Dropping the Labels - Long (Caution!)

[sm=lol.gif]
 
So, you do know me?[]

They chain the jump and the sit together

Then "they" did it wrong. The problem was not with the method but the practioner. When I get Shadow to sit, it is not preceded by a jump. If were to jump, the response is off. That's one situation. The sit command is separate. I didn't realize that I needed to be clearer on that.

As for the use of semantics to sway a debate rather than discuss scientific facts, you proved my point with your take on my statement about "the sun is blue." Introducing the idea that the sun might appear a different color through a different filter is to change the tack of the debate and has nothing to do with the actual color of the sun, as shown by scientific evidence since the at least the Apollo Program. As well, your implication that I am not scientific or that my statements lack scientific credibility. And I'm not offended that you would respond as you did. Nor upset. Nor did anything you wrote cause me to shift one iota what I said.

Please, try again.[]

I guess I'm not much of a debate person, per se. I rely on science and logic which still maintain their integrity in spite of slippery semantics and amphigory. But I'm also the first to admit when I know I am wrong.
 
ETA:
The bit about shifting paradigms based on heretofore unseen factors was the reason for experimental theories of relativity. But Einstein was wrong. And he proved himself wrong with Quantum Mechanics. Now, there's a hairy subject for you.
 

Nicely done, as always. Goodness, you raise your own bar with each post.
 
Thanks!

ORIGINAL: ron2

They chain the jump and the sit together

Then "they" did it wrong. The problem was not with the method but the practioner.

Koehler used to claim that his training methods would produce happy enthusiastic working dogs with any dog with a proper temperament. Of course, then by definition, any dog who wasn't a happy enthusiastic worker using his methods had an improper temperament. Your circle of reasoning parallels (or, I guess is concentric to) his. Your method of choice will work for anyone who does it properly. Therefore, if it doesn't work, the practioner did it improperly. Cool. That's one way to achieve infallibility.

ORIGINAL: ron2

When I get Shadow to sit, it is not preceded by a jump. If were to jump, the response is off.

But, of course the dog doesn't innately know what "off" means. So, we're back to the chaining problem again. If you say "off" in a truly neutral tone of voice, it does nothing to stop the unwanted behavior, and if you follow that by a "sit" that gets rewarded, some clever dogs figure that jumping up is a great way to train you to say sit so they can get their reward. Sometimes the tone of voice that accompanies "off" is enough to eventually discourage the dog from from jumping. In that case, the tone of voice caused a decrease in the behavior, and hence was a punishment. A very mild punishment, but a punishment nonetheless. Sometimes tone of voice isn't sufficient, and one might also step toward the dog and invade his space enough that he can't really jump up without falling over (knowing how things tend to get exaggerated I want to be very clear here: I'm not talking about kneeing the dog, kicking the dog, pushing the dog, stepping on toes, or making physical contact at all. I'm talking about changing position so that the dog who is gathering himself up to jump suddenly is presented with the knowledge that angle of the jump is now so steep that he can't do it without losing balance). So, he learns that attempts to jump up will be unsuccessful, so better to just go directly to the sit and get the reward. The step into the dog's space decreased the jumping behavior by making an attempt to jump uncomfortable - a punishment. My position throughout this debate has been that sometimes mild punishments such as these do not result in a shut down, fearful dogs. In a lot of ways, this jumping up discussion is pretty trivial. All of my dogs have learned not to jump on people, and I can't even tell you how I teach that. I just don't allow it to happen. But I'm faced pretty regularly with students who've tried "no punishments under any circumstances" techniques, and after several months their dog is still knocking the kids over. We can tell them that well, they are obviously just bad trainers, just like Koehler used to tell his failed students that their dogs just had lousy temperaments. But, they are still going to be the trainers they are. If we present using any other technique as shutting their dog down, creating a zombie that is afraid to do anything but stand there, and that anything other than pure R+ is physically and psychologically battering the dog, well, most people won't try any other alternative method. But they'll still have a dog that jumps on the kids and annoys the guests. Add in that the dog still hasn't learned to walk politely on a leash, won't hold a wait command for more than two seconds, and has learned that barking is a great way to get you to tell him to hush so he can get a cookie, and they'll start locking the dog away when guests arrive, crating the dog where the neighbors won't hear the barking, and shoving him out the back door for excecise. And when that creates new problems, well, suddenly the kids will become "allergic" to the dog and we know where he's going to end up once those "allergies" kick in.

ORIGINAL: ron2

Introducing the idea that the sun might appear a different color through a different filter is to change the tack of the debate and has nothing to do with the actual color of the sun, as shown by scientific evidence since the at least the Apollo Program.

The Apollo Program, as well as many many other means of observation have shown what wavelengths of radiation are emitted from the sun. Color is a function of not only the nature of the emitter but also the nature of the receiver and the nature of the matrix through which the radiation travels. But that's a bit off topic.

ORIGINAL: ron2
As well, your implication that I am not scientific or that my statements lack scientific credibility. And I'm not offended that you would respond as you did. Nor upset. Nor did anything you wrote cause me to shift one iota what I said.

I stated that scientific reasoning always has to allow for the possibility of new evidence, and always has to allow for the possibility that new evidence may alter what we previously accepted as scientific "fact". Whether that describes your way of thinking or not is up to you.

So, back to my main question - isn't there some middle ground between pure Clicker Training and training that is pain/fear based, and inevitably leads to shut down fearful dogs?

Koehler used to claim that his training methods would produce happy enthusiastic working dogs with any dog with a proper temperament. Of course, then by definition, any dog who wasn't a happy enthusiastic worker using his methods had an improper temperament. Your circle of reasoning parallels (or, I guess is concentric to) his. Your method of choice will work for anyone who does it properly. Therefore, if it doesn't work, the practioner did it improperly

 
Now, that's an entertaining juxtaposition. Me and Koehler. So, I'll try one more time. I've trained Shadow to sit. It does not usually follow a jump. For him, off is also a trained command. I have also trained him to jump on me. By putting it on cue, I can lower the occurence by not giving the command. Nor am I against negative reward markers (NRM). And I think your move into the space to make the jump impossible is an elegant immediate solution that could work. I also think that the students you saw in the clicker class could create an accidental jump-sit chain possibly in another method. Regardless of the method, don't let that become the chain. Turning away and not aknowledging the jump, -P, also has an affect. The dog is deprived of your direct attention, which is something he wants. And so he tries a different behavior, such as keeping all fours on the ground and sniffs your knee.
 
Comparing me psychologically to Koehler does not invalidate clicker training or what I have said about it or the statement I made that the students you saw did it wrong. I once read a story about a woman who used clicker training to stop her dog from barking and accidently rewarded his barking. I overlooked her problem and tried it anyway. People can make mistakes. Even I make mistakes. When I wanted to try free-shaping, I made a simple mistake and couldn't get a response. I re-read the material, found my mistake, corrected it, then achieved real progress. Also, to begin with, my timing was quite off, because my timing until then was based on basic lure-reward and physical correction. Even now, I have been learning more fully the four quadrants of operant conditioning, of which +P is a part. And any training method uses one or more of the quadrants.
 
And your other points about the blue sun are just for the sake of arguing. Of course, it wasn't fair of me to introduce the rest of what you're talking about and then, perhaps, get away from your comfort zone by mentioning QM and imply the EPR Event. But that's what I get for reciting facts rather than just debating for the sake of debating. I thought it was kind of cute, you trying to teach me something about physics. Next, you'll start teaching me about electrical theory.[] You're new here and don't really know me so this can be entertaining. Others will hopefully keep quiet while you teach me elemental or sub-atomic physics.
 
I have pointed several times that there is a range of +P from extremely mild to near death. I usually have to muzzle Shadow for a vet visit and may still have to in the future and I will accept that.
 
Now, if you're all that up on science, perhaps you can tell me your views on Pryor's study from the 70's, archived in a link that was given here.

Koehler used to claim that his training methods would produce happy enthusiastic working dogs with any dog with a proper temperament. Of course, then by definition, any dog who wasn't a happy enthusiastic worker using his methods had an improper temperament. Your circle of reasoning parallels (or, I guess is concentric to) his. Your method of choice will work for anyone who does it properly. Therefore, if it doesn't work, the practioner did it improperly. Cool. That's one way to achieve infallibility.

 
That's great. [][]
Dogs usually learn what we teach them; unfortunately, often what we intend to teach them and what we actually teach them are not the same thing. The dog's behavior is our only guide to what the dog actually learned.

ORIGINAL: buster the show dog
Funny thing about science though.... A scientific approach has to ALWAYS allow for the possibility of rejecting a previously held belief in the light of new evidence, and one always has to be open to the possiblility of that new evidence existing. One always has to consider the possibilty that one's empirical observations under a given set of circumstances may not apply once one starts altering some of the specific circumstances under which the observations were made.

[sm=wink2.gif] Thanks for your thoughtful posts, Buster! I was writing something about observation, assumptions, belief and inquiry, and figured here would be a good place to put part of it. []

Clicker training comes to us by way of science, good science, at that! What is good science?

Science is experience. To think that a “lack of education” makes a person unqualified to have accurate “observations” and conclusions is to miss the pudding for the packaging. Some assume that preexisting knowledge must be applied to a current event, in order to shape correct observation. Others defend a “science-based” program by claiming it is unproductive to inquire into the mechanics of one#%92s “belief” about them.

Science begins with observation. Observation of phenomena in the physical world leads us to ask questions, form hypotheses (tentative conclusions), then devise experiments to test the hypotheses. If we go into “observation” already with an answer, then we are not practicing science, we are practicing belief and faith. Forms of logical argument that would accept “having the answers first,” use different types of inference, and include logical arguments regarding content other than science, like ethics (i.e., legal argumentation) or metaphysics (i.e., theology).

What we call the scientific method, is the basis of human learning:
see it > ask about it > draw a conclusion > test the conclusion
In scientific thinking, the fourth step (testing) is required. For other forms of logic, like legal argumentation or philosophy, it is not.

Now, if we go to school, or a class, or read a book, we receive the conclusions of someone else who did the science, well or poorly. So, as students, we are really consumers of other people#%92s logical processes. If we are practicing an action in the world with a form and a history, say dog training, and we simply follow the form, then we are not doing science, we are not engaged in “inquiry”, we are applying the products of inquiry: practicing a form, and a highly subjective one at that.

So, how do we develop great skills of observation? I like to think about a few particular groups of people who practice something very close to the scientific method: artists, buddhists, scientists, and people who still rely on direct relationships with the earth for sustenance. Farmers, bushmen, meditators, researchers, and the landscape painter are some of the people who truly work and practice to develop their “powers of observation.”

Extensive hours spent sitting together, watching and comparing notes, makes the Nunamiut hunters able to you much more about the particulars of an individual wolf than a PhD in animal sciences could ever hope to know. A painter will identify dozens more shades of green in the canopy of a forest due to his practised eye than an MFA web designer who#%92s visual expertise is limited to RGB screens. A farmer can tell you far more about that eastern horizon he#%92s studied for 1,460 evenings, than a guy with a B.S. in Meteorology.

And then, the proofs are in the pudding - repeated testing brings validity. Skinner devised a chamber for repeating his now famous experiments. Theoretical scientists, who are confident in their conclusions, offer them in papers for review by the community, where flaws are drawn out. Millan spent part of his childhood performing the scientific method on a farm in Mexico, observing, asking questions, drawing conclusions, and testing them. These are the standard forms of testing hypotheses: experimentation and peer review (i.e., trying stuff or talking over the fence with a neighbor). Think of a courtroom, the lawyers do not practice the scientific method, they already desire a specific outcome. But the jury must be "prudent", must be open to hear both sides, must argue and defend with each other the merits of the arguments.

So, how do *we* practice science? The good news here, is that most anyone *can* express this potential of human intelligence! It starts with being open, and curious, and asking questions! Yes, you have to be brave to let go of your biases, and admit you don#%92t know it all. But in exchange for your humility, a bountiful world awaits you, and it begins at the outer surface of your eyeballs, at the tip of your fingers, in the silence between the sounds, in each breath you draw, and on the tip of your tongue. []

This is why I applaud the sentiments Kim has expressed in her initial post:

ORIGINAL: Kim_MacMillan
Besides, creating a “philosophy”, in writing, for myself is going to be much fun, it will really press my own abilities as a writer, and to test what I really know, and it will even I#%92m sure send me back to some good books for references. In the end, it#%92ll keep me learning. And the best thing, is that I#%92m sure it will be adapting over time, I#%92ll add things, change things, remove things. That#%92s the great thing about philosophies, for me they are not set in stone (like labels are), they are very fluid, dynamic, complex.

I find these sentiments brave and wish Kim a bountiful experience in this spirit of inquiry! She has laid a path of experimentation before her that will offer her the chance to often see the world with new eyes!!

some clever dogs figure that jumping up is a great way to train you to say sit so they can get their reward

YES! That is what happens when I try to get Lucy to stop an unwanted behavior by "asking for an incompatible behavior" that was recommended in all my books and all over the message boards. If I asked for a down-stay to get her to stop chasing the cat more than two or three times - she increased the unwanted behavior in order to get that reward for the alternate behavior. I also asked for and rewarded the alternate behavior at random times and tried alternating the alternate behaviors and that didn't help either. I had to add a body block and stern "out" command to "punish" the unwanted behavior consistently *then* ask randomly for another behavior. It was so bad when we were camping, that she'd run out and bark- then run back to me and lay at my feet before I could even ask for the down-stay. She learned that after she barked she was supposed to go into a down-stay. She did not learn that she wasn't supposed to bark.

It starts with being open, and curious, and asking questions! Yes, you have to be brave to let go of your biases, and admit you don't know it all.

That's exactly what I do and it is why I tried clicker training, a method with a proven scientific basis, as well as neighbors talking over the fence kind of common sense. And now I have a newbie telling me I'm not scientific and trying to parse my posts into convenient non-sequiturs in order to "debate" my logic, rather than actually debate or discuss the science.

I think discussion is great and we can each state our opinion why we think our method is best but we have to be mindful that someone else may not agree with that and that's okay, too. But I cannot, and this may be my personal defect, be able to compare CM to Pryor. Certainly, he has lots of hands on experience. It doesn't always make him right. Just as having a certain set of credentials doesn't guarantee correctness. But what if we accept a combo of creds and experience?  How about quantifiable results with methods anyone can use, no disclaimers?
 
When I started clicker training, the results were so impressive to me that, IMHO, there is no debate. And that is why I quit debating and concentrated on results. And I use the scientific method, sometimes to a fault. Notice, I said, IMHO, in my humble opinion. Now, I expect to take some heat for that one. But that's okay. I'm a big boy and I have concerns far more pressing than whether or not someone likes it if I don't find much reason to compare Pryor to Koehler or CM or any of the more predominately physical and +P trainers.
 
 
As for creds, a behaviorist has the academic record to go along with whatever experience. Anything else is self-defined and even more subject to debate.

But the question still remains (and right on topic):

So, back to my main question - isn't there some middle ground between pure Clicker Training and training that is pain/fear based, and inevitably leads to shut down fearful dogs?

It's like in one thread I started, I asked numerous times what's pack drive got to do with training, and the conversation stopped. I am still waiting to hear if it's relevant...

So, we're back to the chaining problem again

I thought of another response to this, debating tactics or discussing debating tactics aside.

You picked apart my comparison of behavior chains from one style to another.

In training with a clicker, you cue the behavior and wait for completion. When completed, click and treat.

In a style with corrections, such as leash pops, physical manipulations, not all of which are bad, then lure, sit, reward, you have more events happening in the scenario. I chose the correctional scruff, but it could just as easily be a leash pop, finger bite, a fairly firm push on the behind. Each additional step in creating the sit could become part of the behavior chain.

That is, rather than accusing me of setting up a straw boss (you'll find that definition in one of your debating books) or pick apart whether I said scruff, pop, or some other physical manipulative move, let's simply concentrate on the number of steps involved in training and achieving the sit. My preference is for the shorter, simpler version. Less stuff gets involved in a possible chain. Nor will every dog automatically chain a jump to a sit. I think working on sit in various conditions would prevent it from becoming chained to another particular behavior. When trained by itself, it can, in fact, be an incompatible behavior to a jump, either random or chained to something else.

You are saying that in achieving a SIT a number of steps are involved. Do you mean, initially training a dog to SIT? Or *once a dog learns to SIT* (learns the mechanics), teaching it to respond to a cue SIT when asked in certain situations? Figuring out this distinction will help me understand what steps you are talking about in "the number of steps involved". And more importantly, your statement: "My preference is for the shorter, simpler version. Less stuff gets involved in a possible chain."

ORIGINAL: ron2

So, we're back to the chaining problem again

.... Nor will every dog automatically chain a jump to a sit. I think working on sit in various conditions would prevent it from becoming chained to another particular behavior. When trained by itself, it can, in fact, be an incompatible behavior to a jump, either random or chained to something else.

Of course not every dog will automatically chain a jump to a sit. I've rarely encountered anyone who had a problem teaching a dog to sit, whether they were Clicker Trainers, trainers with a clicker, or lure trainers. My concern isn't with the ability of Clicker Trainers to teach dogs to do all sorts of useful and/or entertaining (for both the dog and human) behaviors. It's been my experience, based on 17 years of teaching dog training classes for two different training clubs, that Clicker Training doesn't work for everyone as a means to teach a dog NOT to do some behaviors. I chose dogs jumping on people as one example, but I could have chosen pulling on the leash, barking, grabbing food off the ground, as other common problems. The Clicker Trainer answer to these problems is to teach an incompatible behavior, and that works great. For some dogs and trainers. For others, not so much. Just above Jenhuedepohl describes her experience with teaching an incompatible behavior. Teaching the dog to perform the other behavior isn't the problem. The problem is that the incompatible behavior doesn't always automatically eliminate the unwanted behavior. The dog simply learns to perform both behaviors. It seems to me that from the dog's point of view this is a win-win situation: he gets to do something that he finds internally rewarding (jump, sniff, bark, pull...) AND then he gets to follow that up with another behavior that gets rewarded by the handler. Unfortunately, the dog doesn't figure out that the jumping is why he's now locked away from the kids and guests, that the pulling is why he's being taken on fewer and fewer shorter and shorter walks.... Sometimes, ignoring the unwanted behavior will lead to extinction, but it's been my observation that this isn't always the case.

One can argue that if only the trainer had better timing, or was more patient, or whatever, that Clicker Training would work to eliminate these unwanted behaviors. It's been my observation that at least some of the people I encounter in my classes have already been going the Clicker Trainer route for as much as a year or more, and the technique isn't working for the them with their level of skill, with their dog. One could certainly argue that they just need to become a better Clicker Trainer, but they are the person they are, and at some point I think it's better to recognize limitations and seek an alternative. It is my position, that for this person, and for this dog, it would be better to provide them with some other alternatives like perhaps harsh verbal correction, or a poke with a finger, or even a leash pop because the handler isn't going to magically become a different person with more skill. But if the handler isn't provided with some alternative to a method that isn't working for them, they are going to interact with their dog less because they find some of their dog's behaviors to be unpleasant at best.

Nowhere did I pick apart whether you said scruff or pop. What I have objected to is the use of terms like "smack, wrestle, instill pain and fear, hang, yank & spank, zap" being presented as only alternative to Clicker Training. I have also stated that it has been my observation that using some mild aversives in some situations does not result in a dog that is shut down, afraid to do anything but stand there, or repressed in his ability to make any decisions on his own.

I apologize to the board in general and to Kim in particular that this thread has been deflected from it's original purpose. I know that I contributed greatly to that deflection although it wasn't my intention at the time. But my purpose really isn't to win a "debate". I just want people who have found that one training method doesn't work for them to be able to try other methods without being "labeled" as practicing some highly exaggerated and extremist version of what they are actually doing.

ORIGINAL: TinaK

But the question still remains (and right on topic):

So, back to my main question - isn't there some middle ground between pure Clicker Training and training that is pain/fear based, and inevitably leads to shut down fearful dogs?

It's like in one thread I started, I asked numerous times what's pack drive got to do with training, and the conversation stopped. I am still waiting to hear if it's relevant...

Training is simply another term for teaching, and being trained is simply another term for learning..  Sometimes the teaching is done by the "student" observing behaviors and thus learning from them, sometimes the "teacher" goes out of their way to actively *teach* the behavior.

All social creatures must learn certain behaviors, either by being taught them directly, or picking them up from more experienced members of the social group, in order to live peacefully in said group. 

When you socialize a puppy around a variety of people (thus helping it learn to get around in the larger human social group), you are teaching it a lesson as sure as if you were teaching it how to sit politely front of the door before going outside

From Ron2:
When I started clicker training, the results were so impressive to me that, IMHO, there is no debate. And that is why I quit debating and concentrated on results. And I use the scientific method, sometimes to a fault. Notice, I said, IMHO, in my humble opinion. Now, I expect to take some heat for that one. But that's okay. I'm a big boy and I have concerns far more pressing than whether or not someone likes it if I don't find much reason to compare Pryor to Koehler or CM or any of the more predominately physical and +P trainers.

 
No kidding! Some of these threads, and posts, remind me of nothing but naval gazing. Hello! It's summer! Go outside and barbecue! Go do some Rally O with your dogs. Go for a hike. Anything.
 
And then, when you all want to post, add in what you've found to be true for you . . .