and under the rider’s leg. That’s the holy trinity of contact when you are riding a horse. That’s how you use feel to communicate to the horse. We now have technology that can actually explain and reveal what’s going on when a good rider gets a nice tune out of a horse. One day I predict that we will be able to have coaches working with riders internationally through the web saying, ‘You need less tension in the right rein and more pressure through the left stirrup.’ That would be a dream, because we can then access some amazing coaches using this technology.
We also use things called accelerometers on the horse’s legs to work out stride characteristics, and that can tell us stride length, stride gate and stride frequency. And most amazingly we can now use the same accelerometry technology to work out who is pulling who, because we know that a lot of tug of war can happen and develop in the ridden horse. We have developed a system now to help us work out who is actually initiating that tug of war, which is really important.
The other thing we use is ethology - that is the study of natural horse behaviour. So we are using technology; we are using ethology; and we are using psychology. Over 100 years of psychology studies have informed how we train animals. Of course, the focus was on the rat in all of those psychology studies. There is a lot we can take from those studies that tell us how other animals learn but, as you might have noticed, you can’t ride a rat. The horse is actually the best model for understanding how pressure can be used in training an animal. So this journey is now being carried forward with the horse.
Why are vets becoming interested in this? We know that the behaviour of a horse is going to affect its value, and the value of a horse is going to affect how much veterinary costs are going to be rolled out to look after the horse. So vets are become more interested in this. But they also recognise that the sustainability of horse use depends on doing it ethically. Now we have this conversation going on about what ethical equitation looks like and what ethical racing might look like. The whole area of ethical horse use is being underpinned by this sort of scientific framework.
We know that we can begin to study the horse/human bond. We can understand how being nearer a horse can raise your heart rate or lower your heart rate. How your heart rate can affect a horse’s heart rate when you are riding it. This is something we are becoming especially interested in, and it keys in with what we are finding in dog/human bond studies.
We also know that horses that are not fearful are better, safer patients. Horses kill more Australians than any other animal. We can forget sharks and killer bees. Horses are, unfortunately, killing more people in developed countries than any other animal. So the calmer we can make them, the safer they are to be around. And of course for a horse that needs treatment, the calmer you can make it, the safer it is for the veterinary personnel around it. It is extremely important for vets to understand this.
There is this awful word ‘wastage’, which describes how we discard animals that are no longer useful to us. The figures that you see bandied around about wastage in different sports are horrific because it tells us that we are getting it wrong a lot of the time; we are training badly a lot of the time; we are also breeding animals that will be never destined to perform. So we need to use science to ensure that wastage is brought down year by year. All of this keys in to welfare because, when people don’t value their horse, they don’t actually want to spend much time with it; they don’t want to spend much money on it; and that’s when we get horses just turned out into paddocks and forgotten. All of this science should ensure that we see fewer horses discarded in this way.
What I wanted to remind you about is that science is an antidote to unsubstantiated belief systems, but science can also reveal the good stuff about belief systems. It can also show you why what your grandfather did with horses works; it can also challenge some mythologies and make sure the horses are getting the benefit of an evident-based approach.
The reason I think that’s a good idea is that, if we think about human medicine, can anybody tell me what has happened to this person? Trepanning, yes. For centuries it was a way of treating mental illness by letting the evil spirits out of this girl. [image shown] You can see this poor person had several therapies. They went very quiet after this treatment, unsurprisingly. In fact, in the short term people thought ‘Oh it’s working’, but you can see how centuries of human care were based on belief systems that just don’t stack up. We need to have the bravery to challenge what we have always done.
I am going to talk a little bit about dressage because dressage comes from the French ‘dresser’, to train an animal. That doesn’t mean I have an obsession with dressage. I think it’s actually the best horse sport to study if you want to look at horse learning because it’s basically supposed to be about training. But if you think that I am coming across as being dressage basher, please don’t think that. I think it can be the most elegant form of horse use if it’s done well, but I am going to show you some ways in which it could have deviated off the path in some ways.
We know that when we change animals’ behaviour we are making some responses more probable and some unlikely. We can use either aversive stimuli including pressure or attractive stimuli including social contact, food, scratches and so forth. [slide shown] If you are using those attractive stimuli, you are in this quadrant called positive reinforcement. If you are trying to get rid of a behaviour using aversive stimuli, you are using punishment. That could be quite mild. You could be using very mildly aversive stimuli. You don’t have to be bashing the horse around the head with a piece of 3x2.
If you are taking an aversive stimulus away to make a behaviour more probable and more likely, then you are using negative reinforcement. That means you are removing something, this aversive stimulus, to tell the horse it’s done the right thing. Those of you ride regularly or have ridden will know the importance of negative reinforcement in horse training. So much of it is about pressure release. The technical term in learning theory is negative reinforcement, because we are subtracting that aversive stimulus.
We can even use negative punishment where we remove something that is attractive, and I will give you an example. If you have a horse after you have ridden him with an itchy head, a sweaty head, he might want to rub his head on your shoulder or on your body, you can actually keep him still by scratching him. The moment he moves, you stop scratching, and you’ve got a horse that very quickly says, ‘I’ll keep my head still. I know that.’ What you are using is negative punishment. You are taking away that head movement by and you are making that less likely - his head is less likely to move - because every time it does you punish him slightly with the removal of that reward, that scratching.
These terms like punishment - the prospect of people punishing their horse with removal of a scratch seems awful, doesn’t it? But those are the technical terms that have come from psychology. That’s why we stick with them. But you need to know that, because some people believe that negative reinforcement sounds awful. It’s just an arithmetic term; it’s removing something.
We know the importance of learning theory. It underpins all animal training. I have had the opportunity to understand animal training through sabbatical periods away from the university, and it’s clear to me that the mechanisms are all the same. The species differ and what they value differ, but the mechanisms are all the same.
The pressure principle is also used in camel training with that nose peg, and with the nose ring in a bull. Any pressure applying device so these head collars on these cattle [image shown] – they are Bill and Ben, by the way. They are Short-horns. This is a Doberman. Poor old Dobermans attracted this sort of costume jewellery with very thin choke chains. It’s the same principle. You have to hope that whoever is applying pressure with that device knows the importance of pressure release.
The same goes for when you put a sharp or twisted bit in a horse’s mouth. Any fine surface is going to apply harsher pressures so you want to know that, whoever is applying them, knows how to use the most minimal and subtle pressures and is certainly going to remove the pressure when the horse has behaved appropriately. Putting more weaponry into a horse’s mouth is not the answer. Generally, horses that have developed habituation to pressure, they are damned as having hard mouths. What people don’t realise is that that was manmade and just putting harsher bits in is not the answer. They need to be better ridden.
[image shown] These two guys are reminding us that even animals can train each other using negative reinforcement. This is called the hound couple. When the older hound responds to the huntsman, the younger hound will get pressure. And when the younger hound follows the older hound, then the pressure is removed. So that’s negative reinforcement.
Gentle leaders and head collars in dogs and horses are using the same principle of negative reinforcement. You apply pressure until you get the response and then you take it away.
Elephant training involves an ankhus or elephant hook, and that’s the same principle: you apply pressure until you get the responses you want. [image shown] That’s a modern ankus and an older antique ankus. One may look worse than the other, but what matters is how they are used.
Negative reinforcement is based on the release of the pressure as an instantaneous reward. This is a colleague, one of the council members at the International Society for Equitation Science, who jumps horses without a bit. He is determined to strip any sort of suggestion of magic away. [image shown] It looks incredibly impressive, but he can boil it down to the use of learning theory. He starts the horse off with a bit and then he transfers the pressure to a rope around the neck.
Negative reinforcement underpins all equitation and it can be very subtle but it relies on the immediate release of pressure. It is also going on in a round pen, as I have mentioned. If the horse is not doing what you want, you can put pressure on it. The moment it turns and looks at you, you can take the pressure away. I am not talking about physical pressure; I am talking about the postural pressure that you can put on a horse that is related to pushing it around the pen. So it’s going on all the time.
What you need to be aware of is that, if you dwell too much on this aversive stimulus section, if that’s all you ever use and you are not good as it, then you will use more of it. You will have to because the horse will habituate. If you end up using lots and lots of aversive pressure, the horse will start to shut down, and that’s called learn helplessness. It’s a miserable state to be in. You can also think of it as learned hopelessness - awful.
We are beginning to understand the roles of attachment, arousal, and affective states or mood. What I am giving you is a framework that we are building through the International Society for Equitation Science and through some of my PhD students - we are understanding the role of all these different parties.
The notion that horses can be trainable or untrainable is a really interesting one. We know that early handling is really important, but that can be overdone. Some people advocate putting the horse on its side in the first 24 hours of life and holding it down while it struggles. The same people might argue that the horse has to have its ears poked, its nostrils poked, its mouth poked and even its anus poked 50 times - not 49, not 51 but 50 times - to create this state of calmness. The science doesn’t support that. In fact, the science says that you spend time with the mare and groom the mare and feed the mare, the foal learns more about humans that way than anything to do with wrestling the horse to the ground. Science is beginning to do horses a big favour, I would suggest. Mares don’t like to see their foals being pinned to the ground, and foals are very fragile. They can be damaged in the process.
What we are trying to do is move the bell-shaped curve [slide shown] so that more animals are trainable. That’s our goal. We know that different breeds have different tendencies to react and also different breed tendencies to habituate. It won’t surprise you to know that those amazing racehorses that you see at Thoroughbred Park are reactive. They are reactive to stimuli. You don’t want a dopey, slow, quick to habituate racehorse. So we have bred horses to be reactive. All of the racing breeds are reactive.
Generally the draught horses - the shires, the Clydesdales and the Standardbreds - because they are towing a piece of metal work, you don’t want a reactive animal; you want a horse that is going to be quick to calm down. You can imagine in the days of our forefathers a runaway cart horse was an extremely dangerous missile. So we have probably put a lot of pressure on breeders of draught animals to make sure their product is calm.
I wanted to show you some of the effects of breeding because I believe it can affect the brain. This is something that my colleague from Canada has been working with me. Dr Katrina Merkies from Guelph has been looking at this possibility with me recently. We know that the horse’s vision is affected by the way its head is held, so a horse in this situation is not able to see anywhere near as much as that horse [image shown]. We know that they see the world in a different way. This is a human-eye view of Kings Park in Perth; this is a horse-eye view. You can see that photographer was standing beside the horse. They have wrap-around vision and quite good visual acuity so they can see good detail in their visual field but it’s 320-degree vision wrapping around their head. We can’t really imagine what that’s like.
This is the distribution of cells in the retina once it’s been flattened out in a dead horse [image shown]. You can count the number of ganglion cells that are sorting houses for the rods and cones that pick up light and transmit it to the brain. What we find is that, in dogs the distribution of these cells changes with skull shape. The pug has vision very similar to a human, but greyhounds have vision rather like a wolf. [image shown] The distribution of cells is a streak across the retina; whereas the pug has this area centralis just like us. They see the world more like humans and these greyhounds and the long-skulled dogs see the world more like the wolf.
But, of course, horses have different skull shapes. This is a cheat because that’s a foal and that’s an adult. [image shown] That’s an Arabian and that is a lovely standard bred. So they have different skull shapes. What we have discovered is that horses have different retinas depending on the skull shape. That might begin to explain why they behave differently. If they have different skulls they are seeing the world differently, so they are likely to respond to those stimuli differently.
We know that horses were bred for different jobs and so were dogs - so the influence of breed selection for human jobs. The Arabian breed standard in the United States says that the horse should have a prominent eye, large nostrils and small teacup muzzle. Unfortunately some breeders have interpreted that that the horse’s muzzle must be able to get into a teacup. [image shown] These aren’t photo-shopped. Clearly some people are very pleased with their work. It’s not a giraffe. We know that that’s going to affect the dentition of the horse but we don’t know yet exactly how it’s changing the brain. That’s part of the journey that we are on with Katrina. It’s a horse not a seahorse.
The reason that I think we need to study the brain more is because that can affect the way the horse works. There is a thing called laterality that you might have heard of as handedness. Left- and right-handedness in humans was thought to be a distinguishing feature of humans, but it’s actually been found in every species of animal that has been studied. I have been fortunate enough to study it in horses looking at the way they smell poo because that’s the standard stimulus. We froze a whole lot of stallion poo and then offered it to 186 horses, working out which nostril they were using to smell and they seemed to use their right nostril more than their left. That tells us that the majority of these animals were right brain dominant. Horses tend to examine scary stimuli with their left eye, and that may be why we have the equipment where we tend to approach from the left and we tend to have the straps on the left. It may have been that early horse trainers spotted this.
When we think about the way the horse moves - the trot and the canter are a symmetrical and an asymmetrical gait respectively. So looking at the way horses move is a really important lesson. If you are going to spend a lot of money on a dressage horse that must be balanced in both directions, then I strongly suggest you spend time watching it graze because what we have determined is that horses can have a preference for one leg forward over the other. There is a method I can give you to study this in your own horses. The most balanced horses use the left and right legs equally, but about 50 per cent of horses don’t have that balance so they tend to graze with one leg forward. They start that as foals and they lock it in. We see that 50 per cent don’t show a significant preference - 40 per cent are grazing with the left leg forward and 10 per cent are significantly favouring the right leg. That’s also breed dependent. You see most of it in thoroughbreds and less of it in quarterhorses. I think that’s because of the leg length.
We have studied it with pedometers just putting pedometers onto horses’ legs and letting them graze and we found the same thing. The follow-up to this study was a Dutch group that looked at foals and how they acquired that preference. They found that short-headed foals with long legs were more likely to develop this bias. When we breed horses for sport work, we need to be very careful we don’t obsess with getting long legs and short heads because we are therefore breeding animals that are unbalanced because they graze with one leg forward rather than the other.
It even affects hoof shape. So the way they graze - you can see there is a club foot here. [image shown] The Dutch study show showed that the hoof shape was affected because the horse was putting more weight through the foot that was not advanced, they were putting more weight through the foot beneath them. You find some remarkable things when you look closely enough.
This is hopping to horse sports more generally. This is a baroque breed, a Lusitano. Short legs, would you agree? The old baroque horses that were used for dressage always tended to have short legs and quite big banana-shaped heads. Maybe the ancient horse breeders knew that this is what you get if you breed for a squatter animal.
Let’s just look have a look at the results of a slow motion video analysis. It showed that none of the horses at the Barcelona Olympics achieved the correct amount of flexion for it to be a technically correct passage. You can see these descriptions of what a movement in dressage is supposed to be, but only when you use the right technology do you actually see what’s going on. We know that we are breeding horses for dressage and showjumping more like this beauty and less like this [image shown], but at what cost because we could be breeding in asymmetry, a lack of symmetry.
The sport of dressage has changed enormously over this time period from 1984 to 2008. This is Reiner Klimke winning a gold medal; this is Edward Gal on Totilas; and there is a world of difference in the way these horses are trotting. With this horse – can you see the foreleg that catches the judge’s eye but it will land after the diagonally opposite hind leg? Technically that means it is not a trot any more, because a trot should be a two-time beat. This is a gorgeous horse Matine who died suddenly at only nine years of age [image shown] but to ride her you had to get your legs in quite interesting places. That meant his crotch is in an interesting place as well because he is putting his weight through his wedding tackle.
This search for these extravagant gaits means that the foals are being born with different gaits as well. You can see that that foal is being advertised as a top foal because it has that advanced foreleg. When we do this in German shepherd dogs we end up with hip dysplasia, so I think we need to be very careful about where this is taking us.
The more we understand about the nature of the horse, the better we will be able to suit horses for riders. We will actually be able to match them, because riders are asymmetrical - 90 per cent of us are asymmetrical - and we know that science can begin to get a better match. I think that is part of what ethical equitation will be because, instead of horse forcing the horse to meet our needs, we might be able to meet it halfway.
I would question whether certain humans should actually be allowed to ride horses if they have such a poor match. Here are some questions for you: are some novice riders a danger to horses or a potential threat to their welfare? Should sentient animals be the balancing device on which novice riders learn to sit straight? Is it better to use technology to solve that problem instead of hoping that the horse’s mouth is not being used a balancing aid? I think science will begin to unpick all of this and in the process we will become safer as riders. That will help us move to this more trainable horse future that I dream of.
That is some of the ideas of what we do in equitation science, and I have only really talked about the horse. I could come back and talk about the rider one day. I would love to do that, because the asymmetries and the personalities that we bring to the saddle are really important. Ok folks, I will leave it there and we will take some questions. [applause]
MARTHA SEAR: This is the part of the show where Paul will take some questions from the floor. We have three questions from audience members. I will start with the person who asked the question about modern dog trainers and positive reinforcement.
QUESTION: My name is Rachel and I just bought a horse. My question was that modern dog trainers typically use positive reinforcement in their training. What are the differences between using positive reinforcement with dog training and horse training? What should horse trainers be mindful of?
PAUL McGREEVY: That’s a great question. The horse and dog have evolved in different ways. The horse has evolved to value safety. It also of course does value food. We know when we rattle the bucket the horse will come. But dogs are opportunists. They are looking for opportunities all the time. So they are closer to the rat in that sense. There is more flexibility in their behaviour and how we can capture it with positive reinforcement.
I use food rewards when I am riding. If I want to get my horse to approach something scary, if I get a step towards that I will whistle and then give a food reward. To receive the food reward, the horse has to stop. What I have also created is a horse that will slam on the brakes when I whistle, which can be quite helpful but if you are riding at speed past a whistling person, it’s not so helpful.
The thing to remember is that, when horses find something tasty to eat in the paddock, they stay still, they don’t keep looking around for more so they have evolved to actually slow down and stop. The most elegant use of food rewards in the ridden horse are for calming, not for increasing activity.
There is a good old Aussie inventor who has developed a trick delivery devices so we can put molasses up a line through the mane into a hollow bit, which is something we helped develop. But it does slow the legs down. It will be coming on the market in the next year or so, but we need to be aware of its limitations. They have evolved in different ways so you will get different results. But you should also remember that positive reinforcement in both horses and dogs can involve scratching. That’s another tool in your tool kit. You will become aware of how helpful it is when you start to use it and actually use it to say thank you.
MARTHA SEAR: The second question we had was about horses’ short- and long-term memory. The question was: do horses have a short term and a long-term memory or just memory that they commit certain behaviours to? Tom Roberts trained horses with a ‘this will profit you not’ technique; Andrew McLean uses instant release of pressure. Do you agree with these techniques or has your research found another technique? The second part is: I handle my foals from birth. Some foals are only handled when necessary until broken in. Is there a difference in their learning or training ability?
PAUL McGREEVY: It’s a great question. They do have long-term memory and short term memory. In fact, their long-term memory is equivalent to elephants. We think of elephants as being these incredibly long-term memory animals, but horses are up there with them. That’s why we should always dissociate fear from our training. We don’t want them to ever associate us with fear.
The reason I say that is if you ever get your foot caught in the stirrup and come off, you want the horse to stop and not just keep running. It is so chilling to think we have maybe instilled fear in horses that it’s ended up killing people. We sometimes unfortunately dress that up as a lack of respect on the horse’s part. I never use that approach because I don’t need a horse to respect me. I want the horse to be trained - just keep it really simple. Yes, long-term memory, yes, short-term memory.
When I have spent time studying Andrew McLean’s methods and Tom Roberts, I think they are basically the same. The exquisite timing that both of those practitioners show is the magic. Tom Roberts – he said, ‘If you sit on a pin what do you do? You get up. Why do you get up? Because the pin hurt or because it stopped hurting when you got up? It’s because it stopped hurting when you got up.’ It’s the exquisite timing of that pin because it was so consistent that allowed you to learn the right response. I think Andrew McLean and Tom Roberts have led Australia in understanding the importance of timing, and that’s going to affect the short-term memory; and they have also been careful to avoid fear responses, and that’s going to affect the long-term memory.
But as far as your foals are concerned, if every interaction is a good one, then you are putting money into the bank for when you eventually start foundation training or what we used to call ‘breaking in’. You have to accept if you are going to vaccinate them or restrain them for picking up their feet or putting a microchip in them, that’s a bit of money that goes out of the bank. I think of putting investments into the bank - if every time I go past a foal I can put some money in the bank, then it allows me to have less of a struggle when I have to put a microchip in or do some sort of veterinary procedure. That’s the way I see it. You can certainly keep your hands off but see them as a bank in which you can invest your time.
MARTHA SEAR: Terrific. Thank you for that question. The third question: is horseracing an unnatural activity for horses even racehorses that are specifically bred? The recent Melbourne Cup controversy caused by the tragic deaths of two racehorses resulted in calls to ban all forms of horse racing on the grounds of cruelty. The inquirer has framed the question around getting a veterinary and behavioural science point of view of that kind of question.
PAUL McGREEVY: It’s a great question and it’s very topical. I am sure everybody felt quite sad about what happened at the Melbourne Cup. It seems to have galvanised people’s thinking. Some people will always say that animals should not be used, and that’s an animal rights perspective. If you say that animals can be used but their welfare should be optimised - if we are going to use them we have to ensure we are doing it ethically and humanely - then you are taking an animal welfare approach.
A lot of people don’t understand the difference between animal rights and animal welfare. And this is true also of racing journalists. They are branding anybody who criticises any aspect of racing as being an animal rights activist, and that’s not the case. For my part I would like to see racing continue but I think there is a different way of doing things. If we think about the use of the whip, for instance, we have to accept that tired horses are being whipped and whipping tired horses in the name of sport is very difficult to justify ethically. It’s very tricky especially since in Norway they have had whip-free racing for 30 years; in India they have feature races that are whip free - and people bet on those races and horses win in every race.
If the purpose of the exercise is to find the fastest horse then you actually can do it ethically. The racing industry, I think, is appreciating the importance of innovating. Every sport has to innovate. Every sport is innovating. I think we need to help the racing industry understand there is an interest in their sport and there is an ethical future. I suspect the pressure will eventually come from sponsors of races, because sponsors have to report to their shareholders. If shareholders start saying, ‘I want my money invested in ethical sponsorship,’ then a debate or discussion will emerge about what ethical racing looks like. We don’t have time to discuss all the facets of what ethical racing looks like, but I think science is going to actually help the industry.
MARTHA SEAR: We now have an opportunity to take questions from the floor.
QUESTION: My name is Dick Rowe. Thank you, Dr McGreevy, for a wonderful presentation. I wonder if you could comment on the [Kel] Jeffrey method. As a veterinary student 50 years ago I had an interest in the Jeffrey method but, of course, the science of equitation wasn’t about then. What has equitation told us about the Jeffrey method?
PAUL McGREEVY: Great question. Again, it’s negative reinforcement. You are taking the pressure off when you see the animal responding appropriately, and Andrew McLean started his equitation science journey with the Jeffrey method. There are still people using the Jeffrey method and what they show is, again, amazing timing. If you hold on to ropes too long, you set the horse into a panic and create a flight responses. Knowing when to take the pressure off is the key. It’s the same for all animal training: knowing when to deliver the rewards. If the horse pulls away and it gets release of the pressure that way, if it actually tears the rope through your hand, it has learned a very important lesson. So they are learning all the time even when we are not going to out to train them. The Jeffrey method still stacks up. It’s great that you were taught that at vet school.
Unfortunately, our students are not taught that because we just teach with the quietest horses. You can imagine with naive novice students, often from overseas, we can’t really afford the risk of giving them a tricky horse. We have these wonderful standard breeds that are extremely tolerant, and the students learn how to handle them. But then on day one of practice they could be dealing with a horse upside down in a float on the M23. I think it’s a great shame that we haven’t been able to retain that level of training that you were fortunate enough to receive. It’s interesting times.
QUESTION: I would like to ask your opinion about re-education from the horse’s point of view. For example, a seven-year-old ex-racehorse that had 50 starts, knows the game inside out and been very successful; new owner thinks he would like to have an eventing horse and wants to re-educate that horse. I would like to hear your opinion about that process.
PAUL McGREEVY: It’s a great question because those ex-racehorses are, let’s face it, among the cheaper horses on the market. It’s beguiling to think that for $500 you can get a riding horse for your daughter, but they are, unfortunately, trained in a different way to riding horses. A lot of racehorses are trained to go forward when you take the pressure off the reins. We want horses to slow down when we apply tension through the reins. So we have to re-educate all of that. It is certainly not impossible, but a lot of the horses that we see in this situation are also have also been whipped for trying hard and whipped when they are trying hard so they are often in a slightly frenzied state. If we can peel them off the ceiling we can re-educate them.
The value of safety and calmness is something that the horses are teaching us all the time. The ex-racehorse brings a whole series of challenges, but they are not insurmountable. As it happens, thoroughbreds are becoming perhaps less popular as event horses because the times are less critical, so warm bloods are rising in eventing because fast times are not so critical in cross country. Yes, they certainly can be re-educated. The main thing is getting that deceleration cue back, because the last thing people really want in a racehorse is deceleration.
QUESTION: I have broken horses; I have Friesians and they have a different style of brain as far as I am concerned. Sometimes they can have a lot more attitude or maybe I am just biased - actually I am a lot biased. When training them, sometimes you cannot push to get positive reinforcement. How do you go about getting that positive? You say scratches and that works well. But if you are trying to get a trot you don’t want the horse to stop and say, ‘Yeah I love you,’ and they do. How do you look at that?
PAUL McGREEVY: You can use positive reinforcement to train a trot. You only give the horse the whistle followed by the reward, or a click and a reward, when you have seen the best trot for the day. Part of training is learning to reserve the rewards until you see an improvement - we call that shaping. It’s a battle of wills sometimes, because the animal is saying, ‘Surely that was good enough,’ and you have to say, ‘Well actually, we would like a bit more.’
I have Friesian crosses and as you know they have these huge canters. My horses will canter on just a verbal cue. I know that I have reinforced that positively because if they are in certain parts of the arena they just cantering on the lunge because that’s where they last got a reward. You can use positive reinforcement. But I think they are sensitive souls and a lot of the baroque breeds are. That actually demands more horsemanship on our behalf, on our side. So they are an interesting challenge and I am sure you are up to it, especially if you have had one or two. Each breed comes with its own “instruction manual”. We just need to know how to read it. There isn’t really an instruction manual for Friesians. We have to be students of horse behaviour to read them - that’s what I am saying.
QUESTION: I am a dog trainer and I am interested with horses as well as dogs about whether there is any research into emotional arousal level and in what state they make the change. So in that moment where you might be getting a trot and you want a little bit more, you generally get a heightened level of arousal before that occurs. Is there something that we can use so that our timing can become a little bit more effective as riders and trainers?
PAUL McGREEVY: There is an excellent student of mine who has just qualified with a doctorate, and she and I developed this idea that there is an optimal level of arousal and an optimal level of affective state or mood in the animal for each type of training. Positive reinforcement requires a certain style of arousal and affective state; negative reinforcement might work with a completely different landscape. What we have done is we have put this paper online. If you go online and do a search for Starling and McGreevy, you will be able to download the paper. [‘Conceptualising the Impact of Arousal and Affective State on Training Outcomes of Operant Conditioning’] http://www.mdpi.com/2076-2615/3/2/300
We have horse examples in there of these landscapes. You can actually move them around on the screen to see where you think the optimal level of arousal or affective state is for the tool you have reached for from your tool kit. Again, the animals are telling us where they are working best in those different quadrants of that operant conditioning framework. I would encourage you to read the paper and by all means email me if once you have read it you have some questions. It’s conceptualising the two things that you have identified: the optimal arousal for the behaviour you want but also the techniques that you are using.
QUESTION: Do you want find generally that riders or trainers tend to not give soon enough because of that -
PAUL McGREEVY: I think unfortunately contact has increased. The latest evidence is that a lot more horses are competing behind the vertical than they were 12 years ago at the highest levels, and the marks have gone up. So there is something wrong with the judging system. If I am ever found dead outside the vet faculty it will be because a dressage judge has taken me out. I think it’s really important we get dressage right, and then all the other sports could benefit.
The technology is now exposing some serious problems with the way we are being shown to ride by the elite riders so I think we need to look to technology to get us back on track. We are working with a Swedish group that hope to have reins that change colour when they are pulled too tight so that a coach on the ground could actually say ‘that’s way off the scale’ and nose bands that change colour when they are too tight. I think there are fashions in the way animals are trained. We need to make sure that science at every stage does what it can to keep people honest, to remove the bullshit and to give the animals the best chance of having a happy, working environment.
MARTHA SEAR: Thank you. Can I thank everyone for their participation today and invite you to thank Paul for all that he has contributed to us. [applause]
Thanks everybody. Just before we head off I wanted to let you know that some of the books that Paul referred to in his talk are available for sale today. If you want to have a look they are over there on the table so feel free to go and have a look at that.
You probably are conscious if you have seen the Spirited exhibition that Paul is featured in a couple of videos in the exhibition. We were really thrilled to be able to work with Paul to make those. They are also available online. http://www.nma.gov.au/online_features/horses/features/foundation_training There is a half hour long video that illustrates some of the principles behind foundation training, which many visitors have said they found really interesting.
We would really like to encourage you to visit Unbridled tomorrow at Thoroughbred Park, a terrific event. Hopefully some of you are going. It is on for the first time here in Canberra. It is a horse festival that everyone will find something to enjoy and interest them. There’s a flier on your seats. If you want any information about that, you can have a chat to some of us here. We hope that the event goes spectacularly well tomorrow. Paul will be speaking so, if you would like another chance to hear Paul talk, you have an opportunity tomorrow.
This is the final lecture in the Spirited series of talks. But there is still lots more to come, lots more events associated with the exhibition, including the Spirited Horse Show, which is coming up in January with Tim O’Brien, who is a fourth generation cattleman from Tumburumba. There will be a great family-oriented show around horsemanship with the Museum’s Spirited Horse Show coming up in January as well as our Australia Day Festival, which will be horse themed next year if you interested in coming along.
It’s time to thank you all for coming, thank you to the events and public programs team for putting this all together and thank you to Paul again. Thank you so much. [applause]
Disclaimer and copyright notice
This is an edited transcript typed from an audio recording.
The National Museum of Australia cannot guarantee its complete accuracy. Some older pages on the Museum website contain images and terms now considered outdated and inappropriate. They are a reflection of the time when the material was created and do not necessarily reflect the views of the Museum.
© National Museum of Australia 2007–24. This transcript is copyright and is intended for your general use and information. You may download, display, print and reproduce it in unaltered form only for your personal, non-commercial use or for use within your organisation. Apart from any use as permitted under the Copyright Act 1968 (Cth) all other rights are reserved.
Date published: 01 January 2018