Clicker training?

[valdes43]

Originally posted by mark_sherry
Once a dog knows what's expected of him, he will test you and refuse.

I don't agree with this unless you're having a serious rank issue. Like I said before, dogs are selfish. They're in it for themselves. So if the heat of a hot summer day is more undesirable than the desirable tasty liver in your hand then the dog will simply choose to stop working. It's not worth it. That's why there should be a layer of compulsion in the program.

It sounds like you're implying that he will intentionally test you. I don't believe that's the case.

[mark_sherry]

No, I don't mean that she'll intentionally test me. Forgive my colloquialisms. I mean that they'll test your nerves. In other words, on a hot day when the treat isn't enticing to them, they'll test you (your nerves). By testing, I meant anything that constitutes their refusal or lack of motivation to do something. I didn't mean the pecking order issue. With the way that I wrote that statement, I can see how you got that. But that's not what I meant. The foundation of any good working dog begins with the relationship between the dog and the handler, so the "testing" part should already be a non-issue for the serious team.

[DNeff]

It's not easy to keep that dog from disobeying when you do positive-only training.

It's not easy to keep any dog no matter which method they are trained with from making a mistake if they haven't been trained and proofed for it.
It all has to do with the foundation you lay down. I train my dog to enjoy working even if they are distractors present. Yes, I'm sure if something really major happened my dog might refuse. Dog's who are trained on prongs refuse sometimes when something major happens. Your training is what prevents it from happening again. This is where your foundation comes in. Part of good training is letting your dog make a mistake and then teaching them the proper way to handle it next time. Whether you use a collar pop or voice and food, to teach your dog it is better to pay attention than it is to be distracted. Dog's do what works. Period. If getting food and praise matters, or avoiding a correction matters, the dog will work through it next time. And of course that varies for each dog.

I get my dog back and focused using my voice, our relationship and/or treats and toys. Not by making a physical correction. I proof my dog also, just because I use positive methods doesn't mean I don't proof. I just proof differently. The dog still learns to work through the distractions. The dog learns that there is more in it for him to pay attention and be with me than to engage in something else. I have spent time developing that, very few dogs are born feeling that way.

Considering how new most positive training methods are on the scene, and how few people know the proper ways to implement them, I sure it will be a while before there are top competitors out there doing it. Like anything it takes time and experience. Most trainers that used compulsion are just now (1997 or so) switching to positive methods. It takes time to become adapt at using any technique especially when the resources and support are limited.

Dawn
U-CD Cammcastle's Mystic Riven CD NA NAJ NAC NGC NJC CGC TT
http://members.aol.com/dplantier
http://members.aol.com/rottweilerworld

[mark_sherry]

Dawn,
It certainly is not easy to keep a dog focused at all times. That's the joy and challenge of training for anything. If it were easy, then few people would do it with the intensity that they work in. Please see that you and I are not that far off in what we believe. You make a comment that you redirect your dogs' attention using a variety of methods -- relationship, treats, toys, etc. There's variation in your training method, and that's my point. It requires a varied approach. Whether you and I agree on the use of the prong isn't the point. It's a tool that I use, and I've learned that, if used correctly, it doesn't have to be used much at all. From there, I use the same variety of methods that you do. And I usually teach from a positive-approach from the beginning. It's the variety and relationship that counts. You're using a variety. It doesn't sound like you're relying solely on the clicker either. And that is my point. When it boils down to it, what works for you is what you should stick to, and I think you agree that a one-method approach can limit a dog's progression to the next level.

[DNeff]

Mark,

You make a comment that you redirect your dogs' attention using a variety of methods -- relationship, treats, toys, etc. There's variation in your training method

Actually those are reinforcers, I would not consider them different methods. Clicker training can have hundreds of different reinforers. You do use your voice, your body, your clicker, toys, treats, play, etc... all while clicker training. I still consider all of that one method. Variety in reinforcers helps make a highly effective training program, no matter what type of program it is.

No I don't use just the clicker. You have to wean your dog off of your tools when you compete. Just the way you wean your dog off of a prong when you are getting ready to go in the ring.

I will agree that as a "whole" trainer you should have a variety of tools and methods that you know how to use. I know how to properly correct a dog. I know how to effectively use a prong. I am also an instructor so I need to be able to teach my student how to correct their dog effectively, if they should desire to do so. Now I may encourage them to use positive methods instead, but if they are set on doing it that way, I want them to do it effectively.
I also feel to make an educated decision in my training methods I needed to understand all the methods out there. How else can I determine which works best for myself and my dog.

I still feel you can use one method and progress to higher levels. Every method is actually full of a lot of different avenues and possibilities. As I stated in the beginning, clicker trainers can do "pure shaping" or only "marking" the correct behavior. Some clicker trainers refuse to use a non-reward mark like "no" or "oops". Others feel it is essential to helping the dog understand the exercise. All of it falls under one category: clicker training. It can be a broad or as narrow and focused as you want to make it. Yes a narrowly focused method can have issues, any narrowly focused method, not just clickers.

Valdes43,
I'm from NoVa orginially, Manassas in fact. Don't get up there much, but it would be great to meet another rottie.net person sometime.

Dawn
U-CD Cammcastle's Mystic Riven CD NA NAJ NAC NGC NJC CGC TT
http://members.aol.com/dplantier
http://members.aol.com/rottweilerworld

[valdes43]

Originally posted by DNeff
As I stated in the beginning, clicker trainers can do "pure shaping" or only "marking" the correct behavior. Some clicker trainers refuse to use a non-reward mark like "no" or "oops". Others feel it is essential to helping the dog understand the exercise. All of it falls under one category: clicker training. It can be a broad or as narrow and focused as you want to make it. Yes a narrowly focused method can have issues, any narrowly focused method, not just clickers.

Valdes43,
I'm from NoVa orginially, Manassas in fact. Don't get up there much, but it would be great to meet another rottie.net person sometime.

That non reward mark, I feel, is not negative unless of course you give off negative energy through tone of voice and posture, etc... I think Morgan Spector described it best when he called it an "extinction cue".

I like the fact that even though you prefer all positive techniques you have made that educated decision to understand all techniques. I, personally cannot stand the forced retrieve but am currently helping my friend train his dog with the method because he feels it'll give him reliability. I think you run the same risks doing one motivationally (actually both styles are motivational, just one is postive and one is negative) as you would do force.


I'm in Woodbridge. A couple miles away from Potomac Mills. I shouldn't be far from you at all. Maybe soon.

[Mattweiser]

I do not believe that a dog can be trained through SchH protection routines using "Purely Positive" training

having said that, allow me to qualify that:

"Purely Positive" means no negative training period... to include P+ and P-.. Clinically speaking that includes telling a dog "no" or "pfui"… perhaps I understand nothing about dog training, but I would like to watch a dog being taught the H&B, and see a method of teaching where the dog would in no way perceive some action as a correction. Keep in mind that whether or not a stimulus is a correction or not, resides in the interpretation of the affected animal, not the trainer! ;)

[mark_sherry]

Keep in mind that whether or not a stimulus is a correction or not, resides in the interpretation of the affected animal, not the trainer!

This is what I was referring to, Dawn, when I spoke of using a variety of methods. I'm not sure that you can equivocally state that a dog sees these as reinforcers only. I think that they would be considered corrections as well. Again, it's not the implement that counts, but the variety that gives you overall reliability in the dog to perform at its highest level. I do understand clicker training at its core value, and it does have a solid place in the overall training program. We've gotten off of the spirit of the original issue which is "clicker-only." I still don't see clicker-only training taking a dog to a higher level. But like I've said several times before, I'm not an expert, just a willing student of the game. If it's proven to be a better way of training, then I'll put away the prong collar, shut my mouth on verbal corrections, and buy a case of clickers.

[DNeff]

Actually, they way we have been discussing positive training relates to the use of physical corrections.
There are many definitions out there for "purely positive". It depends on who you ask, and how they train.
I agree that any training method that does not include any type of "punishment" in the technical sense ie. positive punishment or negative punshiment would be severely limited.

Dawn
U-CD Cammcastle's Mystic Riven CD NA NAJ NAC NGC NJC TT CGC
http://members.aol.com/dplantier
http://members.aol.comrottweilerworld

[DNeff]

I'm not sure that you can equivocally state that a dog sees these as reinforcers only.

I'm the first to say admit that a voice, petting, playing can be a punishment for dogs if delivered to a dog who does not like them. My dog hates getting kissed. Giving him a kiss is punishment. So I don't use that for a reinforcer. You voice can be a puinsher, so can food if your dog is interested in a ball or a game at that point.

If you can't tell what your dog considers a reinforcer or a punisher you will never be very successful in training. That's part of being a good trainer. Learning what you dog considers a reinforcer and what they consider a punsihment. It may not even be the same from day to day.

Mark, Do you mean using P+ P- , R+ R- as using a variety of methods? You can use all of those in clicker training or any form of training. Clicker does not necessary exlude P+ or P -. Maybe you can clarify.

For those out there that might not know P + is positive punishment, P- is negative punishment, R+ is positive reinforcement and R - is negative reinforcement.

Dawn
U-CD Cammcastle's Mystic Riven CD NA NAJ NAC NGC NJC TT CGC
http://members.aol.com/dplantier
http://members.aol.comrottweilerworld

[mark_sherry]

Dawn,
That's a good question. I think in the due course of our discussion, I've implied (maybe not stated) that I'm referring to both P+/P- and R+/R-. It's all situational for me, so I really haven't defined either in regards to a specified method. Again, I'm working through this as well and am learning a lot through this conversation. To me, the prong collar works best for P+ and P- corrections. And I may be wrong, but the lack of use of the prong when it's on is an R+. To me, this is what would give the dogs a sense of joy to train with it. And I think that's the problem with a lot of novices who use a prong. They don't understand how it is supposed to be used, and they end up screwing up or abusing the dog in the process. But if done correctly, I think that it can be a useful P- tool and a R+ tool. I dunno. I look forward to your input about these comments though.

[valdes43]

Originally posted by mark_sherry
And I may be wrong, but the lack of use of the prong when it's on is an R+. To me, this is what would give the dogs a sense of joy to train with it. And I think that's the problem with a lot of novices who use a prong. They don't understand how it is supposed to be used, and they end up screwing up or abusing the dog in the process. But if done correctly, I think that it can be a useful P- tool and a R+ tool. I dunno. I look forward to your input about these comments though.

The prong is not a R+. The abscense of a prong period and not rewarding undesireable behavior is R-. Prong can be either P+ or P- , never R+.

That gap of non punishment your referring to is since no negativ stimulation is happening then the dog must be right. It's just the absense of the aversive when the behavior is correct.

[mark_sherry]

That gap of non punishment your referring to is since no negativ stimulation is happening then the dog must be right. It's just the absense of the aversive when the behavior is correct.

When you put it that way, I see what you're saying. But can't the absence of a correction while in a prong be construed as a R+ to the dog? Aren't you positively reinforcing behavior by not applying a correction with the implement? If I'm wrong, it's a matter of how I'm seeing these terms in their current use, not in how I use them to train.

[DNeff]

Here some quick definitions that will help somewhat.

Positive Reinforcement (R+): Adding something to the situation that will increase the chances of that behavior reoccuring. Giving a treat to a dog for sitting.

Negative Reinforcement (R-): Removing something from a situation that will increase the chances of that behavior reoccuring. Pinching a dog's ear until they take the dumbbell, then removing your fingers when they have the DB.

Positive Punishment (P+): Adding something to a situation that will decrease the chances of a behavior reoccuring. A collar pop.

Negative Punishment (P-): Removing soemthing from a situation that will decrease the chances of a behavior reoccuring. Walking away from (shunning) a jumping dog.

To be R + it must add something to the situation. You are not adding anything to the situation. There is a lack of anything happening. It can't be considered R + because of that.

Dawn
U-CD Cammcastle's Mystic Riven CD NA NAJ NAC NGC NJC TT CGC
http://members.aol.com/dplantier
http://members.aol.comrottweilerworld

[Mattweiser]

Originally posted by mark_sherry


When you put it that way, I see what you're saying. But can't the absence of a correction while in a prong be construed as a R+ to the dog? Aren't you positively reinforcing behavior by not applying a correction with the implement? If I'm wrong, it's a matter of how I'm seeing these terms in their current use, not in how I use them to train.

No.. the absence of a punisher is not a reinforcement, simply not a punishment P-..