E Collar Question

[Cosmolo]

Could someone explain the advantages (if any) of applying low level stimulation before a command is given as opposed to after the command is given and the dog has not responded? Thanks!

Edited May 14, 2009 by Cosmolo

[Erny]

In negative reinforcement the dog learns that it has the power to turn the stim off. It "empowers" them and I tend to find increases confidence.

Positive punishment promotes an avoidance behaviour, at least in the early stages of e-collar training.

IMO :cool:

I know you use it the second way (P+) Cosmolo .... have you ever tried it using the R- method? And if you did, did you notice a difference?

Edited May 14, 2009 by Erny

[Cosmolo]

But doesn't the dog still have the power to turn the stim off after a command has been given? Still using the RT at low levels of course. Whats the reason not to give the dog the opportunity to respond to the command first?

I understand that some people want the stim to become the cue itself but surely this limits the number of training exercises you can use the RT for?

I use the e collar at low levels, after a command has been given and while i sometimes refer to it as negative reinforcement, this is not technically correct. But i still get significant increases in confidence- i feel like it differs from positive punishment but i suppose technically it is? :cool:

[Cosmolo]

No i haven't Erny. I have a bit of a fundamental problem with not giving the dog the opportunity to respond to the command first- particularly because i am most often dealing with known commands rather than brand new exercises. Watching Bart Bellon at a workshop a few years ago showed us what we do now but i am curious about the differences in responses and learning that may occur using it in a different way.

[Erny]

No i haven't Erny. I have a bit of a fundamental problem with not giving the dog the opportunity to respond to the command first- particularly because i am most often dealing with known commands rather than brand new exercises.

I tend to use the "known commands" to teach the dog about the stim. I find it is easier for the dog as it makes its learning about the stim easier and faster. And I generally like to teach the dog a new skill by conventional methods (eg. positive reward) before I apply any aversive regardless by the method applied.

Watching Bart Bellon at a workshop a few years ago showed us what we do now but i am curious about the differences in responses and learning that may occur using it in a different way.

I would have liked to have gone to his workshop but I was unable to at the time. So he only uses the e-collar as a P+ ?

Edited May 14, 2009 by Erny

[Cosmolo]

Yes he does- but low level stimulation rather than a correction as such. Only once the dog clearly understands the exercises and the stimulation and response does he then use a correction as the final step.

I am the same Erny with regards to your other comment (sorry- i don't know how to quote :cool: ) but i think thats why i don't understand the difference or advantage to applying stim beforethe command when everything else is the same?

ETA i keep missing workshops and seminars lately but we were just lucky at the time to be able to go to that particular one!

Edited May 14, 2009 by Cosmolo

[Erny]

Yes he does- but low level stimulation rather than a correction as such. Only once the dog clearly understands the exercises and the stimulation and response does he then use a correction as the final step.

I am the same Erny with regards to your other comment (sorry- i don't know how to quote :cool: ) but i think thats why i don't understand the difference or advantage to applying stim beforethe command when everything else is the same?

I see the difference as this :

Negative Reinforcement method :

Stim. Command. Dog thinks : "I know what to do to turn the stim off" (positive attitude). Dog complies with command.

Positive Punishment method :

Command. Dog doesn't comply. Stim. Dog thinks : "I better not disobey" (negative attitude). Dog complies with command.

In the R- example, the stim is not reliant on the dog failing and so the dog doesn't fail and the reliability is stronger as "failing" isn't learnt.

In the P+ example, the stim is reliant on the dog failing. Reliability is not as strong given that "failure" has to have occurred first.

With R- it's about the dog thinking about what TO do.

With P+ it's about the dog thinking about what NOT to do.

Ergo, I think R- is more compatible with drive training as well and can therefore be used in combination.

Also, as training using the R- progresses, the "nick" button can be brought into use. I'm not sure that's the case in a situation where the stim is delivered as P+ (I don't use the e-collar for general training using that method so I haven't really thought about the "nick" button in that scenario, but I'm thinking its delivery would be too fast to necessarily be adequate.)

I might not have explained this well nor as fully as I could and I'm sure K9 Force would be able to fill in a lot of blanks in between my explanation, but I'm about done in and on my way to turning off for the night. Perhaps he'll come in here between now and when I next look, to critique what I've written .

Edited May 14, 2009 by Erny

[Cosmolo]

Its late i know so tomorrow might be better for me to construct this response but anyway :cool:

I don't see the difference though- i look at the dog trained with P+ as also saying 'i know what to do to turn the stim off' too. And the dog trained with N- would also say 'better do XYZ behaviour to avoid the stimulation continuing'- so you could interchange both 'thoughts' in the dog with both methods?

Another question- do you always use the RT in that way, throughout the dogs training? Do you stop the stimulation occurring first as the dogs training progresses?

Sorry Erny- i am confused about what you mean with the nick button being brought in with N- and not with P+.Maybe after a good nights sleep i will get it or you can clarify when its not so close to midnight Thanks for the thoughts- its good to throw some ideas around.

[Kavik]

No i haven't Erny. I have a bit of a fundamental problem with not giving the dog the opportunity to respond to the command first

This is why I am not comfortable using the ecollar :cool:

[Cosmolo]

I understand that Kavik. I choose to use it differently rather than not at all though. :cool:

[Steve K9Pro]

K9: On most occasions, the stim applied after the command needs to be applied at a considerably higher level than if it is applied before the command. The reason is that the duration of the stim when applied after is much shorter, giving the dog less time to think.

The higher levels can often startle a dog & or the dog can negativity pair the stimulation (correction) with the command when the command comes first.

Applying the stim first can be applied at the level the dog can just percieve, not the level that generates a reaction in the dog, can mean that just about any dog can be trained effectively with the method.

Also the stim being applied at a lower level doesnt reduce / extinguish drive to the level that higher levels often do.

Another benefit is that the stim being applied will transgress into a communication pre emption, the dog knows after feeling the stim that a command is coming. This is technically known as a "behavioural interuptor". Which can turn a dog off a chase with very low levels of stim.

Kavik I feel that you need some one on one training (which I think I remember offering you free) so that you could learn more effectivley on how to use the collar. Trainers who hang up a tool when it doesnt work for them the very first time will end up with a lot of tools hanging.

I have had hundreds of novices learn to use a collar in this method & gain easy success without problems.

Edited May 15, 2009 by K9 Force

[m-j]

I don't see the difference though- i look at the dog trained with P+ as also saying 'i know what to do to turn the stim off' too. And the dog trained with N- would also say 'better do XYZ behaviour to avoid the stimulation continuing'- so you could interchange both 'thoughts' in the dog with both methods?

I haven't used an e-collar so know very little about the training using one but from the quickie explanation you (Erny) have written this is also my thought as the stim isn't something the dog wants happening, my understanding is that it is working to stop the stim in both methods .

cheers

M-J

[Cosmolo]

I don't feel i have to use high levels at all using the stim after the command? What happens when you don't then have the e collar to tell the dog the command is coming? I haven't seen a dog pair the stim with the command as its not as though the stim is still on when they are complying. I do get more immediate responses to commands which is what i'm after, to the point that i then don't have to use the RT at all. Do you always have to have the RT to stim the dog first to cue then that a command is coming or does this change in later training?

I'm not saying using the e collar in the way you do is 'bad'- although it is not my preference. But i just don't see the advantages in doing so and can see a few little disadvantages.

[Steve K9Pro]

I don't feel i have to use high levels at all using the stim after the command?

K9: I guess the word "high" is a relative term, the low level I use doesnt have an effect when used after the command.

What happens when you don't then have the e collar to tell the dog the command is coming?

K9: What happens when you dont have the collar on & give a command then don't stim? same thing, you can wean the dog off the collar.

I haven't seen a dog pair the stim with the command as its not as though the stim is still on when they are complying. I do get more immediate responses to commands which is what i'm after, to the point that i then don't have to use the RT at all. Do you always have to have the RT to stim the dog first to cue then that a command is coming or does this change in later training?

K9: No of course not the dog would then not be trained but just cuing off the e collar.

I'm not saying using the e collar in the way you do is 'bad'- although it is not my preference. But i just don't see the advantages in doing so and can see a few little disadvantages.

K9: I guess I would put it this way, if your asking questions it would seem that using the collar my way isnt within your experience, so having a preference is based on you prefer the only way you know?

Until you have been through both methods on many dogs, not sure you will be able to make a good judgement.

It inst unlike saying I dont like e collars, never used one, just dont like them...

[Cosmolo]

Fair enough K9- but thats the point of me asking the question. I'm not going to experiment on my clients dogs with something i am not convinced about just for comparisons sake. I think its good to question things you're unsure about- before considering 'trying it out' on dogs. Its not the only way i know- i have just not used your way before.

How can a dog feel a stim before a command but not after? Their ability to feel wouldn't change just because you have given a verbal command?

Can you explain how to wean the dog off the collar when you teach the dog that the stim is the first part of the command/ behaviour process?

Thank you

[Kavik]

Thanks for the offer K9 - will let you know if I decide to take you up on it

At the moment I am still uncomfortable with idea of negative reinforcement.

[Steve K9Pro]

Fair enough K9- but thats the point of me asking the question.

K9: Yep of course, happy to answer anything I can.

I'm not going to experiment on my clients dogs with something i am not convinced about just for comparisons sake. I think its good to question things you're unsure about- before considering 'trying it out' on dogs. Its not the only way i know- i have just not used your way before.

K9: I would happy to show you some time that it would be possible, for us both, the fact that I use the level that the dog can just perceive means that no harm will come from poor timing or poor association.

Any time that we are going to test or try anything new, it will need to be tried on a dog at some time to get accurate assessment. Can be your own dog if need be.

How can a dog feel a stim before a command but not after? Their ability to feel wouldn't change just because you have given a verbal command?

K9: Well, first, the ability to feel will not change based on the command unless the command changes the dogs level of drive. Secondly, applying the stim before, then giving the command, then guide the dog to success then turn off the collar means a longer stim that has a greater effect at a lower level than a shorter stim at the same level.

The reason that I feel that the stim after the command needs a higer level is the duration of the stim is shorter & the stim needs to compel the dog to comply. It is simple avoidance training which is basically adding a correction for not complying with the known command. No different than replace a jerk on teh check chain with an e collar stim.

This of course works but when I find that a higher level (than mine) needs to be used, some dogs show strong avoidance to the stimulation making learning slower & more difficult. It also seems to take some (un neccessary) time with some dogs to over come the fear.

When training others to use the collar you have to make sure their timing is more accurate with avoidance training than with my method. Timing in my method isnt as important.

I beleive the stim at a lower level over a greater time also gives the dog a chance to think through the excercise more clerarly & comply withpout avoidance (fear).

Another added benefit is that when yo aply the stim, it holds the dogs temperament & a dog that is moving on with training knows that a command will be coming.

Can you explain how to wean the dog off the collar when you teach the dog that the stim is the first part of the command/ behaviour process?

Thank you

Some dogs we dont wean off the collar for certain reasons, for example, a dog running away from you in pursuit of someone will respond to a heard command, if there is background noise, head wind etc the dog may not hear the command this cannot comply, the stim will signla the dog that a command is coming.

Others we have a procedue that is started before training starts & progresses through (too many steps to mention here) until the dog is compliant without the collar.

[superminty]

In the R- example, the stim is not reliant on the dog failing and so the dog doesn't fail and the reliability is stronger as "failing" isn't learnt.

In the P+ example, the stim is reliant on the dog failing. Reliability is not as strong given that "failure" has to have occurred first.

With R- it's about the dog thinking about what TO do.

With P+ it's about the dog thinking about what NOT to do.

Just wanted to highlight this bit - for me, the bolded bit is the clearest difference. Seeing as training (again, for me) is about the dog achieving success, the less opportunities to practise failure the better. Later on, once the reliability is trained, the stim becomes a P+ i.e. a correction.

Wouldn't using the collar as P+ only be just like using any other correction, except that you can use it in more situations (i.e. long distances)? Whereas using it as R- teaches the dog to figure it out on their own? I'm interested in feedback from those that think otherwise or have a different perspective. Or from anyone who can explain it better than I can

I am a shaping nut (love it), and I liken it to a dog that is accustomed to shaping - they figure stuff out on their own, through the provision of feedback (i.e. marker) and the learning is retained (I believe) to a higher degree than a dog trained through "traditional" P+ methods (for want of a better explanation). Please note that my dog is used to shaping and has "learnt how to learn" so to speak, which made progression to the e-collar easier, I reckon.

My e-collar trained (well, training) dog is never afraid to make a mistake with the collar on - he has the power to turn the stim off - and the responses I get from him are almost identical to the way he responds to shaping. This is something I've just noticed recently. I used the collar to teach a "new" behaviour - the recall (trust me, it was a new behaviour ) and the process felt just like shaping, with the exception that I showed him what I wanted a few times first (R+) – then moved onto: stim, help dog to get it right a couple of times, release, reward, then once he's got a rough idea, stim, wait, let him try stuff, release when he gets it right, bang - the dog has it.

The difference the stim makes is he works harder to figure out what I want (in order to release it - no such pressures from shaping, unless you count frustration ) and his responses are solid, more so than other P+ methods I have used with him (short explanation - taught "sit" with R+ and moved to P+ to proof, did not get the reliability I need - he is DA). I do use a stim as P+ when proofing commands but the level of response is at a much higher level when I start proofing than with previous methods.

I would consider using an e-collar as purely P+ if I was going to proof a known command – and I agree it would improve reliability. However, the process using it as R- first seems to me to be much shorter – and therefore preferable (as all learning is stressful and my aim is to make things as easy as possible, for me and the dog).

Interested in people's opinions on this.

[Cosmolo]

I just typed a long reply and my computer shut down

Superminty, using the e collar as P+ technically doesn't mean the level is such that it is like a 'normal' correction. I find the same things you do in my e collar training- dog knows how to turn the stim off, dog figures it out on their own (helped to start with of course)- the only difference is that i give a command first and then use stim if needed, whereas K9 and others stim first then give the command.

I agree with all that you said about the similarity to shaping but with added pressure and the dog not being afraid to make a mistake etc. I just think the same can be done while also giving the dog an opportunity to respond to a known command, before using the stim.

[WildatHeart]

[quote name='Cosmolo' I don't feel i have to use high levels at all using the stim after the command?

K9: I guess the word "high" is a relative term, the low level I use doesnt have an effect when used after the command.

[quote name='Cosmolo' How can a dog feel a stim before a command but not after? Their ability to feel wouldn't change just because you have given a verbal command?

Maybe what he was saying was that the level of stim used before the command, that if he used the same level stim after the command for non compliance wouldn't have the same effect (not perceived as a consequence cause its such a low level.) If he was to use it your way then he would have to raise the stim a level or two for it to be punishment......

ETA: oh, K9 has already answered, i should read the whole thread first

Edited May 15, 2009 by WildatHeart