E Collar Question

[Cosmolo]

No Erny- i'd been taught to use the e collar as a high level, on off positive punisher and for that reason i rarely ever used e collars until doing Bart's workshop.

Erny i have used the method with dogs that don't train in drive, don't have high prey drive and/ or are fearful/ timid.

Erny "But if you're only guiding the dog and if the dog is not perceiving the stim as a correction, why teach it that non-compliance might be worthwhile trying? That's the part I don't 'get' in the method that you describe in your use."

Sorry- i find this really confusing When the dog is learning, as the stim is applied, the dog is guided- i think there is limited opportunity for them to fail at this point. At some point, yes this would change and the dog may make a mistake. Just as with NR, if they don't comply straight way, the stim stays on until they do- i am struggling to see the difference in this respect. Doesn't the dog still make a mistake by not complying immediately? At some point you would have to stop guiding?

I think your both right in that the stim is a 'correction'- to be honest i have never called it a correction though because the way i was taught previously (high level stim, on and off) is what i associated as a correction. But i do understand that technically, no matter how low the stim that if it occurs after the command and changes behaviourin the way that it does it must be one.

I do appreciate your replies and discussion on this

[Erny]

Sorry- i find this really confusing When the dog is learning, as the stim is applied, the dog is guided- i think there is limited opportunity for them to fail at this point.

Ok .... so now you have me confused .

I thought (we'll call it "you're") method was this :

Dog commanded known command.

Dog doesn't comply (failure).

Dog stimmed (low level stim).

Dog complies (guided or not).

So in the above instance, which I think is what you described earlier in this thread, I'm still seeing where the dog is able to learn to fail. And if "failing" means he got an extra few seconds to sniff the smelly patch on the ground, or whatever, then this is reinforced. The stim comes on - low level, so it's not too much of a worry to him and worth it for that extra few seconds of 'sniff' that he gained, and he's then guided (you mean by leash or hands on?) into the position that he should have been in. I don't recognise the "classically conditioned" response (ie command compliance) that you can get through the "no failure" method R- produces.

OR

by what you've just said in the above quote, do you mean that (in the learning stages) you apply the stim and then guide the dog to doing what it should do and then the stim comes off?

If so, that sounds to me like you are not using the low-stim/P+ method you describe, but rather, the low-stim/R- method that I use.

Edited May 18, 2009 by Erny

[superminty]

K9: With the stim being switched on first, then you have the diog attention, then you give a command & help the dog shape the movment & then the stim turns off. This means the dog completing the correct movement is marked at the time of success, just like in clicker training.

I know of some trainers whom have take the next step & press the collar stim, & at the removal supply a food treat, to replicxate their prevous clicker training.

Aha, that "marking" of the behaviour was what I was trying to get at but couldn't articulate I am a clicker fan and use it extensively. I think the timing I've developed through using a clicker has really helped with the e-collar stuff - it's about knowing what you want and reinforcing the precise moment that occurs. Makes sense to the dog too from my perspective. He really "gets" the work we do with the e-collar, just like he "gets" clicker work.

The second part of the quote above is what I do (kind of) - I also reward with food/tug (drive currently being developed) when teaching exercises as I found it speeded up the learning process. While training I still use food rewards from time to time but tend to use the release from pressure (stim off) as the main reward. When I want to sharpen something up, out comes the food again, when I've got the response I want, I go back to relying on the pressure release.

[RealityBites]

This is completely off track, but something I have been thinking about after reading this thread. I have a very boisterous 3 year old dobe, that at present CANNOT be an inside dog due to his high energy levels, and the fact that he destroys anything he gets his paws on BUT I am planning on using the e collar as a correctional tool when he is inside to teach him some manners, as I would like him to be an inside dog. I have been using K-9's method of stim before command when using the e collar in training, so am wondering if there will be any adverse effects of using the e collar as a punishment, maybe on a higher stim than I would use in training? Or do I still use it on a low stim, stim him BEFORE his mouth wraps around the teddy bear/cushion/clothes and then release when he leaves the item?? Sorry if I am confusing things

[Erny]

I have never used the e-collar on my boy as a P+ and I'd be inclined to use more 'hands on' training for your situation. This is just my own opinion, RB, and using my own boy as personal "in-house" experience, as I would not liked to have risked a possible pairing of correction with him taking things into his mouth due to me wanting to build his drive and enthusiasm for the tug. But then I have a RR and they are wooses, and not necessarily dogs with obsessive drive and inclinations that we seek for drive training as other dog breeds might have.

[superminty]

If you use the stim on a barely perceptible setting, and use rewards as well as using the e-collar, then the stim should be a positive experience for the dog. The stim becomes a conditioned signal that the dog has just gained the opportunity to comply with the coming command and earn a reward. If anything, the dog should become excited and focused when it feels the stim start, as the stim simply means: "heads up! here's a chance to earn something cool!"

So IMO in that case it isn't at all unfair or unkind to stim before the command, as the stim isn't a negative experience for the dog, in fact is it positive, as it signals the availability of earning reinforcement.

Just my 2 cents, from what I understand.

Agree with this. Although I should point out that this is the first dog I've trained using an e-collar.

The application of the stim signals to the dog that "something" will be reinforced - it has to be, coz the stim has to come off at some stage. The release of pressure is the reinforcer, and it's a powerful one - and when you need the reliabilty that I do, having a dog with "issues", a low desire to please and with high prey drive that has proved near impossible to channel, that's pretty darn important.

I just find it hard to comprehend that there is such a vast difference between applying a stim on a low level 1/2- 1 second before the command vs 1-3 seconds after the command. Applying the stim after the command does not have to = a correction IMO.

Been thinking about this a bit - now I remember why I stay away from DOL, so distracting

K9 said something similar earlier (somewhere): Applying the stim before the command gives the opportunity to mark the precise behaviour (by releasing the stim). This adds clarity to the dog's learning. The dog now knows *exactly* what turned the stim off (if your timing is spot on).

I have two perceived issues with using the stim after the command while teaching an exercise:

1. Applying stim after a command has been given (and not complied with, otherwise the stim would not occur) gives the dog an opportunity to fail. In fact, the dog has to fail to get the stim. I prefer to let my dogs learn at a level where they win the majority of the time in teaching phase and to some extent training phase. Once we get to the upper level of training/moving into proofing, I'm happy for the dog to fail, however at this level I imagine/expect failure would not happen very often. Perhaps a "perfect world" scenario, but hey, that's my aim.

2. A lot can happen in 1-3 seconds. I have erred in shaping with a clicker by being out with my timing by 1/2 a second - although I have trained a very nice head-down-with-your-nose-touching-a-precise-part-of-the-couch...oops

What happens in the time between the command, the non-compliance, the application of the stim and the eventual compliance can vary greatly - and I believe this could affect clarity and precision of communication, potentially confuse the dog and increase the time it takes to get the behaviour you want.

Really enjoying this discussion by the way, thanks for kick-starting my brain, it's been on holidays lately

[Cosmolo]

I agree that the release of pressure is a reinforcer- i think this is the case regardless of whether the stim comes before or after the command though.

Erny- by what you've just said in the above quote, do you mean that (in the learning stages) you apply the stim and then guide the dog to doing what it should do and then the stim comes off?

Yes- i just give a command, then apply the stim and guide simultaneously. There might be a very short time where the dog can make a choice but hey would then be guided to teh right behaviour. I think some of these things are really hard to explain via ritten word- it'd be great to have an e collar workshop with lots of different trainers who use the collar differently and getting to see exactly what happens!

Edited May 18, 2009 by Cosmolo

[Erny]

...

Edited May 18, 2009 by Erny

[Cosmolo]

Sorry Erny- that was me quoting you

[Quickasyoucan]

As I have said, both work, no disputing that. But I use the collar as a teaching tool not a replacement for a check chain,I also use them a communication tool, the stim cues the dog that a reward is available, the dog knows I give commands in these circumstances that helps the dog achieve the reward & switch off the stim.

Just bumping this up because I was thinking of something on the way to work, it might be simplistic, but if the stim becomes a cue that something is about to happen, say if you were using it for recall, you would stim, command, comply stim off, is there a danger that the stim then becomes so paired with the command so that if you need to use the recall command and the dog doesn't have the collar on, because there is no "heads up" and the sequence is not followed compliance with the command alone may not be 100%?

Also what happens using this method if you want to wean the dog of the collar?

[huski]

I have a very silly question to ask about e-collars.

It is generally accepted that low stim training is the way to go, so why do some collars have such high levels? i.e. the one with 100+ levels of stim... does anyone actually need to go as high as level 100? Why do they still make them to stim so high? I know - stupid question but I'm curious

[Steve K9Pro]

Q: Just bumping this up because I was thinking of something on the way to work, it might be simplistic, but if the stim becomes a cue that something is about to happen, say if you were using it for recall, you would stim, command, comply stim off, is there a danger that the stim then becomes so paired with the command so that if you need to use the recall command and the dog doesn't have the collar on, because there is no "heads up" and the sequence is not followed compliance with the command alone may not be 100%?

K9: in all cases but a few the dog is weaned off the collar, meaning no stim will be needed once the dog is solid on each command under distraction, in my experience the dogs not weaned off comply anyway.

Also what happens using this method if you want to wean the dog of the collar?

K9: The weaning protocol removes the stim being needed or the collar being needed.

[Quickasyoucan]

Q: Just bumping this up because I was thinking of something on the way to work, it might be simplistic, but if the stim becomes a cue that something is about to happen, say if you were using it for recall, you would stim, command, comply stim off, is there a danger that the stim then becomes so paired with the command so that if you need to use the recall command and the dog doesn't have the collar on, because there is no "heads up" and the sequence is not followed compliance with the command alone may not be 100%?

K9: in all cases but a few the dog is weaned off the collar, meaning no stim will be needed once the dog is solid on each command under distraction, in my experience the dogs not weaned off comply anyway.

Also what happens using this method if you want to wean the dog of the collar?

K9: The weaning protocol removes the stim being needed or the collar being needed.

Thanks

[Steve K9Pro]

I have a very silly question to ask about e-collars.

It is generally accepted that low stim training is the way to go, so why do some collars have such high levels? i.e. the one with 100+ levels of stim... does anyone actually need to go as high as level 100? Why do they still make them to stim so high? I know - stupid question but I'm curious

K9: Low stim training is accepted but not practised by many, many still use the higher end of the scales. Some things such as food refusal & snake proofing are still trained by many on the highest level of stim they can find.

Another reason is that all dogs feel the stim at different levels, so a variabnce is needed. I use the dogtra because the models we recommend have 127 levels. I am not interested in higher levels, just the fact that each level has less than 1% change of stim level at the collar, meaning I can find the perfect level of stim for every dog.

level one is so low, never met an animal that coulkd eve feel less than about 5.

[huski]

I have a very silly question to ask about e-collars.

It is generally accepted that low stim training is the way to go, so why do some collars have such high levels? i.e. the one with 100+ levels of stim... does anyone actually need to go as high as level 100? Why do they still make them to stim so high? I know - stupid question but I'm curious

K9: Low stim training is accepted but not practised by many, many still use the higher end of the scales. Some things such as food refusal & snake proofing are still trained by many on the highest level of stim they can find.

Another reason is that all dogs feel the stim at different levels, so a variabnce is needed. I use the dogtra because the models we recommend have 127 levels. I am not interested in higher levels, just the fact that each level has less than 1% change of stim level at the collar, meaning I can find the perfect level of stim for every dog.

level one is so low, never met an animal that coulkd eve feel less than about 5.

Thanks Steve, that makes total sense

[Erny]

I have a very silly question to ask about e-collars.

It is generally accepted that low stim training is the way to go, so why do some collars have such high levels? i.e. the one with 100+ levels of stim... does anyone actually need to go as high as level 100? Why do they still make them to stim so high? I know - stupid question but I'm curious

Huski - Just to add to K9 Force's reply, a collar of 127 levels would be equivalent to a level of about (eg) 12.7 on a collar with only 15 levels. But as K9 has said, the difference is that the 127 levels gives you increments of "1" between every level that a 15 stim or 10 stim collar would have.

What you asked is not a silly question.

[huski]

I have a very silly question to ask about e-collars.

It is generally accepted that low stim training is the way to go, so why do some collars have such high levels? i.e. the one with 100+ levels of stim... does anyone actually need to go as high as level 100? Why do they still make them to stim so high? I know - stupid question but I'm curious

Huski - Just to add to K9 Force's reply, a collar of 127 levels would be equivalent to a level of about (eg) 12.7 on a collar with only 15 levels. But as K9 has said, the difference is that the 127 levels gives you increments of "1" between every level that a 15 stim or 10 stim collar would have.

What you asked is not a silly question.

Thanks Erny - I think it was the 127 level collar that I let Midol stim me with, I do remember it had to get to level 15 or so before I even felt the slightest sensation.

[Erny]

Thanks Erny - I think it was the 127 level collar that I let Midol stim me with, I do remember it had to get to level 15 or so before I even felt the slightest sensation.

And if that had been say a '10 level' collar, you probably wouldn't have felt anything on level 1, but would have jumped to level 2 to feel it and it would have been a stronger sensation than was necessary, if that makes sense.

[Steve K9Pro]

K9: another way to look at it is, a four level collar jumps 25% every time you go up a level, a ten level would be 10% & a 100 level would be 1%.

Think of the lowest level on each collar as the same & the highest level on each the same, important difference is the amount of steps.

Some dogs on say an Innotek 9 level collar cant feel one but level two is too high for them, using my just perceivable level of stim, so that collar has to be modified or replaced.

[huski]

Thanks Erny - I think it was the 127 level collar that I let Midol stim me with, I do remember it had to get to level 15 or so before I even felt the slightest sensation.

And if that had been say a '10 level' collar, you probably wouldn't have felt anything on level 1, but would have jumped to level 2 to feel it and it would have been a stronger sensation than was necessary, if that makes sense.

Makes total sense. I figured the stim levels would have to go up in bigger incriments if you were working with a collar that had less levels, makes sense when you put it like that!

ETA: Ah gotcha Steve - so that's why the collars with more levels tend to be more popular with trainers who do low stim training? because more levels gives you more opportunity to 'find' the right working level for each dog.

Edited May 20, 2009 by huski