Pinch Collars

[malsrock]

'bedazzledx2' date='15th Sep 2010 - 02:51 PM' post='4814411']

The reliability is there if you train it right....can be done with a clicker.

Taking into account that my training background is aversive methods, so I am relatively new to positive reinforcement methods and I have never used a clicker, my 12 months old Malinois has terrible sits without a reward. His sits are perfect with a toy or food reward to follow and without, I get one maybe two sits, in other words unless he sees what's on offer, he won't work for free

Disobeying a known command like he does, I would have in the past administered a leash correction, I can train the behaviour I need, but without using an aversive which he has never experienced to date, where do I go from here, how do I fix it without an aversive to get a reliable sit without a reward on offer:confused:

Fiona :D

Edited September 15, 2010 by malsrock

[Kavik]

It looks like you haven't faded out your reinforcer well. You can start by having the rewards somwehere the dog can't see (eg pocket) and then reward, and move to having the rewards off your person in a different location (up high and you go get it, or send to a reward somewhere else). You want to make it so that he doesn't know where the reward will come from so he doesn't have to see it and you don't have to be holding it for him to work.

[Erny]

'bedazzledx2' date='15th Sep 2010 - 02:51 PM' post='4814411']

The reliability is there if you train it right....can be done with a clicker.

Taking into account that my training background is aversive methods, so I am relatively new to positive reinforcement methods and I have never used a clicker, my 12 months old Malinois has terrible sits without a reward. His sits are perfect with a toy or food reward to follow and without, I get one maybe two sits, in other words unless he sees what's on offer, he won't work for free

Disobeying a known command like he does, I would have in the past administered a leash correction, I can train the behaviour I need, but without using an aversive which he has never experienced to date, where do I go from here, how do I fix it without an aversive to get a reliable sit without a reward on offer:confused:

Fiona :D

Hi Malsrock - you directed your post to Bedazzledx2, I think .... so I hope you don't mind me throwing in my tuppence .

One of the ways I would work to "fix" this problem is that I would have a tug toy on me, but not in sight. I would ask for sit and if I did not get the response, I might correct the dog to the sit. But once corrected to sit, I would then release into the tug reward. Or it might be that I got sit, but it was crappy, but I'd still reward with (then) hidden tug. What I'd be teaching here is not the sit, but that the possibility of reward with the tug is still there, even if the tug is not seen to be on offer. Once your dog realises this, I think you would find the 'gambler' in him would start having him work better for you and you can then begin to use this 'gambling addiction' you've created, to produce better sits.

I tend to randomise with my boy with regards to whether his tug is hidden or not.

ETA: Kavik posted whilst I was typing .

Edited September 15, 2010 by Erny

[Steve K9Pro]

K9: A much better way to look at this is, if you dont start with the reinforcer as the trigger, you never need to fade it out...

[malsrock]

'bedazzledx2' date='15th Sep 2010 - 03:58 PM' post='4814431']

How about teaching an alternative behaviour? ie..see a distraction, focus on me. Lots of ways to teach this before the need for a correction should ever come into the picture.

I don't see the purpose in saving a dog the experience of learning there are consequences to an action good and bad. Behaviour reinforced with positives is good, but behavior reinforced with positives and negatives IMHO is better and amounts to increased reliability.

Fiona

I assume we're talking about behaviour that is rewarded when performed on cue and punished when not performed on cue rather than a single behaviour that is reinforced with positives and negatives? What's the difference in reliability between rewarding a behaviour when it is cued and rewarding a behaviour when it is cued... and then punishing whatever behaviour the dog chooses instead of the original one when it is cued?

Ok, I will give an example:

I teach my dogs a "stop" command. We walk along on leash and the dog is out front for instance and I say "stop", the dog stops I catch up in a footstep and reward which is working well in low distraction atmospheres and the dog knows the stop command well. We up the distractions where the dog moves forward on the leash focusing on another dog. We say "stop" and the dog ignors the command, I will then correct him and the dog stops, catch up and no reward. The dog from my experience learns that unless he obey's the stop command a correction follows and has the choice of which action he wants to take. Stop and be rewarded, don't stop, get a correction and no reward. Im my way of thinking, there is a "double" reinforcer in place, a correction and no reward for disobeying, no correction and reward for obeying.

Does it work??? ABSOLUTELY, my 8 year old boy's "stops" are bullet proof off leash even with my OH calling him at the other end although it was 7 odd years ago I trained him to do this

Fiona

Ok an interesting alternative to the situation you just described . . .

walk on leash and dog is out in front, I stop, give dog no command, the dog stops and looks at me, I click/treat, I stop and the dog comes back to me and looks at me, click/treat. Up the distractions to where the dog is focused on another dog. I stop, the dog stops and looks at me, click/treat, I stop and the dog comes back to me and looks at me, click/treat. The dog has the choice of which action to take. :D Same result

Kind of the same result, but will your's reliably exhibit the behaviour off leash without a clicker and treat???

Fiona

[huski]

Malsrock, it's easy to train the dog to respond to a trigger word or phase instead of using the sight of food/toys as the trigger. Think of dogs who flip out when they hear you call out 'dinner!' or jingle the car keys or like in my house... put my runners on

Edited September 15, 2010 by huski

[malsrock]

'bedazzledx2' date='15th Sep 2010 - 02:51 PM' post='4814411']

The reliability is there if you train it right....can be done with a clicker.

Taking into account that my training background is aversive methods, so I am relatively new to positive reinforcement methods and I have never used a clicker, my 12 months old Malinois has terrible sits without a reward. His sits are perfect with a toy or food reward to follow and without, I get one maybe two sits, in other words unless he sees what's on offer, he won't work for free

Disobeying a known command like he does, I would have in the past administered a leash correction, I can train the behaviour I need, but without using an aversive which he has never experienced to date, where do I go from here, how do I fix it without an aversive to get a reliable sit without a reward on offer:confused:

Fiona

Hi Malsrock - you directed your post to Bedazzledx2, I think .... so I hope you don't mind me throwing in my tuppence .

One of the ways I would work to "fix" this problem is that I would have a tug toy on me, but not in sight. I would ask for sit and if I did not get the response, I might correct the dog to the sit. But once corrected to sit, I would then release into the tug reward. Or it might be that I got sit, but it was crappy, but I'd still reward with (then) hidden tug. What I'd be teaching here is not the sit, but that the possibility of reward with the tug is still there, even if the tug is not seen to be on offer. Once your dog realises this, I think you would find the 'gambler' in him would start having him work better for you and you can then begin to use this 'gambling addiction' you've created, to produce better sits.

I tend to randomise with my boy with regards to whether his tug is hidden or not.

ETA: Kavik posted whilst I was typing .

Thanks to both Kavik and Erny, much appreciated

I will try those suggestions..........I can get one good sit without a reward in sight and work on that. I have done a lot of command and release for reward work with him which that part is going well with food and toys but it's when trying to wean him off that system is where I am struggling a bit. I did get four sits yesterday with hand in pocket..........and then without rewarding, he packed it in :D I will give the suggestions a try, thanks!!.

Fiona

[Kavik]

'bedazzledx2' date='15th Sep 2010 - 03:58 PM' post='4814431']

How about teaching an alternative behaviour? ie..see a distraction, focus on me. Lots of ways to teach this before the need for a correction should ever come into the picture.

I don't see the purpose in saving a dog the experience of learning there are consequences to an action good and bad. Behaviour reinforced with positives is good, but behavior reinforced with positives and negatives IMHO is better and amounts to increased reliability.

Fiona

I assume we're talking about behaviour that is rewarded when performed on cue and punished when not performed on cue rather than a single behaviour that is reinforced with positives and negatives? What's the difference in reliability between rewarding a behaviour when it is cued and rewarding a behaviour when it is cued... and then punishing whatever behaviour the dog chooses instead of the original one when it is cued?

Ok, I will give an example:

I teach my dogs a "stop" command. We walk along on leash and the dog is out front for instance and I say "stop", the dog stops I catch up in a footstep and reward which is working well in low distraction atmospheres and the dog knows the stop command well. We up the distractions where the dog moves forward on the leash focusing on another dog. We say "stop" and the dog ignors the command, I will then correct him and the dog stops, catch up and no reward. The dog from my experience learns that unless he obey's the stop command a correction follows and has the choice of which action he wants to take. Stop and be rewarded, don't stop, get a correction and no reward. Im my way of thinking, there is a "double" reinforcer in place, a correction and no reward for disobeying, no correction and reward for obeying.

Does it work??? ABSOLUTELY, my 8 year old boy's "stops" are bullet proof off leash even with my OH calling him at the other end although it was 7 odd years ago I trained him to do this

Fiona

Ok an interesting alternative to the situation you just described . . .

walk on leash and dog is out in front, I stop, give dog no command, the dog stops and looks at me, I click/treat, I stop and the dog comes back to me and looks at me, click/treat. Up the distractions to where the dog is focused on another dog. I stop, the dog stops and looks at me, click/treat, I stop and the dog comes back to me and looks at me, click/treat. The dog has the choice of which action to take. :D Same result

Kind of the same result, but will your's reliably exhibit the behaviour off leash without a clicker and treat???

Fiona

I really only do this while on lead, so haven't tried off lead. And I am never without my clicker and treats when out walking the dogs - you never know when you will have the opportunity to train

I train primarily agility with a little bit of competition based obedience exercises (I compete in agility, but my obedience dog has allergies so we only train for fun in that)

[Erny]

I will try those suggestions..........I can get one good sit without a reward in sight and work on that. I have done a lot of command and release for reward work with him which that part is going well with food and toys but it's when trying to wean him off that system is where I am struggling a bit. I did get four sits yesterday with hand in pocket..........and then without rewarding, he packed it in I will give the suggestions a try, thanks!!.

Make the sessions SUPER short. Even with the first session just that one "good" sit .... out comes tug/play; win; and finish (whilst he still REALLY wants it). Push for two if you think you can achieve it. But finish on that. You can return to it later. "Lick of the ice-cream .... " . Let the dog's gambling addiction build up first, before you go for more. That's what I'd suggest, anyway. I feel a bit humbled given K9 Pro's presence in this thread. No doubt, he'll have a better suggestion :D.

Edited September 15, 2010 by Erny

[malsrock]

Malsrock, it's easy to train the dog to respond to a trigger word or phase instead of using the sight of food/toys as the trigger. Think of dogs who flip out when they hear you call out 'dinner!' or jingle the car keys or like in my house... put my runners on

We have those trigger words too :D My older GSD will go out into the garage and sit by the car with a rattle of the keys, I know what you mean.

Fiona

[malsrock]

I will try those suggestions..........I can get one good sit without a reward in sight and work on that. I have done a lot of command and release for reward work with him which that part is going well with food and toys but it's when trying to wean him off that system is where I am struggling a bit. I did get four sits yesterday with hand in pocket..........and then without rewarding, he packed it in I will give the suggestions a try, thanks!!.

Make the sessions SUPER short. Even with the first session just that one "good" sit .... out comes tug/play; win; and finish (whilst he still REALLY wants it). Push for two if you think you can achieve it. But finish on that. You can return to it later. "Lick of the ice-cream .... " . Let the dog's gambling addiction build up first, before you go for more. That's what I'd suggest, anyway. I feel a bit humbled given K9 Pro's presence in this thread. No doubt, he'll have a better suggestion :D.

Thanks Erny, I will try

[MelissaS]

I have used one - they're not illegal in Australia, but they are in Victoria (I'm not sure about other states - they're legal in NSW).

I rescued my dog at 2.5 years old. He was fearful of everything (he has a very sad back story from before I got him ) - I couldn't go near him. We worked on his fears but when we started being able to go for walks, he would react with extreme aggresion whenever we saw a dog. It got to the point where he would slink around corners, just in case there was a dog around there. On day he even went around a corner and tried to attack his own shadow because he was so on edge about other dogs.

I took him out to see 4 trainer, 3 of them purely positive trainers/behaviourists and every one of them said I shold put him to sleep because I'd never be able to cure him and he was a danger and a menace to other dogs. A few more trainers wouldn't even see us. Then I took him out to see K9 Pro, who assessed him and recommended we try a Prong collar (a pinch collar). Within 6 weeks (by now, zero was 4 years old and i'd been dealing with his problems for 18 months), Zero was able to be around other dogs. For the first time ever, instead of trying to deal with everything himself, if there was a problem, he would look to me. That's what we taught him with the prong collar. There were no harsh corrections, he was given the choice to either walk away with me or to pull on his collar. He chose to follow me. :D He has never felt the full force of the collar - he wags his tail and gets excited when he sees it because it means we're going out somewhere fun. I would rather have used it on him and had him get better sooner rather than go through all the purely positive stuff we did that did nothing.

Do i think that all dogs should use one? No, they should only be used under the supervision of a properly qualified behaviourist. I don't use mine anymore and would happily give it back to Steve but Zero's collar was never supposed to be used forever - it is a training tool, not a cure all. He acts the same on his flat collar now (and it's a ruthless leather one so I want him to show it off!) so I haven't used his prong in over a year.

This is him now (taken on australia day this year by terranik):

I'm not going to get into the politics of the collar but quite frankly, it saved Zero's life. I would never have put him to sleep but what that collar did in 6 weeks, no other training tool did in 18 months. Zero lived in constant fear and now he's one of the most happy dogs you'll ever meet. I can't thank Steve at k9 Pro enough for that!

Here's a video terranik made about Z and I - it makes me cry to watch it.

http://www.youtube.com/user/TerraNik#p/u/18/Y6F7eQyXJkY

Troll or not, I'm happy to tell people I've used one. I had a lot of s*** thrown at me in the past about it and I think the results speak for themselves. Attack away!

I watched this Video of you and Zero and was very very moved. So touching

Great work to the both of you and awesome that Steve at K9 Pro yet again helped another dog.

[malsrock]

K9: A much better way to look at this is, if you dont start with the reinforcer as the trigger, you never need to fade it out... :D

Hmmmm, What I used to do to teach the sit was physically (gently I might add) place him in the sit position Koehler style through repetition, with the command as the trigger. This time, I didn't it with a food lure above his head then let him eat it as the reward....it's got me thinking now

Fiona

[Steve K9Pro]

Malsrock, it's easy to train the dog to respond to a trigger word or phase instead of using the sight of food/toys as the trigger. Think of dogs who flip out when they hear you call out 'dinner!' or jingle the car keys or like in my house... put my runners on

We have those trigger words too :D My older GSD will go out into the garage and sit by the car with a rattle of the keys, I know what you mean.

Fiona

K9: This is actually called a concomitant cue, something the dog has built in to help him predict reward that we dont always have control of...

[Steve K9Pro]

MR: Hmmmm, What I used to do to teach the sit was physically (gently I might add) place him in the sit position Koehler style through repetition, with the command as the trigger. This time, I didn't it with a food lure above his head then let him eat it as the reward....it's got me thinking now

K9: How you teach him to sit isnt really that important as long as it isnt taught by force if you want the cue of sit to predict a drive reward. Once he knows sit you stimulate him with the toy and dont let him catch it, but stop moving it, even put it behind your back...

Watch what he offers you...

The sit will come out and when it does release and reward.

Take a look on pur Facebook page, in the discussions, loads of people training in drive in my distance package, you can see their videos there that can help.

[m-j]

I believe you can use whatever technique or tool you fancy with dog aggressive dogs, but you can never rush the natural process of desensitisation which occurs slowly and over time with regular exposure to other dogs. Everything else, whether it be clicker, prong or whatever are just tools and techniques to help us handlers handle our dogs whilst the process of desensitisation is taking place naturally and on it's own time.

I agree.

[huski]

Malsrock, it's easy to train the dog to respond to a trigger word or phase instead of using the sight of food/toys as the trigger. Think of dogs who flip out when they hear you call out 'dinner!' or jingle the car keys or like in my house... put my runners on

We have those trigger words too :D My older GSD will go out into the garage and sit by the car with a rattle of the keys, I know what you mean.

Fiona

K9: This is actually called a concomitant cue, something the dog has built in to help him predict reward that we dont always have control of...

Steve, when you say, a reward we don't always have control of, do you mean, it's something that we unknowingly/unintentionally did to teach the dog a reward they value might be coming? i.e. when a dog gets excited by rattling keys because they think it means a trip in the car, even though they don't necessarily get to go in the car every time they hear the keys rattle?

Edited September 15, 2010 by huski

[Cosmolo]

What Aidan said- thats why i prefer that method to teaching traditional focus.

[Skye GSD]

Geez - I am really having a hard time taking in all the food lure training. I can think of nothing worse than having to continually have food on you in order to ensure your dog follows your directions. After suffering thru structured classes (that insisted on food and toy training only) with a high drive/dominant pup that was not interested in either rewards in a class situation, I have returned to the choke chain method with probably 95% accuracy. We are still a work in progress but what we have achieved in 3 weeks I struggled to achieve in 9 months using purely positive methods. Only wish we had K9 training in Perth.

Shelle - your video reduced me to tears - what a fantastic result

[Steve K9Pro]

H: Steve, when you say, a reward we don't always have control of, do you mean, it's something that we unknowingly/unintentionally did to teach the dog a reward they value might be coming? i.e. when a dog gets excited by rattling keys because they think it means a trip in the car, even though they don't necessarily get to go in the car every time they hear the keys rattle?

K9: Yes, sometimes it isnt something we "do" but anything that can help the dog predict reward. Like a treat pouch, putting on a training jacket, sneakers, grabbing a leash etc.

These are all concomitant cues the dog has valued as a trigger when we may not even know they existed. They can be a problem in training as we dont know the dog is counting on them to predict reward, so when they arent in play, we may not have the drive in the dog.

Like sometimes, something extra may be added that compromises the trigger as far as the dog is concerned, not just distractions either, even changes in handlers persona can have a great effect....