[wvasko]

Originally Posted by Curbside Prophet

This too is not proof the collar is effective. What the dog prefers, is what the collar predicts...a walk. This is an example of classical conditioning, no different than how a dog can be classically conditioned to a click. Your dog gets excited about the walk; the collar just happens to be the antecedent for a walk. There's no way on Earth you can argue a dog gets excited by pain...that's what these collars are meant to do, deliver pain as a punisher. So no, you shouldn't interpret it that way.



But you still love me so that's all that matters.

CP
But you still love me so that's all that matters.

Love ya and need your input to keep us on the straight and narrow. All sides of dog training have to be discussed. I'm sure not going to knock somebody because of a humane attitude towards dogs. How stupid would that be. Trust me, I dislike dog abuse also and many years ago ran a few owners out of class because they thought if one correction was good, 10 would be better. If a light correction was good a hard correction was much better. I did decide to stop classes because of owner stupidity. Understand also a hard correction for the right dog is sometimes needed, but you try other methods first. As I have stated before you do not use an elephant gun to shoot a flea.

[blunder]

And what is that noise associated with? If it's not pain, you aren't using it to punish behavior. This is a different argument.

The only correction the dogs have ever had from a choke chain is the directional jingling caused by the index finger against the side of the lead. By the time they are put on a choke they already know that it is their responsibility to ether maintain a loose lead or a tight lead depending on what the dog is being trained for. (we train our show dogs to gate on a tight lead)
Need to know tho' I am a firm believer in using both positive and negative punishment in my training programs, just as I also use positive and negative reinforcement.
Most notably, punishment (as properly defined in operant conditioning) is used to artificially create stress or distraction and has nothing what so ever to do with the dog doing something right or wrong.

[Curbside Prophet]

Originally Posted by blunder

The only correction the dogs have ever had from a choke chain is the directional jingling caused by the index finger against the side of the lead.

So are you using this noise to cue attention? If so, this is not a punishment, it's a cue - an antecedent to a desired behavior - not a consequence.

Need to know tho' I am a firm believer in using both positive and negative punishment in my training programs, just as I also use positive and negative reinforcement.

And I'm a firm believer punishments need not be physical, otherwise I too use all four corners of the quadrant - as they are defined in OC.

Most notably, punishment (as properly defined in operant conditioning) is used to artificially create stress or distraction and has nothing what so ever to do with the dog doing something right or wrong.

This is not how punishment is defined in OC. The definition says nothing about causing stress or distraction; punishments need not be any of these, they only need to be effective. The definition only says a punisher *is* punishing if a behavior is weakened by the consequence. It does not define what that consequence is.

Furthermore, if your criteria (what *is* right or wrong) does not define what that right or wrong behavior is, you are indiscriminately punishing or reinforcing your dog's response. Avoiding the punishment can't be learned if you use your punishment indiscriminately.

I don't believe this to be true, however, if you're simply cuing attention with a noise. For this noise to have any punishing value it must be classically conditioned to an aversion. With a choke chain this aversion *is* pain; you said you don't use this collar other than to make a noise.

'Corrections' are not defined in OC, so I try not to use this term. It's too confusing and can lead us to talking about two different learning processes as illustrated here.

[blunder]

Originally Posted by Curbside Prophet

So are you using this noise to cue attention? If so, this is not a punishment, it's a cue - an antecedent to a desired behavior - not a consequence.


And I'm a firm believer punishments need not be physical, otherwise I too use all four corners of the quadrant - as they are defined in OC.


This is not how punishment is defined in OC. The definition says nothing about causing stress or distraction; punishments need not be any of these, they only need to be effective. The definition only says a punisher *is* punishing if a behavior is weakened by the consequence. It does not define what that consequence is.

Furthermore, if your criteria (what *is* right or wrong) does not define what that right or wrong behavior is, you are indiscriminately punishing or reinforcing your dog's response. Avoiding the punishment can't be learned if you use your punishment indiscriminately.

I don't believe this to be true, however, if you're simply cuing attention with a noise. For this noise to have any punishing value it must be classically conditioned to an aversion. With a choke chain this aversion *is* pain; you said you don't use this collar other than to make a noise.

'Corrections' are not defined in OC, so I try not to use this term. It's too confusing and can lead us to talking about two different learning processes as illustrated here.

Exactly, rattling the chain is nothing more than using a clicker (marker) that has 4 side inorder to tell the dog which direction to go.

On page 2 of this article (teaching swim-by) you will find a good example of using positive punishment to create stress/distraction. We call it FTP (force to pile)
http://www.retrieversonline.com/swimby.htm

[Curbside Prophet]

Originally Posted by blunder

Exactly, rattling the chain is nothing more than using a clicker (marker) that has 4 side inorder to tell the dog which direction to go.

No, this is not accurate. When you use a clicker, the click occurs when the behavior is performed. You do not use the click to cue a behavior. You can, but that would ruin the value of the click. The click is classically conditioned to the reinforcer. In most cases this reinforcer is food. So rattling the chain is not a marker, it's a cue. 'Sit' is a cue, 'down' is a cue, 'out' is a cue, etc.

On page 2 of this article (teaching swim-by) you will find a good example of using positive punishment to create stress/distraction. We call it FTP (force to pile)
http://www.retrieversonline.com/swimby.htm

What was described in the article is not positive punishment...it's negative reinforcement. If the e-collar were used as positive punishment, you would cue 'back', and nick the dog at a high level, one time, when the dog drove away. What the article described is a lower level nick, set continuously and removed (negative) once the desired behavior is performed. This reinforces the behavior of 'back'. You are not punishing 'away'.

[blunder]

Originally Posted by Curbside Prophet

No, this is not accurate. When you use a clicker, the click occurs when the behavior is performed. You do not use the click to cue a behavior. You can, but that would ruin the value of the click. The click is classically conditioned to the reinforcer. In most cases this reinforcer is food. So rattling the chain is not a marker, it's a cue. 'Sit' is a cue, 'down' is a cue, 'out' is a cue, etc.

What was described in the article is not positive punishment...it's negative reinforcement. If the e-collar were used as positive punishment, you would cue 'back', and nick the dog at a high level, one time, when the dog drove away. What the article described is a lower level nick, set continuously and removed (negative) once the desired behavior is performed. This reinforces the behavior of 'back'. You are not punishing 'away'.

NO-NO-NO
better read it again. You nick the dog either with the collar set on momentary or with a short duration nick with a collar that has continuous l,m,h (the difference between a Pro-100 and a Pro-200 or a collar that only has cont.) and the nick is applied AFTER the dog is in motion, not before. Sometimes well after the dog is in motion (already in the water)
The level of the nick has no bearing what so ever on it being positive or negative anything, it is only the fact that you did or did not nick and when the nick is applied.
What you described is what we do when "proofing" a basic command (recall for instance) with negative reinforcement.
In the article you are not TAKING AWAY (negative) anything with a nick, you are ADDING (positive) something
It is when you send and don't nick that it could be classified as negative reinforcement, even tho' it really isn't.

BTW, what is "sit-nick-sit"? and what is "nick-sit-nick"?

[Curbside Prophet]

This is a quote from the article...

This step involves the use of the e-collar to reinforce the ‘back’ command both on land and in the water. Start the session by marking the pile and giving 2-3 freebies. Put your dog in a remote sit front finish position 10-20' from the water’s edge. Command ‘back’and just after the dog has turned, nick and then repeat the command ‘back’. Use momentary mode or a very quick nick on the continuous mode.

A definition of negative reinforcement from here.

In an attempt to increase the likelihood of a behavior occurring in the future, an operant response is followed by the removal of an aversive stimulus. This is negative reinforcement.

The operant response is back and the aversive stimulus in this case is a continuous nick. It should be at a lower level so as not to be a positive punishment. The author went out of his way to warn against this and appropriately calls it a reinforcer in his quote.

Let me help you explain the difference. There is a simple equation to use that helps in understanding learning theory. This equation is Antecedent (A) + Behavior (B) = Consequence (C). Also know as the ABC's of learning. What is an antecedent? Anything that occurs before or during a behavior. What is a behavior? What the dog physically does. What is a consequence? A reward or punishment for a behavior.

In the forced retrieve example, your antecedent is both the cue 'back', and the continuous nick. The behavior is the dog retrieving the bumper back to you, and the consequence is the continuous nick is ceased once the behavior is fully performed. This is distinguishable from a positive punishment because the negative aversion occurs while the behavior is being performed, not after.

What about an example of positive punishment? Say your dog likes to chase rattle snakes. You don't want your dog to chase rattle snakes. So, what would an antecedent be for your dog to chase rattle snakes? A rattle snake. The behavior is your dog chasing one, and now the consequence in this case would be a high level nick. This nick occurs after the behavior begins. Why a high level? You want the dog to associate rattle snakes with severe pain, and you want him and the behavior to cease immediately (no need for a continuous nick, one jolt should be enough). On the other hand, you do not want your dog to associate 'back' with severe pain, thus you set the e-collar to a lower level. The author said to do just that.

[blunder]

Originally Posted by Curbside Prophet

This is a quote from the article...


A definition of negative reinforcement from here.

The operant response is back and the aversive stimulus in this case is a continuous nick. It should be at a lower level so as not to be a positive punishment. The author went out of his way to warn against this and appropriately calls it a reinforcer in his quote.

NO-No You are still miss reading it. You first send the dog AND THEN nick it AFTER it is in motion, AFTER the nick you then repeat the verbal command. You DO NOT hold the button down making a continuous correction. You get your finger back off the button giving the dog as short a duration NICK as is possible if all you have is a cont. only collar. His only reference to continues is for collars that do not have a momentary setting (nick).

For it to be negative reinforcement you would give the command DURING continuous (longer duration) stimulation.

[Curbside Prophet]

I am not misreading it. You are not using the terms of learning theory correctly. How can you deny that the author said the e-collar is used to reinforce back? If it is used as a punisher, what behavior are you trying to punish? Remember, reinforcement gets behavior. Punishment decreases behavior, so what behavior are you attempting to decrease? Certainly not back, that's what you're trying to train.

[blunder]

Originally Posted by Curbside Prophet

I am not misreading it. You are not using the terms of learning theory correctly. How can you deny that the author said the e-collar is used to reinforce back? If it is used as a punisher, what behavior are you trying to punish? Remember, reinforcement gets behavior. Punishment decreases behavior, so what behavior are you attempting to decrease? Certainly not back, that's what you're trying to train.

"Remember, reinforcement gets behavior. Punishment decreases behavior,"

That is absolutlly not true.

OK,, let me use a different approach with you.

It is STILL using a correction for no other reason than to create stress/distraction, and is totally independent of obeying or disobeying the command! (my original statement), and!! creating anything is STILL ADDING SOMETHING TO (positive).
In the old days before the e-collar the same thing was accomplished by giving the dog a swat on the butt with a riding crop after it was in motion (for those that think the e-collar is some kind of monster)

[DogGoneGood]

*Raises hand* Question!...

And I'm a firm believer punishments need not be physical, otherwise I too use all four corners of the quadrant - as they are defined in OC.

What is "OC"? Last I checked it was both "Orange County" and an apparently hit TV show

[pamperedpups]

OC = Operant Conditioning

[Curbside Prophet]

Originally Posted by blunder

"Remember, reinforcement gets behavior. Punishment decreases behavior,"

That is absolutlly not true.

This will be the last attempt I'll make to try and help you understand the terms of learning theory. I don't know what more I need to do than define the terms for you again. These terms are not questionable, and you can't make up your own definitions as they suit you. Either we both use a common language (I'm suggesting those as they are defined in academia), or we will continue to not understand what each other is saying. If you didn't read the link I provided earlier, here it is again: http://www.psychology.uiowa.edu/Facu...forcement.html (for reinforcement) and this one http://www.psychology.uiowa.edu/Facu...unishment.html (for punishment).

I'll now quote reinforcement and punishment from these links, and you tell me how this is any different than my reinforcement gets behavior, punishment decreases behavior statement, and why these definitions are not usable.

Reinforcement is defined as a consequence that follows an operant response that increase [gets] (or attempts to increase) the likelihood of that response occurring in the future.

Punishment is defined as a consequence that follows an operant response that decreases (or attempts to decrease) the likelihood of that response occurring in the future.

Now prove to me you are a reasonable person and not just arguing to be 'right'. At the risk of sounding pampas...I'm already convinced of what I know, I don't need your confirmation that I'm right. If you don't see that I'm trying to be helpful, our conversation need not go any further.

[wvasko]

Originally Posted by Curbside Prophet

This will be the last attempt I'll make to try and help you understand the terms of learning theory. I don't know what more I need to do than define the terms for you again. These terms are not questionable, and you can't make up your own definitions as they suit you. Either we both use a common language (I'm suggesting those as they are defined in academia), or we will continue to not understand what each other is saying. If you didn't read the link I provided earlier, here it is again: http://www.psychology.uiowa.edu/Facu...forcement.html (for reinforcement) and this one http://www.psychology.uiowa.edu/Facu...unishment.html (for punishment).

I'll now quote reinforcement and punishment from these links, and you tell me how this is any different than my reinforcement gets behavior, punishment decreases behavior statement, and why these definitions are not usable.


Now prove to me you are a reasonable person and not just arguing to be 'right'. At the risk of sounding pampas...I'm already convinced of what I know, I don't need your confirmation that I'm right. If you don't see that I'm trying to be helpful, our conversation need not go any further.

CP
Way to go, even I understood the Reinforcement/Punishment. Now I'm getting scared some of you is rubbing off one me and I'm getting smarter.

[blunder]

Originally Posted by Curbside Prophet

This will be the last attempt I'll make to try and help you understand the terms of learning theory. I don't know what more I need to do than define the terms for you again. These terms are not questionable, and you can't make up your own definitions as they suit you. Either we both use a common language (I'm suggesting those as they are defined in academia), or we will continue to not understand what each other is saying. If you didn't read the link I provided earlier, here it is again: http://www.psychology.uiowa.edu/Facu...forcement.html (for reinforcement) and this one http://www.psychology.uiowa.edu/Facu...unishment.html (for punishment).

I'll now quote reinforcement and punishment from these links, and you tell me how this is any different than my reinforcement gets behavior, punishment decreases behavior statement, and why these definitions are not usable.


Now prove to me you are a reasonable person and not just arguing to be 'right'. At the risk of sounding pampas...I'm already convinced of what I know, I don't need your confirmation that I'm right. If you don't see that I'm trying to be helpful, our conversation need not go any further.

Also my last post on the matter.
Lets just agree to dissagree.

[Curbside Prophet]

Originally Posted by blunder

Lets just agree to dissagree.

I still love you too blunder.

[French Ring]

It appears like I'm too late to join the debate; however, I tend to start with marker training first. I prefer to start with something very positive because dogs don't know what I want and why should they get harsh corrections for something they don't understand at first? Marker training/click training is a good way to teach them something new and positive. I have a reward that they want like food, praise, or toy.

If my dogs chose to ignore the given command, (yes) there will be corrections. I only give corrections when a dog chose not to listen to me and know the command. (More likely if he is being stubborn and ignoring my commands then there will be corrections) I have seen owners giving out harsh corrections to dogs who don't understand the commands and thought their dogs know or understand it.