I've trained my JRT with "yank and crank" methods for over a year. I'm beginning to work on marker training him now and he responds quite well.

I'm still working on the "charge up". Here's the deal: he figured I like when he sits and look me in the eyes. I mark, then give him a threat. Then he sits again and waits for the next one.

My options here are:
1)Lengten the exercice "good boy, good, good... YES!". He's not a puppy (his attention span is quite good if I keep him interested). That's what I've been doing so far.

2)Ignore him until he tries something else. I.e. stand there looking at him until he lies down, then mark+reward.

3)Lure him into something he already knows. Like if he's sit, lure him into a down, and mark it. This is what I plan on doing. Even after reading several articles, I'm not sure how to shape these exercises (especially since he already knows them), which could be an even better idea. If this is your recommendation, can you please instruct me on the exact difference between luring a sit/stay/down/stand and shaping them on a dog who already knows them through other methods ?

4)Ask him something he already knows, and mark it. This is the option I like the least because he sometimes gets so excited that he throws all his tricks at me instead of listening to the command.

Every dog is different, and there is an element of tailoring training to what works for you and your dog.

One thing that jumped out at me, and only because I've made this mistake with past dogs, is putting the dog into a down, from a sit.
If you always put the dog into a sit, and then a down, the dog will always perform the command in that order.
That makes downs from a heel, and downs in motion, a real bear to train.

Much better to train downs from a standing position first.

Like I said, this dog is already trained, although not with markers.

He will lay down from any position, and on pretty much every surfaces including sewer grates and rope bridges.

The fact that I can't really find a good "cookbook" for my situation makes me kind of uncertain on the next step to take. Right now it seem I'm charging "sit and look at me" more than the marker itself.

I'm reading various articles to get a good grasp on shaping and targetting, but I'm still very confused. For example, my dog knows how to sit on command, but also when I raise my hand with palm upward. He downs on command, but also when I lower my hand with palm face down.

Are these non-verbal cues what people call "targetting" ? My dog actually learned these cues AFTER knowing the commands.

No, bridge and targeting methods are very different-
What you are doing is hand signals, which can be very nice when you have a mouth full of dinner and the dog is acting like a spaz. LOL

If the goal is to charge the mark, you simply randomly mark and reward all good behavior, just to have an excuse to mark and reward. The goal is simply so that the dog sees that the mark is followed by the reward.

The only reason we pick behaviors to mark and reward at this point, is just so we don't inadvertantly reward negative behavior- at this point we aren't trying to teach the dog anything- just that mark is followed by reward.

For fun, I'd start teaching the "hand touch"- it's a good way to move foward from here.

 Quote: alyssa myracle

If the goal is to charge the mark, you simply randomly mark and reward all good behavior, just to have an excuse to mark and reward.

Except "all" is only "sit and and look at me" so far.

When he knows it's training time, he sits and looks at me. I mark, reward, he sits again, I mark, reward, etc. He figured quite quickly that I'm a treat dispenser when he sits there with his "bring it" face.

He's eager to please, so if I wait it out for a moment (let him in a sit position), he will offer me something else. I obviously mark it, but for this to happen I have to ignore the sit. Is this a good method or it will only confuse him ?

So, the marker is loaded.

Unless you ask him to sit, don't mark or reward.

Start teaching the hand touch.

Francis,
It doesn't have to be so structured like a formal obedience session. Just carry the clicker around with you whenever the dog is around you and have treats within reasonably easy access. If he is being good (like lying quietly or walking nicely on a leash) click and mark that. He knows more commands than just sit and look for attention already, right? If you ask him to do any of those, mark those too.

Some people may disagree with this, but he actually doesn't have to be doing anything specifically good at all (though you wouldn't want to reward bad behavior, obviously). You can click for breathing. The point is that he learns that a treat comes after the click.

As an example, my parents have a very soft lab mix that has been trained yank and crank style. If he even suspects that he is being trained (like being leashed in the house, even on a flat collar, with or without treats) he shuts down. This has been the only method that has worked for him.

I agree with Kristin. We got our 2.5 yo GSD 5 months ago, she was already obedience trained.

We learned about marker training a few months ago from this site. I started implementing it into her obedience training, just to make it more fun for her. Instead of praises only, I started giving her treats for desirable behavior.

I have treats at different parts of the house. I would put her on a stay anywhere in the house, go to another room, grab a treat, and give it to her, before going on to other commands. I don't necessarily carry the treats around with me.

I do reward her intermittently for following me around the house, but not if she were to just come up, sit next to me, expecting a treat.

Not sure if this answers your question.

With our mutt it worked well to just wait for the dog to offer something else. Probably because she had very little patience (she could do over 1 minute stays after she was 2,5-3 years). With the dobermann it was very different because she is naturally very patient and I had taught her some OB without the clicker (plus she was an adult when I got her). When she offers me a sit or down I just say "Wrong" and the release command "At ease." If she insists in staying in that position I touch her paw with my foot that for some reason has always made her stand up. Basically something to just make her move.

Francis,

From reading your posts, I'm confused<g>.

I think the first thing you need to do is be clear about what your training goal is for any specific session. What, exactly, are you working on, where you want to go with it with your dog.

When you charge a marker, like a clicker, once the dog understands that "click" means "treat coming," the marker is charged. That's it. The clicker is charged, you don't have to continue to charge it. Done.

After that, it's up to you how you want to use the marker. If you want to do free shaping, that's one approach. That's when you wait for the dog to do something, move incrementally in a direction you desire, click/reward, gradually raising the bar/waiting for more of the same behavior, or a progression of it, before marking/rewarding.

Lure/reward is just a different way to train that also uses markers. You are showing your dog what you want by luring, clicking when you get want you want, then gradually fading the lure so that your dog responds to the verbal or physical cue for the behavior.

When you get out your treats and clicker, and just stand there looking at your dog, of course he's going to start offering behaviors. If you click/treat the first thing he does (sit and give you eye contact), you're just telling him that's what you want from him. He will continue offering that behavior as long as that's what you are rewarding.

When you train a lot using shaping, you tend to get dogs who will run thru a whole repetoire of behaviors that have gotten them rewards in the past. Some dogs don't have temperaments that enjoy this kind of training, where they don't have a clue what you want, but just keep trying things until they find something you do want. This can lead to a lot of frustration on the dog and trainer's parts<g>.

It is also an art to be good at shaping behaviors, because you have to recognize even the smallest move in the direction you desire (moving a foot even the smallest bit in the right direction, for example, because you want your dog to eventually move towards some object across the room), and then you have to continue rewarding only moves in that direction, and moves that are incrementally more than the earlier one. You can't shape by waiting for your dog to move from a stand or sit to a down...you have to break it up much, much smaller than that.

Personally, I don't have the patience or personality to use shaping as my main training tool. I play with it some, but it's just for fun. My dog doesn't care for it either. Her opinion is, "show me what you want, damn it! This is making me nuts!"

So, I have always had a lot of success with the lure/reward method. I start with a specific goal in mind of how I define what I want my dog to learn, break it down into very small pieces, and go from there. Once your dog learns a specific behavior, such as "sit," then you ask for the behavior, and use the marker when you get what you want (quicker or straighter or longer sit, etc.).

I do agree that it's a good thing to let your dog know that you're happy with what s/he is doing at any given moment, such as resting quietly on his rug...so you go up to your dog, say "good rug" or "good rest" or "good" whatever, maybe pet or give a treat, and thus are rewarding the dog for just behaving well around the house.

But, when it comes to training, I don't think it's a good idea to reward unsolicited behavior. If I ask my dog to sit, and she sits, fine, reward that. But, if the dog comes up to me and sits in solicitation of a reward, I may tell her good girl, or good sit, but I'm not going to reward unsolicited behavior with treats.

So, I think you may want to back up a bit and decide exactly what you want from your dog...what are you trying to train? Until you have that figured out, you're only going to confuse your dog and yourself, regardless of which method you use.

leih