Allow the Right, correct the wrong: a primer on praise

[Sekah]

Thanks for the article, Donald. Very interesting. I think it boils down to having the opportunity (or luxury) of being able to provide a natural outlet for your dogs that in and of itself brings them joy. That, and the magical focus on the handler that defines a border collie can make for some spectacularly mannerly dogs.

 

I also agree that so much of the chatter we toss at our dogs can distract and that praise is not inherently meaningful to our dogs, especially when it is removing their attention from the joyful task at hand. The opportunity to continue a behaviour can be a powerful reinforcer to our dogs.

 

I have one of those Skinner dogs (who also is lacking some of the soft biddability of a border collie). She does pretty well, all told. Now, especially that she's mature, she's become a go-anywhere type dog that I trust with strangers and all polite dogs and she lives life predominantly off leash. However, I've been too distracted working on fun things like tricks and sports, so her ability & desire to stay anywhere more than 30 seconds is ... not what it used to be. I'd have to work with her to make her one of those Sturgis hotel lobby dogs. The tricks are fun, but the exposure to the real world has made her into a dog that can handle the real world in stride.

[Donald McCaig]

Dear Doggers,

 

Thanks for your thoughtful remarks. One difficulty many people have with Border Collies is the extremes of their attentiveness. I have had Border Collies run full-tilt into the only big tree in a pasture while focusing on sheep. Sometimes and in some relationships they don't respond because they are on another planet. When they are listening, however, they can be painfully acute. I had been working Gael on a plastic whistle but used a bone whistle for the first time at a trial. Fortunately a friend was videoing because plainly, when Gael had completed her no command outrun and lift and I started whistling, she didn't understand the commands that had been so clear just yesterday. Plastic si, bone no.

 

Do they know they are pleasing or displeasing us? We can't help telling them. How far away might your spouse be before you couldn't know he/she was angry or upset?

 

Donald McCaig

[Root Beer]

Recently I discovered - quite by accident - that Dean loves it when I applaud and call out "yaaaaaay, Dean Dog" when he catches his frisbee. When he doesn't catch it, he gets a "good boy" or "nice, bring it!". He's not "wrong" for not catching it. Usually when he doesn't catch it, it's because my throw was awful, and we are playing a game so even if my throw was perfect and he didn't catch it, I don't care.

 

But he does love that "round of applause". He just glows and he has an extra spring to his step as he brings the frisbee back. I like amusing my 9 year old boy.

 

However, when he's going out to catch it, I had better keep my mouth shut or I will distract him!

 

There are definitely times and places in sports when it is much better to keep quiet and let the dog do his or her job. Knowing when to encourage, when to keep quiet and let the dog do what he or she is supposed to do, and when to praise or offer feedback is part of developing one's skill as a handler. And it can very from dog to dog and sport to sport and even day to day . . .

[Maja]

Maja,

I would like to hear more about how you transitioned from clicker training to herding (with the lovely and talented "she wolf" Darinca... ).

 

You must have made many decisions regarding the kinds of feedback that worked best for you and your dog on sheep (body pressure, restraint, verbal correction, aversive noise, praise, encouragement etc.)

 

Also, are there positive hunting dog trainers in Poland? Here, many use e-collars, but this trainer/group is in high demand:

Rebecca

I don't have the information about the actual hunters, I know that e-collars among people who do field trials are not at all popular.

 

My first BC (second dog) was Kelly, and when she was a puppy someone advised me to use clicker for the off-sheep training. I knew nothing about any training so I tried it and liked it. I somehow fell into the silent session type, which later was helpful for me in sheep herding when it started. So I have always liked clicker training, and I still do, but my third dog changed my mind about some things. My third dog is a berner. She was a fast learner with the clicker as a pup, but I noticed that she didn't care so much for the treats, but that she was doing it for me. That got me thinking.

 

To make a long process short, I came to the conclusion that border collies more than most dogs have been bred to work for the human not only without an external reward, but that they were bred to give up on the self rewarding behavior (doing something with sheep after their own desires) for the sake of cooperation with the human (so basically - I am doing something and my reward is that I gave up something more fun, in order to do that something) . That lead me to realize that early training should promote this sort of bond and desire, that it should help the pup develop the sense that the grandest thing ever is doing things together for the sake of doing it together, not treats attached.

 

So with my third dog, border collie Bonnie, I gave no treats or play reward to her ever for doing anything. She got treats for doing nothing at all, just to share. But teaching her manners and anything else I needed of her like walking, sitting, staying etc. was rewarded only with praise. She learned her speedy recall by 5 months, and it stayed that way. At the age of 6 months she was introduced to the sheep, and if I was doing with her some other activity, this is the time when I would introduce clicker for training. Bonnie went on to become an excellent sheep dog and a super, bomb-proof companion.

 

I did the same with Darinka and the puppies born in our home.

 

In sheep dog training in the beginning I praise the dog if the dog needs it (and often they don't) and if it has no negative effect on its work (and often it does). Later, because I use Derek Scrimgeour's tones, a command merges with the praise through the tone. Often silence is the praise. And in the end for the trained dog, there is just business, no fussing, quiet business-like tone, which is also sometimes gruff, or impersonal whistle, and yet we have a bond in our work that is very deep. It is so deep that if I ever am forced to give up on sheep I won't take another border collie.

 

I think that many sports trainers utilize child training paradigm in their method - they talk with a high voice, exaggerated intonation, and a lot of encouragement. And I think it is good for many other sports (though by no means the only way to go). But sheep herding is adult work from the point of view of the dog's behavior patterns, and it requires, in my opinion an approach like towards an adult.

 

hum, I dunno, I dunno......praise yeah I praise but I find for my own work the less said the better.

How do you praise a dog when working a pissed off young range bull away from heifers- it is dangerous ( he/she works because they love it).....I say nothing.....if I could say something that would break the dogs concentration....he could get squashed

[pressing an imaginary "like" button]

 

 

 

I think that there is a pretty big distinction between jobs where dogs can use instinct and ones where they need to be taught every step. They can figure out jobs that take instinct and people need to keep their mouth shut and let them work while they're figuring it out. Any sort of scent work is best set up so the dog is always working independently of the handler. Otherwise you can cue false alerts if the dog looks to the handler for direction.

 

But training the agility and obedience portions needed for SAR work and I talk and encourage quite a bit because I'm directing what I want them to learn.

[pressing an imaginary "like" button]

[Alfreda]

A stream of speech (or any sensory info) can be a distraction. I just want to point out another possibility: If the word "good" is used as a discrete marker, like a clicker, then some dogs are conditioned that that means "that's it, come get a treat."

 

In the example of "praise" stopping the dog in the middle of the weave poles, that could be distraction, or it could be a marker that's been delivered too early (before weaves are complete). It could also be a dog that needs experience with a longer lag time between the click/good (marker) and the reward/end of action.

 

I had a livestock person tell me that for her "good" was a "release" from a requested behavior. The dog stops, the person says "good," the dog is then free to move. For me, "good" is more like "right, keep going." I use "release," and/or another command to free the dog from an enduring behavior such as heel, lie down. I use "YES" as an instant positive marker. Even in the use of simple words, there are a lot of nuances, and differences not only among different sports, but also different individuals.

 

Going back to the OP, as far as I can tell Donald's list of criticisms re: "praise" vs "correction," are really about poor or unskillful USE of praise, so the same flaws would apply to the use of "corrections" (verbal or otherwise).

 

In the discussion on "corrections" many of us thought the word included techniques that were not just verbal. "Praise," however, is generally thought to be verbal. Even if we were to limit our discussion to verbal signals alone, we would still need to consider aspects of delivery: timing (too soon/late, frequency (too often/rare), intensity (too loud, hard,much/quiet, little, soft), does it interrupt dog's concentration or learning process?, is it even necessary...? I believe these aspects would apply not only to both "praise" and verbal "correction," but also to any other tool whether it is aversive or reinforcing.

 

Another consideration is the intended function (vs the actual function) of the message. As Denise and others have said, a word can have an emotional function (to encourage, to discourage, to warn, to attune with, to convey urgency,) a word can also be used to simply label an object or action ("Away" when the dog is already going counter clockwise), or to mark an instantaneous event "Yes!" The same word can be used in different ways for different purposes. And that's only a word-- there are visual signals, and positional signals too.

 

Tone and non-verbals change the function of single words for people as well as dogs. I can say "ball" while pointing to name it. I can say "ball?" to ask about it's location. I can say "ball?" with my hand out to request it. I can say "ball" with my hand out to demand it. I can say "ball" in a directive way while looking at you and then at the ball to tell you to get it. I can say "ball!" with my eyebrows up while looking at you to invite excitement.

 

Point is, issues about using "praise" vs "criticism" may be actually be issues of delivery skill and/or function, and not "correction" vs "praise" per se.

 

Donald also said that praise is complicated and correction isn't. Not sure about that-- If so- how so?

 

I'm sure that *I* can make "praise" complicated by over-thinking it , but again, that's not specific to praise. I can do that with almost anything!

[Donald McCaig]

Dear Doggers,

Sorry to be getting back to this so late but Ms. Mum writes:

 

 

"Just one misconception to correct to start with; Sheepdog training is not unique in beginning off lead (and does it always?)

 

Many positively inclined trainers also train off lead preferring to enable the dog to choose to cooperate rather than be compelled."

 

It may be different in the UK but in the US I have never seen any pet dog trainer (whether plus R or traditional) begin without the dog under physical restraint. Not once. I've seen famous +R trainer's dogs on short cables attached to walls and, in Kristine's beginning class in small pen enclosures but off-lead, able to run off should they wish to? Never.

 

In fairness I should note that many small sheepdog training rings are hard to get out of but that's many square yards of freedom not the running space a thirty inch cable allows.

 

I believe this reliance on physical control is why the default for pet dog trainers is the leash (or ecollar). It's how the handler feels safest.

 

Donald McCaig

[GentleLake]

Donald also said that praise is complicated and correction isn't. Not sure about that-- If so- how so?

 

Good question. Correction can be as simple as a quiet "ahh" or as painful as a shock from an e-collar or a beating with a stick or rubber hose (I once had a clinician recommend the last for an 11 month old pup just starting out.)

 

Some of us asked for some clarifications from Donald in the original post on corrections. I'm still hoping for some answers.

 

 

[CptJack]

Dear Doggers,

Sorry to be getting back to this so late but Ms. Mum writes:

 

 

"Just one misconception to correct to start with; Sheepdog training is not unique in beginning off lead (and does it always?)

 

Many positively inclined trainers also train off lead preferring to enable the dog to choose to cooperate rather than be compelled."

 

It may be different in the UK but in the US I have never seen any pet dog trainer (whether plus R or traditional) begin without the dog under physical restraint. Not once. I've seen famous +R trainer's dogs on short cables attached to walls and, in Kristine's beginning class in small pen enclosures but off-lead, able to run off should they wish to? Never.

 

In fairness I should note that many small sheepdog training rings are hard to get out of but that's many square yards of freedom not the running space a thirty inch cable allows.

 

I believe this reliance on physical control is why the default for pet dog trainers is the leash (or ecollar). It's how the handler feels safest.

 

Donald McCaig

 

 

I've never seen an agility class that begins on leash - at least not once obstacles are introduced (rather obstacles are introduced off leash). I'm not saying that counts as a pet trainer, but frankly speaking the few agility places I've heard of that DO begin dogs on obstacles on leash are by and large looked down upon because that's *dangerous* and doesn't really teach the dog what it needs to know.

 

And even foundations classes before obstacles (ie: puppies) where those are available focus heavily on being able to do things like recall past other dogs, food, toys, distractions and stay with the handler.

[mum24dog]

The reason you haven't seen it Mr McCaig is because you haven't been in the right place. I'm sure you don't do all your training with an audience and neither do many of us.

 

Old style classes here tend to be the ones that do most work on lead. And I certainly haven't heard of any pet class in this area where a shock collar would be allowed, whether old school or not. I promise you word of anyone using one would spread like wildfire.

 

Just in case you are under the misapprehension that some of us attend classes where all dogs are off lead all the time, I doubt that that is the case. Commonly dogs will be on lead while waiting for their turn for safety reasons but much of the actual training is not done on lead.

 

I wouldn't take my dog to a class if it couldn't be trusted to work off least.

[Sue R]

I think that there is a pretty big distinction between jobs where dogs can use instinct and ones where they need to be taught every step. They can figure out jobs that take instinct and people need to keep their mouth shut and let them work while they're figuring it out. Any sort of scent work is best set up so the dog is always working independently of the handler. Otherwise you can cue false alerts if the dog looks to the handler for direction.

 

But training the agility and obedience portions needed for SAR work and I talk and encourage quite a bit because I'm directing what I want them to learn.

[Sue R]

What Mara Lynn said, and I couldn't quote properly.

[Donald McCaig]

Dear Doggers,

 

Yes, corrections are simpler than praise. When I say "Ack" to Jake he understands "Don't do what you're doing." The tiny part of his brain which philosophizes may puzzle over what part of what he's doing is being corrected but Jake has learned it only involves his activity: he is putting himself where he should not be. Easy-peasy.

 

In contrast, praise may have many meanings to the dog and sometimes quite different meanings to its owner. "You're doing swell." "I love you", "That's what I want/mean.", "Isn't this really really exciting!!!" Why aren't you happier?" come to mind.

 

Donald McCaig

[Maralynn]

I would propose that corrections are only simpler than praise when the handler *makes* them simpler than praise.

 

My dogs understand a myriad of one word commands. Why would they not understand that a calm, upbeat "good" as praise?

 

Just because many people make it complicated or ineffective doesn't mean that it inherently is. I've seen plenty of people make corrections confusing and ineffective.

[Root Beer]

It may be different in the UK but in the US I have never seen any pet dog trainer (whether plus R or traditional) begin without the dog under physical restraint. Not once. I've seen famous +R trainer's dogs on short cables attached to walls and, in Kristine's beginning class in small pen enclosures but off-lead, able to run off should they wish to? Never.

 

In fairness I should note that many small sheepdog training rings are hard to get out of but that's many square yards of freedom not the running space a thirty inch cable allows.

 

I believe this reliance on physical control is why the default for pet dog trainers is the leash (or ecollar). It's how the handler feels safest.

 

Donald McCaig

 

Sure, I have my confidence and self-control students work with their dogs in enclosures. That is for partially for safety, but it is also because most of the dogs in that particular class need to start off in a limited safe area where the dog knows that none of the others will intrude. My students are always welcome to work with their dogs off leash in the enclosures, as long as the dog isn't going to try to bust out. And, as the class goes on, the dogs begin to work off leash out on the floor.

 

That said, I've worked with my own dogs off leash in beginner classes. I haven't tethered a dog to the wall since Speedy's very, very early days, and I would so seriously not do that now. That was back before I knew anything about knowing anything about anything!!

 

I don't agree that pet dog trainers use leashes because they rely on physical control to train (yes, some do, but it is not universal). I would say they rely on leashes because they have to learn what they are doing before they can actually train their dogs. What are they supposed to do? Allow a bunch of untrained German Shepherds, Chihuahuas, Shelties, Boxers, Pit Bulls, Mutts, Papillions, etc. - some of whom do not have the world's greatest temperaments or dog skills - run roughshod in a group while they are trying to listen to an instructor who is explaining how to convey the concept of "sit"? That would be a legal liability, for one thing. And for another - it's just not practical.

 

I would agree with you that first time pet dog trainers in group classes use leashes in class.

 

However, most experienced +R performance dog trainers start their training off leash and use leashes mainly where required by law or training facility policy.

 

I learned a principle about leashes from my first Rally instructor. "The leash is a decoration". When I train a dog on a leash, the leash is exactly that - a decoration. If it needs to be there for some reason, fine. But I'm not using it to physically control the dog to train.

[Alfreda]

Dear Doggers,

 

Yes, corrections are simpler than praise. When I say "Ack" to Jake he understands "Don't do what you're doing." The tiny part of his brain which philosophizes may puzzle over what part of what he's doing is being corrected but Jake has learned it only involves his activity: he is putting himself where he should not be. Easy-peasy.

 

In contrast, praise may have many meanings to the dog and sometimes quite different meanings to its owner. "You're doing swell." "I love you", "That's what I want/mean.", "Isn't this really really exciting!!!" Why aren't you happier?" come to mind.

 

Donald McCaig

Disagree!! This has more to do with the imprecision in both words (praise & correction). You have a concise understanding of what *you* mean by that particular correction (ACK!) and Jake understands that based on the context and his experience with you.

 

In this case the word "ACK," is your example of the form the correction takes: it is a word. The positive analogy would be a 1-word marker as in (for example) "YES," which could be used when you see Jake do something you like.

 

A "correction" could also be in the form of a leash jerk, a fart sound, a frown, a flag, a thumbs down. Another form of praise could be a pat, a smile, a thumbs up. That's one complication. The form. Both praise and corrections can take different forms.

 

But "correction" (as we saw in the other thread) like "praise" also has different functional meanings and uses. You are only seeing the different possible meanings of "praise." As you say, a person saying ie: "Good boy," might be expressing affection, reinforcing a single action (ie: a lie down), encouraging ongoing activity. "Good boy" might mean "keep trying" or "You're o.k." or "I love you no matter what."

 

But, it's the same with "correction." (ACK) can mean stop that, leave that, try again, do something else, not there, not now, I'm grumpy/anxious, This isn't working, listen, focus, shit! etc.

 

The point, in both cases is for the handler to be aware what exactly they intend to communicate and then to be consistent.

 

A problem with the word "good," is that people over-use it for many purposes. (Which is the benefit of using a non-word sound like a clicker to mark.) But the word "No has the same problem. "Good," and "no" are kind of like "right." In conversation we say "right" a lot, it might mean: Go right vs left. It might mean right, as in correct. It might mean right like "yeah, I heard you," or "I understand," or, "You heard me: you understood what I said. "Right," could be a sarcastic "Really? For sure?" "Yeah right" expresses --doubt, cynicism. Using the context, and the tone, the listener must infer the intended meaning. (Luckily Border Collies are able to do this as well! ;-)

 

"HEY!" or "NO" are just as capable of being used for many different intended meanings as "good!" or "right." Try an exercise: count every time you use the word "no" (or know) in a day. Write down the sentences and see how many different possible meanings it has.

 

It seems to me tho, that it's an excellent question to ask ourselves (or students) this: "When you say..." (choose your own example correction, praise, or command word) "Boo, what do *you* mean?" and then: "What do you think it means to your dog?"

[geonni banner]

I think corrections are simpler - at least verbal corrections, because they are almost always delivered in the same tone of voice - one that suggests the behavior the dog is engaging in at the moment of the correction isn't what I want, so it needs to try something else. If subtext starts creeping into corrections, it's time to hang up the leash and chill for awhile.

 

I have a tone of voice that I use when the dog has really pissed me off. But that's not the tone of voice I use for a training correction. It's a voice that says the wrath of doG it coming if you (Step off the curb, snap at the puppy, steal my filet mignon, etc. or, I would imagine, if you don't spit out that sheep's neck right now.)

 

Tone of voice with praise usually carries a subtext, sometimes more than one. ("Yes, but slower/faster", "almost", etc.) And there are multiple tones of voice used to covey endorsement of the animal's action, as well as whatever subtext the handler inserts. It is unavoidable.

 

I'm not saying that the subtext of praise in training is indecipherable to the dog, but it is a potential distraction and affects concentration on performance.

 

In general, I agree that the less said, the better, in training. So much so that when I was training other people's dogs, I would sometimes forbid verbal commands and teach them hand signals only, if they were the kind that were unable not to babble. Once the dog had the luxury of peace and quiet for learning what the owner wanted, I would gradually allow the owners to introduce spoken commands. I generally would have to keep lessons very short, and give the owner frequent breaks to vent all the frustration, performance anxiety, and generally habitual blather they had been confusing the dog with.

 

(Incidentally, I found that the experience of keeping quiet was a real eye-opener for the babblers. It gave them a whole new view of themselves and their communication skills. I did not charge extra for this )

[Alfreda]

I think corrections are simpler - at least verbal corrections, because they are almost always delivered in the same tone of voice - one that suggests the behavior the dog is engaging in at the moment of the correction isn't what I want, so it needs to try something else. If subtext starts creeping into corrections, it's time to hang up the leash and chill for awhile.

 

I have a tone of voice that I use when the dog has really pissed me off. But that's not the tone of voice I use for a training correction. It's a voice that says the wrath of doG it coming if you (Step off the curb, snap at the puppy, steal my filet mignon, etc. or, I would imagine, if you don't spit out that sheep's neck right now.)

 

Tone of voice with praise usually carries a subtext, sometimes more than one. ("Yes, but slower/faster", "almost", etc.) And there are multiple tones of voice used to covey endorsement of the animal's action, as well as whatever subtext the handler inserts. It is unavoidable.

 

I'm not saying that the subtext of praise in training is indecipherable to the dog, but it is a potential distraction and affects concentration on performance.

 

In general, I agree that the less said, the better, in training. So much so that when I was training other people's dogs, I would sometimes forbid verbal commands and teach them hand signals only, if they were the kind that were unable not to babble. Once the dog had the luxury of peace and quiet for learning what the owner wanted, I would gradually allow the owners to introduce spoken commands. I generally would have to keep lessons very short, and give the owner frequent breaks to vent all the frustration, performance anxiety, and generally habitual blather they had been confusing the dog with.

 

(Incidentally, I found that the experience of keeping quiet was a real eye-opener for the babblers. It gave them a whole new view of themselves and their communication skills. I did not charge extra for this )

 

O.K. try this:

Toss out the word "praise" and replace it with "right." Toss out the word "correction" replace it with "wrong." Or maybe it would be better to just use "hot" and "cold." (You don't say what you're correction word is).

 

Choose the non-word "blawz" to mean "wrong" and the word "yoop" to mean "right." A good consistent trainer will be clear and consistent using those "markers" for behavior without any subtext. A less-skillfull trainer may blur the meanings, but blurred meaning is possible (and common) with either the positive or the negative.

 

So, the dog pulls on the leash "blawz." The dog gives (leash goes loose again) "yoop." For best results always say "blawz" with a deep, firm voice and always say the word yoop with rising intonation (to further distinguish them from one another). As Geonni says, be consistent. Both reinforcing (positive) markers and aversive (negative) markers can be used with the consistency that you describe.

 

A clicker is better than words tho, for the very reason you describe. It is a non-verbal signal without baggage. It's just a sound that's unique in the environment, and very salient to the dog. You could use it to mark "wrong" just as easily as you can use it to mark "right." No subtext comes automatically attached to either one. The positive or negative meaning is attached to it in the initial stages of training in which the marker sound is paired with a reinforcer, or, if you chose, a punishment.

 

Babbling is a separate issue. I agree that at times, too much babble is just empty noise. Let the dog concentrate, process. Make the few words that you *do* use, (and the behavior you are marking) really stand out. You can eliminate babble training with mostly "no" feedback. You can eliminate babble training with mostly "yes" feedback. In fact, that's what the "silent" clicker trainers *are* doing. Click for what you like, reward, shut up and wait.

 

A person can also misuse a clicker. Sometimes beginners will go click-click-click-click indescriminantly without matching it to one precise moment of behavior.

 

Plenty of correction based trainers are "babblers" too. They sound something like this: no no uh-uh, no, no no no fluffy no, no fluffy fluffy! ah-ah no! no no fluffy! fluffy! no fluffy! No NO!

 

(One can overuse a dog's name as well. In fact, a correction based trainer once told to use my dog's name in a warning tone AS the correction. This seems to happen in herding quite a bit).

 

It's just as possible to overuse "no" as "yes." Just as possible to overuse "wrong" as "right" Just as possible to fill the gaps with lots of extra stuff.

 

I totally understand if "wrong" seems simpler to individuals because that's what they're used to using. But "Wrong" is not inherently less complicated than "right."

[Tea]

Hum going back to Blood Sucking Flyman.

 

His best corrections came from not me...but the cattle he worked.....

 

It was then he realized that the reasons I was asking him to do thing had to do with getting the job done without getting kicked or smashed.

[D'Elle]

Many times I will praise my dog for doing what is expected simply because I was brought up to say please and thank you and not take people for granted. I don't see my dogs as furry people but the attitude is deeply ingrained.

 

Much of the time it is more of an acknowledgement that I have noticed they are there and what they are doing.

^ this.

and.....

 

I would propose that corrections are only simpler than praise when the handler *makes* them simpler than praise.

 

My dogs understand a myriad of one word commands. Why would they not understand that a calm, upbeat "good" as praise?

 

Just because many people make it complicated or ineffective doesn't mean that it inherently is. I've seen plenty of people make corrections confusing and ineffective.

^^ this.

I have noticed that invariably people who criticize +R training, or positive reinforcement training, are people who have very little or no experience in that kind of training.

[Alfreda]

Tea- I loved reading your last post about all your different dogs and the different training challenges/lessons they’ve presented! Blood Sucking Fly Man indeed! (Also got teary reading your other post about the choices you had to make re: who to use in the midst of fire evacuations. I hope you guys are getting, or will soon get, some rain, and that the fires are under control now!)

 

Thanks for clarifying your view about how tone conveys info about how to approach things rather than “praise” exactly. In that context, the work itself is it. Makes sense.

 

It’s fascinating how our emotional states are more contagious to some dogs than others. (I think this might be a separate issue from just plain noise sensitivity.) With my current dog, I can be in another room and stub my toe and curse under my breath “boody f-ing hell!” and my dog will dash in all growly, ready to join “the fight” on behalf of said toe. I can also whine from the next room and he’ll coming running in anxious and worried. (I try to minimize both). I think he’d be a good therapy dog because he has what I would call a lot of empathy. In trying to start him herding I’ve also seen “corrections” (verbal and physical) in high-pressure pens do nothing but wind him up, or make him go closed/hard in confusion. It basically blocked learning. Praise might be a distraction at times, however, since he knows that “language” in other contexts, my current idea is that calm verbal reassurance and some positive marking of particular (correct) moments might make more sense to him (if he’s not too amped up). Guess we’ll see.

 

I’m glad you brought up horses. There are a lot of horse people in the border collie word. Maybe those more experienced will set me straight here, as I go out on a limb… I believe there is some clicker training in horses, but I doubt they’ve gotten rid of reins, bits, and heels. When I rode and jumped as a kid tho, we were encouraged to talk to and praise our horses.

 

I was thinking about pressure-release (in both horses and herding) and how it seems like the highly skilled trainers use the lightest amount of pressure possible. If I were to understand it in (Skinner’s) behavioral terms, the bit, reins, and leg pressure start out as aversives, and they’re released or removed as in the “negative punishment” quadrant (removing something unpleasant).

 

To me, it’s interesting to think about crossing that line where the mildly unpleasant thing (the pressure) becomes just an aid- when it is no longer unpleasant, it’s just a signal. The lightest tactile pressure is pure communication, and it's not about coercion or control at all. (Am I crazy?)

 

But it can also work the other way: a horse (or dog) can become habituated (resistant some might say) to mild discomfort, and a trainer can then feel forced to escalate to something even more uncomfortable. I would like to understand the difference between “correction” based training that escalates in order to work, and “correction” based training in which the aversive fades and disappears— Maybe praise plays a role in that, or maybe it’s just the combination of a good relationship between a trainer and a trainee who also has a lot of inner drive and abilities.

 

Interesting too about the hound guys using mostly praise!

[Tea]

Thank you, I've enjoyed reading about your toe and your dog running to defend it!!!

 

The fires are 75% contained we've been home for two weeks, trying to prepare for winter.

 

Horses- In the best riding- it is as if you think and the horse can feel your mind and executes the movement.

In the best riding- your are so connected you are really one body. You feel solid, attached into the horse.

It is amazing. It is communication that is done through touch- The orchestra of the body you feel instead of hear, from your calf and seat and weight to the skin and back and mouth of the horse. In great riding there is no push or pull or kick or jerk, always you try to stay light as possible. It was designed to be used to fight on horseback with a sabre. You could not be worried about the horse as you were trying to cut off someone's head.

 

No you are not crazy.

 

Oh the hound guy- he was old but he was a good man. He loved his hounds.

 

What I love is to send a good dog and say nothing, even turn away while he runs up out of sight and does his job.

 

That is beauty.

[Tea]

Great riding- you cannot see how the rider communicates to the horse. The aids are invisible.

[Alfreda]

^ this.

and.....

 

^^ this.

I have noticed that invariably people who criticize +R training, or positive reinforcement training, are people who have very little or no experience in that kind of training.

Yes, it seems so. And so we need translators so we can exchange ideas and see where we agree! So many of us are using some sort of hybrid system and it can get confusing to integrate it all :-).

 

One common misconception I hear a lot about R+ is that it's a "bribe" rather than a reward. Another is the assumption that R+ people have to use treats and praise forever, rather than fading them out.

 

I don't know if the mission in correction-based training is to fade out corrections for behaviors once they are established, but I assume so?

 

Once established, tho, I bet both schools of trainers still need to randomly "review" once in a while...?

[geonni banner]

Yes, it seems so. And so we need translators so we can exchange ideas and see where we agree! So many of us are using some sort of hybrid system and it can get confusing to integrate it all :-).

 

One common misconception I hear a lot about R+ is that it's a "bribe" rather than a reward. Another is the assumption that R+ people have to use treats and praise forever, rather than fading them out.

 

I don't know if the mission in correction-based training is to fade out corrections for behaviors once they are established, but I assume so?

 

Once established, tho, I bet both schools of trainers still need to randomly "review" once in a while...?

I routinely use both reward-based approaches and corrections. I don't think one is intrinsically better than the other, however, one or the other will be preferable, for a given dog on a given day. The trick is to read the dog and know which is appropriate or, said another way, which will get the job done with the least amount of fuss, and the best result for both dog and handler.

 

I believe that the fading out of treats is rather common knowledge, and as for "bribes" vs "rewards" I believe there is a large element of "you say tomato" here.

 

Corrections, if delivered skillfully, don't need to be faded out, because there is no further need for them once the "light goes on" in the dog's head.

 

Review is always good. If a precision job is involved, the response can be sharpened, thereby generating pride in performance and a good work-ethic in both trainer and dog. It isn't only the dogs that can get lax and sloppy.

[simba]

Mum24dog- I have been known to say 'thank you' to the cash machine. Or 'sorry!' to the stool that I knock over. Needless to say, the dogs get thanked.