Punishment/negative Reinforcement/both?

[pgm]

just about. I don't 'reduce' the reward to verbal praise - that stays constant thoughout - (except for the first week where I don't use any praise or reward at all - at least not extrinsic reward).

I think of corrections as stimulating and increasing the dog's focus, drawing him more and more into the game/activity for its own sake - rather than for external reward.

[Staranais]

Can I ask then, what if the dog isn't particularly interested in verbal praise?

Obviously, the stronger the bond you have with your dog, the more interested he will be in your approval. However, some dogs are more independent than others, and don't overly care for praise. And others are so driven that, even though they might like praise, it's just not the highest priority for them (for example, if all I had ever offered my dog was praise, I really think that if you bounced a tennis ball near him he'd be off like a shot. He stays in this situation because he has a history of prey drive satisfaction and correction in response to my commands, not because he wants the praise).

How would you deal with this type of animal?

[pgm]

dogs are social animals - all dogs look to the leader or to whomever they percieve as leader. The first week of training I don't use any external praise or reward at all - everything the dogs earns in terms of reward and correction in the first week is internal to the activity and his own actions. The first week is more or less establishing the proper relationship between the handler and dog of leader/follower. This is the basis of what follows. The dog looks towards you for direction because you are the leader, not because you give it rewards. Praise in this sense is better described as encouragment, not external reward.

The idea of external rewards offered to the dog is one that I think is misplaced. Alpha dogs don't reward their inferiors - the inferior simply looks towards the alpha for direction.

Operant conditioning is not the best langauge to describe this process because its perspective is as an observer. Whereas the kind of training and the philosophy it works from (and the logic that pertains to it) that I am working with is all about developing the internal (and social) relationship between dog and handler. OC describes things from an external perspective and I think misleads a great deal as to what is going on in training.

Training develops the bond - it isn't dependent on it. That is why the question of what happens if the dog isn't motivated by praise is misplaced.

[Staranais]

Very interesting philosophy.

I have not heard of this style of training before - most training methods I have heard of rely heavily on extrinsic rewards (whether they are prey drive, food or social acceptance based rewards). Are there any books you could recommend to me to get a better overview of your training style, just for my own education?

[pgm]

There are two books, both of which usually draw negative comments whenever they are mentioned. The first is the Koehler Method of Dog Training. The second is Vicki Hearne's Adam's Task: Calling Animals by Name. Hearne's book is a philosophy book which looks at our relationships with animals. Hearne was herself a Koehler trainer (and also academic, poet, and horse trainer) and gives a very good philosophical account of the method and what it is trying to achieve. I recommend it to anyone, even if your not interested in Koehler.

[Staranais]

OK, thanks pgm, I'll put them on my xmas wish list (along with the other gazillion dog training books on there... )

[Steve K9Pro]

K9: heard my name called so I will just jump in the middle, I have had a quick read through & would like to make some comments...

H:

Personally Amhailte, I think it is playing a dangerous game, one that could certainly lead to learned helplesness, handler aggression etc.

K9: If your starting with a dog that isnt showing handler aggression, & your in control of the training program you will never get either learned helplessness nor aggression. In fact if the dog was suffereing from Learned helplessness, the symptoms are shutting down.

Through this discussion people are spending a lot of time talking about punishment, I never ever punish the dog. The dog may suffer consequences for its own actions made possible by me. There is a difference & I will tell you, a huge one at that.

The other thing that must be looked at is the way in which I apply a correcting, my body language & vocal tonation dont portray anything but positive. The corrections are so minute as I dont train the dog under a distraction level it hasnt worked up to.

A:

If you're aiming for a speedy response, wouldn't it be better to either:

1) use negative reinforcement (e.g. e-collar set VERY low), turned off when dog completes requested task so that it is in dog's best interests to comply very quickly; or

2) selectively positively reinforce only the fastest responses with a prey drive type reinforcer?

Why do people think this 'active training' is a better route than these two options?

K9: The fastest way to get a dog to do anything is by placing the dog in high drive.

You have separated the two above but thye way I train, I combine them both.

The dog is looking for prey drive satisfaction & the e collar is telling the dog that he is going the wrong way about it, I use the e collar most often as a communicator, not a correction device.

Kavik:

I think active trainers (feel free to correct me if I am wrong K9) would say that the first option with the e-collar IS one sort of active training. If I understand it correctly.

Active - Command - Correct - Comply - Reward

K9: the action line you describe is avoidance training, which can be completed with any correction tool including ane collar but I dont use them that way.

H:

I personally think that the whole point of operant conditioning is to teach the dog the consequences of his/her behaviour. So the dog doesn't comply to a command, they get a correction. They comply, the get reinforcement. Some consider this an inferior way to train, I think it is the whole point of training.

K9: as do I, however I feel that when explaining someone how to apply this method, they always end up in consistent in the corrections & rewards.

The dog starts to wonder " do I have to comply? is this the time I get a correction".

It also lends itself to me asking, do we want the dog complying to avoid punishment, or competitively complying to beat consequence.

There is a whole different mid set to the training on both handler & dog parts & as I am always seeking drive, this method always (IME) produces more consistent results in higher drive.

As I start with low / zero distraction, the dog is only recieving constant corrections for a short period of time, then the dog through my program starts to beat the corrections, thus avoiding them, once I see this increase in speed & drive, I add more distraction & the dog knows there is a process to winning (avoiding corrections).

A:

I understand how 'active training' would work with the e-collar, it's simple negative reinforcement (escape training).

K9: in the beginning yes, in the end no. In the end the stim gives the dog the hint that a command is on the way, thats command completed will give drive satisfaction. Its a c ommunicator, tap on the shoulder.

The dog experiences zero negative from it, thus no loss in drive.

Drive is a dynamic that most people dont consider until its gone.

A:

It is in his interest to obey quickly, because that means the collar turns off quickly. Right?

K9: only in the beginning, in the end, he see's it as he is going away from the drive satisfaction that he seeks.

A:

But how does this apply to a prong collar correction, for example? You give the command, prong the dog, he complies. But what has he learned?

K9: it isnt the same, but you can get close, the dog wont learn anything from one rep, but over time with timing, you can teach the dog to comply faster until he outperforms the correction.

When the dog see's this, which ideally should happen between the 10 & 15 rep, he gains pleasure from his success.

I recently had a LGD attend training, with no high desire for food, praise or prey, the only motoivator left is avoidance of correction.

In 15 minutes the dog was positively motivated bia this method to comply quickly & beat the correction.

A:

And what if you give the command, prong the dog, and he does nothing? Just prong him again? Then what was the point of pronging him the first time?

K9: its the first of a never ending series that will continue until you comply. Guidance is also applied & the dog also must "know" the command & not be highly distracted.

H:

If the aversive comes before the behaviour and is then removed as a result of that behaviour it is -reinforcement.

K9: yes, but if the e collar isnt considered an aversive by the dog, what does the stim become? A motivator... the key to drive satisfaction.

H:

I have issues with using corrections on dogs during the teaching phase.

I have issues with using corrections without giving the dog the opportunity to comply.

I have issues with giving a punisher and reinforcement for the same command.

K9: low level e collar stims are not corrections, punishers or painful, so I dont have an issue using them when teaching.

H:

I'm not saying dogs can't be trained this way, I just don't think it is the most ideal for most dogs and most handlers and could have potentially disasterous consequences.

K9: I agree, but only in theory, I have run workshops over most of Aust & people see a lot different reactions than they were expecting...

A:

Hmmm, so you're basically giving the dog numerous small corrections, like repeatedly hitting the nick button on an e-collar?

K9: yes but forget the e collar for a moment, multiple corrections allow you to apply - R over a greater period of time & use a lot lower level corrections.

Instead of choosing a correction level to over come, cease & desist the behaviour, you apply lower level corrections that dont stop until the dog offers to quit the behaviour or comply with command...

A:

Is this in the learning stage, or the proofing/correcting stage?

K9: Learning is teaching mechanics of command, no corrections, correction/ training stage is practice with consequences.

Proofing is whe your all done & dog requires no corrections to comply under thye level of distraction you have trained to.

So answser is, in training/correction phase.

I will continue...

[Staranais]

Please do, this is really interesting.

Can I ask one question in the meantime, though?

A: I understand how 'active training' would work with the e-collar, it's simple negative reinforcement (escape training). 

K9: in the beginning yes, in the end no. In the end the stim gives the dog the hint that a command is on the way, thats command completed will give drive satisfaction. Its a c ommunicator, tap on the shoulder.

If you always stim the dog before issuing a command, doesn't this result in the dog becoming habituated to the stim so that he needs the 'tap on the shoulder' to work? Or do you eventually wean him off the 'shoulder tap' somehow?

Edited August 25, 2005 by Amhailte

[Steve K9Pro]

Erny:

Imagine if I was under pressure to get somewhere really fast ... for example, the airport - assume I was running late and if I missed my plane I blew my overseas trip.

Imagine also my thought process if I knew I'd lose my licence if I got caught speeding and that getting caught was a good probability - assume I need my licence for employment purposes and that I'd be losing my job if I lost my licence.

The "threat" (or "possibility", as Amhailte so appropriately worded it) of the latter would be negative reinforcement - that's what stopped me from speeding. But the fact that the "threat" of the punishment (ie negative reinforcement) is there, doesn't make me dislike driving because I know how to avoid the punisher (ie loss of license). And I made my plane anyway (release & positive reinforcement).

K9: now add one more dynamic, no matter how fast you drive, K9 Force has the ticket & you will only fly if you have obeyed the road rules...

By adding myslef as the only constant determining factor, I am the first consideration the dog needs to have.

This is essential when you have a highly trained high drive animal that you require 100% compliance from.

Edited August 25, 2005 by K9 Force

[Steve K9Pro]

A:

If you always stim the dog before issuing a command, doesn't this result in the dog becoming habituated to the stim so that he needs the 'tap on the shoulder' to work? Or do you eventually wean him off the 'shoulder tap' somehow?

K9: All training that I do is "goal dependant". The clients goal dictate the program direction.

With a Law Enforcement dog, I dont wean the dog off the collar or the stim as I always want that tap on the shoulder.

When you think about it, if you have bought the colalr, why not use it?

If your pet is running toward a distraction, its inportant he recalls. Lets add some weight.

Lets say that distraction is on the other side of the highway, & the wind is blowing in your face, dog cant hear nor see you... That tap on the shoulder is a mighty handy thing to have.

The other thing that a tap on the shoulder does is distract from the behaviour, oits called a behavioural interuptor...

Dog chasing cat will need high stim to extinguish prey drive, dog chasing cat that understands this training method feels stim & "needs" to know where you are to give the command... It has forgotton all about the cat without loss in drive.

& of course I have a program for competitions where you cant use the collar....

[Staranais]

When you think about it, if you have bought the colalr, why not use it?

If your pet is running toward a distraction, its inportant he recalls. Lets add some weight.

Fair enough, but just from my own (limited) experience, I do think it's desirable that a pet dog acheives a reliable level of obedience without needing the tap of an ecollar.

Even if the dog is always wearing the collar, life is full of surprises and you don't always have your finger ready on the button. If you're trying to carry the shopping from the car to the house, for example, you're not going to have your transmitter right there in your hand ready to use, so it's desirable that your dog will recall anyway.

Or am I misunderstanding?

[Steve K9Pro]

A:

Fair enough, but just from my own (limited) experience, I do think it's desirable that a pet dog acheives a reliable level of obedience without needing the tap of an ecollar.

K9: as do I, same as any working dog, its just when the dog cant see nor hear you or does make (even a rare) mistake, your not lost.

A:

Even if the dog is always wearing the collar, life is full of surprises and you don't always have your finger ready on the button.

K9: again, its not just about making every episode a training session, its about not loosing out totally.

Your dog is 50 feet from you, it takes a dash after something, your not going to catch that dog on foot, but the 27mhz signal from your remote will.

Think of your dog running up to another, & half way there you give a drop command that "will" be obeyed, everytme.

A:

If you're trying to carry the shopping from the car to the house, for example, you're not going to have your transmitter right there in your hand ready to use, so it's desirable that your dog will recall anyway.

K9: the remote doesn replace training, it raises the compliance level.

If your walking in with the shopping right now, is your dog off or on leash? The bet is that if your dog is trained to a high level of compliance yet, it's on leash, thats not the case with the e collar most times.

Having the remote "is" on leash, without the restraint of the leash...

Remotes fit nicely on your belt, lanyard around neck, utility vest or belt, you learn to find that button real fast.

& remember you will only need to press it as a last resort, if your dog doesnt comply, or you dont think he can hear you.

Commands are completed & the dog weans off the stim, thats like saying I dont tug on my check chain anu longer, but its there if I need it. Same thing only with the remote comes freedom... quickly.

[Staranais]

Commands are completed & the dog weans off the stim, thats like saying I dont tug on my check chain anu longer, but its there if I need it. Same thing only with the remote comes freedom... quickly.

So if I'm understanding you right K9, initially you train with a e-collar stim for every command. When the dog is great at that, you wean him off the stim, so there is a stim only for non-compliance (which should be very rare anyway). Am I right yet?

Your dog is 50 feet from you, it takes a dash after something, your not going to catch that dog on foot, but the 27mhz signal from your remote will. 

Think of your dog running up to another, & half way there you give a drop command that "will" be obeyed, everytme.

Another dumb question coming up - if your dog's dashing off as you describe, he's already in full drive, right? So is an ecollar stim really going to be enough to interupt his focus?

I'm just thinking of my own dog - he's very biddable when he's relaxed, but when he's in full prey drive, I swear I'd have to pick him up by the unmentionables before he broke his focus on the prey item. (I generally try to interupt his focus before he decides to take off, for this reason.)

Thanks for answering all my stupid questions - I'm learning heaps, and I'm sure everyone else is too.

[pgm]

Another dumb question coming up - if your dog's dashing off as you describe, he's already in full drive, right? So is an ecollar stim really going to be enough to interupt his focus?

Can I answer this question - k9 will correct me if I am wrong. The answer is yes, the stim will stop him, even on low stim if the training is good. But with the ecollar you have a fail safe get out card - you can raise the level of the stim. If you work your dog on level 2 for instance in most normal situations, in times of high excitement you can always raise the level up to 4 for instance to get his attention.

Its like tapping on someone's shoulder to get their attention, if they are distracted you simply tap a little harder or a little more insistently.

[Steve K9Pro]

A:

So if I'm understanding you right K9, initially you train with a e-collar stim for every command. When the dog is great at that, you wean him off the stim, so there is a stim only for non-compliance (which should be very rare anyway). Am I right yet?

K9: more or less, yes.

A:

Another dumb question coming up - if your dog's dashing off as you describe, he's already in full drive, right? So is an ecollar stim really going to be enough to interupt his focus?

K9: yes it does, in other methods you have to reach a correction level to overcome & extinguish drive, this is not so with my method & low stim, the dog thinks about something else.

Here is a scenario, Patrol dog trained to bite man in full prey drive, dog is after man & you recall the dog just as he is about to bite, why does he come?

He comes as the training program has taught him that drive satisfaction only comes via me...

A:

I'm just thinking of my own dog - he's very biddable when he's relaxed, but when he's in full prey drive, I swear I'd have to pick him up by the unmentionables before he broke his focus on the prey item. (I generally try to interupt his focus before he decides to take off, for this reason.)

K9: lol, its not breaking focus, thats extinguishing drive, aversives reduce drive, thats one of the pro's & cons of thats style of training.

E collar used my way doesnt loose drive, that's one of the many pro's.

If your dog goes into drive, & you need to correct him to keep compliance, thats because he believes he can get satisfaction with out you, your now only left with taking drive away, making him the "biddable" dog you had before.

In my work I often need that drive, so I cant take it away...

& yes PGM, thats correct, you must have had a good teacher

[Staranais]

If your dog goes into drive, & you need to correct him to keep compliance, thats because he believes he can get satisfaction with out you, your now only left with taking drive away, making him the "biddable" dog you had before.

Yes, I can see that. The dog thinks that the best way to satisfy his drive is to ignore you and run after the prey item. He does not think that obeying you will result in him achieving drive satisfaction, so that is why he ignores you and you have to correct him to keep his attention.

But, with a dog like this, is it too late to retrain him any other way? He's already had many years of learning that he can get drive satisfaction on his own, without the intervention of any person. Would it be to late to try to teach him that drive satisfaction comes only from listening to and obeying commands, since he's already accustomed to satisfying himself?

I'm asking since it seems to me this method would work best on dogs that you have raised since they were puppies, since then you could control their experiences and ensure they never have the experience of achieving drive satisfaction on their own.

[Steve K9Pro]

A:

But, with a dog like this, is it too late to retrain him any other way? He's already had many years of learning that he can get drive satisfaction on his own, without the intervention of any person. Would it be to late to try to teach him that drive satisfaction comes only from listening to and obeying commands, since he's already accustomed to satisfying himself?

K9: never too late, but your overall success may be less, that is largely dependant on your consistency of the new program.

A:

I'm asking since it seems to me this method would work best on dogs that you have raised since they were puppies, since then you could control their experiences and ensure they never have the experience of achieving drive satisfaction on their own.

K9: it actually works best on high drive dogs with solid nerves. But you can start at any time.

[haven]

I've no doubt that people like you K9, are more than capable of assessing if their dog is temperamentally sound for this technique and applying it correctly, to benefit the dog rather than be detrimental. I realise that you run seminars and train people average people in this method succesfully, but I have seen so many people mismanage their dogs in training so often that I would hesitate to teach the method to the general population, because I honestly think there is great potential for damage to the dog (psychologically).

When I use the term aversive, it doesn't necessarily mean a painful stimulus or saomethin that causes fear. I still consider an ecollar stimulus to be aversive, just as someone tapping me on the shoulder would get irritating. The person sitting next to me on the train and whistling inceasantly is aversive, as is a cold shower etc. I don't think there is anything wrong with using -R though, in fact I believe it is not utilised in training nearly as much as it should, because of the misconception general people have about it being pain induced.

So when are you coming to Melbourne? I would like to attend and learn more.

[Lablover]

I have read this thread, which has given me a headache. All these new catch words. I admit I am getting older and could only really understand K9 force. All this theory means nothing unless you can be put to practical use.

Today I was in our front paddock, picking up some firewood, the labs flew by chasing a couple of rabbits. Only by a good training program which included negative reinforcement (correction), did they immediately stop the chase-and sit- with a quick quiet command.

[Denis Carthy]

Lablover

I have read this thread, which has given me a headache. All these new catch words. I admit I am getting older and could only really understand K9 force. All this theory means nothing unless you can be put to practical use.

Denis

New, they are not new. Every time I read all the billions of texts specifically on dog training I have not once heard anyone correctly giving a realistic and easy for anyone to understand interpretation of OC punishment OR reward, except Steve or someone else here. It first came into being in the 1930s, but not as its generally used in ‘dog training’ conversations.

Rather than go into length on what’s a very good post, someone, I think Steve, already said that the way/methods etc he trains it is perceived by the dog as consequence of a behaviour and that is ALL it means and if some people do not know how to apply punishments in a dog training programm, favourable to the dog, then that is a statement of limited knowledge of the person and not of something bad about the use of punishments.

I will go one step further and state quite clearly, no animal, capable of learning, of ANY species, would survive without punishments, punishments are to the benefit of all animals capable of learning something.

What has happened in dog training is that commercial interests, cranks and extremists have been active this past few years and have completely given a misinterpretation of the meaning of punishment AND reward to innocent pet owners who then pass a version of it on as some form of retribution by bad tempered individuals and which has no welfare advantage to the dog.

Aversion is often mentioned, the way many pet owners have been indoctrinated with incorrect meanings to these and a few other words, it is THEM, the pet owners who have been exposed to these words, who show aversion conditioning behaviour not just to the use of punishment, but an aversion to learn anything about punishments favourable to the dogs welfare, its relationship with its human owner and resulting in a well trained well behaved dog which gets ample excecise as a result and easily obtainable recall when the dog is under pressure not to recall, which is all most pet owners want.

Brainwashing is another appropriate term for the process of exposing vulnerable people to incorrect defintions of punishments and rewards, the problem is that they do not know what punishment means, it literally means ANYTHING and has NO specific definition beyond what I have written below.

A positive punishment is ANY consequence/stimulus any animal of ANY species perceives/learns is a non appetitive/unfavourable consequence of a/its own behaviour, it might decrease a behaviour.

A positive reinforcer is ANY consequence/stimulus any animal of ANY species perceives/learns is an appetitive/favourable consequence of a/its own behaviour, the behaviour might increase.

Interpret that into a human behaviour and a murderer experiences the act of killing as appetitive/favourable and he/she might kill again for the reward they experience. So, as reward is the fashionable thing, then who is going to find the poor murderer another victim to save him the punishment of tedium by haveing to look for one himself.

.

Edited August 26, 2005 by Denis Carthy