The Art of Natural Dressage

Something that I did think about in the past two months a lot, but felt like not discussing too much here As I am one of the few, who is not using clicker training/marker training. I did start with the introduction of food rewards, but then decided to try a different approach.

I realize that clicker training is a very powerful tool and I did get very fast results - as I can read in other peoples diaries also. But I know, that I have not yet much horse experience. I could train my horses to do many things with clicker training without that experience, by just rewarding what I like. Yet, I might end up rewarding things, that are not the best for my horses.

Today I did read in another topic something that is along these lines of thought, but I thought to open a new topic to not sidetrack the other one too much:



In dog training, people sometimes teach the dog not to growl/snarl at people. That can lead to very dangerous situations, as the dog then does not give a warning signal anymore.
I did read here and also saw on youtube, people training their horses to put the ears forward, to show a "happy face" and look friendly. I was wondering, could that result in similar problems? The horse is taught to not show the warning signals? I do not know, but just would not want to take such a risk.

Tammi does still move her head away in the way she learned through my "introduction of food rewards" when she hopes I might have some apple or carrot. It is just one more proof to see how powerful that kind of training is.

With great power comes great responsibility

I do not mean to tell people to not use clicker training. If it works for you and your horse - that is the most important. I just thought to maybe open a discussion about what one might want to be aware of when applying clicker training.

_________________
Andrea
http://www.youtube.com/user/FinhorsesAndPinscher
http://basichorsemanship.blogspot.com/

I don't think, in general, that clicker training speeds training up in any significant way. If my horse needs three years to develop piaffe, I can stuff all the cookies into him that I want, and all it will do is make him fat.... it will still take three years to develop the piaffe. Clicker training doesn't make the muscles develop any faster.

Food is also a powerful method to alter (but not mask!) the emotional state of the horse, as the chewing and salivation that goes on while eating is indicative of the "feel good" state of the horse, and if I can contribute to that with this type of training, then all the better.

I had an encounter with a dog who was taught to get what he wanted by giving off calming signals. This was reinforced with food. They didn't teach him to display the signals inappropriately, but they did teach him that by giving them, the threat (a person) would go away. I wasn't told this. So when the dog gave calming signals, and I gave them back, all was well. But when I then took hold of his leash, he bit me. So they didn't teach him to give signals in the wrong context, he simply had different expectations of the outcome than I did. He simply expected that I would then leave. That is what he was taught. He was not taught to give the signals in order to deceive a person so he could bite them.

I don't really believe that you can teach an animal to lie. I have clicked a dog for wagging his tail, but there was a calm acceptance within him that I was looking for accompanying that wagging.

Ears forward in a horse, is not necessarily a sign of calm acceptance that you can wrongly teach a horse to do. I don't believe you can teach a very angry or frightened horse to LOOK calm and happy. You can most certainly have a horse that gives extremely subtle signals. Cisco is very much like that. He is very introverted. That doesn't mean he doesn't communicate. And I didn't teach him to do that with treats. Rather, it was totally ignoring his ideas and showing him that the only way he could find relief was to hide his feelings. That was when I was training traditionally. Food has brought him out of his shell. With or without a clicker (it is nothing but a signal), the key aspect is the food.

And in the case of Tempo, she still did not lie. She wasn't taught to lie. She wasn't taught to hide her feelings. It's just that no one noticed the other signals she was giving.

So whether you are using food or not, you can only teach a horse what he/she is physically able to learn, and at the speed at which you are able to teach it. The basis of all training, traditional, or with food, is good timing. The speed at which a horse will learn hinges on the timing of the handler. If you use pressure (or feel) and release, your horse will learn faster if your timing is good, and your "release" is instantaneous with the behavior your looking for. Same with using food. There is no difference.

What I think you see, in fact, with those using clicker training, is that you see a method by which the person can learn better timing, faster than in other training methods.

We wrap around this topic time and again here. When in reality, the speed of learning has nothing to do with the food.
I just don't believe that you can teach a horse to lie with food. In fact, because of the nature of the training, it's more likely to be the exact opposite. That you are more likely to teach the horse that it's ok to express themselves.

The dangerous behaviors of horses are not masked by food. There is no danger to using food as a training tool. The only danger comes from not being able, or ignoring warning signals a horse may give. That they may be naturally more subtle than we are used to seeing, only compounds the problem. But a clicker alone, or using food, cannot make an angry horse thoroughly display a happy and inviting attitude. Even if you could train the ears to lie (and again, I don't think you can) the tail, the tension in the body, the eyes, the muzzle, the breathing...all would tell the real story.

But I don't think in the videos you are watching of people training the ears to be forward, are showing people that are teaching a horse to show an inappropriate language. They are standing with a relaxed horse and asking the horse to put his/her ears forward? That is very different than attempting to teach a horse to put it's ears forward when he is emotionally in fight or flight mode.

Based on my experience with Cisco, it's far easier to teach a horse to lie...or rather teach the horse that it is in his best interest to hide his emotions...through pressure based training - as long as you can physically overpower the horse if it comes down to a battle of wills. But with positive methods, you may have to deal with a horse that was previously taught to suppress his/her emotions to a dangerous level through pressure, but I don't think you can create that suppression with food.

Can you post some links to the videos?

_________________
"Ride reverently, as if each step is the axis on which the earth revolves"

Karen wrote:


I totally agree.

Food doesn't mask emotions. It can help to change them, or not, depending on the situation. If my horse Circe is really upset about something, a treat can help shift her out of what's seized her emotionally for a moment, but it doesn't replace it. It simply can give us an opening to explore what's happening. And treats can be helpful in breaking that emotional state so she can find calmness -- but they don't hide her emotions if she is upset.

And my horses, at least, don't do things they don't want to do, even if treats are involved. They are pretty darn clear about this! For example, we've been experimenting with whether Circe and I can walk together around the ranch without a lead rope connecting us. She is completely aware that I have treats in my pouch, and she varies from being very attentive and wanting to figure out what I want to she can get one to happily taking off to explore somewhere on her own. Or with Stardust, as we re-invent what it means to be ridden (which has been awful for him in his past) -- all of the treats in the world wouldn't convince him to let me near his back until we began to build some trust around that. The treats have helped me begin to build that trust, but didn't provide a short cut to getting on him.

What I've found is that food works as a reward, in the context of the best version of careful communication and awareness I can muster. Treats don't work as bribes. They don't overwhelm the horse's wishes or instincts.

I find it really interesting that most of us who are wrestling with whether or not food rewards in training are a good idea seem to get caught up in the idea that somehow it's cheating, or at least taking a short cut. That the relationship is only about food...

I don't think horses can be somehow "tricked" (for want of a better word) into not feeling what they feel, not wanting what they want with food any more than people are. It's not that simple, and horses aren't that simple. In fact, I think they're a lot smarter than that.

(And I totally don't mean this in a disparaging way, Andrea -- I had the same wonderings before I really started working with rewards based training. But my resistance was mostly about underestimating my horses' ability to think for themselves.)

As I begin to build my skills with food reward based training, however, I am finding a few things to be true for us...

• It is not the solution for all things, a magic pill, a shortcut. In fact, things can take longer to teach than with other methods, especially anything that is pressure based. For example, as I'm starting to work with Circe about what it means to have me on her back -- if I'd wanted to, I could have used a lot of techniques to convince her to do what I wanted while I was on her that would have us much farther down that road than we are now.

• It does not work well without consistency and commitment. It's a building process, and food becomes a part of how we communicate. And it takes a while for it to become fluid for everyone, and even then is something that continues to take focus and work.

• It brings an energy and openness for all of us that I like very much, and my horses seem to as well. Stardust really didn't like to be touched but loved food -- as I thought about something that was a clear reward, food made total sense to me. A scratch, a stroke, or a hug did NOT feel like a reward to him. That is slowly changing, using food -- I've not "trained" him to like being touched, but have provided an opening for him to trust me differently and think differently about what it means when I'm touching him, and then actually FEEL what it feels like, rather than having an instinctive response about what is going to happen. The food provided the opening, not the solution, if that makes sense. He's now accepting my touch, but he still hasn't translated it to being a clear reward -- hopefully, we will eventually be able to add that to our vocabulary, but we're not there yet. Even for Circe, who is naturally cuddly, didn't immediately equate touch with reward -- this is a human reward, not a horse one, to me. A learned one. Food is basic. I've never met a horse who didn't think food was pretty darn dandy!

• I feel like I'm fumbling through learning how to speak horse -- when I was doing a lot of international travel at performing arts festivals, spending time with people who spoke a lot of different languages but often didn't have a common one, food was a great way to help us communicate. We could be generous, enjoy what was being offered, and it broke the ice. Food (and beer! ) were the two things that were guaranteed to get people communicating when they didn't have a language in common. I feel like treats do the same thing with my horses.

You wrote, Andrea:


In my experience, this just doesn't happen, any more than any other training technique I've worked with, and far less than with most.

I haven't watched the videos that have been discussed, but I think Karen's story about Cisco is extremely helpful. Pushing too hard is pushing too hard, and all of the treats in the world won't change that if we miss signals.

And my sense is Karen's sense of awareness of that situation with the filly has been not about the failure of clicker training at all, but instead a reminder that we can miss signals, even with care and good intentions and gentle training techniques. And Karen is one of the most careful watchers here, and her reminders about this are really helpful!

Any reward based system (whatever the reward is) can inadvertently reward things that aren't good for the horse. But it's far less likely to get the horse to do things that aren't good for them, in my experience, than any pressure based training. They can always say no! And my horses often do, if it's something they don't want to do, don't understand, or don't think they can do.



Again, I don't think anything that happened with this young filly necessarily has anything to do with clicker training. Somewhere, something shifted for her and/or some signals got missed -- this is something that can happen with any kind of training!

Actually, as an aside, when I was reading about it, I found myself thinking of Sue's description of how Sunrise has been shifting as she's been growing physically and emotionally and how much how she processes the world has changed. This was a huge eye-opener for me, and helped me to be better aware of what's going on in Circe's world that doesn't have anything to do with our experiences together. It's helped me to remember that she's got an entire world, reality, sense of self, etc. that doesn't have a darn thing to do with me!

I've gone on -- I think, Andrea, your question is a very good one about helping us to step back and understand that all of the gentle training we can offer, regardless of the specific technique, doesn't replace our own learning and awareness of what our horses are saying to us.

But for me, at least, it isn't clicker training that's the issue. For me, it's about co-learning, not training at all -- we're all learning, all the time.

Best,
Leigh

_________________
"Ours is the portal of hope. Come as you are." -- Rumi
www.imaginalinstitute.com

Is clicker training fast and easy for beginners?
I think so, especially for people like myself, who do not have that much practical experience around horses yet, clicker training does speed up things a lot. I had a horse, that did not like humans too much, that did not know me yet that well and myself having no experience with clicker training at all. Yet she learned in the first clicker session on the pasture (before I had just introduced the food rewards from behind a fence), being without halter or any other utensils, to follow me, stop on signal, lower her head down on signal, do a ramener and to back up without being touched. She also learned to stand without moving her feet beside me. There might be people who can do all that in short time without clicker training, but surely not me.

Here is a little video showing a first time horse person/clicker trainer teaching a horse to allow brushing:
http://www.youtube.com/watch?v=P8bRwQvdDFs
I am working on the same thing with the above mentioned horse (Tammi), but I decided to not use clicker training, so it is very slow going. But I am not in a hurry.

Can one change the bodylanguage of an animal by training?
I think so, but I surely would not call such "to teach lying". A animal can just learn that a different type of bodylanguage gets rewarded and adjust accordingly. It is a smart thing to do, when one thinks about it. Most people, me for sure, have a hard enough time to read a horse or dog as is. A lot of bodylanguage goes just unnoticed to me - so I rely on those very clear signs for now.
Here is an article about "why growling is good" concerning dogs:
http://k-9solutionsdogtraininginc.blogs ... -good.html

We can easily teach a horse or a dog something without us realizing it, that we would rather not have wanted to teach.

Here is the video about someone training the ears forward:
http://www.youtube.com/watch?v=dFdrxM52yaA

Cheating or just food based relationship?
I do not consider clicker training as "cheating" - it is a powerful training method. I just tried to point that out. I think it it important to be aware of how powerful a tool it is.

As I do always feed my horses myself, of course I give them many times a day food (at least 4 times). So even without clicker training or food rewards, there is some kind of food based relationship. That is just fine I like to watch them eat



Thank you for finding these words - it is really what I tried to communicate

_________________
Andrea
http://www.youtube.com/user/FinhorsesAndPinscher
http://basichorsemanship.blogspot.com/

The people training their dogs not to growl at people are usualy not using food and clicker for it - they are using punishement. If you use food - and clicker - then you are not taking the signals away, but the focus. If the dog is able to eat it is usually not that stressed. Plus, eating is calming. And it is reward - adn then you reward for the nice stuff, and that is not to take signals away, but to say how is better.

But that saied - I have stopped training "nice ears" - but I did the first weeks with Lisa, two years ago.....

Hey Andrea:

I've been thinking about this conversation, and trying to get a better sense of what you're asking (as best I can!) and what my responses are to it.

I just went back and took a look at your diary and read the Leslie Dorrance article you referenced...
(http://www.lesliedesmond.com/index.php?id=124 for those people who might not have seen it in Andrea's diary)

I think, for me, it comes down to a difference first between pressure based and rewards based training, and ultimately, training and co-learning.

Leslie's writings, to me (and what I've read of Bill's work), is about a very gentle, very thoughtful pressure based style of training. It is about, at its heart, I think, getting the horse to do what the human wants, getting the horse to respect the human, and trying to find ways to do so gently, with the lightest force/pressure possible.

And I agree with Sue as she said in your diary, I think that horses who are trained this way can be quite happy and gentle. It is a far cry from your descriptions of your trainer who would whack horses with a whip when she thought they'd "misbehaved!"

However, I don't think it's rewards based training. A lot of what she talks about in this article is about finding ways to get the horse to submit to the person's desires, building techniques that will both a) put the horse into a submissive frame of mind, and b)help people clarify what they're telling the horse to do. As I read it it, she's trying to help people work towards a gentle telling -- which is a wonderful step towards a better relationship than harsh training -- but it's still telling, not asking.

For me, this is the biggest difference between pressure based training and rewards based training -- I trying to find ways to ask my horses to join me, not tell me that they have to.

For example, Leslie Dorrance's definition and pictures of how she teaches how to lead up freely, and the "pull" that she describes. I've used versions of this technique effectively on other horses who were being resistant, prior to my stepping into AND and my version of a commitment to working in this kind of partnership. Eventually, these horses would give in and go the way I wanted them to -- essentially it works because you are shifting the horse's balance so their feet have to move underneath them, and that can free them up out of a balk. And generally, if you repeat it often enough, the horse will learn it's not worth fighting with you and will submit to your wishes. It could take longer than harshly demanding, but it still was a demand.

But while she says in her experience that this doesn't cause the horse "mental upset" I can tell you from my experiences with Circe it really did. She felt that pressure and HATED it! She would pull back as hard as she could in response. I could continue to pull until she needed to move her feet, and get her to move, but she was very unhappy afterwards and was looking for the next moment she could plant her feet again.

For us, this leading was one of the turning points for me into a rewards based training approach with her. I had a choice to make -- I could, because I've done it with other horses, push this until I'd achieved power in this situation. It would have taken a fight with her, because she's not inclined to be very submissive -- she's a fighter, with a very strong sense of who she is, and is young enough (and was handled gently enough) she'd not learned to be afraid of what might happen when she didn't submit, like my other horse Stardust. (He learned the hard way what happened when he fought.) But we could have gotten there eventually.

But here's where my sense of co-learning rather than training came in -- I didn't want to train Circe to walk well with me, but instead wanted us to learn together how we could want to walk together.

So I did three things. First, I stepped back and tried to listen to her -- get used to her "feel," to use the Dorrance's term, before I asked her to get used to mine. What she was saying to me needed to be more important first, because I wasn't sure why she didn't want to do this. And in this watching, listening, and learning, I figured out a couple of things. First, she really hates being told what to do (and I do think different horses, like different people, have different tolerance levels for this, based on personality and experience. Stardust was really uncertain when I first started to ask him for his opinion, because he'd been so thoroughly taught he shouldn't have one). Second, I was marching her through territory she really wanted to explore -- she wanted to go look closely at and sniff everything that she was seeing from her paddock every day. That hadn't even occurred to me, and was another eye-opener for me about how little time I'd really spent trying to look at things from my horses' eyes, rather than from mine.

So, we began a period where she got to lead me, rather than me lead her. She got to decide where we went and how long we'd stay there. This was important for both of us -- it gave her the opportunity to be "in charge" and softened her need to essentially yell at me that she didn't like what we were doing, so it softened her ability to communicate. And it gave me the opportunity to really assess what I needed to do to not assume I was in charge all the time -- this was probably the hardest part for me!

Once we'd spent some time doing that, cleaning the slate, I began to think about how I could invite her to come with me, and offer her something that she felt was valuable as a reward for it, rather than announcing she had to come with me in any way.

As I said in my post above, food was the obvious reward. I don't think horses naturally think touch from humans is a reward -- at least not all of them -- certainly most of the horses I've spent time with have not made that link. I think they can get there, but it's learned. I think this for two reasons -- first, I think this is one of the places where horses are very aware of the differences between horses and people, and while physical affection is something they understand with each other, it's not in their natural vocabulary with humans. Second, even with other horses, they don't look for that kind of affection from one another until they've gotten to know and trust each other.

So, for me, food was an obvious tool. I certainly don't think it's necessarily the only one, but I personally couldn't figure out how to do rewards based training without using it -- I couldn't think of any other reward (at least to start) that wasn't a release of pressure that they'd actually see as a reward.

If C/T is easy for a beginner horse person to use, I personally think that's because it's the most obvious way for both horses and humans to understand the shift into rewards based, positive operant conditioning.

I think you can train a horse to change its body language in such a way that it masks its emotions -- horses are VERY good at learning to do this, as are most animals who are not predators. But I actually think it's a lot harder to do with C/T training or rewards based training than it is with pressure based training.

And I personally don't think it's a smart thing to do! Because I think you can teach a horse to mask what they're really feeling so when they hit the point they can't cope, they explode. This is exactly what Stardust taught me about -- he would hold things in so long that when it came up and out, his reactions were huge and dangerous to him and anyone around him. I recognize he's an extreme version, because of his background, but it firmly put me in the camp of not trying to teach specific body language/behaviors to avoid stress.

This is why I've actually not spent a lot of time training "head down" on my horses for example -- I'm not convinced that they physiologically MUST drop all their tension when they drop their head -- I don't think it's that simple -- they may experience some physiological relaxation when they do this, but what ever it was that made them upset is still there and still not dealt with as we move on.

I know that there is a whole school of thought in managing stress for humans as well as horse by doing things that physiologically change the stress levels in our body. As a dancer/performer, this is something that I've worked with a lot on myself, trying to figure out how to work through stage fright, for example. However, what I learned is that while these tools can be helpful, they're not a solution -- ultimately, we need to work through what is causing the emotion and address that directly. And I could cognitively work through all of the nuances in them in a way that I don't think horses can -- I knew what I was going for and why, and could think through what the scary parts were, and knew when these techniques were working and when they weren't!

So I don't try to mask anything -- I would far prefer to have Circe or Stardust tell me softly that they're freaked about something than only being able to say this to me when their fear is so great it breaks past their training to look happy. Again, this is another level of the co-learning for me -- if I am focused, when they are exhibiting signs of being stressed, on teaching them not to exhibit those signals, I'm not focused on why they're showing those signals. In my mind, we treat the symptom but not the disease, as it were, by doing this, and I'm not learning about what they're concerned about and the subtle signals they give off when they are concerned.

But I'm interested, for Andrea or anyone else who's chosen not to use food as a training tool in a rewards based training model -- what do you use as a reward? How does it work for you?

Thanks!

Leigh

_________________
"Ours is the portal of hope. Come as you are." -- Rumi
www.imaginalinstitute.com

Hi Leigh, I had not that much of a question, but I just wanted to build some more awareness by opening this discussion. And also a possibility to discuss some points that came up in another topic about clicker training.



I found the first half of your post interesting and I would like to answer also, as this is what I am working and thinking on at the moment, but I would rather do so in another topic ( viewtopic.php?f=7&t=1689&p=38146#p38146 ), as it is not related to clicker training.

And I opened a new topic viewtopic.php?f=8&t=2007&start=0 for the question at the end of your post.

And the center part is what I would like to leave here for further discussion:

_________________
Andrea
http://www.youtube.com/user/FinhorsesAndPinscher
http://basichorsemanship.blogspot.com/

Interesting topic - I've read it all and forgot what I wanted to quote so I'll just write things that are on my mind after reading all. ;

I do think it's good to be aware of that everything you do does have side effects that you don't necessarily are aware of right from the start. And that goes for clickertraining as well as for pressure-release based systems.

There were a lot of statements starting this topic and I'm not sure if they are questions or opinions or discussion points, but this is what my experience is:

- clickertraining is a short-cut for beginners
I do know that beginners can get results very fast by using clickertraining. However, I don't think that's because of the clicker or just the food, but more because of the fact that the human uses extremely precise timing right from the start. That means that your horse knows exactly what you're asking from him. If you have a rewardsignal and then scratch him on an itchy spot or do something else un-foodrelated that he likes, he will learn quite fast as well. If you just generally always make him feel the same and don't use any (positive or negative) stimulation that tells your horse how you feel about what he's doing, then learning indeed will go very slow as a horse simply doesn't understand what it is that you want then.
If you don't want/need/expect anything, then working without timing and stimulation is fine as well. However, when teaching something makes the quality of your horses life better immediately, then I would want to teach him that as soon as possible. If you can teach a horse not to be scared of something, then that really is such a relief for him and makes his general wellbeing so much better. And if you can make him understand easily how to move in a physical more healthy way, then I wouldn't heasitate to do that as soon as possible either.

- clickertraining is dangerous as it teaches the wrong behavior
It is if you reward for the wrong moments!
And in that it's exactly the same as regular pressure-release based methods. A big problem with training spooky horses in the regular way, is tha when the rider is on board and the horse jumpes away from that scary corner, the rider is thrown off balance and releases the reins. If he then redirects the horse towards the corner, he will tighten the reins in order to prevent the horse from jumping away from it again. The horse spooks again, throws the rider off balance, the reins are slackened for a moment again and the problem slowly increases. The problem is that whenever the horse approaches the scary corner, the reins become tighter, and whenever he spooks away from it, the reins and the control of the rider are lightened. So the rider effectively teaches the horse to spook more and more.
I did read the Leslie Desmonds links you placed here before and read somewhere that she uses 'as little pressure as possible and as much as needed'. That's exactly the same thing: Leslie knows when exactly to push a bit harder, when not to give up or release pressure to get that right result. The fact that she uses the power of timing in her training does speed up the training sessions a lot, because if she didn't use it, the horse would have no idea of which behavior was wanted. And that generally wouldn't be a problem, if Leslie wouldn't be trying to teach the horse something!

You could even say that timing really is the cause of and solution to every problem. That's also the reason why the best (softest) pressure release trainers can get results without using a lot of force: they realise that when they use pressure and release right at the second the horse does(n't do) something, they can use the tiniest amount of pressure simplye because the behavior hasn't escalated so much yet that brutal force is needed. You see the same in classical dressage: you teach things first in halt, then in walk, then in trot and then in canter - simply because in a slow gait you simply have more time to respond to thing correctly.

- teaching ears forward is bad
Well, I've done that with Blacky and I wonder if it has been bad. I haven't taught it as a trick with a cue, but simply wanted to see if he really meant to have his ears back when I approached. What we did essentially was the way of approaching a horse as is described over here: when I went to the gate and Blacky had his ears back, I would stop and walk backwards away again. Whenever he put his ears forwards, I would move towards him. When I got next to him finally, I would thank him a big time and also give him a treat. But the fact is that Blacky only got that treat the first time after he had decided that he wanted me with him anyway.
Another thing that is good to know, is that Blacky came from a bad background and had always hated/attacked humans before, so ears back had become a sort of default mode. When after a year of positive training (with food, without clicker) he still did this, I decided that it was time to see if he really meant what he was expressing with his face.
Apparently he didn't, as when I responded to his angry face by moving away, he started looking more neutral or positive in order to draw me back in.

- Teaching movements teaches horses to lie
I'm not sure about this. I think that what the owners of Karens dog have done, to teach the dog to give off calming signals whenever he felt scared, actually was quite a good idea, because it could very well have been that the alternative past was that the dog attacked everything that scared him. However, it is handy to tell this to people who don't know this. Just as it's very handy to tell people who have no experience with horses how the horse expresses himself naturally. If you can't read his signals, then there is bound to be trouble someday. I did teach Sjors to lower his head when he was nervous, not because of the hormones and stuff, but more because it gave him a chance to express to me in a safe way that he was nervous. The alternative was exploding and running away fast. Which of course also is very sensible from a horses point of view, but not handy in every situation and that also didn't give me a chance to point out that there was nothing to be scared of.

- you can teach horses dangerous things with a clicker/food
Yes, and it's good to be aware of that, I think. It's not a flawless system, simply because the human component is still involved.

- horses get aggressive from food
This is an interesting one, because I feel that it actually is the other way round: you can teach horses to become aggressive around food. Many people unconsciously have done so even before they started rewarding with food - simply because the horse sometimes pulled the reins out of their hands to snatch some grass, because at feeding time at night the horse has realised that if he crowds the human or scares off other horses, he will get more food. And he will have learned that at the end of the ride when the human has an apple in his hand, he can just grab it and eat it without a second thought. So it is only obvious that when you then say to your horse 'I'll work with foodrewards from now on', your horse will add up his previous experiences and will put all his effort in getting the food from you as much as he can.
The only thing you really have to do, is to teach him how to behave around food - just as you've taught him how to behave around you. For some reason horses generally know not to bit, trample and harrass you in general - now all you have to do is ask him not to do that when you wear a red jacket, or carry a bale or hay, or a pocket with carrots as well.

- The horse who kicked is an example of a clickertrained horse gone wrong
I haven't followed that discussion and haven't seen the video either, but you can wonder if the horse was being bad? Was his action actually clicker/or foodreward triggered? I do work with treats, and regardless of that can cause the ponies to play harmoniously with me or kick me in the head. Just the fact that I use foodrewards doesn't mean I've become a saint! I can still be overdemanding, obnoxious, boring and as hyperperfectionist as I was before. I think that response the woman who got kicked will get on the c/t list will go in two ways essentially:
One: teach your horse not to kick/that kicking is wrong (which in my opinion seems like solving the results, not the problem underlying)
Two: realise that your horse has given you valuable feedback on what he thinks of the training (which would be more like AND; use the signals your horse is giving you to improve your own behavior).

My suggestion to her would be twofold: indeed see the kick as valuable feedback and make a list of what might have triggered it and how you can prevent doing that in the future - and also teach your horse that you interpret a kick as such a huge 'go away!!!' signal that you will listen to it immediately by going away and staying away from your horse for several minutes. Often horses just use the bodylanguage they have always needed to get through to humans, and often they have learned to speak very loud because humans are very stupid. If you show your horse that big signals will cause big results in your behavior, he will very soon realise that he probably didn't want you to leave him for ten minutes, but that he just wanted to tell you to stop annoying him by asking that same exercise over and over again. So he will make that feedback more appropriate in size: 'if the human is annoying again, I'll just walk away or stop playing along, as obviously she has started to understand smaller signals as well. There still might be some hope for her.'

So I do agree that just teaching the horse not to kick (or to put his ears forwards always even when he feels bad) doesn't solve the immediate problem - if there is one underlying the behavior.
However, I don't believe in totalitarianism. Just because teaching a horse not to kick can be unwise in some instances (when you teach the horse not to express his anger), doesn't mean that it is always wrong. Just because you can stab a horse in the eye with a hoofpick, doesn't make hoofpicks a tool of torture either.
For me teaching a horse B always involves two simple steps:
1. Ask myself: why does he do A now? Why doesn't he do B? What can I chance already in order to make A less necessary and B more logical to the horse?
(for example, not annoy him so he doesn't kick - not allow him to steal treats because he just looks so cute whenever he does it )
2. Teach him that B works even better than A in getting what he wants
(and with that I mean not just food, but also making the human behave better simply because now you've established a signal together that means 'you annoy me, stop it')

- Clickertraining turns horses into brainless zombies who will sell their soul if it would earn them treats

Well, we're puzzling with treats and the piaffe for 1,5 year now, and I still wouldn't call it a real piaffe and improvement is steady, but sloooow. Partly because I'm not that good at finding the right timing in this movement, partly because it's tough to do for Blacky as well, and also because I'm not the real 'only reward for improved behavior' clickertrainer. I reward for everything, including the lesser options that Blacky gives me a lot because I'm just a big softy and I'm already brimming with joy whenever a pony just halts on my cue. That makes it less likely for the ponies to overstretch their possibilities as they will get the reward otherwise anyway. I'm quite suprised that they actually improve what they do so much, and I really feel that I should blame that on the fact that they simply love to solve puzzles and to surpise me (probably into giving them more food, but still).
If you clickertrain sharper (only reward for improved behavior and by that make the rewards rares) then the horse will rate the rewards higher and will do more to get that reward and then training will speed up, but will also become somewhat more risky as the horse is more likely to overstretch his borders in order to get it. However, it is really, really, really difficult to be such a trainer! If you talk about clickertraining being for beginners, then the general get lured into rewarding almost anything really is simply. becoming extremely specific in what you reward and having the mental and emotional power not to reward anything else that your horse tries in order to please you - that is really, really tough. Horses simply are too irresistable cute. At least I still haven't been able to master it.

_________________

New horse book: Mandala horses!

Never stop making mistakes! Natural Dressage

Wow such a great topic and lots to think about. I am very happy Andrea posted this as I was feeling the same. I am reluctant to startt clicker/food rewards and could have written this:

I see a whole set of new problems being created and just don't want to go there.
However I struggle with which method to use. I don't want to use pressure/release although there are times when I have to like leading across a road, riding out in a group etc. So I find myself avoiding those things so I don't have to use pressure at all. This is great and we are both happy. When it comes to HE type movements it is hard to motivate myself or Morgan as there is no positive reward other me making approving sounds/voice and strokes/rubs. At the beginning it was a novelty to us both but Morgan got bored quite quickly and so we are not playing much. He will offer moves and try different things if he wants my attention and I am outside the camp. He also offers them under saddle as a way to avoid the pressure of riding. Very hard to know where to go to next. I know it would be very easy to train him with food and I have used picking grass for him as a way to get his interest at liberty.........
Has anyone acheived HE movements without treats? I think some of the NHE members did/have? I don't believe in keeping my horse seperated from a herd. This is one of the most important aspects of our relationship. To keep him happy in his environment and to let him live as much a free life as possible. At the moment this means more to me than training or schooling him.

_________________
Annette O'Sullivan

Life is what happens to you while you're busy making other plans. - John Lennon

I loved your way of describing how you taught him to back up and give his feet - maybe you can find a similar way to teach him that other dressage and haute ecole movements can be physically rewarding as well?

I don't really know how to do that myself, but what I do know is that horses really can't get enough of certain exercises: most horses really love standing on plateaus, retrieving things and doing the jambette. The difficulty with those things often isn't to get them to do this, but rather to get them to stop doing it!

What you can do is use those favorite exercises to turn them into other exercises: sure, you can fling that leg forwards, but you can also fling it forwards, make a step forwards and then do another one, as that way you can do hundreds of those wonderful leg-swings after another!
If you have a plateau-loving horse and a plateau that's strong and wide enough to stand with four legs on and that can stand a bit of gravitational violence, then you can also use that as a starting point for haute ecole movements.

All of this is what I actually did to teach the ponies these things - I can't make Blacky rear and can't clicker him into doing something he thinks is pointless. So as he liked lifting his legs (he will always stand with a frontleg in the air ), we started piaffe out of that. I also can't just convince him with food that a really energy wasting low rear (levade) is a fun thing to do. However, Blacky loves standing on the plateau, so instead of asking him to rear and rear lower, I asked him to not walk off the plateau when he stood on it with four legs, but to jump off of it. The entire rear actually came out of another thing: out of jumpin over brances. I would come to s sliding stop in front of each jump and then step/jump out of a standstill over the thing. Blacky copied that with a jump out of halt, and then it turned itself into a rear.

Even for foodrewarded horses the most important motivation isn't the food: it's the fun they have in doing the exercise itself. When I was still using pressure/release next to treats, Blacky really showed the differences between the exercises he liked and the exercises he thought were boring and pointless. The food was the same. So the first thing I learned when I put away the pressure, was that I should try to make sense of the movements themselves because otherwise we wouldn't be doing any exercises at all (exept for picking up things, jambette and standing on the plateau ). So since then my main goal is to make sense of everything. Walking collected for a long straight stretch is pointless, so you should make sense of it for your horse: if you walk parallel to a wall, then turn towards that wall and then walk back to the direction you came from, then he will realise that the turn is easier when he collects his steps. When you give the signal that you will turn longer and longer before the actual turn, then your horse will anticipate longer on the turn: he will walk collected.

And then you can do a lot with turn around the hanches in walk towards the wall! We were doing it as well in a collected walk (but a couple of meters away from it, Blacky just was too lazy to complete the entire circle ) but as I turned much faster than he did on purpose, Blacky finished his turn with a canter-turn so that he could run after me in a canter at once. Et voila, the beginning of a canter pirouette.

Making sense of exercises - giving them sense - is what our training really is all about. If Blacky thinks something is boring or pointless, no carrot in the world will get him to do it. And it's just important to make the sense stay alive! Of course that's easy for exercises that a horse likes already because the movement simply fits him. Blacky loves to canter, Sjors loves to trot, Blacky loves to stand on a plateau or walk over branches, Sjors loves to chase a tiger.

It's the other things that take more energy that require a very elastic brain from the human. So yes, Blacky can do a canter pirouette now, but that doesn't mean I can just ask it ten times after another at the same spot from the same collected gait. The fact that he should stay collected must be of importance to him, so what we do is to make of the entire haute ecole collection thing a waiting game: we move slowly and highly collected for a while - in a pirouette or terre a terre or in a series of levades - and then I'll shout something like 'get me!' and run away as fast as I can, leavin Blacky behind to explode away himself and to chase me. If he was more collected, he will get away sooner and get to me faster. So for him haute ecole and the really high collection is what it really is for the horse when playing with other horses as well: a way to move in one place and still become/stay so high-strung that you can take off at top speed at any second, in any direction. I'm just being that other horse who dares him by boasting that I will beat him when it comes to speed and agility.

And even that isn't the end, because that too can become predictable and boring. So then we use the haute ecole stuff to make the already fun exercises more interesting. Blacky can rear and he loves to stand on a plateau. So I'll then ask if he could also rear off the plateau backwards if he stands on it with two legs. And when he stands on it with four legs, can he rear off of it at the front? Can he stand in front of the plateau and then rear onto it? And onto a small wooden beam? Can he stand on that beam with his frontlegs, do a levade and then land on top of it with his frontlegs again? Can Sjors walk in Spanish walk on top of the plateau? Can he do a jambette and then walk backwards?

The treat for me is just a way of communicating: That's it! You've cracked the puzzle! Getting there! Isn't this fun? However, even though Blacky just swallowed that carrot won't stop him thinking 'well, I liked that piece where you gave me the carrot, but the rest - god, I've never been that bored in all my life!''
The real motivator behind all the real energy costing exercises is succeeding in showing them why collection can come in handy, why it's worth giving it a go, how much fun it is to stand on two legs instead of four and how you can (ab- )use that to do other things as well. That's when they start experimenting more with the tough stuff themselves as well. We had our first courbette our of the canter pirouette - Miriam playing foul by chasing off in an unexpected situation too soon, Blacky still trying to perfect his canter-turn, realising that I cheated and then jumping off with his hindlegs to combine the turn and go in one forwards leap.

So you really could describe our training session as making sense of everything, making sense with obstacles. And now I come to think of that, I think that might be the key to doing Haute Ecole without pressure, without foodrewards as well. I'll stick to the foodrewards for now, if only because the ponies just love them so much and because I just love to give them something in return for all they do with me - but I would love to see others try it this way and see if my theory works for playing without treats as well!

_________________

New horse book: Mandala horses!

Never stop making mistakes! Natural Dressage

Miriam,
Thanks that helps alot. I think for me the problem lies in my motivation to teach him HE movements. Maybe when things go back to normal (kids at school etc) and I am seeing him more regualrly I can get back to asking for more. At the moment I want to please him when I see him so follow what he wants to do and that seems to be lots of liberty grazing at this point!!!!!
I do think he finds the movements fun and the ones I have taught him he offers over and again, so he obviously gets some kind of payoff (joy) from doing them.
I do think the hardest part is always the work on us, it is never the horse, always our inability to communicate or make it fun. It is very hard for me also to see the movements/riding as fun as opposed to "work" and I think that comes across to him too.
Thankyou for explaining what you do as I now have lots more the think about.

_________________
Annette O'Sullivan

Life is what happens to you while you're busy making other plans. - John Lennon