Jump to content
BC Boards

Recommended Posts

Ms. Rushdoggie wonders what part of BF Skinner's theories I don't like.

 

They're self referential, evangelical, discredited science. Rather than writing that essay I'll refer you to the Breland's fifty year old refutation.http://psychclassics.yorku.ca/Breland/misbehavior.htm

 

She also notes: I don't see a whole lot of marketing controversy anymore. Heck, even Leerburg uses a clicker nowadays . ."

 

Ah, can I suppose Karen Pryor has tried an ecollar?

 

In the early days of Cesar Milan's Nat Geo show, a concerted letter writing campaign was mounted by positive trainers to cancel that show. That effort was countered by a letter writing campaign by traditional and ecollar traiers to keep the show on the air. Ratings finally won that argument.

 

I assure you the war rages on with real consequences\. When Mr Traditional trainer complains that "cookie pusher" methods are ineffective and lists dogs Ms Positive screwed up he had to subsequently retrain and when Ms Positive says traditional methods are inhumane and cites "all those trainers helicoptering and hanging dogs" (I've heard those exact words from one of the most prominent behaviorists in the country), these opposing camps aren't just having intellectual disagreements - it's their status and livelihoods. While most pet dog trainers, like sheepdog trainers, could make far more remunerative use of their time, some pet dog trainers I know net six figures. Those who give seminars, workshops and write books often do even better.

 

Yes, the argument is about dogs by people who care deeply about dogs. Yes, it's about marketing.

 

Donald McCaig

Link to post
Share on other sites
  • Replies 263
  • Created
  • Last Reply

Top Posters In This Topic

 

And of course, as they say in the essay, this was not a wholesale refutation of operant conditioning by them, but rather their understanding of it's usefulness on a continuum with other aspects of animal behavior, like instinct. For those interested there is more on Marian (Breland) Bailey and Bob Bailey and the things they accomplished with animals here: http://www.behavior1.com/page8.html

 

B.

Link to post
Share on other sites
Clickers teach humans timing but mostly they clarify for the dog. In addition, there is research going on that suggests that there may be more. There are kinds of stimuli like flashing bright lights and sudden sharp sounds that reach the amygdala first, before getting to the thinking part of the brain. The click is that kind of stimulus. If this is true, another reason the clicker works so well could be that the click is processed by the brain much faster than any word can be. "Even in the most highly-trained animal or verbal person, the word must be recognized, and interpreted, before it can 'work;' and the effect of the word may be confounded by accompanying emotional signals, speaker identification clues, and other such built-in information." (from KP's website). I think thats kind of cool.

 

I'm not fluent with a clicker, but his makes perfect sense to me. I nearly always train a hand signal for basic obedience commands before I teach a verbal command. I just thought that dogs being such consummate readers of body language, that they would pick up a hand signal (or leg cue, etc.) more quickly. Dogs aren't language-oriented, so why use the communication device the two species have least in common? A dog reacts instantaneously to a snapping twig behind it - but will often not react at all to a spoken word. (It's the same with me.) It's a matter of wiring.

 

Hand signals have the added benefit of requiring the dog to look at you before a reward can happen. A dog that is taught with hand signals looks at me more, one that is trained with verbal commands frequently spends part of its time scanning its surroundings, with maybe one ear cocked back for a command.

 

Another benefit of hand-signal training is that I am paying closer attention to my hand/body cues, and thereby run less risk of inadvertently "mis-cueing" physically while trying to impart a verbal impression.

Link to post
Share on other sites
Ms. Rushdoggie wonders what part of BF Skinner's theories I don't like.

 

They're self referential, evangelical, discredited science. Rather than writing that essay I'll refer you to the Breland's fifty year old refutation.http://psychclassics.yorku.ca/Breland/misbehavior.htm

 

What I was getting at was: is it Skinner you don't like or do you discredit the concept of operant conditioning?

 

I've read and agree with the article you linked (its a classic), but the Baileys don't discredit OP as a whole. What suggest is that there are more factors when training in the world than one can replicate in a lab including instinct and intellect. I doubt any "positive" or "clicker" trainer would disagree.

 

I understand that BF Skinner was kind of a tool as a human being. But being a jerk doesn't discredit the basic tenets of OP: that R+, R-, P+ and P- have an effect on learning. This idea is a very basic framework that has been built on and integrated into other theories by Thorndike, Lorenz, Kohler (not the dog trainer) and the Baileys.

 

So I guess what I meant to ask was: do you discredit the concept of operant conditioning as a whole? is OP what you feel has been discredited? or just Skinner's narrow "black box" interpretation?

 

I am genuinely curious...not trying a sneak attack.

 

Ah, can I suppose Karen Pryor has tried an ecollar?

 

Oh, Mr. McCaig, sometimes you are such a Smartypants. :rolleyes:

 

KP didn't start out as a clicker trainer you know...

 

In the early days of Cesar Milan's Nat Geo show, a concerted letter writing campaign was mounted by positive trainers to cancel that show. That effort was countered by a letter writing campaign by traditional and ecollar traiers to keep the show on the air. Ratings finally won that argument.

 

And why did many people want to have the show taken off the air? Because many of the things CM does are a REALLY bad idea for JQP to try. I really don't want to turn this into a CM bash thread (theres enough of those), however I have a conflicting opinion regarding the "anti-Cesar" campaign. My perception is that wasn't about marketing, it was about the perception of abuse.

 

I assure you the war rages on with real consequences\. When Mr Traditional trainer complains that "cookie pusher" methods are ineffective and lists dogs Ms Positive screwed up he had to subsequently retrain and when Ms Positive says traditional methods are inhumane and cites "all those trainers helicoptering and hanging dogs" (I've heard those exact words from one of the most prominent behaviorists in the country), these opposing camps aren't just having intellectual disagreements - it's their status and livelihoods. While most pet dog trainers, like sheepdog trainers, could make far more remunerative use of their time, some pet dog trainers I know net six figures. Those who give seminars, workshops and write books often do even better.

 

While I would guess that the 'six figure' folks are really in the minority, I won't disagree with what you wrote here (shocking!! <vbg>). It might also be the part of the country where I live that I simply don't see too much controversy so I don't hear it as a loud chorus.

 

edited to try to fix the alarming number of spelling and grammar errors...yikes.

Link to post
Share on other sites

Fundamental dog training requires a basic skill set:

 

--Accurate observation

 

--Correct timing

 

--Appropriate feedback adjustment (The ability to process feedback from the subject and adjust method based on what is truly happening, rather than simply robotically imposing a pre-conceived system: a roundabout way of saying "judgment" or even "listening to the dog")

 

The vast preponderance of human students need to develop their basic training skills, whether their dog has a solid gold temperament or otherwise. The clicker is a very useful device for shaping observation and timing skills in humans. That "click" makes the human's opinion about when a behavior occurs public and unmistakeable: to the person clicking, to the other people present (peer pressure being what it is, a powerful learning stimulus), and to the human teacher. A clicker has the attractiveness (particularly to beginners) of making the process seem to be a matter of black and white, "yes" and "no." (A tempting oversimplification, but a place to start.) Provided the clicker itself isn't distressing to the dog (I've known many who would not tolerate the noise) it's mechanically consistent, and it's relatively neutral (compared to some other cues) even when misapplied.

 

But chiefly, the clicker clarifies FOR THE HUMAN exactly when and where observation and timing can be improved. People think they're getting better results because the dog learns better from the clicker; but in fact, it's generally the improvement in the handler's skills of observation and timing (i.e., training) which are producing better results. And yes, the clicker can often have excellent results in that regard: most humans who get as far as taking in a clicker class are highly responsive subjects.

 

Once you've learned what a clicker has to teach, you can get equally good results without it. Or continue to use it. It's a fine mechanical marker. It's also just another tool.

 

Unfortunately the clicker doesn't address matters of judgment and mutual feedback between handler and dog. The top percentage of trainers have that extra gift which takes them beyond the binary code. But whether or not they use a clicker, it's their gift which took them to the top, not one specific tool.

 

Liz S in South Central PA

Link to post
Share on other sites

I'm just curious, Liz - how much clicker training have you done? :rolleyes:

 

What did you use it for, exactly? Pet manners? Tricks? Sports? Behavior modification?

Link to post
Share on other sites

I've been to a couple of classes and at least one weekend-long seminar by one of the early proponents. (Sorry, I don't recall which one -- I've been training dogs, taking classes, reading books, and going to seminars and camps for for 32+ years.) And I've read (and own) a couple of the books. What's been done with clicker-training captive marine mammals is amazing; but like the hungry, sensory-deprived rat in a classic Skinner box, the situation doesn't translate well to the richer environments where I work with my own animals.

 

I've taken a clicker for a test spin with one of my dogs, because I could see definitely the value of it as a training aid, and he was a shy dog who loved learning tricks. It worked fine but wasn't superior to using an associative "yes!" leavened with favorite treat rewards. I've recommended a neutral auditory marker for associative reward-based training more than once, where it seemed appropriate to the circumstances. Whatever it takes to get a person to focus carefully on what their dog is *really* doing, and when, and the timing of their own actions. But for me, the clicker wasn't a superior tool for improving anything I needed improved, be it for fearful, oblivious or aggressive dogs; basic obedience; trick training; or competitive flyball (NADAC), rally (APDT and AKC) or agility (USDAA, NADAC and AKC). I muddled around with freestyle moves (fun for the right dog!) but never competed -- sheepwork came along, and I toddled off down the dark path without a backward glance. Traditional methods are working fine for me on stock. If my dogs learned the sheep dance any faster, they'd be training me. Come to think of it, they are.

 

For me, personally, the clicker has so far been a redundant tool. I still have a couple of them around but the noise annoys the heck out of me (even with middle-aged hearing loss) and now I have two dogs who are strongly averse to the sound as well. They're there if I need them but I have other tools I'd prefer to try first.

 

The high-powered obedience club I used to belong to in Washington, DC (Capitol Dog Training Club) actually banned clickers (during group class hours) despite having gone the cookie-power route in general. But I think that was because you can't have three rings worth of people and dogs under one resonating warehouse roof all clicking away without a major gonzo breakout erupting. It wasn't an objection to the soundness of the theory as such.

 

Not uninformed in South Central PA,

 

Liz S

Link to post
Share on other sites

I have to say, I am impressed by your breadth of knowledge on this subject.

 

I've been to a couple of classes and at least one weekend-long seminar by one of the early proponents. (Sorry, I don't recall which one -- I've been training dogs, taking classes, reading books, and going to seminars and camps for for 32+ years.) And I've read (and own) a couple of the books. What's been done with clicker-training captive marine mammals is amazing; but like the hungry, sensory-deprived rat in a classic Skinner box, the situation doesn't translate well to the richer environments where I work with my own animals.

 

I've taken a clicker for a test spin with one of my dogs, because I could see definitely the value of it as a training aid, and he was a shy dog who loved learning tricks. It worked fine but wasn't superior to using an associative "yes!" leavened with favorite treat rewards. I've recommended a neutral auditory marker for associative reward-based training more than once, where it seemed appropriate to the circumstances. Whatever it takes to get a person to focus carefully on what their dog is *really* doing, and when, and the timing of their own actions. But for me, the clicker wasn't a superior tool for improving anything I needed improved, be it for fearful, oblivious or aggressive dogs; basic obedience; trick training; or competitive flyball (NADAC), rally (APDT and AKC) or agility (USDAA, NADAC and AKC). I muddled around with freestyle moves (fun for the right dog!) but never competed -- sheepwork came along, and I toddled off down the dark path without a backward glance. Traditional methods are working fine for me on stock. If my dogs learned the sheep dance any faster, they'd be training me. Come to think of it, they are.

 

For me, personally, the clicker has so far been a redundant tool. I still have a couple of them around but the noise annoys the heck out of me (even with middle-aged hearing loss) and now I have two dogs who are strongly averse to the sound as well. They're there if I need them but I have other tools I'd prefer to try first.

 

The high-powered obedience club I used to belong to in Washington, DC (Capitol Dog Training Club) actually banned clickers (during group class hours) despite having gone the cookie-power route in general. But I think that was because you can't have three rings worth of people and dogs under one resonating warehouse roof all clicking away without a major gonzo breakout erupting. It wasn't an objection to the soundness of the theory as such.

 

Not uninformed in South Central PA,

 

Liz S

Link to post
Share on other sites

Thanks Liz. That makes our difference in perspectives on the clicker much more clear.

 

I wouldn't say you are uninformed, but our personal experience with the tool and the effect that we have seen it have on the dogs has been as different as night and day.

 

I can honestly say that there is no way I could have done some of the things I've done with the clicker - both with Dean and with Speedy - without it. Not even now, after learning what I needed to know about the clicker to make those transformations happen. Even after helping Speedy get from extremely fearful to very confident with the assistance of this tool, I find there is still more to learn than I have had the opportunity to learn to this point. And even after helping Dean transform from a highly motion triggered dog to one who can sit and watch pretty much anything go by calmly (yes, even when the clicker and treats are not present) with the assistance of this tool, I find there is still more to learn that I have had the opportunity to learn at this point.

 

Sure, teaching basic manner stuff and foundation concepts to a dog with a clicker is a snap now that I've done it a couple of times. That certainly doesn't make me want to ditch it for that purpose, though! :rolleyes: Quite the opposite. And for more advanced clicker concepts, there is still a lot that I'm looking forward to learning.

 

I would find training pretty much any higher level precision move for Freestyle without the clicker to be extremely burdensome. Possible? Probably. More trouble than it's worth - likely. When I need to communicate to the dog, "THAT is perfect backing" or "THAT is exactly the shift of position that goes there", I'd be hard pressed to give the same information with a verbal. Sure, it could work, probably with more repetitions and a slower raise in criteria. But I like the instant speed with which the click can convey that kind of information.

 

I have some students who prefer not to use the clicker in my classes and I don't make them. But it is a fact that, by and large, their dogs rarely "get" the concepts that the dogs that are given the introductory information through the clicker do. I can't think of a single student who chose a marker word instead of the clicker whose dog progressed at anywhere near the rate that the dogs of the handlers who used the clicker (even the completely green newbies who have never used a clicker before) have.

 

So, it really makes sense that I do view this as a tool that can actually convey information to the dog - just as a book can convey information to a human. No, a book never takes the place of the dynamic and personal exchange of information that takes place through spoken communication and the nonverbal component of that communication. But books, when used to their advantage, can help speed such learning along and provide information to the learner that goes beyond the scope of the spoken lesson. That's how I've seen the clicker work - over and over again.

 

And, like I said above, I do see it as an exchange of learning. Through clicker work, when it is done well, the dog has as much to teach the handler as the handler has to teach the dog.

 

I can understand why you don't see it that way.

Link to post
Share on other sites

I've wondered sometimes whether the reason clicker training is indeed often very effective for developing tricks isn't because dogs hate the racket. You'd get a terrific mixed reinforcement cocktail: the anticipation of an unpleasant stimulus the dog tries to avoid through the previously learned activity of offering random behaviors; the click itself (ouch! ick!); and then the withdrawal of the negative stimulus (click now over) PLUS the bonus (or anticipated bonus) of the associated treat and praise.

 

Dogs have to be darned smart. There's no other way we could play so many different complex headgames with them and still get the results we think we want.

 

Liz S

Link to post
Share on other sites
I've wondered sometimes whether the reason clicker training is indeed often very effective for developing tricks isn't because dogs hate the racket. You'd get a terrific mixed reinforcement cocktail: the anticipation of an unpleasant stimulus the dog tries to avoid through the previously learned activity of offering random behaviors; the click itself (ouch! ick!); and then the withdrawal of the negative stimulus (click now over) PLUS the bonus (or anticipated bonus) of the associated treat and praise.

 

I don't think so...I have only trained one dog who found the click of a clicker to be an uncomfortable stimuli. If you walk into my kitchen, pluck a clicker off of the fridge and press, you will have 3 dogs stampede each other to get to the kitchen. I was recently in an Open/Utility class with 6 people clicking 6 dogs and every dog was looking happy, animated and unstressed.

 

Thats not to say that some dogs don't like the sharp sound, but its a small percentage, I think.

 

I ask this question sincerely: Do you think there is any chance you don't get the results you like from a clicker because you personally find the sound offensive?

Link to post
Share on other sites

Dear Doggers,

 

Ms. Rushdoggie writes: "But being a jerk doesn't discredit the basic tenets of OP: that R+, R-, P+ and P- have an effect on learning. This idea is a very basic framework that has been built on and integrated into other theories by Thorndike, Lorenz, Kohler (not the dog trainer) and the Baileys.

 

So I guess what I meant to ask was: do you discredit the concept of operant conditioning as a whole? is OP what you feel has been discredited? or just Skinner's narrow "black box" interpretation?"

 

Do rewards and punishment have an effect on dog learning? Duh.

 

The problem with Skinner based theory isn’t that rewards and punishment effect learning it’s Skinner’s central assertion that (primarily food) rewards and punishment explain all mammalian learning. That isn’t true.

 

I don’t doubt that dog trainers have modified Skinner’s behaviorism. I am unfamiliar with Thorndike and Kohler but certainly Lorenz’s ethological investigations/posture has profoundly affected how rational dog trainers of any persuasion understand dogs.

 

But these modifications are erected on a oversimplified foundation. If you’ll permit a Mac user’s smug comparison: it’s like Windows 7. You can prop up a bad system and do work arounds and patches but if the foundation is inadequate, at some point its probably better to abandon the system and start over.

 

I often hear how far positive trainers have evolved from Skinner and how modern Behaviorism is different from Skinner’s “Radical Behaviorism”. But these same trainers use many of the same tools and make the same overblown claims Skinner did.

 

“This book is about how to train anyone – human or animal, young or old, oneself or others – to do anything that can and should be done. How to get the cat off the kitchen table or your grandmother to stop nagging you. How to affect behavior in your pets, your kids, your boss, your friends. How to improve your tennis stroke, your golf game, your math skills, your memory. All by using the principles of training with reinforcement.

“These principles are laws, like the laws of physics. They underlie all learning, teaching situations as surely as the law of gravity underlies the falling of an apple. Whenever we attempt to change behavior, in ourselves or in others, we are using these laws, whether we know it or not.”

 

Doubtless Ms. Rushdoggie recognizes the author.

 

Donald McCaig

Link to post
Share on other sites

I apologize in advance, but the scenario you just described would be a circle of he double tooth picks to me. How did all those dogs compete pre-clicker? I know one obedience instructor who doesn't use a clicker, but who is VERY good with timing, and a simple YES! is just what she uses. The dogs learn to love her voice, which is a side benefit, and look from approval from her, and not the clicker. At demos one time, someone gave away clickers to kids. OH MY LORD. There were about 150 clickers going off........

 

I don't think so...I have only trained one dog who found the click of a clicker to be an uncomfortable stimuli. If you walk into my kitchen, pluck a clicker off of the fridge and press, you will have 3 dogs stampede each other to get to the kitchen. I was recently in an Open/Utility class with 6 people clicking 6 dogs and every dog was looking happy, animated and unstressed.

 

Thats not to say that some dogs don't like the sharp sound, but its a small percentage, I think.

 

I ask this question sincerely: Do you think there is any chance you don't get the results you like from a clicker because you personally find the sound offensive?

Link to post
Share on other sites
I ask this question sincerely: Do you think there is any chance you don't get the results you like from a clicker because you personally find the sound offensive?

 

Actually, what I said was that I can get fine results using a clicker.

 

However, I can get the same good results without one. (Possibly because I already pay close attention to my dogs and I have good timing.) If I were training captive dolphins, or chickens, or some of the other situations where clicker training has had what appears to be unique success, I'd definitely be inclined to reach for a clicker or similar device. With dogs, I have a host of other reinforcements available.

 

The fact the standard clicker noise is an aversive to me does play some part in my decision. Probably others as well. (And yes, I'm aware that there are clickers designed to make a different, softer sound.) But I don't particularly care for having my hands (etc.) coated with liver, cheese, natural lanolin or stockyard effluvia either. None of which stops me from contact with those stimuli in a training environment on a regular basis. So I'll reiterate that I personally find other training tools preferable.

 

If you're happy with 'em, use 'em.

 

Liz S in SCPA

Link to post
Share on other sites
Duh.

 

Good one. Groundhog Day meets Wayne's World! In all seriousness, clarification hit me when I read that. It may be that there is less objection to operant conditioning, than to the "clicker training will bring world peace if only those unsaved souls would catch up" tone used by some of it's practitioners. Well, I have no quarrel with that objection. I might just add that, like the sound of the clicker, it may be annoying, but perhaps that is not enough reason to throw the baby out with the bathwater. (If you'll pardon the violent metaphor.)

 

And, if I am getting the hang of this thread, I might now look forward to a volley of quotes and overblown claims from big names of various methods. As Wayne (a bit of a philosopher himself if I recall), would say: Party time! Excellent!

 

Barbara

Link to post
Share on other sites
I apologize in advance, but the scenario you just described would be a circle of he double tooth picks to me. How did all those dogs compete pre-clicker? I know one obedience instructor who doesn't use a clicker, but who is VERY good with timing, and a simple YES! is just what she uses. The dogs learn to love her voice, which is a side benefit, and look from approval from her, and not the clicker. At demos one time, someone gave away clickers to kids. OH MY LORD. There were about 150 clickers going off........

 

 

I'm a "pre-clicker"! I'm an "old timer trainer" - last OTCH was attained in 1994 and last in the obedience ring was 1997. I started in obedience in 1983 with my first border collie. Believe it or not there were positive training back than, much more then you would think. Yes, there were the old school pop and jerk and forced retrieve methods as well. Probably still around today, I would assume. I think the key is finding the right balance between positive and correction and being comfortable with your training. I was never comfortable with ear pinching and never did it even though it was taught in a class I took, I just refused, just as I refused other nonsense training methods. My first dog was a non retriever too. But also, all positive was not in the picture either. Because it's called "Correction training" it does not have to be negative. It's all in the way you train correction training. I've never abused my dogs physically. Clickers came on the scene in the 80's, I attended a seminar about it and videos started being available. I just didn't care, or "need them" maybe a better word for them. I used my voice, YES! WONDERFUL! YEAH! - For me a clicker just got in the way and it was clumsy to use. Some of my training partners did use them and liked them. Could they of gotten the same results with the use of their voice and good timing, I would bet yes. I drop in training classes from time to time just to see what's going on, and I see a big decline in training with all positive training. I attended an obedience trial recently and it was really sad to see.

Link to post
Share on other sites
Liz S said:

 

Fundamental dog training requires a basic skill set:

--Accurate observation

--Correct timing

--Appropriate feedback adjustment

 

I've been following this discussion with interest and staying out of it lest I display my profound ignorance but, to me Liz has hit the nail on the head.

 

Those three things are the difference between Jack Knox and me; he has them, I lack them. I'd add one more to Liz's list

 

-- Experience to interpret one's observations (you can be accurate in your observation but if you don't know the significance of what you observed, it won't help you much).

 

Which, I believe, was part of Donald's original point. There are far too many people hanging out their shingle in the dog world who lack those four qualities and deluding the credulous or the desperate into believing they know more than they do, and who hide behind the jargon of a field they hardly understand.

 

I think that the technique used matters much less than the experience of the trainer, and that different techniques are appropriate in different situations or to achieve different ends.

Link to post
Share on other sites
I apologize in advance, but the scenario you just described would be a circle of he double tooth picks to me. How did all those dogs compete pre-clicker? I know one obedience instructor who doesn't use a clicker, but who is VERY good with timing, and a simple YES! is just what she uses. The dogs learn to love her voice, which is a side benefit, and look from approval from her, and not the clicker. At demos one time, someone gave away clickers to kids. OH MY LORD. There were about 150 clickers going off........

 

 

Hi Kelpiegirl;

 

I'm not sure what your question means: how did they compete pre-clicker?

 

My point was that I have known very few dogs who had demonstrated that they found the click aversive. Thunderhill made a post suggesting that most dogs found the clicker aversive, and I disagree. She also suggested more than one person in a class can't click and I know from experience its not true. Dogs can hear the subtle differences in "their" click and they tune into their handler and become unaffected by the other sounds. The humans don't find them aversive either...I probably have a nice associative thing going on myself as I *really enjoy* training dogs so the click is a positive thing for me.

 

Other people feel differently and I understand that. I absolutely can't stand air blowing on my face...even when I am hot. I don't know why, it irritates me very much. Other people love the a/c blowing right in their face. We're all different.

 

Can a dog learn without a clicker? Of course they can! is a neutral, sharp marker and improvement of a language? I believe it is, and in my experience it is. Thunderhill has had a different experience.

 

We disagree is all.

Link to post
Share on other sites
Do rewards and punishment have an effect on dog learning? Duh.

 

OK, thats helpful. A frustration I have with these threads is at times I feel like we are talking about different things based on language and it makes it hard to understand replies sometimes.

 

The problem with Skinner based theory isn’t that rewards and punishment effect learning it’s Skinner’s central assertion that (primarily food) rewards and punishment explain all mammalian learning. That isn’t true.

 

Actually, he did not ever write that the rewards had to be primarily food, in fact he had a long dissertation about what a reward was and how rewards are only rewarding if the subject found them to be, but thats a digression.

 

 

But these modifications are erected on a oversimplified foundation. If you’ll permit a Mac user’s smug comparison: it’s like Windows 7. You can prop up a bad system and do work arounds and patches but if the foundation is inadequate, at some point its probably better to abandon the system and start over.

 

OK, so lets keep this very simple: What system is it that you are using when you (and I mean you, personally, Mr. McCaig) teach an untrained dog to lie down? How is your dog learning what you want? What is your technique for teaching this? If you don't mind, I'd like to keep this about a basic, novel behavior as I know too little about sheepdog training to have a competent discussion with an expert such as yourself. Lets keep it a simple human-requested behavior.

 

I often hear how far positive trainers have evolved from Skinner and how modern Behaviorism is different from Skinner’s “Radical Behaviorism”. But these same trainers use many of the same tools and make the same overblown claims Skinner did.

 

So your real objection is the "overblown claims." That makes a lot of sense.

 

 

“This book is about how to train anyone – human or animal, young or old, oneself or others – to do anything that can and should be done. How to get the cat off the kitchen table or your grandmother to stop nagging you. How to affect behavior in your pets, your kids, your boss, your friends. How to improve your tennis stroke, your golf game, your math skills, your memory. All by using the principles of training with reinforcement.

“These principles are laws, like the laws of physics. They underlie all learning, teaching situations as surely as the law of gravity underlies the falling of an apple. Whenever we attempt to change behavior, in ourselves or in others, we are using these laws, whether we know it or not.”

 

Doubtless Ms. Rushdoggie recognizes the author.

 

You bet I do.

 

Mr. McCaig, have you ever heard of TAGteach? What do you think about it?

Link to post
Share on other sites
Fundamental dog training requires a basic skill set:

 

--Accurate observation

 

--Correct timing

 

--Appropriate feedback adjustment (The ability to process feedback from the subject and adjust method based on what is truly happening, rather than simply robotically imposing a pre-conceived system: a roundabout way of saying "judgment" or even "listening to the dog")

 

[snip]

 

Unfortunately the clicker doesn't address matters of judgment and mutual feedback between handler and dog. The top percentage of trainers have that extra gift which takes them beyond the binary code. But whether or not they use a clicker, it's their gift which took them to the top, not one specific tool.

 

Liz S in South Central PA

 

This is one of the nicest articulations I've seen of this sensible position. Thanks, Liz.

 

Your last paragraph in particular captures, IMHO, the limits of an approach focused entirely on operant conditioning (or classical conditioning) since it highlights the importance of other factors (cognition, interaction, etc) that are necessarily independent of the stimulus/response chain assumed through that paradigm.

Link to post
Share on other sites
Your last paragraph in particular captures, IMHO, the limits of an approach focused entirely on operant conditioning (or classical conditioning) since it highlights the importance of other factors (cognition, interaction, etc) that are necessarily independent of the stimulus/response chain assumed through that paradigm.

 

I couldn't agree more.

 

B.

Link to post
Share on other sites

Dear Doggers,

 

Ms. Rushdoggie asks: " What system is it that you are using when you (and I mean you, personally, Mr. McCaig) teach an untrained dog to lie down? How is your dog learning what you want? What is your technique for teaching this? If you don't mind, I'd like to keep this about a basic, novel behavior as I know too little about sheepdog training to have a competent discussion with an expert such as yourself. Lets keep it a simple human-requested behavior."

 

Okay. Usually although not always I teach Border Collies to liedown on stock. The dog must be five months old or older. Carrying a four foot plastic wand I walk into a 75 foot circular ring, often but not always with the dog dragging a light ten foot catch line. I will either release the dog or excite it vocally so it goes for the sheep. Some circle, some bust 'em up.

 

Arms and wand extended, I control 10-12 feet of space in that ring. The dog is much faster than I am but he is drawn to do SOMETHING with the sheep and I can read his intentions and the sheeps'. I place myself, like a gate on the side of the sheep I don't want the dog to go and he will always take the line of least resistance. I can move him off if he gets too close to think. Sometimes not the first session but almost always by the second (hours afterward, often next week), the dog is circling the sheep with me between dog and sheep and bodily, wordlessly directing his movements.

 

I aim to teach the dog two first lessons (1) it's neat to work sheep and (2) everything goes better when we work them together. The down is primal;the dog tells me when its time to ask for the down. With timid or less keen dogs I may not ask for it for weeks. With hot dogs, I'll get it by the third session.

 

He's quicker than I am but since I'm next to the sheep I have the advantage of leverage. I can block him by moving a short distance. So after he's run out his piss and vinegar - four minutes? - I block both sides of the sheep. He moves left, I cover; right - I cover. If he hasn't before, he is focusing on me because I'm blocking his heart's desire.

 

If the geometry isn't perfect I wait until it is. When I ask for a down, I want to get a down.

 

Until now I may have used growls or cries to check or encourage him. Now I say, sharply. "Down."

 

By this time, my small gesture or tapping the wand on the ground blocks him. I step toward him.

 

Change voice. "Down."

 

I step a bit closer, looming, but not so close I lose my leverage. Change voice. "Lie down."

 

If the dog goes down on his elbows or (infrequently) sits or drops flat I immediately step aside, and release him to circle the sheep. If I get one down that first or second session, I'm satisfied, catch the dog and praise it while leashing it. If I can I'll crate the dog for an hour or two after: so he can think about it.

 

 

I don't think what I do (much like what other sheepdog trainers do) is a "system". It's pragmatic anecdotally transmitted strategies to train most sound young collies and cure some of the commonest problems. It depends on knowing sheep, knowing what the dog 's genetics inform him about stock and pack structure and using the dog's energetic but confused and partial understanding to my (and ultimately his) advantage. When my dog goes off his feet he is submitting to a leader. Much much later, when the dog has run four minutes and half a mile uphill through rock and brush to get behind sheep he trusts are there, and I whistle "Down", he is submitting.

 

 

I have put downs on other breeds by cornering them and looming and some traditional trainers have tried "looming" on pet dogs and I think that's interesting but wouldn't recommend it. Try it on a terrified or aggressive Boxer and you'd likely get bit.

 

But it works time after time and year after year with young collies beginning their life's work.

 

Donald McCaig

Link to post
Share on other sites
Dear Doggers,

 

Ms. Rushdoggie asks: " What system is it that you are using when you (and I mean you, personally, Mr. McCaig) teach an untrained dog to lie down? How is your dog learning what you want? What is your technique for teaching this? If you don't mind, I'd like to keep this about a basic, novel behavior as I know too little about sheepdog training to have a competent discussion with an expert such as yourself. Lets keep it a simple human-requested behavior."

 

Okay. Usually although not always I teach Border Collies to liedown on stock. The dog must be five months old or older. Carrying a four foot plastic wand I walk into a 75 foot circular ring, often but not always with the dog dragging a light ten foot catch line. I will either release the dog or excite it vocally so it goes for the sheep. Some circle, some bust 'em up.

 

Arms and wand extended, I control 10-12 feet of space in that ring. The dog is much faster than I am but he is drawn to do SOMETHING with the sheep and I can read his intentions and the sheeps'. I place myself, like a gate on the side of the sheep I don't want the dog to go and he will always take the line of least resistance. I can move him off if he gets too close to think. Sometimes not the first session but almost always by the second (hours afterward, often next week), the dog is circling the sheep with me between dog and sheep and bodily, wordlessly directing his movements.

 

I aim to teach the dog two first lessons (1) it's neat to work sheep and (2) everything goes better when we work them together. The down is primal;the dog tells me when its time to ask for the down. With timid or less keen dogs I may not ask for it for weeks. With hot dogs, I'll get it by the third session.

 

He's quicker than I am but since I'm next to the sheep I have the advantage of leverage. I can block him by moving a short distance. So after he's run out his piss and vinegar - four minutes? - I block both sides of the sheep. He moves left, I cover; right - I cover. If he hasn't before, he is focusing on me because I'm blocking his heart's desire.

 

If the geometry isn't perfect I wait until it is. When I ask for a down, I want to get a down.

 

Until now I may have used growls or cries to check or encourage him. Now I say, sharply. "Down."

 

By this time, my small gesture or tapping the wand on the ground blocks him. I step toward him.

 

Change voice. "Down."

 

I step a bit closer, looming, but not so close I lose my leverage. Change voice. "Lie down."

 

If the dog goes down on his elbows or (infrequently) sits or drops flat I immediately step aside, and release him to circle the sheep. If I get one down that first or second session, I'm satisfied, catch the dog and praise it while leashing it. If I can I'll crate the dog for an hour or two after: so he can think about it.

I don't think what I do (much like what other sheepdog trainers do) is a "system". It's pragmatic anecdotally transmitted strategies to train most sound young collies and cure some of the commonest problems. It depends on knowing sheep, knowing what the dog 's genetics inform him about stock and pack structure and using the dog's energetic but confused and partial understanding to my (and ultimately his) advantage. When my dog goes off his feet he is submitting to a leader. Much much later, when the dog has run four minutes and half a mile uphill through rock and brush to get behind sheep he trusts are there, and I whistle "Down", he is submitting.

I have put downs on other breeds by cornering them and looming and some traditional trainers have tried "looming" on pet dogs and I think that's interesting but wouldn't recommend it. Try it on a terrified or aggressive Boxer and you'd likely get bit.

 

But it works time after time and year after year with young collies beginning their life's work.

 

Donald McCaig

 

 

OK, first of all, thank you very much for taking the time to write all that out...its really neat. I didn't expect such a detailed response, and honestly wish I lived just a tad bit closer to some really skilled sheepdog folks so I could spend more time watching.

 

Second, I want to take a second to reiterate that while I disagree with you dramatically about many things, I have the utmost respect for you as a dog lover and I admire your skills training sheep. The argumentative nature of this discussion is what it is, and just because we disagree doesn't mean I am "superior" etc etc. I actually *like* arguing, and I do make every effort to keep the discussion neutral (and fail spectacularly sometimes) and engage in the arguement because in some way I *want* to be challeneged and have my mind changed. If you find these discussions to be too much effort, or are insulted by my questions I will be happy to end it here and move on.

 

OK, as an aside: I wish you would tell me how you would have trained Jack, my puppy foster dog who is a 14 lb (!!) 14 month old Papillon. Would you put Jack on sheep? Jack came in my front door whinging on the end of his leash like a yo-yo, barking maniacally. He had been living by himself in a fenced backyard and was basically ignored by owners who popped him outside when he turned out to be big and energetic. He had no leash skills, and no idea what sit and down meant.

 

He's quicker than I am but since I'm next to the sheep I have the advantage of leverage. I can block him by moving a short distance. So after he's run out his piss and vinegar - four minutes? - I block both sides of the sheep. He moves left, I cover; right - I cover. If he hasn't before, he is focusing on me because I'm blocking his heart's desire.

 

If the geometry isn't perfect I wait until it is. When I ask for a down, I want to get a down.

 

Until now I may have used growls or cries to check or encourage him. Now I say, sharply. "Down."

 

By this time, my small gesture or tapping the wand on the ground blocks him. I step toward him.

 

Change voice. "Down."

 

I step a bit closer, looming, but not so close I lose my leverage. Change voice. "Lie down."

 

If the dog goes down on his elbows or (infrequently) sits or drops flat I immediately step aside, and release him to circle the sheep. If I get one down that first or second session, I'm satisfied, catch the dog and praise it while leashing it. If I can I'll crate the dog for an hour or two after: so he can think about it.

 

As I read this I am seeing the application and release of pressure as positive and negative reinforcement. Reinforcement does not always equal cookie. I also that "blocking his hearts desire" until he acknowleges you as a form of reinforcement.

 

Are you using a clicker or whistle? not at all. Are you chucking cookies? Heck no. You are using reinforcement that addresses his internal instict to do something with that stock, or the natural urge of any animal to remove themselves from pressure to teach him what you want. "You do A, you get B."

Link to post
Share on other sites
As I read this I am seeing the application and release of pressure as positive and negative reinforcement. Reinforcement does not always equal cookie. I also that "blocking his hearts desire" until he acknowleges you as a form of reinforcement.

 

Are you using a clicker or whistle? not at all. Are you chucking cookies? Heck no. You are using reinforcement that addresses his internal instict to do something with that stock, or the natural urge of any animal to remove themselves from pressure to teach him what you want. "You do A, you get B."

 

Hi Rushdoggie,

 

I am not sure what the statute of limitations is on butting in on an exchange on the Boards, and I don't mean to butt in, just to point something out from my own experience. This may even belong in a new thread.

 

As someone who is newer-than-newbie to the working sheepdog world, I thought I would say what a huge difference it has made for me to watch a lot of, and particularly to get in and try, working a dog on sheep. I have had a much-appreciated chance to work an Open dog. In one of my very first lessons, the instructor commented to me that it was like a web of operant conditioning, with the sheep, human and dog each having their behaviors reinforced or punished by the interactions with the other, including interactions between dog and sheep that were no doing of the person (I use reinforce and punish in learning theory definitions----reinforced behaviors you get more of, punished you get less of). However, she in no way (I hope she'll jump in and correct me if I'm wrong) implied that this is the only (or even primary) dynamic going on, and in fact I suspect she used that terminology b/c she knew in my dog trainer geek way I would immediately know what she meant. It isn't just about the handler getting the dog to do something and then reinforcing it, but rather all three species are involved and giving and getting feedback. Honestly, it is mind boggling.

 

But the piece that has really struck me, more so than in other dog training endeavors I've done (although before long it's going to get me rethinking those, too :rolleyes: ), is this. From what I've observed in training a dog to work stock, I do not think it is possible to accurately boil down what's happening in terms of operant conditioning alone. I am not making any assessment about whether it's possible to achieve success using that as one's main assumption, I don't have near the experience to assess that. What I am saying is that there are things at work, very important, integral things, that make sheepdog training what it is. Otherwise I could get the same results on sheep from my feist, who I am pretty sure could be shaped to do virtually anything, and look cute doing it.

 

The things I think are missing in an attempt to boil it down to operant conditioning might include: what calms sheep and what upsets them (specifically because they are sheep and not something else), the dog's ability to think and problem solve (even quite abstract things he's never been taught), the genetic influence on behavior unique to working-bred border collies as a breed, whether a dog is loose-eyed/sticky/etc., intangible things like power, individual temperament (of all involved!), and my hunch is there are probably more but those come to mind.

 

Speaking of being a training geek, the first pressure/release combination you referred to sounded to me like P+ followed by R- (which of course we'd only know if we could see the dog's subsequent behavior). This may be stating the obvious, but it reminds me that there seems to be a great range of rewards and corrections used; not all sheepdog training is the same, not all sheepdog trainers use the same approach (to teach a down or other things). Having said that, while those of us who use primarily reward-based training with dogs may wonder, 'why not just use it alone to train sheepdogs?', at this point in my experience I guess I'd say this: if it is possible, then perhaps it would be at the expense of the other variables I listed, or at least not letting those other dynamics shine through as they otherwise might. This was no small thing for me to get my head around when I first started, but it has been worth it to try, as it is stretching my training brain, and really making me even more in awe of dogs, if that was possible.

 

Barbara

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...