When should you lie to your Border Collie?

[Shetlander]

I propose that a properly timed reward system communicates to them *when* they have it right and that this clear communication increases the internal drive. Personally, I feel that that is the beauty of (shudders) clicker training (which *shudders* incorporates treats). It's a clear communication system that says "you're right" at the instant the dog is right. Is it the only effective way to train? no. Does it cause the internal motivation to wither? Again, I would say no.

 

Exactly. Clicker training and treats communicate in a way that, done correctly, is very clear and enjoyable to my dogs. It works well for me. My dogs have fun learning. And I have fun watching them learn. And because they are well trained and obedient, our life together is more harmonious, which is my internal motivation.

[Maja]

Eileen's question - I wondered about it myself - the vehemence with which some people were opposed the idea, although I never opposed the idea of treats - I just suggested another option and walked the walk with my dog.

 

And I don't know why clicker training should make any one shudder? It is a very nice, straight forward training tool, which I used in the past with excellent results.

 

Julie,

Exactly.

[Maja]

"Does it cause the internal motivation to wither? Again, I would say no."

 

I might have missed a post or two, but I have not noticed anyone actually saying that. I think the general tenor has been " The internal motivation is not as high as it (easily) could be".

[Donald McCaig]

Dear Doggers,

 

Philosophers as different as Plato, Marx and BF Skinner yearn for universal, immutable answers - what was true yesterday for kumquats will be true for copper sulfate the day the earth falls into the sun. One of the confusions I sense in this discussion is the assumption that training methods which are perfectly reasonable for a pit bull will suit a Pomerian or a Border Collie and I don't think that's true.

 

My dog experience - unlike some on the forum - has been entirely with Border Collies and my training experience has been training Border Collies for sheepwork and mannerliness in a broader range of conditions than is usual - I think - with most companion dogs.

 

With the possible exception of still evolving ecollar training and (I am told but haven't enough evidence) drive training, training methods exist before we trainers take our first dog out. Koehler methods were developed during and right after WW2, Skinner published The Behavior of Organisms in 1938 and Lorenz's insights were taken up by (mostly) German Drive trainer in the 1950's.

 

All these methods claimed to work equally well for every breed of dog.

 

My objection to treat training isn't that it's harmful or doesn't work, it's that, for Border Collie sheepdog work and mannerliness treats are unneccessary. Unlike the universal philosophers I named I believe some theories work pretty well sometimes and that how you train a dog bred to fight other dogs or slip into a burrow to kill a fox isn't the same "how" you use to train a dog breed selected for centuries for its work ethic and eagerness to work with human beings.

 

As this discussion has progressed, the images of a priest and a young Goldman trader have come to mind. They are motivated very differently. Sure, as Pope Francis has reminded us the motivations can blur and overlap but that isn't how you get good priests.

 

Some years ago, my friend Vicki Hearne (Adam's Task, Animal Happiness) bought a pup from me. Vicki was a fine trainer who'd studied with Bill Koehler. Her heart was with the big terriers, Airedales, American Bulldogs, Pit Bulls etc) who, she told me, liked to "overcome". She often spoke of "gameness" as a dog's primary virtue.

 

So, one morning Vicki took Border Collie Kep out to start training with the longe line (standard Koehler Method). In brief, when the dog is walking one way, ignoring you, you turn and walk the other way. Line tightens, dog is jerked. After enough repititions the dog learns" Hey, I better pay attention to what this nut case is doing".

 

Kep got jerked. Once. He figured it out and trotted attentively at Vicki's side. Used to big terriers, Vicki wasn't used to this and hated it: where's the "overcoming", the "gameness", the resistance the trainer had to overcome?

 

Kep said, "Just tell me what you want and if I can I'll do it."

 

Vicki said something unprintable.

 

That's why I don't like treat/clicker training a Border Collie. It's an unnecessary waste of time and may be (cf Pope Francis and the NYT article) corrupting.

 

Donald McCaig

[teresaserrano]

Love it, Mr. McCaig. I've been following this thread and not participating as I just arrived at the forum, but I've loved your posts.

 

My previous dog was a Cão da Serra de Aires (portuguese sheepdog). She was sensitive and easily startled, suspicious of anything new and with not a lot os self confidence (when she was 4 mo I was even told she would never amount to nothing because of her weak character). But she loved to work.

 

That dog ended up climbing the 6 feet wall, jumping jumps on fire, eagerly doing bite work, even surfing. She did those with great joy and an evident pride in herself. I did use treats when she was a pup, not exactly to teach her to do stuff but to teach her how to learn, which is a different thing. Then I stopped using treats and it made zero diference in her learning rate and willingness to work with me. She liked working and she liked doing stuff with me.

 

Now with Tess, I started out using treats. I stoped almost imediatly. She is motivated by us doing things together and doesn't even like food much. As you say, Mr. McCaig, she's of the mind "Just tell me what you want and if I can I'll do it". The obstacles at training, all of them, she did first try just because I asked her to. And she thought it was super fun.

 

Obedience training, I have of course to be very clear about what I'm teaching, and about giving her feedback for her performance. And motivation is essencial, keeping things fun and interesting. But right now, at almost 1 yo, her best reward is by far us making a party together, jumping and yapping and running around, because she did awsome. And if there's a game involved, all the better.

 

There's another aspect here. I do understand that clicker/treats can make it clearer for a dog what is expected of him. But honestly, personally, I don't find it hard making myself clear and never felt that my dogs found it hard understanding me. Tess, now, is a quick learner, she understands things fast. Not saying it doesn't work well, just that I don't feel the need.

 

Is this a good aproach for every dog? God no! I see plenty of dogs at training that couldn't care less about their owners, or at least couldn't care less about doing stuff with the owner. I tend to think that in many cases the bond between dog and owner is kind of week and that's where they should start working, but many dogs just aren't wired to work closely with a person and must have an extra incentive.

 

For me and my dogs, though, not working with a cliker/treats works very well, better I think than working with them. I've heard that one of the assets of clicker/treat work is that it makes training more impersonal. I do understand the logic here, but personally I want training my dog to be very very personal. But that's of course just an individual experience, nothing more.

[Shetlander]

That's why I don't like treat/clicker training a Border Collie. It's an unnecessary waste of time and may be (cf Pope Francis and the NYT article) corrupting.

 

An unnecessary waste of time for you and your Border Collies. A fun, effective and relationship building approach for others. As long as we are happy with the results of our training methods and enjoy a deep bond with our dogs, isn't that what is important here? The quality of our relationship with our dogs, regardless of the lives we lead with them?

[Pippin's person]

I think Mr. McCaig's point concerning the presumed universal applicability of training methods is one to keep in mind. One of the things I've learned and continue to learn is how to engage the intrinsic motivations (as mentioned in the NYTimes article but also in terms of the internal make-up) of each of my individual dogs. The dogs I work with are primarily border collies and, while there is much they share, they are also really different, needing different kinds of information and strategies of communication. The best bred ones are the least motivated by stimulus/response (behaviorist) based training (I make no claims about cause and effect there, just note it as an observation). Since that's how I learned to work with a dog, it's taken me a lot of time and thought to revise my approach. Still very much a work in progress :-)

 

Patrick Shannahan posted this article (http://www.patrickshannahan.com/Correction.asp) about corrections in stockdog training that I thought was pretty interesting and in line (both in tone and content) with this discussion.

[teresaserrano]

That is something I completely agree with, Liz. I don't usually give much thought to HOW a dog is being trained, if it works for both dog and owner, creating a happy productive team.

 

I do however react a bit sometimes, not often but sometimes, when people that clicker/treat train assume I must be at least kicking/yelling at my dog in order to get the fast reliable responses I get, if'm not clicker /treat training.

 

I have NOT felt this in this forum. I do remember one instance a few years ago when I was talking to a PP trainer about border collies, because I thought that might be my next dog, and he said, knowing where I train: "But if you get a border collie, you can't train him the way you train now, because they're very sensitive dogs". I laughed and said, "Well, but the way you think I train is NOT the way I train". But it was ofensive, because it's assuming that if one is not clicker/treat training, then one is not working towards clear fluid communication and building a strong bond of trust and partnership.

 

Turns out that Tess is much much less sensitive than Sara was. She's a though little dog always wanting to DO stuff, that just needs orientation in order to learn the right things.

 

Sorry about the rambling.

[Root Beer]

For me and my dogs, though, not working with a cliker/treats works very well, better I think than working with them. I've heard that one of the assets of clicker/treat work is that it makes training more impersonal. I do understand the logic here, but personally I want training my dog to be very very personal. But that's of course just an individual experience, nothing more.

 

That's actually what I was addressing above. While there are some clicker trainers who do want it to be that way, more and more of us are incorporating the clicker (and other modes of +R) into a much more personal approach.

[Root Beer]

That's why I don't like treat/clicker training a Border Collie. It's an unnecessary waste of time and may be (cf Pope Francis and the NYT article) corrupting.

 

Having used a clicker/treats to completely transform a Border Collie who started out terrified of all dogs and people (except my husband and myself) into one who met the world with joy, danced in front of 500 people and loved it, and played off leash with other dogs and children, actually came to love meeting people, was able to go into the back at the vets alone and be fine, and enjoyed a quality of life that seemed impossible when he was young, I have to disagree with your assessment of "waste of time".

 

No degree of telling him anything could have achieved those results. No amount of praise would have done a thing for him. No correction could have gotten those results. It was the clicker that literally changed the way his brain worked in the situations that were originally crippling for him. I watched it happen - one instant at a time. It was an incredible thing to witness.

 

I can't categorize a life lived with joy and enthusiasm and a hearty sense of adventure instead of a life lived hiding and cowering as a "waste of time". Nor as a "corruption". And I would be very surprised if anyone here - really and seriously - would say that it was, and would assert that it would have been better if I had just let him remain a severely fearful dog, hiding at home his entire life.

[Shetlander]

I think Mr. McCaig's point concerning the presumed universal applicability of training methods is one to keep in mind

 

That isn't the point I hear him making in these conversations.

[rushdoggie]

I don't think the internal motivation of a dog is to eat. As Maja noted, the internal motivation of the dog is to please the human. They've been bred to work *with* a shepherd to reach a common goal.

 

I assume you are talking only Border Collies here?

 

I teach pet dog classes, with all breeds/mixes. Its often very clear to me there are types of dogs who work to work, work for their humans and those who do not.

 

We teach basic manners with food and clickers, but also talk about other options for rewards such as touch, toys, genuine praise.

 

Some types of dogs thrive on the touch and praise, melting into it. Some look annoyed.

 

There's a Border Collie puppy and a Malinois puppy in my current basic manners class. They both have the attention span of a housefly but both genuinely thrive on dog-human interaction and praise. The terrier mix, the Akita mix and the Mastiff? not so much. They have minimal interest in their humans in the face of distraction and require that extrinsic reward to make the effort.

 

My own BC loves touch and praise but not in a context of training. He seems to thrive on hearing the click but often doesn't care that much about the actual reward (he takes it and it falls out of his mouth). If I try to pet him in that context he flicks an ear at me ("ok fine hurry up theres things to do"). So for him, to a degree getting it right *is* enough. Unless...I break out a tennis ball. He will do anything for a tennis ball.

 

My Papillons love praise but they also love the cheese reward, and my younger one ADORES personal play. I have to watch myself because he finger and arm wrestles ferociously and with glee, and that crap will get me thrown out of an obedience trial.

[HJTRAS]

Making this whole thread clear. seems it went in so many directions . Lets give the simple question with a simple answer. I am really curious the whole thing was hard to make simple. question and answer.

[juliepoudrier]

Rushdoggie,

Speaking for myself and my comments on this thread, yes, I am talking about border collies (and I don't think Donald and Maja are talking about any other breed either). I'll be even more specific than that and say that my comments relate mainly to working border collies. As this discussion has unfolded I have noted that the people talking about not training with treats are largely (though not exclusively) people who use dogs for stockwork, and I thought that some of the disconnect (yet again) is that border collies are being bred for any number reasons now, besides stockwork. As that happens, it's possible that the dogs as a whole may move away from the biddability (meaning working solely from a desire to please) that has been a hallmark of the breed.

 

I am not anti-treat training; as I stated earlier, I use treats to train tricks where it makes sense to do so. But I do firmly believe that the working border collie has been bred to want to please, to want to work with the shepherd, and this is a trait that is important, certainly for those of us who want to work stock with our dogs. We always hear that the work itself is the reward and so of course we are also rewarding our dogs, just in a different way--with stock rather than a click or a treat. That argument loses strength when you realize that yes, the dogs like to work, but a dog working well is not the same thing as a dog pleasing itself, and both can happen while the dog is working stock, the supposed reward. (I'm sure I'm not saying this well, but what I'm trying to say is that you can't immediately take the stock away if a dog is being a jerk, nor can you immediately make the stock be a perfect reward, because of course they are sentient beings with minds of their own and so even when the dog is doing the right thing--or trying to do the right thing--it's getting feedback from the livestock that might not be entirely commensurate with what it's doing, not to mention that the human is also giving feedback. There are times when the feedback from the stock is more compelling than the feedback from the human, as well as times when the stock issue a correction that is harsher than any a human would use, and so it's a more involved/complex training situation, rather than a situation where we can say "if you offer this correct behavior, there will be a reward or that here are the sheep, working them is your reward, or conversely, if you don't offer what I want, you will not get the reward. For me, verbal praise and corrections are important mainly because they can be given when the dog is at a distance. I realize the dog can hear a clicker from a distance, too, but you may still be asking for actions at a distance that aren't requiring the dog to process feedback from you and other beings.)

 

Anyway, when some of us say pleasing the human should be enough, it's this POV we're coming from. I don't believe treat/clicker training somehow ruins or cheapens a person's relationship with their dog, and in fact I think such training methods have enabled many people to do things with their dogs they might never have done before, simply because they find that they can easily train that way, but I do worry that in all of this is a kernel of going down a road where we start to lose the type of border collie that has that strong desire to work for the human just because it pleases both to do so.

 

Just my opinion of course.

 

J.

[G. Festerling]

Christine....and I am singling you out but not with ill intent....how do you know that another approach would not have worked? Admittedly, it may not have worked for, very specifically, the pair of you! Not because you are not a dedicated and good trainer or the dog a mess. I would like your statement a lot better if it did not seem to exclude the possibility, that there were no other ways with other people. I see people do amazing things with methods and ways that I know, I would struggle with greatly. And it is always a great remind ....that, the approach is as individual as the pair working. It has to be.

 

Shaping is great. But I find it funny how so much focus is always on the animal. When I teach it, I find it much more helpful with the humans. Because it makes THEM have to focus on the act of training and making thinks clear. Therefore the animal benefits. Not sure that I am wording this right.

 

I also shape using no treats. But praise, giving other rewards beyond just taking pressure off for the right choice. Too numerous to list.

[G. Festerling]

And I just saw Julie's post....a huge YES....to her statement that above all, pleasing should be enough.

 

I will dare to go further...pleasing is easy, especially when the one making the demands does his or her due diligence in presenting the situation. Being acknowledged for doing the right thing is genuine. It is comfort. It continues. It breeds trust and respect. On both parties.

[Shetlander]

Anyway, when some of us say pleasing the human should be enough, it's this POV we're coming from. I don't believe treat/clicker training somehow ruins or cheapens a person's relationship with their dog, and in fact I think such training methods have enabled many people to do things with their dogs they might never have done before, simply because they find that they can easily train that way, but I do worry that in all of this is a kernel of going down a road where we start to lose the type of border collie that has that strong desire to work for the human just because it pleases both to do so.

 

Thanks for the explanation, Julie. Lots to think about there.

[Mark Billadeau]

From what I can tell from working with dogs and working with college students, not one teaching method can be applied to all. There will be a large group that can learn with any one method; but there will always be individuals that need other teaching methods to learn (or learn at the same rate). Proponents of only one teaching method either have not worked with enough of the other learners (those that do not learn or learn much slower by their method) or are only interested in working with those that can learn by their method.

[teresaserrano]

I really like your post, Mark. I'm also a teacher and find that to be very true.

 

I assume I'm mostly interested in working with inteligent sheepdogs. They just are my kind of dogs. But you can have 10 dogs that are all, a) inteligent, and sheepdogs, and yet have such distinct temperaments that the modus operandi when is comes to teaching them will have to be vastly different among them.

 

I don't think clicker/treat training is wrong. Far from it, it has been a huge advantage for many many people training their dogs, and it's what works best for many dogs. But in my very limited personal dog training experience, I haven't found much use for it.

[Sue R]

From what I can tell from working with dogs and working with college students, not one teaching method can be applied to all. There will be a large group that can learn with any one method; but there will always be individuals that need other teaching methods to learn (or learn at the same rate). Proponents of only one teaching method either have not worked with enough of the other learners (those that do not learn or learn much slower by their method) or are only interested in working with those that can learn by their method.

Wow, well said.

[Mark Billadeau]

Is the success of clicker training due to the rate at which dogs learn with it or due to the rate at which dog owners learn to train their own dogs with it?

 

I often wonder if those dog trainers that are successful with clicker training would also be successful with other training methods because they know dogs.

[G. Festerling]

To me, I already answered it. Because, it is all about finding communication. And as the human, when we stop and care enough to really isolate and listen and get on the level of who or whatever....it is us that holds the answer. Just as I stated before.

[Root Beer]

Christine....and I am singling you out but not with ill intent....how do you know that another approach would not have worked? Admittedly, it may not have worked for, very specifically, the pair of you!

 

I'm going to give a twofold answer to this question. (BTW . . . Kristine. No "Ch". )

 

First, I knew this dog. I spent just over 12 years from him. He got me started on the path of training and performance. He taught me what kind of trainer and handler that it is in my heart to be. He taught me what really matters.

 

I also know what I tried that failed and I know what I tried that worked.

 

So, when I say that a particular approach would have failed with him, I base that on my knowledge of this particular dog, which was first hand and extensive.

 

But . . . in a way that doesn't matter. The fact is - I used the clicker and more general +R training approach with him. That is what I did. And the result was a dramatic transformation that was good for him.

 

So, in answering the assertion that clicker training is a "waste of time" or a "corruption", all that really matters to me is the actual difference that clicker training made in this dog's life. One could say "shoulda, woulda, coulda" regarding other approaches with him, but that was not our path.

 

 

I would like your statement a lot better if it did not seem to exclude the possibility, that there were no other ways with other people.

 

Well, suppose, in theory, another person could have made an identical transformation simply by praising this particular dog (in reality, that really wasn't happening - he was way too far gone for that back in the early stages)?

 

I would maintain that theoretical possibility does not in any way render the actual results of the actual method a "waste" or "corruption". The bottom line is - in the end the dog was able to move beyond fear to have an excellent quality of life.

[Maja]

Excellent posts Julie and Mark.

 

This also touches on another aspect: not every method works for every teacher/trainer. So the best method is the resultant vector of the dog's needs/desires/predispositions/etc on the one hand, and the teacher's needs/desires/predispositions/etc. on the other.

[Mark Billadeau]

Maja, now consider teaching novice sheepdog handlers to train their dogs. The teacher needs a method that they can execute which the novice handler can learn to perform in a way that teaches their dog to deal with other animals.

 

Everything is in constant motion: student, dog, and livestock.