So my one-dog class didn't show up, and I was there with Madigan. I figured I'd use the time teaching her a new trick.
I took a box of biscuits off the shelf, got out the clicker (she knows that means she's going to learn something new), and decided to teach her to knock it over.
So I tapped on the box, said "Knock it," and prepared to click if she glanced towards it.
Instead she stared into my eyes. Then she deliberately walked over to the box, and knocked it over.
Nah. Had to be a coincidence. So I tapped it again, and said "Knock it." She did it again. Then I grabbed another object, tapped it and said "Knock it." She did. I practiced with her several more times. She did it every time. After a while, I didn't have to tap the object. Just pointed to it.

How did she KNOW??
PetsMart Pet Trainer
My Kids, My Students, My Life:
http://hometown.aol.com/dfrntdrums/myhomepage/index.html Last updated June 27 at 10:00 a.m.
1 2 3
Granted this had a happy ending since Madigan got the new behavior right away, BUT you're doing clicker training all wrong.

FIRST shape the behavior. Only after Dog is reliably performing behavior do you add the cue (command). Otherwise, Dog is hearing cue, not performing behavior and associating cue with no behavior.

In other words, don't start with "knock it." Clicker training is different in the order of things than other training.

Lia
BUT you're doing clicker training all wrong.

I disagree.
There is nothing wrong with adding the cue when she did. It worked.

Take care,
Maggie
FIRST shape the behavior. Only after Dog is reliably performing behavior do you add the cue (command). Otherwise, Dog is hearing cue, not performing behavior and associating cue with no behavior.

And don't tap. Just wait until she glances towards the box, then click & treat. One of the things you're working for in clicker training is a dog that tries to problem-solve on their own: "OK, what's new about this situation? Is it that? Oh COOL! I was right!"
The ideal dog for clicker training is one that already spontaneously offers up a variety of behaviors on their own. However, initiative on the part of the dog is definitely something that clicker training reinforces.

I used to play the clicker training game regularly with a group of my dog friends. One person the "dog" would leave the room and everyone else would agree on something they wanted the dog to do. It might be touching the doorframe, crawling under the table, doing jumping jacks very odd things. Then the "dog" would come back into the room and had nothing but clicks from the designated "trainer" to guide him/her into performing the behavior that was wanted. Sort of like the old game of hot & cold without the cold, and far more complex and challenging.
We all took turns being both the dog and the trainer, and while we certainly became better trainers, our improvement in being dogs was far more marked. It
didn't take us long to start offering up more and
more creative behaviors, because we knew our mean
ol' friends were constantly thinking up with more
and more difficult challenges.
JFWIW,
Dianne
In other words, don't start with "knock it." Clicker training is different in the order of things than other training.

I was taught to do it the way you describe. But in practice, it seems that the timing of introducing the cue word is irrelevant. Especially with Madigan. She knows that the sight of the clicker, plus a new cue word, means that I want her to do something new.
I was taught to withhold the cue word during food-lure training, also. But again, in practice - as long as it's only said ONCE and at the same time as the lure - it works. IMHO, it works better.
Commands are given in a certain tone of voice. I believe that a dog who is trained this way associates the new word said in that tone of voice, along with the lure, as a cue that they are expected to do something new to earn that reward.
Of course, if the command is repeated as a nag, all bets are off. Then the word itself becomes irrelevant - tuned out - by the dog.

As always, correct me if I'm wrong, but without compelling testimony I'm probably not going to fix this one. It works well, at least for me and my students.
By the way, I can't recall who gave me the idea of having the owners train each other, but I find this makes a WORLD of difference with their perception of how their dogs learn.
Around the 3rd week, I pass out index cards. Each one has a nonsense word at the top (like MORP or PLOO), and then the description of a behavior (like jump up and down). I tell them that the only words they can use are the command (nonsense word), praise, or "eh" correction. And they must use a food lure. It's a fun exercise, with a lot of laughing, but it's also frustrating. And then if I hear them saying, "Sit, sit, sit" to their dogs, I can come up behind them chanting, "Ploo, ploo, ploo!" They get the point. :}

I also found a good way to teach very young children how to approach dogs. If I see a kid coming from behind a dog and grabbing it, I make him stand with his back to me and suddenly grab him. I ask him how it felt, and tell him it feels the same scary way to the dog. Same thing with putting out a hand to pet a dog's head. I come at his head with my hand, then ask him how it felt.

Moms seem to love this. :}
PetsMart Pet Trainer
My Kids, My Students, My Life:
http://hometown.aol.com/dfrntdrums/myhomepage/index.html Last updated June 27 at 10:00 a.m.
In other words, don't start with "knock it." Clicker training is different in the order of things than other training.

I was taught to do it the way you describe. But in practice, it seems that the timing of introducing ... compelling testimony I'm probably not going to fix this one. It works well, at least for me and my students.

I'm not a clicker trainer, but it's my understanding that the way you're using the clicker and the command bypasses one of the purposes of clicker training - - to get the dog to offer new behaviors, and then shape those behaviors to what the trainer wants. Tapping an object, giving a command, and then clicking if the dog guesses correctly is not clicker training as I understand it. In my book most things are fine if they do no harm and if they work, but IMO it would be wrong for you to give your students the idea that what you're teaching is clicker training.
Mustang Sally
I was taught to do it the way you describe. But in practice, it seems that the timing of introducing the cue word is irrelevant.


Especially with Madigan. She knows that the sight of the clicker, plus a new cue word, means that I want her to do something new.

Leah, the sight/sound of the clicker alone accomplishes that (eventually). Adding a cue at that point is totally premature (at least as to how it pertains to clicker training).

If you introduce the cue word (e.g., "Spin!") before* you've gotten anywhere near the actual behavior you're trying to get, for example, you're trying to get a double spin (720 degrees), and at a really fast speed, you'll run out of cues (yes, even if you use a *reallythick dictionary Emotion: smile) long before you've ever shaped her behavior enough to get her to spin totally around twice, and really fast. Plus, you'll also by then have taught her perhaps several dozen totally useless commands. I.e., all those other behaviors you clicked/treated for before she finally got around to the 720, and at the proper speed.

"Note: Don’t bother saying the name of the behavior at this point. Dogs learn through association, and you want them to associate the cue (or command) with the COMPLETE, PROPER, and PROMPT action. Besides, you might distract your dog or yourself!"
http://www.wagntrain.com/OC/GetStarted.htm
(Emphasis added by moi)
I was taught to withhold the cue word during food-lure ... as the lure - it works. IMHO, it works better.


It's classical conditioning (that's why it works), but it's not clicker training, in the literal sense.
But you're not the first person to come up with her own version of "clicker training." Emotion: smile
Tapping an object, giving a command, and then clicking if the dog guesses correctly is not clicker training as I ... would be wrong for you to give your students the idea that what you're teaching is clicker training. Mustang Sally

Ditto.

Handsome Jack Morrison
*gently remove the detonator to reply via e-mail
I'm not a clicker trainer, but it's my understanding that the way you're using the clicker and the command bypasses ... IMO it would be wrong for you to give your students the idea that what you're teaching is clicker training.

I haven't had a clicker class. This isn't what I teach (at least as yet), this is what I use with Madigan.
As I see it, the clicker has more than one viable use. It can be substituted for voice when using the food lure method. It's a quicker, more consistent way to mark the behavior. Or it can be used to capture behaviors and put them on cue, in which case you give the dog no clues and allow him to offer behaviors.
Some of my students use them to call their dogs. It works for them.

When I was taught clicker training, I was taught to use it in conjunction with food lures.
PetsMart Pet Trainer
My Kids, My Students, My Life:
http://hometown.aol.com/dfrntdrums/myhomepage/index.html Last updated June 27 at 10:00 a.m.
I'm not a clicker trainer, but it's my understanding that ... students the idea that what you're teaching is clicker training.

I haven't had a clicker class. This isn't what I teach (at least as yet), this is what I use ... for them. When I was taught clicker training, I was taught to use it in conjunction with food lures.

I have a question which has nothing to do with any of the above. You mentioned that you're going to be starting a class in Orlando with Madigan. You've also mentioned that you've done very little training with MacKenzie, and less with Murphy than Madigan. Why aren't you taking one of them to this class?
Mustang Sally
Show more