Training With No Compulsion?

[Diablo]

I trained heel without a lead, compulsion, or correction. Then I proofed it without corrections.

Very nice heel and performance Bedazzledx2, you have done an excellent job with great focus in your dog. With my dog in the same circumstances, his focus would be to chase and attack the other dog running around on field and no matter what reward you offered him to behave, nothing other than a correction would redirect his focus and drive to attack that other dog or the other person for that matter if he didn't know them, in fact, without the training he has completed, I would have to block him (cut off his air) to settle him down to work.

Dogs like mine can't be manipulated with treats to override aggression, but at the same token, dog's that can be overidden with treats can't work in a security/protection role as good as mine can which results in dogs at different ends of the spectrum with a common factor that both dog types still must be able to be trained effectively as an end result.

[Cosmolo]

Kavik, i have a dog here who was 'compelled' to go through a tunnel after other methods had failed (this was also under the guidance of people more experienced in agility than myself) and after 2 compelled repetitions with high value rewards at the end- she was fantastic and now loves the tunnel.

My only regret is not compelling her earlier as the other methods (which had worked fine for mine and many other dogs) caused more stress over a longer period of time for her. I understand that this would not be the 'norm', but it is an example of different things for different dogs- i knew my dog well enough to know it would work well, i just didn't have the confidence to speak up to those more experienced than me at the time when we were going through the other processes.

On the other hand, my kelpie x would be seriously p**sed if anyone tried to 'compel' her to do anything pertaining to agility- combination of luring and shaping is the best option for her!

[Diablo]

You're missing the same point MonElite is missing, Diablo. You can't teach a dog what a red light is without letting them run it and work it out for themselves from the consequences. That is how animals learn, but if you deliver positive consequences instead of negative consequences than at least they don't have to deal with the stress of trying to make the negative consequences stop without knowing what will make them stop.

You correct a dog for disobeying a command that has been previously learned that the dog knows. The stress of correction is the consequence of disobeying a leaned command. The positive reward is the pleasantness of obeying.

It's not about yanking the dog around by the throat, it's about teaching the dog a command and teaching what is required from the command. Simple example the casual heel: You start of and command heel and praise the dog for being in the right place. The dog surges ahead.......aghhh.......a couple of leash pops and "heel" then praise the dog for returning to the correct position etc etc. When the dog has learned in bullet proof fashion to heel without distractions off leash.........the dog knows the command.

Add some distractions like another dog approaching. Heel......the dog is in position and praised, then he see's the dog approaching and surges ahead, aghhh heel and he pulls to the end of the leash ignoring the heel command and a leash correction is admistered for ignoring the heal command. The correction is not the response for surging towards the other dog, the correction is for ignoring a known command which the dog has the choice of compliance or not. The dog learns that ignoring commands results in unpleasantness and complying with commands results in positive reward. The dog in this instance has learned right from wrong and the consequences of each chosen action.

The dog will stay behind and heel without pleasure. Period. If you teach your dog like that you never have dog that wants to work. ( No drive) The dog will learn to expect punishment or wait for the ".......aghhh.......a couple of leash pops"

That's not training a dog to work Misha, that's training obedience which is two different concepts, but without obedience as the foundation you create a solo tasking dog. It's like the GSD that was trained specifically for Schutzhund competition that got hit and killed by a car chasing a cat across the road. The dog had a SCHh3 title but the handler couldn't stop the dog chasing the cat through lacking in "off field" general obedience that the dog had never been taught.

Work is taught in drive...........a different situation and different commands. However, you could train a casual heel in drive with a release word for the dog to chase and bite it's target being another dog approaching as the reward, but I doubt that too many people would be happy with such a training concept

Wrong wrong wrong

Even the info that you have that SCH3 dog will do that is questionable or the SCH3 title is questionable. Training SCH only for competition???? You raise dog for SCH not train. You raise dog for sport. I personally can not distinguish training obedience and training a dog to work.

It's not wrong at all, sport is a specific routine for which it's trained to do. A police K9 will have a higher level of general obedience than a dog trained for Schutzhund competition and will be workable over a wider range of situations. Police K9's are definitely not trained in purely positive methods which I think Jeff Jones explained to us earlier.

[MonElite]

You're missing the same point MonElite is missing, Diablo. You can't teach a dog what a red light is without letting them run it and work it out for themselves from the consequences. That is how animals learn, but if you deliver positive consequences instead of negative consequences than at least they don't have to deal with the stress of trying to make the negative consequences stop without knowing what will make them stop.

You correct a dog for disobeying a command that has been previously learned that the dog knows. The stress of correction is the consequence of disobeying a leaned command. The positive reward is the pleasantness of obeying.

It's not about yanking the dog around by the throat, it's about teaching the dog a command and teaching what is required from the command. Simple example the casual heel: You start of and command heel and praise the dog for being in the right place. The dog surges ahead.......aghhh.......a couple of leash pops and "heel" then praise the dog for returning to the correct position etc etc. When the dog has learned in bullet proof fashion to heel without distractions off leash.........the dog knows the command.

Add some distractions like another dog approaching. Heel......the dog is in position and praised, then he see's the dog approaching and surges ahead, aghhh heel and he pulls to the end of the leash ignoring the heel command and a leash correction is admistered for ignoring the heal command. The correction is not the response for surging towards the other dog, the correction is for ignoring a known command which the dog has the choice of compliance or not. The dog learns that ignoring commands results in unpleasantness and complying with commands results in positive reward. The dog in this instance has learned right from wrong and the consequences of each chosen action.

The dog will stay behind and heel without pleasure. Period. If you teach your dog like that you never have dog that wants to work. ( No drive) The dog will learn to expect punishment or wait for the ".......aghhh.......a couple of leash pops"

Perhaps it depends on the original level of drive in the dog? Have you thought of that?

Check out this thread, first dog there http://www.dolforums.com.au/index.php?show...tzhund&st=0

I perosnally know this dog and I ahve never seen anything more keen to work then him. And I can tell you that he has been corrected many times for missbehavior. He is also clicker trained and a SCHIII now.

Nordernnstam Harley is his pedigree name, if one day you get an opportunity to see this dog work its pure pleasure. The willingness to work is just over the top.

Again - this dog has been corrected but not over corrected.

Mind you what level of correction this dog can receive is a lot higher then any of my dogs could.... And he springs back from the correction in a split second.

Edited January 11, 2010 by MonElite

[Kavik]

Kavik, i have a dog here who was 'compelled' to go through a tunnel after other methods had failed (this was also under the guidance of people more experienced in agility than myself) and after 2 compelled repetitions with high value rewards at the end- she was fantastic and now loves the tunnel.

My only regret is not compelling her earlier as the other methods (which had worked fine for mine and many other dogs) caused more stress over a longer period of time for her. I understand that this would not be the 'norm', but it is an example of different things for different dogs- i knew my dog well enough to know it would work well, i just didn't have the confidence to speak up to those more experienced than me at the time when we were going through the other processes.

On the other hand, my kelpie x would be seriously p**sed if anyone tried to 'compel' her to do anything pertaining to agility- combination of luring and shaping is the best option for her!

I have also seen people use compulsion to get a dog go through a tunnel and even helped (not competition agility training). However especially after an agility instructing seminar I attended last year where the tunnel was specifically mentioned, I feel that you can find a way to make a small enough step so that you should not need to use compulsion. Breaking the tunnel down into a hoop even for the first step if the dog is really scared.

My main point was that in agility you want independence and a lot of the training is about CHOICE. Shaping is very important in agility.

[Cosmolo]

I completely agree with you- for the most part But providing 'choice' stressed this dog and the small steps we had tried didn't impact the real thing.We probably could have persisted for longer but i decided moedrate stress for a very short time was better than low stress for a long period of time for this girl. She is an unusual dog though- certainly not the norm!

[Aidan]

With my dog in the same circumstances, his focus would be to chase and attack the other dog running around on field and no matter what reward you offered him to behave

Without judging anyone for their methods, and never having met your dog, I would dispute this as a fact. I have no doubt that my working-line GSD would be perfectly suited to security work (all her littermates are working), and certainly if I were to "wave food around under her nose" in the early days with another dog around it would have been a waste of time, so I'm not doubting you on that point. Yet I have got her to a high level of obedience around other dogs using food, even playing with some other dogs and off-leash walking with selected dogs. How you set the dog up to learn is very important, you can't take a ham-fisted approach with a fistful of ham and an aggressive dog (sorry, I had an urge to pun).

And again we come back to the strength of reinforcers ("no matter what reward") - it is the history of reinforcement that matters, the strength of a reinforcer is just one of the small variables to be accounted for in the dog's history of learning. You can take the best reward your dog has ever come across and still not reinforce anything with it because you didn't set the dog up to learn.

[ZEPING]

You're missing the same point MonElite is missing, Diablo. You can't teach a dog what a red light is without letting them run it and work it out for themselves from the consequences. That is how animals learn, but if you deliver positive consequences instead of negative consequences than at least they don't have to deal with the stress of trying to make the negative consequences stop without knowing what will make them stop.

You correct a dog for disobeying a command that has been previously learned that the dog knows. The stress of correction is the consequence of disobeying a leaned command. The positive reward is the pleasantness of obeying.

It's not about yanking the dog around by the throat, it's about teaching the dog a command and teaching what is required from the command. Simple example the casual heel: You start of and command heel and praise the dog for being in the right place. The dog surges ahead.......aghhh.......a couple of leash pops and "heel" then praise the dog for returning to the correct position etc etc. When the dog has learned in bullet proof fashion to heel without distractions off leash.........the dog knows the command.

Add some distractions like another dog approaching. Heel......the dog is in position and praised, then he see's the dog approaching and surges ahead, aghhh heel and he pulls to the end of the leash ignoring the heel command and a leash correction is admistered for ignoring the heal command. The correction is not the response for surging towards the other dog, the correction is for ignoring a known command which the dog has the choice of compliance or not. The dog learns that ignoring commands results in unpleasantness and complying with commands results in positive reward. The dog in this instance has learned right from wrong and the consequences of each chosen action.

The dog will stay behind and heel without pleasure. Period. If you teach your dog like that you never have dog that wants to work. ( No drive) The dog will learn to expect punishment or wait for the ".......aghhh.......a couple of leash pops"

Perhaps it depends on the original level of drive in the dog? Have you thought of that?

Check out this thread, first dog there http://www.dolforums.com.au/index.php?show...tzhund&st=0

I perosnally know this dog and I ahve never seen anything more keen to work then him. And I can tell you that he has been corrected many times for missbehavior. He is also clicker trained and a SCHIII now.

Nordernnstam Harley is his pedigree name, if one day you get an opportunity to see this dog work its pure pleasure. The willingness to work is just over the top.

Again - this dog has been corrected but not over corrected.

Mind you what level of correction this dog can receive is a lot higher then any of my dogs could.... And he springs back from the correction in a split second.

i can also vouch for this dog it brillent

[Diablo]

'MonElite' date='11th Jan 2010 - 09:58 PM' post='4250228']

Perhaps it depends on the original level of drive in the dog? Have you thought of that?

Check out this thread, first dog there http://www.dolforums.com.au/index.php?show...tzhund&st=0

I perosnally know this dog and I ahve never seen anything more keen to work then him. And I can tell you that he has been corrected many times for missbehavior. He is also clicker trained and a SCHIII now.

Nordernnstam Harley is his pedigree name, if one day you get an opportunity to see this dog work its pure pleasure. The willingness to work is just over the top.

Again - this dog has been corrected but not over corrected.

Mind you what level of correction this dog can receive is a lot higher then any of my dogs could.... And he springs back from the correction in a split second.

My GSD is the same MonElite regarding a recovery from corrections for misbehaviour, in fact, it makes no difference to his drives whatsoever afterwards. The trainers at our Schutzhund club "all" teach and use corrections for misbehaviour and blocking with unwarranted aggression which is standard practice. Treat and marker training is used also to teach the dog to do something. I have never seen a good Schutzhund prospect shut down from corrections yet.

[Kavik]

Yes but not all dogs are good Schutzhund prospects and just because your GSD doesn't shut down doesn't mean someone's Whippet wouldn't shut down or even that my Kelpie wouldn't shut down.

[MonElite]

Yes but not all dogs are good Schutzhund prospects ;) and just because your GSD doesn't shut down doesn't mean someone's Whippet wouldn't shut down or even that my Kelpie wouldn't shut down.

This is why its important to know what level of correction can be issued to the dog.

Last night I watched a rabbit chasing dog that jumped 1.5 meter fence and gone through barb wire in a frenzy to catch it, in my opinion that dog could not be taught to recall off the prey with pure positive methods. Mind you the dog has a pretty good recall as is. But maybe Im talking about re-training here not training from puppy, again.

Hey I even watched a human in a frenzy of rabbit catching (barefoot) I wonder if he could be recalled He eventually came back by himslef but later then the dog

Edited January 12, 2010 by MonElite

[Diablo]

Yes but not all dogs are good Schutzhund prospects and just because your GSD doesn't shut down doesn't mean someone's Whippet wouldn't shut down or even that my Kelpie wouldn't shut down.

Exactly Kavik..........that's why each particular dog needs to be assessed by someone competent enough to determine the "best" methods to train that individual dog. Neither compulsion or positive reinforcement can be used successfully across the board in all cases and depends on the dog and the end result that needs to be achieved.

Edited January 13, 2010 by Diablo

[corvus]

Perhaps it depends on the original level of drive in the dog? Have you thought of that?

...And I can tell you that he has been corrected many times for missbehavior.

One has to wonder just how punishing "corrections" are to these dogs. Especially if it happens many times. Maybe it's a NRM.

Although I do agree that what is punishing to one dog is not so much to another. And if the dog has become over-aroused where he just wants so badly to do something that everything around him fades to grey, you'll have a hard time being noticed without something punishing or doing a lot of conditioning and management for a while. Prevention is better than cure, though. If you pace your training sensibly, you can build drive and control at the same time. I'm discovering this for myself.

[bedazzledx2]

This article is interesting http://www.clickertraining.com/node/2451

Perhaps it depends on the original level of drive in the dog? Have you thought of that?

...And I can tell you that he has been corrected many times for missbehavior.

One has to wonder just how punishing "corrections" are to these dogs. Especially if it happens many times. Maybe it's a NRM.

Although I do agree that what is punishing to one dog is not so much to another. And if the dog has become over-aroused where he just wants so badly to do something that everything around him fades to grey, you'll have a hard time being noticed without something punishing or doing a lot of conditioning and management for a while. Prevention is better than cure, though. If you pace your training sensibly, you can build drive and control at the same time. I'm discovering this for myself.

[corvus]

I thought that article was very valuable, bedazzled. It articulates what I was suggesting more clearly:

We need to separate the “thing”—the cookie or the stick or the click or the cold water or whatever—from its outcome. What defines its function is not what it looks like to common sense, but how it changes the behavior. If that stick doesn't slow the dog down, then no matter how scary it looks to us, it's not functioning as a punisher.

[Staranais]

i dont believe you can reliably train a dog under high distraction without compulsion as if there desire to chase the cat across the road is higher than the want to eat that food or treat in your hand its gunna run and attack the car ??

I used to think that too, now I'm not so sure.

I certainly don't think you can teach a recall off a running cat by bribing a dog with food or with treats. A high prey drive dog will view the food as completely irrelevant in that circumstance.

However, with some dogs, correcting the dog to within an inch of its life is not necessarily going to stop it chasing either.

Positive or corrective, the solution to that type of prolem must be done intelligently. In my experience, the correction must be timed appropriately (before the dog is fully spun up in drive). IMO the timing of the correction is in many ways more important than the magnitude or type of correction. With my old boy, I had far more success in correcting him and redirecting him gently as soon as he started to think about doing something inappropriately predatory, compared to correcting him harshly when he was already adrenalised and focused on his prey.

However, positive solutions can be really helpful too, either as stand alone solutions, or to augment the corrections. You can work on desensitising the dog to the cat, you can work on prey drive games so the dog understands that obeying you leads to drive satisfaction, etc. It's not just about waiting until the dog is taking off after the cat, and then waving a cookie around.

Also, in my experience, many people whose dog won't recall off a running cat, also have dogs that won't recall in many other circumstances. They just don't have a good recall, and jumping straight to correcting the dog for a hugely distracting recall isn't the answer. The answer is to go back and train all those intermediate steps that they missed, and then work on the huge distractions like running cats.

But hey, what would I know?

[corvus]

Also, in my experience, many people whose dog won't recall off a running cat, also have dogs that won't recall in many other circumstances. They just don't have a good recall, and jumping straight to correcting the dog for a hugely distracting recall isn't the answer. The answer is to go back and train all those intermediate steps that they missed, and then work on the huge distractions like running cats.

But hey, what would I know?

Plenty, if that last paragraph is anything to go by.

It always surprises me how my dogs become more and more capable of hearing me in highly distracting situations the more I practice. It's like they become attuned to cues they have been heavily rewarded for taking notice of.

[Diablo]

i dont believe you can reliably train a dog under high distraction without compulsion as if there desire to chase the cat across the road is higher than the want to eat that food or treat in your hand its gunna run and attack the car ??

I used to think that too, now I'm not so sure.

I certainly don't think you can teach a recall off a running cat by bribing a dog with food or with treats. A high prey drive dog will view the food as completely irrelevant in that circumstance.

However, with some dogs, correcting the dog to within an inch of its life is not necessarily going to stop it chasing either.

Positive or corrective, the solution to that type of prolem must be done intelligently. In my experience, the correction must be timed appropriately (before the dog is fully spun up in drive). IMO the timing of the correction is in many ways more important than the magnitude or type of correction. With my old boy, I had far more success in correcting him and redirecting him gently as soon as he started to think about doing something inappropriately predatory, compared to correcting him harshly when he was already adrenalised and focused on his prey.

However, positive solutions can be really helpful too, either as stand alone solutions, or to augment the corrections. You can work on desensitising the dog to the cat, you can work on prey drive games so the dog understands that obeying you leads to drive satisfaction, etc. It's not just about waiting until the dog is taking off after the cat, and then waving a cookie around.

Also, in my experience, many people whose dog won't recall off a running cat, also have dogs that won't recall in many other circumstances. They just don't have a good recall, and jumping straight to correcting the dog for a hugely distracting recall isn't the answer. The answer is to go back and train all those intermediate steps that they missed, and then work on the huge distractions like running cats.

But hey, what would I know?

Personally Staranais, I think you have a very good understanding of general dog training and in fact, you have provided an excellent post

Correction and positive re-direction with crucial timing of events was the basis of my GSD's obedience training with predatory behaviour which was not attainable easily without some compulsion. It may be possible to achieve results with "expert" positive reinforcement but on the average of what I have seen from dogs trained without compulsion at our Schutzhund club, the unreliability rate is high in those dog types. For the average person, it appears more successful to train with a combination of both methods. Too many people I believe are clouded with the thought that training without compulsion is always best for the dog, but depending upon what you need to achieve, complusion in some behaviour in some dogs should be be regarded as just another training tool in the box.

With my dog especially, positive conditioning proofed him against predatory behaviour with certain distractions, but corrections proofed him against every distraction with far better reliability. It's a case of having an open mind and access to all methods of training and using what is most appropriate. Neither purely compulsive nor purely positive is the best method to successfully train every dog.

[bedazzledx2]

I guess it depends on what you are training for. I am the first to put my hand up and say I know nothing about training Schutzhund or protection work. I do know a fair bit about training competative Obedience and Agility. For those sports I do beleive it is possible to train a very high level of reliability and style without using compulsion. For the average dog and owner it does carry over into real life. My obedience trained dogs will call off a cat or other animal, they will sit, down, stay on command out of the obedience ring context but that is probably due to training and positive proofing. Staranais said it all in the last paragraph....completely agree!

i dont believe you can reliably train a dog under high distraction without compulsion as if there desire to chase the cat across the road is higher than the want to eat that food or treat in your hand its gunna run and attack the car ??

I used to think that too, now I'm not so sure.

I certainly don't think you can teach a recall off a running cat by bribing a dog with food or with treats. A high prey drive dog will view the food as completely irrelevant in that circumstance.

However, with some dogs, correcting the dog to within an inch of its life is not necessarily going to stop it chasing either.

Positive or corrective, the solution to that type of prolem must be done intelligently. In my experience, the correction must be timed appropriately (before the dog is fully spun up in drive). IMO the timing of the correction is in many ways more important than the magnitude or type of correction. With my old boy, I had far more success in correcting him and redirecting him gently as soon as he started to think about doing something inappropriately predatory, compared to correcting him harshly when he was already adrenalised and focused on his prey.

However, positive solutions can be really helpful too, either as stand alone solutions, or to augment the corrections. You can work on desensitising the dog to the cat, you can work on prey drive games so the dog understands that obeying you leads to drive satisfaction, etc. It's not just about waiting until the dog is taking off after the cat, and then waving a cookie around.

Also, in my experience, many people whose dog won't recall off a running cat, also have dogs that won't recall in many other circumstances. They just don't have a good recall, and jumping straight to correcting the dog for a hugely distracting recall isn't the answer. The answer is to go back and train all those intermediate steps that they missed, and then work on the huge distractions like running cats.

But hey, what would I know?

Personally Staranais, I think you have a very good understanding of general dog training and in fact, you have provided an excellent post

Correction and positive re-direction with crucial timing of events was the basis of my GSD's obedience training with predatory behaviour which was not attainable easily without some compulsion. It may be possible to achieve results with "expert" positive reinforcement but on the average of what I have seen from dogs trained without compulsion at our Schutzhund club, the unreliability rate is high in those dog types. For the average person, it appears more successful to train with a combination of both methods. Too many people I believe are clouded with the thought that training without compulsion is always best for the dog, but depending upon what you need to achieve, complusion in some behaviour in some dogs should be be regarded as just another training tool in the box.

With my dog especially, positive conditioning proofed him against predatory behaviour with certain distractions, but corrections proofed him against every distraction with far better reliability. It's a case of having an open mind and access to all methods of training and using what is most appropriate. Neither purely compulsive nor purely positive is the best method to successfully train every dog.

[corvus]

I guess it depends on what you are training for. I am the first to put my hand up and say I know nothing about training Schutzhund or protection work. I do know a fair bit about training competative Obedience and Agility. For those sports I do beleive it is possible to train a very high level of reliability and style without using compulsion. For the average dog and owner it does carry over into real life. My obedience trained dogs will call off a cat or other animal, they will sit, down, stay on command out of the obedience ring context but that is probably due to training and positive proofing. Staranais said it all in the last paragraph....completely agree!

I'm the first to put up my hand and say I know nothing about Shutzhund or any kind of competetive dog sports. But I have two dogs that go pretty much everywhere with us and spend a lot of time off leash in unfenced areas full of exciting stuff. They've got to at the least recall reliably. And they can, even off a chase. They can also do anything else we have taught them in any situation we ask them to do those things. We have done crap all proofing with those dogs, and training has been haphazard, yet they are more reliable than I ever set out to make them. It doesn't matter where we are or what is happening around them, if we ask for their attention we get it. I never set out to achieve that, but it happened on its own. The correction thing didn't work out so well with our last dog. You could tell what she'd been taught with corrections and what she'd been taught with rewards and what she'd been taught with classical conditioning by how reliable they were. The things she'd learnt with corrections were unreliable, frankly. I didn't do them very well, but I did them the way I was taught by a trainer. If a trainer can get it so wrong, a layman certainly can.