Life is full of corrections

[KelliePup]

Spock would train very logically, scientifically, and understand Newton's First Law: For every action, there is an opposite and equal reaction. (BTW, thank you soooo much Tea for bringing Star Trek into this discussion. I love the Original Series!)

 

Fair warning, I'm about to become technical...

 

With both training modules, you are still using operant conditioning, and that consists of four quadrants: positive reinforcement (R+), negative reinforcement (R-), positive punishment (P+), and negative punishment(P-). If we accent the premise that positive means adding something while negative means removing something, then the four can be paired together like this: R+ = P- ; P+ = R- (so the = sign isn't the best, I'm a bit limited in my symbols).

 

What this basically means is that you cannot operate positive reinforcement with out negative punishment, and vice versa. The same is true with positive punishment and negative reinforcement. Some quick examples, and we'll use teaching the same behavior for consistency's sake (and I'm not saying this one or that one is better at this point, I'll get to that later):

 

Problem at hand: Dog is reactive (barking and charging at the end of the leash) when within sight of another four legged animal (insert dog, cat, squirrel, sheep, or other four legged species)

 

R+ = P- method (one of several, I know, but I'm using extremes for these examples):

 

Allow the dog to flop about at the end of the lead without either party moving closer(you or the other animal assisting you, moving away from each other is okay) until the dog calms down (P- because the dog wants to go to the other animal and is being denied access. The access denial is the punishment removal in this scenario, and is exaggerated if the other animal goes out of sight). Once the dog becomes calm, give treat, toy, or move a step closer (R+ because the dog gets something for being calm). And repeat until the dog consistently remains calm.

 

P+ = R- method (again, one of several that come to mind):

 

Physically force the dog to the ground and hold without either party moving until the dog stops thrashing (P+ because you are adding to the punishment). When the dog becomes calm, remove the pressure you were using to hold the dog down (R- because you are removing the unpleasant sensation of restraint). And repeat until the dog consistently remains calm.

 

 

Granted, the above is a simplistic representation of what could potentially be a more complex problem, so here's another that really is simple:

 

Problem at hand: Teaching the dog to sit

 

R+ = P- method

 

You have something the dog wants, you withhold that something (P-) until the dog sits. When the dog sits, you give him a treat, toy, open the back door, etc. (R+).

 

P+ = R- method

 

You say "sit," wait a second, and then pull up on the dog's collar (P+) until the dog sits. Once that happens, you release the pressure on the collar (R-).

 

 

Really, the reinforcements and punishments (or corrections) go hand in hand. There are consequences in both for the dog's actions. On that premise, I would argue that there really is no such thing as a "positive-only" trainer because, even through environmental manipulation, that trainer is still employing negative punishment. A correction based on what the dog is not getting, if you will.

 

Herding is really a combination of all four of the quadrants happening almost simultaneously in their pairings. When we put pressure on the dog as a correction, we're really doing two things at once: denying access (P-) and putting on pressure (P+). Likewise, when the correction is over we're removing the pressure (R-) and allowing access/work to continue (R+). So, I guess unless you're just beating your dog and then not letting him work after every infraction, or letting your dog to just flop on a line until he calms down before moving closer to the stock, herding really employs a balanced method of training.

 

 

Now for the kicker, which training method is best. My opinion might be a little unorthodox, but here it is: The best training method is the one that works for that particular dog.

 

I think, all too often, we forget that dogs are individuals, not a collective or a hive, and, as such, each individual dog might have his own way of learning. I've certainly seen it where a dog couldn't give rat's bum about treats, toys, environmental manipulation, or anything really for certain behaviors. The dog needed a sudden shock (not talking about e collars, BTW, but a sudden noise or poke) to her system to get her brain working and back on the "job." I suppose I'd equate it to daydreaming... continuously. That one was a very interesting case, and difficult to say the least because the dog just didn't care; meanwhile, the owner thought the dog was just being stubborn, but it had a happy ending after a lot of work that involved both adversives and building value in reinforcers. I kind of wish I had been allowed to tape those sessions...

 

Anyway, back on track, IMHO, a truly skilled trainer will use the method that is appropriate to communicate the skills being taught to the dog. Add the owner into the mix, and now you have to consider not only the dog's learning method, but the owner's too and figure out how to communicate to both as a unique team, which is it's own special challenge.

 

Basically, if one way isn't getting through to the dog, try another, and then another if it didn't work, and so on in varying degrees. Training is a very personal experience, it will vary from dog to dog, so personalize it and customize it to fit your goals. You should not be bound by merely one way of training, that's like running an experiment over and over again without changing the variables, and if you don't change the variables, then just how are you going to figure out how to grow the biggest tomatoes on the moon?

 

I really hope I explained this well, still a touch sick with the flu...

 

And McCoy says...? (Aside from get off the internet and back in Sickbay)

[ThunderHill]

Ah. Que sehlat, sehlat?

 

 

Really, I don't think Spock would train a dog, now that Vulcan bear like creature, maybe!

[KelliePup]

If I remember right, Spock had one as a pet... name started with and "E" and was featured in the Animated Series during a Guardian of Forever episode...

[ThunderHill]

Returning aaaaaaalllll the way back to the original post, I have some reservations about how the (indirectly) cited conceptual framework is being interpreted.

 

First, is the "resistance" an infant experiences in the formative stages of self-identification necessarily negative? The polarization of this thread into positive/corrective training factions assumes that it is. However, the word "resistance" seems to have been carefully selected for minimal negative connotations. (At least as far as dog training is concerned.)

 

And second, is the process of an infant's self-establishing of personal identity/boundaries through autonomous exploration really comparable to training imposed from the outside for purposes of behaviorial programming?

 

Just askin'.

 

Liz S in South Central PA

[geonni banner]

I find this whole thread very entertaining

 

So far, and someone correct me if I am wrong , noone has mentioned that what works for one dog might not work for another??????????

Well... Should we correct you with a leash-pop or simply withhold the cookie until you are right?

 

And akshually I did mention the business about dogs responding differently to the same method, way back on page 2.

 

That was before Spock arrived, and before pork was discussed and before the cat-fight started. It is an entertaining thread, isn't it? Except for the Jerry Springer overtones. Oh well. I think we all knew that nothing was likely to be learned/solved by this thread.

 

Does anyone know if there are dogs that work tribbles? Or would they just get squashed by the cascades of baby tribbles? Anybody for BBQed tribble with a side of Italian Dog Pasta?

[Carson Crazies]

And second, is the process of an infant's self-establishing of personal identity/boundaries through autonomous exploration really comparable to training imposed from the outside for purposes of behaviorial programming?

 

This ^^^. Also, can we really compare the dog's sense of self to a child's sense of self? Some would argue that dogs don't have a sense of self vs. other.

[geonni banner]

Uh-oh... Is Spock a CAT person?

Or is he a Sabre-toothed Rottie-bear person? Who would have thought Spock would go for a designer dog?

[juliepoudrier]

I think, all too often, we forget that dogs are individuals, not a collective or a hive,

Well, unless the Borg decided to assimilate dogs too.

 

J.

[KelliePup]

Well, unless the Borg decided to assimilate dogs too.

 

J.

 

"Biological and technological distinctiveness..."

 

I guess that narrows it down to Border Collies

 

But, awww... I thought we were just going with TOS, now we have the Next Generation!

[Tea]

"Dam it Jim! I'm a Doctor not a dog trainer!"

 

 

 

Sorry....

 

 

 

had to

[juliepoudrier]

Tea,

:lol:

 

KP,

I'm an equal opportunity Trekkie!

 

J.

[geonni banner]

 

Where did the sheep-geezer wander off to?

[ThunderHill]

Those pesky peripatetic subterranean silicates just don't flock worth a d*mn.

 

Liz S

[beachdogz]

 

Now for the kicker, which training method is best. My opinion might be a little unorthodox, but here it is: The best training method is the one that works for that particular dog.

 

Anyway, back on track, IMHO, a truly skilled trainer will use the method that is appropriate to communicate the skills being taught to the dog. Add the owner into the mix, and now you have to consider not only the dog's learning method, but the owner's too and figure out how to communicate to both as a unique team, which is it's own special challenge.

 

Basically, if one way isn't getting through to the dog, try another, and then another if it didn't work, and so on in varying degrees. Training is a very personal experience, it will vary from dog to dog, so personalize it and customize it to fit your goals. You should not be bound by merely one way of training, that's like running an experiment over and over again without changing the variables, and if you don't change the variables, then just how are you going to figure out how to grow the biggest tomatoes on the moon?

 

 

Finally! This gets my vote for best post of the thread (and it has been really fun to read this whole thread...training, pork, and Spock!) Your explanation re operant conditioning was spot-on!

 

Years ago, I attended a well-known obedience seminar and the presenter began it by saying, "First thing everyone asks me is what method do you use? My answer is 'the whatever works best method'...that is my method." I learned more from her seminar than from any other I attended. I sometimes think people come up with "methods" and "gimmicks" to make a buck and make them famous. In the end, it all resorts back to just what Kellie Pup says. Me, personally, I like to stay away from extremes, and I like to land somewhere in the middle. You kid yourself if you think one method will work for every dog (or you haven't trained enough dogs!)

 

Thanks, Kellipup. Well done!

[terrecar]

The access denial is the punishment removal in this scenario, and is exaggerated if the other animal goes out of sight). Once the dog becomes calm, give treat, toy, or move a step closer (R+ because the dog gets something for being calm). And repeat until the dog consistently remains calm.

 

Incredibly well explained. The only questionable sentence I can find is the one above, where I assume you meant "The access denial is the punishment "by" removal in this scenario.."?

 

IOW, punishment by contingent withdrawal.

 

It could be that I'm misunderstanding...

 

In any case, great explanation and good points. Go straight to Walden II and do not pass go

[Heather Rae]

"Really, the reinforcements and punishments (or corrections) go hand in hand. There are consequences in both for the dog's actions. On that premise, I would argue that there really is no such thing as a "positive-only" trainer because, even through environmental manipulation, that trainer is still employing negative punishment. A correction based on what the dog is not getting, if you will."

 

This was something that I was thinking about too. Just semantics because I get what Positive Only means, I do think it's incorrect/misleading term though.

 

 

 

Beyond that though. I've read this whole thread and found it fascinating. I personally don't see anything wrong with Ah-ah's, No's, partly because I feel the need to teach my dog what "no" means as it's my first reaction when I'm afraid for my dog. I, however, don't see anything wrong with positive only training, if the trainer reaches the level of obedience that they want, whatever that may mean.

 

Frankly, as long as the dog seems to still enjoy being with the trainer and what ever behavior desired is offered, I don't typically frown on it, even if I don't use it.

 

Myself I tend to avoid physical punishment just out of preference, more over I know some dogs simply don't respond to physical punishment (or at least physical punishment most people are willing to dole out: leash pops, ear pinches). Even if I was sure they would have some response to physical correction, I avoid it because I rather not risk damaging my dog because of my lack of experience and knowledge.

 

But... one thing... It was a while back that I thought I read someone implying that saying "ahh, ahh" or "no" isn't giving useful information to the dog. I believe it is. In fact I consider it as the prelude to "You're not going to get the reward". I actually consider it some what gentler than no treat at all. It is possible that this prolongs the learning time for the dog but its too natural for me to warn the dog they are heading in the wrong direction in attempts to help it be successful.

 

Either way, corrections or not... as long as the owner and the dog are happy, I can't disagree, but I have my preferences and boundaries.

[KelliePup]

Incredibly well explained. The only questionable sentence I can find is the one above, where I assume you meant "The access denial is the punishment "by" removal in this scenario.."?

 

IOW, punishment by contingent withdrawal.

 

It could be that I'm misunderstanding...

 

In any case, great explanation and good points. Go straight to Walden II and do not pass go

 

Thank you

 

Yes, that. I did mention that I'm still sick, so my typing and proof reading skills aren't the best today... Too much brain activity required.

[Tea]

What is the Ah ah?

[Root Beer]

Problem at hand: Dog is reactive (barking and charging at the end of the leash) when within sight of another four legged animal (insert dog, cat, squirrel, sheep, or other four legged species)

 

R+ = P- method (one of several, I know, but I'm using extremes for these examples):

 

Allow the dog to flop about at the end of the lead without either party moving closer(you or the other animal assisting you, moving away from each other is okay) until the dog calms down (P- because the dog wants to go to the other animal and is being denied access. The access denial is the punishment removal in this scenario, and is exaggerated if the other animal goes out of sight). Once the dog becomes calm, give treat, toy, or move a step closer (R+ because the dog gets something for being calm). And repeat until the dog consistently remains calm.

 

I realize you were deliberately using an extreme, but the vast majority of +R trainers would not actually recommend this as a +R method.

 

What you describe is very close to CAT, and while some +R trainers do recommend it (I know of exactly one), most would go with some form of systematic desensitization/counter conditioning where the dog is always kept sub-threshold and reinforced for remaining calm in the presence of the trigger with a gradual decrease in distance between dog and trigger, moving closer only once the dog's actual response to the trigger starts to change.

 

Using an increase in distance from the trigger for remaining calm as a reinforcer is becoming more common, but very few +R trainers would recommend putting the dog over threshold to begin with.

 

Again, I know you were using an extreme to make a point. I just wanted to point this out lest anyone get the impression that what you describe above is "the norm" for rehabbing a reactive dog through reinforcement. Most +R trainers would not choose to handle it that way.

 

On a personal note, I shudder to think of what most likely would have happened to Speedy if someone had recommended helping his reactivity by the method you describe above and I had gone that route. He may well have ended up having to be put down. Thankfully, I learned DS/CC instead and it worked so well that I actually forget he was reactive at one point if I don't stop to think about it. He is happy, very well adjusted, and is more or less normal in almost all situations.

 

So, if I seem nitpicky, I beg pardon! This is something important to me personally, and not just as a trainer.

[terrecar]

Thank you

 

Yes, that. I did mention that I'm still sick, so my typing and proof reading skills aren't the best today... Too much brain activity required.

 

You did a helluva' lot better explanation than I could have attempted, irrespective of the typo. I 2nd the vote for best post!

[KelliePup]

No worries, Kristine. I can think of, and have used, dozens of other techniques for this type of reactivity. The most extreme R+/P- one being with my own dog, Kellie, and it wasn't the one I put above. The reasons behind the decision to do it was that nothing else was working in any way, shape, or form. It was painful and personal for me, but it worked because of Kellie's value system. I don't typically advertise it because it is extremely difficult on the owner and the dog, and most people don't have the constitution to reset the relationship like I had to with her. To my knowledge, no trainer recommends the procedure and protocols I put into effect with her.

 

That experience, for me, formed my belief in canine individuality and the need to be, as others have put it, flexible in training methodology. Other dogs since have only solidified it.

[terrecar]

What is the Ah ah?

 

Tea, if you've ever heard a pissed-off parakeet, the sound is kind of the same, only it's the human version. Like, ah ah, stop that!

 

That is of course not to be confused with Skinner's rats' "Aha!", as in, "Aha! I could have had a V8!"

[Tea]

I have never heard a pissed off parakeet

 

Sadly.

 

 

 

Ah Ah Stop that???????

 

 

hum this may be a cultural thing. I am going to let my brothers read this and see what they think of it.

 

 

 

Skinners Rats???? Lost me.... Aha I could of had a V-8 I know that one, Dont know why you say this to dogs. But maybe you don't and so thats why I wasn't supposed to get confused.

[Heather Rae]

"Ah, Ah" is just a noise made by a person that literally sounds like "ahhh, ahhh." Occasionally it sounds like "Uhh-uhh", but basically it means, "No, no, that's not right, try something else"

 

It's a verbal correction that warns a dog that they are doing something wrong. Some people don't use this verbal correction because it doesn't tell the dog what it is supposed to be doing.

[Donald McCaig]

Dear Doggers,

 

Kristine writes (in part): "The truth is, I'm not a dog. I don't communicate with dogs as if I were a dog, I communicate with them as a human with a dog.

 

Whether that is "classical" or not, it is simply the truth.

 

Personally, I am grateful that I am not limited to the way that dogs communicate with one another to train. As humans, we have quite a spectrum of ways of interacting and communicating with our dogs that are unique to us as humans.

 

So, the argument that "dogs correct each other" really doesn't make the use of correction any more attractive, nor even sensible, to me."

 

With due respect to Kristine who I know as a dedicated pet dog trainer, this reminds me of the English speaker getting his point across to the French speaker by raising his voice: "No, no I mean the train to Berlin. WHERE CAN I BUY A TICKET?!!!!"

 

Understanding dogs, how they see the world, what they value, what they want, hope for and fear is the key to all dog training. That understanding may not be and often is not articulable. I know many brilliant trainers who can't really explain how it is they do what they do. Others resort to more or less useful metaphors: "orey drive", "R+/P", "responsibility for his own actions" or the sheepdoggers' "He's a good listener".

 

I know sucessful intuitive trainers who advocate theories which when applied by ordinary pet owners with poor timing and no comprehension of what their dog is thinking must and do produce damaged dogs.

 

And I have met articulate humans, in full command of one or another training theory, who could never, on the best day of their lives, train a dog to come when called.

 

To paraphrase a philosopher, "Dog training theories without understanding are empty."

 

Donald McCaig