Friday, February 24, 2012

Yea, get negative on that horse!

The previous post has some fantastic comments that have influenced what I'm going to say here - if you're finding this topic interesting, make sure you read the comments on the last post!

Shannon said something that really piqued my interest:
I don't really like the article on positive reinforcement that you linked. It relies far too heavily on B.F. Skinners original 1950 work. We have made great progress in the study of behavior since then, particularly in identifying the neurologic pathways involved in operant conditioning. There is little evidence that negative and positive reinforcement are any different from a neurologic perspective. Many behaviorists are actually pushing to drop the words "negative" and "positive" completely and simply go with "reinforcement". Both negative and positive reinforcement are reward based. The only difference in the two is the application of the stimulus: Negative = stimulus removed to gain desired behavior, positive = stimulus applied to gain desired behavior. Realistically, arguing negative vs. positive reinforcement is arguing semantics.

Both negative and positive reinforcement are part of operant conditioning. Operant conditioning is a method by which a behavior is "shaped" (trained) to a stimulus which is irrelevant to the behavior. So, a rat running on a wheel when he hears a bell and a horse knowing to stop when he hears the word "whoa" are both examples of operant conditioning and have been trained by reinforcement, either positive or negative.
...
Clicker training is the use of a "click" as a stimulus. Whether or not it is positive or negative reinforcement is a matter of semantics. In the end, the "click" is no different from the "aids" that conventional trainers use. We are all using a stimulus and a reward to shape a desired behavior.

So - I acknowledge that it may be outmoded for me to differentiate between positive and negative reinforcement. I definitely see the point, that biochemically there might not be a difference between a click followed by a reward versus a release of pressure, but I don't think that totally invalidates what I wanted to talk about today.

Negative reinforcement definitely has an image problem, that's for sure. It just sounds mean, and many horse owners don't want to be mean. I mean, hell, I don't want to be mean! As firm as necessary, and fair, yes, but I don't want to be the mean hateful human. So let's call it -R (as opposed to +R, usually but not exclusively clicker training).

Still, I think -R is the most natural way to train horses. (Oh man I'm gonna get some hate mail for saying that!) Hear me out, though - we touch our horses more than we touch any other being aside from human family members. How often do you touch a stranger? Almost never. How often do you touch your dog or cat - and you can't count "the cat climbs on my lap every night" in this instance? You pet them, yes, and pick them up as necessary, but I bet you hardly ever lay hands on your house pets like you do your horse. Horses are one of the few creatures that we manipulate with our hands every single time we interact with them. The most intuitive, even "natural" way we have to communicate with horses is through direct contact.

Direct contact is -R. It can also be +R: scratching an itchy spot or massaging a tight muscle feels immediately good to a horse. And it can even be a form of clicker training - if your horse has learned to associate a pat on the neck with positive feelings, that's +R in my book (but just like the clicker noise, it's a learned association). But most of the time, when we touch our horses, it falls in two camps: "don't move, let me do this to you" or "move away from my pressure." Almost everybody indicates "move from pressure" with, well, pressure, repeated as necessary.

Here's where I think -R training diverges from clicker training. All the c/t stuff I've read says "don't nag." Ask (verbally or physically) and wait, and eventually, when the behavior happens, reward. -R behavior teaches your horse new behaviors (or reinforces old ones) by, in essence, nagging. If I want my horse to back up from the ground, I give her the verbal cue ("shhhhh"), the body language, and the physical cue (jiggling her halter backwards) - and I keep that up, escalating the physical cue, til I get the response I want. I start with a cue that's as subtle as I can manage, because I want to train her to respond to the lightest touch, but I ramp it up geometrically if she's balky or not paying attention. For some behaviors (i.e. moving the hindquarters at a tap), I don't ramp up the intensity of my request, but I keep asking over and over til I get what I want.

Clicker people: Is there a way to ramp up the request? Because if there is, I'll buy the "C/T is equivalent to -R training" argument.

So, back to my main point: the easiest way for me to ask my horse to do something is physically. I reach out and literally touch her and indicate "please move away from this pressure." She'll try different things to get a release from that pressure, and when she tries the right thing, I release to reward her. Sometimes, she doesn't get it and gets frustrated, so I go to my #1 Plan B: clicker training.

(I might write yet another long post about why I like clicker training, which will no doubt piss off both the hardcore c/t'ers and the traditional horsepeople. We'll see how tired I am tomorrow night!)

+R behavior modification - clicker training - is the only way to work with animals that you cannot regularly touch, like captive elephants or dolphins or even chickens (god bless all four brain cells). It's fabulous for working with animals that you do not want to touch all the time, like dogs doing liberty work (agility, off-leash training, etc.) It even works for horses - but why start there? We touch horses more than we touch any other creatures, so why not start with the intuitive -R pressure and release method? Sometimes it's ok to be negative.

All right, smart readers: rip it apart!

24 comments:

  1. I got this!........Yea! Thanks for that Shannon. I agree with you! We cannot speak as in words to horses, but we do use the tone of our voices etc. So ergo, its either a sound, or visual cue or a physical cue? Or am I missing something?

    I have trained my firt horse using Prof.Millar`s methods, of course some what altered!
    It worked. I have tried the Parelli method, for a while anyway. My younger horse got somewhat bored and dare I say it? I got bored.

    However, whatever I did, I always ended up using a touch! Whether a finger, palm of my hand, or a
    my whole body weight. if thats a negative way, I`m all for it. Surely if you can "talk" to your horse so to speak, by a physical touch, that cant be bad?

    ReplyDelete
  2. I couldn't agree more. There are dog trainers who are "pure positive" (I know that's not a proper term, but it effectively describes what they do.), and for obedience and the like, it works just fine. However, when you're training dogs on stock, you don't have time to "wait" for the dog to offer the behavior, like when the dog has a leg o' lamb in its mouth and you want it to let it go, meanwhile, they are running 100 mph down your pasture. Just like with horses. They are 1,000+ pounds, and they have the potential to hurt us. I am not going to wait to c/t for a good behavior while a bad one is happening. Pressure and release has worked for all this time ... why fix what's not broke?

    ReplyDelete
  3. I went back and read the link in your original post last night Funder. Apologies for not doing so prior to commenting yesterday. My comment would have been better informed.

    Frankly, I was loathe to read something tearing Mr. Brannaman down - no matter how well written, whether he deserved the criticism or not. It sounded like he may have, although we only heard one side of the story.

    Also, I wasn't sure I was qualified to comment - my educational background is in art, and my training experiences limited. (two dogs and one horse - flying by the seat of my pants generally)

    What the heck - here goes:

    Firstly - I think a bone of contention in the +R/-R vs operant conditioning (clicker training) may be the specificity of the cue.

    The OC camp would say that the cue - the click - has higher specificity related to the action being cued than gooood boys or pats on the neck.

    Even more so when -Ring... how can we be sure that ceasing the tap, tap of the whip is directly associated (in the horses mind) with increasing energy, or moving our outside leg back definitely means to the horse, keep hindquarters there?

    I can see how clicker training could be considered a more direct form of communication. I am wary however, of schools of thought that are too rigid.

    What is happening when for instance when I ask my horse to yield to my leg, he does, and I reward him both by removing the pressure (my leg), and praising him? And maybe the next time my horse responds more quickly, maybe bigger, and hopefully I used a more subtle aid.

    +R/-R seems to give a richer palette of options for the complex task of training a 1000 lb + animal.

    Another thought - when horses communicate with each other, there is plenty of -R going on...

    ReplyDelete
  4. I totally agree, CFS - I think they both have their place in horse training. The -R release means "you did something right" but the laser precision of the +R click means "you did that exact thing right."

    It's only 8:30 here so I can't think of anything more profound to add yet :)

    ReplyDelete
  5. I thought the whole point was to have fodder for argument. Click!

    ReplyDelete
    Replies
    1. Well, yeah, but I'm willing to be swayed in my opinions!

      Delete
  6. Okay, so I read the letter, I read your last post, and I read this post... I tried to compile my thoughts into competent sentences so I could comment, but I know I'm going to fail miserably. Everyone has their thing, and I don't like to comment negatively on methods that are used that are different than mine... I prefer to keep my yap shut, but I wanted to chime in because I could probably learn a thing or two.

    I don't like clicker training in the traditional sense of the word and have never used it on any of my horses. Maybe it's because I don't totally understand how it works, but I've seen CT done from the back of a horse and I find it ridiculous that every time the horse does something right, it hears the click and then receives a treat. That would annoy the crap out of me. Does the horse always receive a treat after every click, or do the treats disappear after the training evolves and eventually all you're doing is clicking? And how would you clicker train for a flying lead change? Click when they change and then stop to give the treat?

    Maybe I actually do a bit of "clicker training" myself by using my voice during training (by saying "good girl"), but I'm definitely a -R kind of trainer. I feel that the release is just as instantaneous as a click and I don't have to be bothered with clicking and treats. My horse is a pleaser and I feel that she gets more out of the -R training than she would get out of being offered a treat. Instead of thinking about how to get the treat, which seems kind of "mindless", she's focused on the job and is hunting for the release from me.

    I don't think I'm explaining it well, and I don't want to sound completely anti-clicker, but it sounds like clicker training would work well on animals you can't physically touch or who might have specific issues that need special attention. It's just not the method I want to use on my mare.

    Like CFS said, there's no +R going on in the horse's herd...

    ReplyDelete
    Replies
    1. Ooh ooh, a question I can answer!

      I use clicker training like it's just another training tool. Many clicker trainers, I think, do c/t for the behavior every single time, but I fade out the clicks and treats and then it's just like a traditional cue. I have never understood how you could effectively RIDE and c/t - maybe if you were trying to really reward a nice sliding stop or something? LOL.

      Here's a good useful thing I did with clicker training: I taught Dixie to line up at a mounting block. This was very early in our relationship and she was just a giant ball of nerves. I can think of three ways to teach "line up at the mounting block and stand still"; maybe you can think of more? Teach the horse to ground tie and "park" her by the block. Walk the horse near the block, and use a whip or hand gestures to line her up perfectly, then have her stand there while you walk around and climb up. Use a clicker and shape her behavior in two main parts: "put your body here" and "don't move."

      2012 Funder is like Reiner Klimke compared to 2008-09 Funder. (Not that I'm very good now, but I was WAY WORSE of a trainer back then.) I couldn't get her to halt using any traditional stuff I tried - whips freaked her out, she moved from hand pressure toward the block but moved from my presence when I darted around to the other side, and ground tying was about as likely as teaching her to fly. So I got a big bag of frosted mini-wheats and clicker trained her to stand at a block. It took like four cups of mini-wheats and 30 minutes to get her exactly right the first time. After that, I mounted and dismounted (using lots of treats) several times every time I went out to ride. Eventually she got pretty good at lining up, and I dropped it to one treat. Now, a lot of the time, when I offer her the treat after I'm up, she's like "nah no thanks" - and I don't even offer when we're out on the trail. I'll climb up on a rock and point to my left and she looks at me like "you are so useless" and lines up by the rock and I climb on and off we go.

      That's how I clicker train - if I can't figure out how to teach something traditionally, I'll try it c/t. Once it's a good rock-solid cue/response, I just drop to occasionally offering treats (and that's largely because I enjoy feeding treats to my horse).

      Does that make sense?

      Delete
    2. Oh - and it's not like I just couldn't figure out how to get her to stand still. She'd never been trained to stand still for mounting. When I bought her, getting on board was a minimum two-person operation - one standing in front firmly holding her bridle, one person climbing on, and sometimes another one blocking her from swinging sideways. She'd been ridden like that for three years.

      Delete
    3. I can see what you're saying about training for specific things, like mounting. I just can't fathom using it for things you're teaching once you actually are mounted, and I really don't like my horse working for treats. I want her to do it because it's her job and she's expected to do it... she's such a pleaser that I don't feel like I need to do anything special. Each horse is so different though. I wouldn't count it out for a different horse if I had run out of ideas.

      I pretty much agree with Bif, although I can't say it with words nearly as eloquently

      I can just imagine the look Dixie gives you, though!! LOL

      Delete
  7. I think my reasoning for being in the anti-clicker camp is similar to something Knotty Dogs said: sometimes you can't "wait" for the behavior to be offered. Clicker purists think any negative reinforcement is bad.

    OK. I can understand positive only reinforcement in working with wild animals, like orcas and dolphins. You really don't want to get into a fight with something that can easily kill you. But you know, the orca related deaths at Sea World, etc, are with positive conditioned animals.

    I prefer teaching animals to work off of praise, or their own enjoyment of the activity. Treats are OK, and definitely useful.

    I "get" how clicker training is great for people who do agility. And if I did agility, I wouldn't use clicker training.

    Perhaps I still think a bit too much of clicker training as Pavlov's dog. The dog gets conditioned that the bell means food. Dog starts to drool when the bell rings. WHERE IS THE INTELLIGENCE? I want the animal to CHOOSE the behavior I am teaching, not respond like an automan. I think clicker training at its root works to take away the thinking aspect, and seeks to make a response automatic.

    I know I am probably not correct in this, but it is how I feel. I like the concept of "free will". My animals have all learned "obedience equals freedom"; I didn't use a clicker. If you can teach an animal to be a willing partner through working your own mind to figure out how to get them to want to do the desired behavior, how much more interesting and rewarding than just shaping responses in what is basically a puppet? Clicker training, IMO, makes puppets.

    A nervous animal needs to learn where to draw its confidence, and what behavior is and isn't acceptable regardless of level of fear.
    You want the horse to look to the human for leadership any time he needs leadership, not just in previously "shaped response" situations.

    My old horse Cap never ONCE in his life barged into me. When he was crazy from stall rest and being hand walked, he would leap and rear and act the fool, but he never went ahead of me, never put tension on the line, and when he reared would only flail the front leg that wasn't near me.

    Put yourself in a similar situation with a positive only trained horse... And obviously I do A LOT of positive reinforcement with my animals. You have to keep a "positive bank balance" in any animal where you may need to do occasional negative things, whether due to the sport, training, whatever. Back to my aced and still crazy months long stall restee, going out for his daily in hand walk:
    How do you discipline that when he is already so full of excess energy any actual reprimand would make him explode? How can you clicker train any of that aberrant behavior away? Instead, his respect of our partnership kept him within line. I don't think one can accomplish that level of thought and respect with clicker training.

    ReplyDelete
    Replies
    1. You clicker train an aberrant behavior away by reinforcing the opposite behavior. Have a horse that barges into you? Train the horse to back away from you.

      If you'd like to see a clicker trained horse thinking, take a look at Alexandra Kurland's Lesson 9 DVD, "Overcoming fear and the power of cues." You can clearly see a little icelandic mare making a decision how to react in response to a stimulus she doesn't like.

      Delete
  8. I've always trained my horses with a "the release is the reward system" and been really successful. It kinda fell apart with my puppy and started doing a clicker based positive training system and OH BOY - it was WONDERFUL. I got FAST results and a REALLY motivated puppy, and taught her tricks and tasks that I couldn't dream of getting any other way but her offering them. However, after a couple of months, I found that 100% positive training doesn't work for her - so I incorporate what they call "non-reward markers" - ie "no", "enough", and even some forced down stays - but try to balance this out with positive clicker type training, as this seems to work best for her. After ~ a year long absence from my horse (busy with new puppy among other things) I came back to riding and toyed around with trying to do more "positive" training with Farley - and it just doesn't work for me. The "pressure/release" system works so much better and seems to communicate so much clearer. Tess (the dog) needs a creative outlet that allows her to offer behaviors that she can feel proud about and make her mind work at problem solving - my horse (Farley) DOES NOT. I do not want my horse to problem solve beyond "where do I put my feet". I DO want my dog to problem solve since some of the competition events that I'm planning on depend on her ability to offer behavior and put together complex ideas.

    I hope this was clear enough. Maybe I need to put a blog together? Or just link yours :).

    ReplyDelete
  9. Great series of posts here. I will just pick one point that I find interesting. The point you make on "being mean". I would agree that there is a definite distinction between being firm and asking till you get "a change" and being mean. From an outsider looking in that doesn't know what might be going on, some natural horsemanship methods when done correctly, could easily be mistaken for being mean or even abusive. I would also add a couple of other points that are equally important in my training. Specificity and timing. In training a stallion, for example, I had to be very firm with JB, to keep us and everyone else safe. I definitely use a system of pressure and release with my horses. When I ask something , I have to be clear in my mind(this is key and is often missed)what it is I am asking. Then follow through with the necessary pressure to get a change. I may not always get exactly what I after the first try but I reward THE TRY. If I am on the ground asking Maggie to release her hind quarters away from the pressure in her side, and she braces and stands there coming into the pressure, I continue to ask. However, I don't keep asking at the same intensity level, I up the pressure, until a get a change.Granted , it might just be that she shifts her weight the first time but that pressure gets released at that moment (if my timing is accurate!). Upping the pressure is where one might think it's being mean..Sometimes depending on the circumstance it can look that way. I watched my mentor take a stallion and just get what I call "big" with him. It was interesting because I could see the stallion wasn't completely listening to him but he wasn't out of control.There was alot more going on between the trainer and the stallion than what I could see. The trainer was asking nicely , once, twice, and then bring up the pressure when the stallion ignored him the third time. Eventually, the stallion knew that if he did what was being asked the first time, it didn't have to be so bad for him on the thrid attempt. It made sense. The other thing is that once a horse understands what is being asked of him/her it should be the trainers goal to set things up so that the horse is "allowed" to find that sweet spot. A horse wants to do what is being asked. A horse being taught to get soft in the bridle for example. It may take several attempts of the rider/trainer being in their mouth a little more than they would like to be before the horse understands it doesn't have to be this way. Be specific , allow the horse a break at the right time and pretty soon, that horse will start hunting for that on it 's own. That is when I know I am making a connection in my training..

    ReplyDelete
  10. This is a great discussion, Funder. Thanks for starting it. As I've mentioned to you, I am firmly in the camp that thinks training a horse is founded on the horse being taught to "respect" the handler/rider and I believe that respect can only be taught through negative reenforcement. This sounds well, negative, but I think many of us understand that its a fundamental underlying truth when it comes to making a truly reliable riding horse. I'm not going to go on about it in the comments cause we've already discussed it.

    But you have taught me where there may be a good niche for clicker training and positive reenforcement (I have never used these methods and never felt I needed them). Fear issues and refining complicated behaviors ("tricks" such as presenting at the mounting block when asked) seem to be an area where this sort of training could be really helpful, particularly on certain types of horses. In my opinion, the clicker training would need to be used on top of a foundation based on traditional/negative training (the horse knows he must respect you), or the horse would never become truly reliable--by my standards. But that's just my opinion.

    ReplyDelete
  11. Hey funder - I posted a video on Tess's blog showing a fairly recently behavior I go through clicker training - http://underbakedbrit.blogspot.com/2012/02/showoff-moment-roll-over.html

    Since it's new, you can see her offering alternate behaviors and "figuring it out". This is useful and cute in a dog. Do I want this in my horse - NO!!!!! I don't want Farley to offer a behavior that gets me hurt - I want as much control as I can get over teaching a new thing. I contemplated doing some clicker training with my horse - but decided I was just more comfortable with the pressure/release way of training in that situation.

    ReplyDelete
  12. I have found that my horse responds best to - R. He's a pleaser, and he likes to think up the correct response. He is learning to be lighter and to search for the release of pressure to indicate that he did something well. When I do release the pressure, I also say "good boy" and rub his neck, which he knows is meant to be praise, or +R. However, for most training, I start out using -R.

    My dog on the other hand is balanced pretty much the opposite way. He LOVES +R and I use a lot of praise, petting, and treats to get my desired responses. It's worked out really well for him. Occasionally, I do need to use -R to show him that what he is doing is the wrong thing to do, and quickly revert back to +R when he does the right thing.

    This may have something to do with prey vs. predator mentality.

    ReplyDelete
  13. Hey, thanks! I'm glad I didn't nerd things up too much. I do that sometimes (a lot).

    You made some very interesting points. As I said before, I know little about clicker training as it is practiced outside the lab. From the comments, I gather that there is a contingent of clicker trainers with a superiority complex towards their method of training. I get that, I see the same thing in Natural Horsemanship and Classical Dressage. For myself, I use many different stimuli to achieve the response that I want from my horse.

    I think that is is very short sighted to ascribe to one single "method" of training. The very earliest work in operant conditioning showed that different species responded better to different stimuli, and that, even within a single species, different behaviors needed different types of stimuli to be trained. For example, you can train a rat to run on a wheel using a mild shock, but you can only train a rat to stand on his hind legs with a food treat. The method of training required is entirely dependent on the type of behavior desired and the inherent nature of the trainee.

    I am thoroughly enjoying this series of posts, by the way. I think you've done a really good job presenting an educated, unbiased review of the material.

    Oh, and just to play Devil's advocate: It can very easily be argued that clicker training is negative reinforcement. Negative reinforcement = removal of stimulus. Stimulus = desire for treat. The animal desires a treat. When the target behavior is displayed, the treat is given and the animal is satiated. Thus, clicker training is negative reinforcement. See how it is arguing semantics?

    ReplyDelete
  14. I couldn't imagine only using positive reinforcement with a horse. We can use clicker training with more traditional methods. In fact, we should. A horse needs to learn the typical cues for riding--just so other people can ride him, too.

    With my clicker horse, Cole, I use the traditional negative reinforcement and I click. If I am teaching a move, I will use my leg pressure. When he gets it right, I remove my leg pressure and click. He learns the cue in the traditional manner, but much faster because of the click. The click is a bridge to faster learning.

    When he is understanding it well, I will only click improvements--that way he knows when he is doing better than before. Just negative reinforcement can't tell him when he improves quite the same way. Since he gets clicked for improvements, he keeps trying to improve on his own. Sometimes, he gets it wrong, but I just ignore it, and he won't try that, again.

    He is far from a robot--responding just because he was conditioned and only because he might get a treat. He can be quite a joy to ride because of his enthusiasm. He is incredibly focused on the training, learns fast and we both have fun. And, if I drop something when I am riding, he will pick it up for me.

    I have found that clicker training is best when you have a specific problem that needs to be solved. The first horse I used it on, started to have serious health problems that we never could figure out. Anyway, he was having trouble lifting a hind foot. Bad luck would have it that he got an abcsess in that hoof. He fought any attempt to lift it for examination, soaking or wrapping. He didn't even want me to touch the leg.

    I got the clicker and some mints. I started clicking for letting me touch his leg, moved on to shifting his weight off it and 10 minutes later, he was lifting his hoof to a light tap on the leg. It was a miracle. He never gave me a problem with it, again, desptie the pain. He realized he could trust me--and that fighting me hurt his leg more than cooperating. I could have solved the problem in other ways, but none would have been as kind and gentle.

    ReplyDelete
  15. I think the other issue for me is I find the sound of the clicker EXTREMELY ANNOYING. You'd have to give me a hell of a lot of Godiva to make positive associations with that 'clicket' sound.

    Bleg.

    ReplyDelete
  16. Yes, it is an annoying noise--I replaced my clicker with a tongue click. Much more pleasant and way more convenient.

    ReplyDelete
  17. I wrote the letter that Funder originally linked to, and I'm really grateful to her for keeping the discussion open (in every sense). I hope it's clear to anyone who's read my letter that I did *not* intend to tear down Buck Brannaman or his work. Just the opposite - I was deeply impressed by his skill (it's no exaggeration to call it his art) in communicating with horses and enlisting their tremendous intelligence in partnership with his own. Which is why I found it heartbreaking to hear him contemptuously dismiss a whole body of trainers whose ideals and methods actually overlap significantly with his own.

    I think Shannon is absolutely right on a couple of counts. First that a dogmatic devotion to any specific method amounts to a dangerous abdication of our independent judgment. I love the approach she describes on her blog of auditing all the clinics she can and staying in a questing, learning frame of mind at all of them. I think if more so-called “clicker trainers” (more on that dubious phrase in a minute) attended Buck’s clinics and more “natural” trainers attended Alex Kurland’s, our conversations would be enriched beyond measure. It’d also be great if we made a common commitment to speaking out when we hear trainers of any label unfairly caricatured. As Funder demonstrates, we can be critical to great productive ends as long as we stay curious, and as long as we resist the temptation to vilify the people with whom we disagree.

    Second, Shannon makes the excellent point that a lot of our most heated debates are really about semantics. I strongly prefer the term “marker training” to “clicker training” not only because the latter is loaded for so many people with negative associations, but also because it’s badly misleading. It puts the focus on that annoying doohickey and not on the broader principles that guide the effective use of markers of every kind (verbal, tactile, visual, etc). The sound that clickers make can be a highly effective marker thanks to its precision, consistency, and clarity. That clarity can accelerate the learning of new behaviors (including a ready response to tactile pressure). For the same reason, it makes an excellent entry point to the whole concept of marking and rewarding for the novice trainer. But the best “clicker trainers” are those who see clickers as just one tool in a big box. Same thing with treats. Good marker trainers, whether or not they use clickers, are creatively engaging the whole range of an animal’s appetites: for food, yes, but also for freedom, for play, for calm, for connection, for competence.

    ReplyDelete
  18. Sorry, had to divvy up my too long comment. One more thought expressed in too many words ;-) :

    As Shannon notes, current research has all but dissolved the scientific distinction between R+ and R-, but I think Funder is right that the philosophical and practical training distinction may still have some substance. I’ve ridden horses maybe ten times in my life, and the main reason I haven’t ridden more is that my pleasure has always been overshadowed by my pity for the poor horse who has to suffer through my insensitivity and incompetence. When I do learn to ride, I’ll have a big learning curve ahead of me, and I’ll want to inflict as little of my struggle as possible on the innocent horse who has grudgingly agreed to carry my imbalanced weight. As long as my control over my own tactile signals remains poor, I’ll want other means to communicate clearly with that horse. As long as I lack the ability that Buck Brannaman possesses of inviting the horse into a mutually pleasurable dance, I’ll want other ways to reward him for his patience and good will. If I were prepared to invest as much care and time as Buck does in finding a feel on the ground before I ever presumed to get on a horse’s back, and if I stayed attentive to the highly particular things that motivate the horse and established to his satisfaction that I could provide those things, I’d have good leverage for getting us both past our limitations of understanding and skill. That’s what R+ means to me in the practical sense: keeping the question ‘what’s in it for the horse?’ always in mind.

    Gretchen

    ReplyDelete

Feel free to comment!