Dropping the Labels - Long (Caution!)

And more importantly, your statement: "My preference is for the shorter, simpler version. Less stuff gets involved in a possible chain."

 
I'm not sure I can make it any more plain than that. Firstly, my preference, which you can do nothing about. Secondly, my view that what I'm doing with sit should involve fewer actions that could possibly be in a chain. Though I did generalize that view into a principle of training. It seemed logical. Fewer steps, especially for basic obedience, would have less chances to go wrong, like a machine with only one or two moving parts. Not much can go wrong. I'm not saying that you can't train your dog to sit with leash pops, hands on the rump, or do a redirection that leads to a sit. Many people have. If it helps you feel better, I saw plenty of people at the Dog Day popping leashes, using prong collars, and one guy smacked the rump of his dog but not too hard. So, you're not alone. In fact, people with a physical bent on training out-number me, I suppose.
 
Do you think I just generalized? My bad. I'll just let others go on thinking I'm an unscientific treat-dispensing butler without a clue to dog behavior because I don't pop a collar.
 
Here, in the clicker training section, many of us prefer clicker training and do have science on our side, though we do allow dissension, unlike the other section, where dissenting against CM is against the rules. As for experience, Koehler was a trainer for decades and yet his style is very much out of favor with many on this board, who have found his ways to be wrong. You just happen to have me in your gunsights at the moment.[]

And now I have a newbie telling me I'm not scientific...

I'll just let others go on thinking I'm an unscientific treat-dispensing butler without a clue to dog behavior because I don't pop a collar.

Sounds like being perceived as "not scientific" is your ultimate nightmare. [sm=biggrin.gif] In attempts to get you to consider the middle ground, you reply that such and such is more logical, "science on our side", etc. But science doesn't endorse anything, it investigates. Where in this forum did we ever have a scientific debate? It's just not a place for this kind of discourse. Or, if we did want to be "scientific", first thing we'd want to do is to consider the results that contradict your theory...

ORIGINAL: ron2

And more importantly, your statement: "My preference is for the shorter, simpler version. Less stuff gets involved in a possible chain."

I'm not sure I can make it any more plain than that. Firstly, my preference, which you can do nothing about. Secondly, my view that what I'm doing with sit should involve fewer actions that could possibly be in a chain. Though I did generalize that view into a principle of training.

Why would I want to do anything about your preference? I am not a missionary who is trying to convert. I am just trying to get you to acknowledge the existence of people who are neither CM-traditional or Clicker trainers.



Pssst, Ron, so is there a middle ground? [8|] [impatiently waiting on answer]

double

In attempts to get you to consider the middle ground, you reply that such and such is more logical, "science on our side", etc. But science doesn't endorse anything, it investigates. Where in this forum did we ever have a scientific debate

 
Semantics designed to discredit or devalue my scientific approach or the value of the science offered and it doesn't actually discredit or disprove the efficacy of clicker training. It just serves to steer the debate, hopefully in your direction. That is, when you can't discredit the science, call into question the proponent (me in this particular case.) This has more to do with debate tactics than actual science or understanding of operant conditioning. And just because not all of us are scientists or  dog-related professionals doesn't mean we can't discuss scientific topics. For a while in this forum, before you got here, we had briefly gotten into the habit of discussing peer-reviewed publishings, pros and cons, how we thought the study could be better, such as independent problem-solving with a string pulling experiment.
 
Not all corrections are purely physical or intentionally painful. Even I have pointed out training depending on lifestyle may be different than what we achieve in clicker training. As I have pointed in other threads, authentic sled dog training is quite different than our standard obedience training and the needs of that job require a dog that does not make a good housepet. And that quite a bit of the training is provided by other dogs on the team and these dogs don't carry clickers.
 
So, how many here are working their dogs? Shadow has a sled dog ancestry but we are not yet pulling carts, or bikes, or skates. (Though I plan on that in the future.) As a sled dog, he is a natural. Until I started working on the heel, I have fostered his innate ability to pull at least 4 times his weight. That's great if we are pulling a cart, not so great in a standard public setting. Even then, I was training him with mushing commands but not using collar pups, as I don't walk on the collar, I use a standard walking harness, the barrel shape or I-back. So, he knows his commands, including "about," which means to reverse direction. Sometimes, I would simply utter a directional command when he went in that direction, to connect the behavior with a cue. So even in the pre-clicker years, I was looking for the least invasive way. But it took a while to train that way. 1.5 to 2 years. With the results I get with clicker training, I could have achieved the results in a month. It has to do with motivation. And, IMO, a simplification of the teaching process and, for my part, better timing.
 
Nor have I always been averse to some physical training methods. I've read of where a properly trained and adjusted remote collar has helped in field dogs, who will be too far away to hear a click or whistle when they are focused on a target. Nor will one always use the clicker. It will fade as the behavior gets on cue. Whatever marker you use must be used to mark the successfully completed behavior followed by a reward. In time, the behavior will become classical or pavlovian. And you can keep it strong with an occasional reinforcer.
 
But, all in all, I think the more positive approach yields better results for the majority, who are primarily interested in companion dogs. I still think that marker training can be used to train working dogs, Rebecca's thread notwithstanding, my own example of traditional sled dog training notwithstanding. Though let it be said, with some credence to DPU, that sometimes, dogs can train other dogs very quickly. There are general rules and there are exceptions and we could dicuss those, ad infinitum.
 
Though the intial point of this thread was dropping labels, it morphed in a debate of clicker versus aversive. I think we'll always have labels and we can't escape some need for semantics, even if it's just to determine what we mean by alpha, leader, teacher, whathaveyou. Part of the scientific process is labeling, bean-counting they used to call. Quantifiable results with a meaning attached to them. Labeling and quantifying is what brought us are understanding of marker training.
 
Even the notion of projecting "calm, assertive energy" requires some meaning for those terms, even if it's in allegory or metaphor.
 

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.

 
Sorry, I hate to sound like so many of those CM threads, but this isn't a legitimate criticism of +R-only-training, it's user error.  An incompatible behavior doesn't FOLLOW the bad behavior, it takes its place. If the dog chased the cat, then lay down, why would you reward that? You would only reward the dog lying down when cued by the cat running away, without doing any chasing first, and without you ever having to issue a verbal command. I'd have the dog on leash so he couldn't practice chasing, put him in the presence of the cat, and click and reward any approximation of lying down in the cat's presence. Once the dog is habituated to lying down whenever a cat moves, you can gradually phase out the rewards.
Same with jumping up, then sitting, why would you reward that? you want the arrival of a new person to be the cue to sit.  Use environmental management (leashes are so handy) to prevent the dog from practicing the bad behavior while you train the incompatible behavior.
 
 

I unintentionally, though hilariously, created a chain with Marlowe when I was eating popcorn (which I do a lot!). Well, it was unintentional at first because I wasn't really paying attention what he was doing, I was just tossing popcorn at him for any behavior that was not all up in my grill begging. But then I realized what was happening and I had a running bet with my husband about how many behaviors I could add to the chain and I think we got up to 4 or 5 when it just started to get annoying (stand up, move to opposite side of couch, wipe paw, lay down, put head down). So I stopped rewarding for the chained behaviors and instead just began rewarding for a sustained down with no other behaviors tossed in. It really didn't take that long to change from the chain back to one simple behavior.

ORIGINAL: mudpuppy
An incompatible behavior doesn't FOLLOW the bad behavior, it takes its place.

Yes. The mechanics of this is an incredibly determined sticking point, and one that underscores the unhelpful nature of labels and "camps." Tina keeps asking about this, jen has a thread on it .... it's being asked about and evaded/glossed over/ignored contiuously in the forums! It put me on my soapbox a couple of months back (regarding the limitations of "reactivity" programs [:@]).

Let's allow ourselves to relax and talk about how positve we aim to be, by offering a seat at the table to the ghosts in the closet!

Let's stop having "dirty little secrets" revealed through "confessing" and "admitting" our real behaviors with shame and guilt. Let's simply acknowledge the fullness of life and how we behave in it!

In the spirit of Kim's thread, wouldn't it be great if we all typed less and self-reflected more?

I don't understand why people even talk about "camps" or "labels". Regardless of what you call yourself or believe about your training methods, every training method uses the same underlying principles of behavior modification. These are clearly laid out in Karen Pryor's don't shoot the dog.
 
That said, most training failures are due to user error-- poor timing, inadequate feedback, inconsistency-- rather than flaws in the methods.

ORIGINAL: nfowler

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

Thanks!

I was just thinking the same thing.  Heck, with Kim, ron, and you on this thread, I can just quit typing, eh?

ORIGINAL: mudpuppy

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.

Sorry, I hate to sound like so many of those CM threads, but this isn't a legitimate criticism of +R-only-training, it's user error.  An incompatible behavior doesn't FOLLOW the bad behavior, it takes its place. If the dog chased the cat, then lay down, why would you reward that? You would only reward the dog lying down when cued by the cat running away, without doing any chasing first, and without you ever having to issue a verbal command. I'd have the dog on leash so he couldn't practice chasing, put him in the presence of the cat, and click and reward any approximation of lying down in the cat's presence. Once the dog is habituated to lying down whenever a cat moves, you can gradually phase out the rewards.
Same with jumping up, then sitting, why would you reward that? you want the arrival of a new person to be the cue to sit.  Use environmental management (leashes are so handy) to prevent the dog from practicing the bad behavior while you train the incompatible behavior.

Exactly.  I've seen people make this type of error using lure/reward, and using correction - they accidentally correct a right response that follows a wrong one.  mudpuppy's point about how you reward, and with what timing is right on the money IMO.  And, the part about management is what sends this forum into a separate but equal tizzy sometimes.  Some people just can't see why they should have to manage or why they should not correct, but rather allow a behavior to extinguish.

Here's an example of someone who thought she was rewarding the sit, and rewarded the sit and paw.
[linkhttp://www.youtube.com/watch?v=za5IO00OAt0&watch_response]http://www.youtube.com/watch?v=za5IO00OAt0&watch_response[/link]
This is not a disaster at this early stage of learning, but the solution is to stop rewarding the sits that are accompanied by the paw raise.  There's really nothing wrong with her method, and provided she realizes that the dog is offering the chained behaviors rather than a it", she can fix it quickly.

Yes. The mechanics of this is an incredibly determined sticking point, and one that underscores the unhelpful nature of labels and "camps." Tina keeps asking about this, jen has a thread on it .... it's being asked about and evaded/glossed over/ignored contiuously in the forums! It put me on my soapbox a couple of months back (regarding the limitations of "reactivity" programs ).

Let's allow ourselves to relax and talk about how positve we aim to be, by offering a seat at the table to the ghosts in the closet!

Let's stop having "dirty little secrets" revealed through "confessing" and "admitting" our real behaviors with shame and guilt. Let's simply acknowledge the fullness of life and how we behave in it!

In the spirit of Kim's thread, wouldn't it be great if we all typed less and self-reflected more?

 
[sm=asking03.gif]
 
It may be a long time before people quit talking about what they like about clicker training in the clicker training section of this forum. Just a theory of mine, anyway ....
 

Hi Ron,

I'm not sure what your response means!
Isn't clicker training a part of "the fullness of life and how we behave in it"?
I think it is! []

Why would connecting the dots in all-things-dog hurt our enjoyment of clicker training?

I couldn't figure out the intent of your post. It seemed, to me, to be a non-sequitur, at best. At worst, are you implying that a dirty little secret for Dunbar is that one guy says he talked about a balance of punishment and reward back in the 90's? Now, I want to research and see if there are another sources of that besides that one page.

ETA:

Here are some words from the man himself.
[linkhttp://www.clickersolutions.com/articles/2001/macho.htm]http://www.clickersolutions.com/articles/2001/macho.htm[/link]
[linkhttp://www.clickersolutions.com/articles/2001/whycant.htm]http://www.clickersolutions.com/articles/2001/whycant.htm[/link]
 

brilliant links ron.

"Dogs are dogs lets treat them like dogs"