Originally Posted by
Wysiwyg
I agree it's much better when owners understand why a dog is doing something, I can help them to be patient and also to notice things more, with widened eyes if you like.
How would you then, teach a dog not to jump up? I'd probably use a clicker but some might put it on cue or teach a solid Sit.
Would you use a shock collar for this sort of training (ie household manners).
Wys
x
Ok Wys, your turn, finally!!
The process would depend to some extent on the individual dog, so I will pick a brandy-new pup for my explanation, one which knows nothing of the stay command, and is acting on an innate drive.
1. I initially prevent this behavior from ever being rewarded with attention. I do not allow boneheads who encourage jumping access to the pup, and I start to "de-program" the behavior, if you will, by showing the pup that he is not indeed the center of the universe every time someone walks in the door.
2. I begin reward/play training of all commands at 7 weeks, earlier in litters I have bred. Because a static exercise is compulsive by nature, I do not use food as a reward as I find it causes too many mistakes when the dog is over-stimulated by it, and I want the pup to succeed. I am using a secondary reinforcer in combination with the primary one in my teaching of dynamic commands, and use that same reinforcer for static commands.
3. Remember that when I am teaching and conditioning, I am doing everything possible to make the right thing easy and wrong thing hard. In the case of stay, I work distraction-free, keep time and distance from the dog absolutely minimal, and use two physical aids to help the pup. This lessens the chance that the pup will fail; when he does, and notice I say when and not if, I merely give a mild no and gently replace him. As the pup exhibits increased understanding/acceptance of the exercise, I begin to increase time and distance as rapidly as
that pup allows, ultimately at this stage to 30 seconds and six feet.
4. Up to now,
I am taking the responsibility for making the pup successful. At some point I will judge that it is time to transfer the responsibility for "being right" to the pup, and; i.e., the pup demonstrates that he clearly understands the correct response, and that the correct response generates a pleasant result; he knows what he
can do, now it is time to show him that he
must do what he has learned, no matter what the venue or distraction level. He also learns that he has a choice: Do the right thing and all is copacetic; choose the wrong thing and the result is an unpleasant one, in the form of a correction tailored to the individual pup (it may be something as simple as a growl), and the circumstances of the failure. I am gradually raising the distraction level to the one I believe is strongest for pup, increasing time and distance, and moving to new venues to encourage generalization of the learning. This is the proofing stage of training.
5. With regard to jumping: At some point I will begin exposing the pup to opportunities to indulge in jumping. I mostly ignore them, in some cases I will be a bit more proactive by taking a step forward just as he leaves the ground, or standing on the leash at a point that leaves the pup just enough slack to produce a bit of a shot (read unpleasant result) when he hits the end of it; in other words, I am introducing the concept of "unpleasant result" via
self-correction. I also reward the pup which decides to try something other than jumping.The pup begins to learn that some behaviors get "corrected" and alternatives get rewarded, that he can avoid that correction by choosing to not repeat the offense (thus
he controls the correction), and correction is not the end of the world.
6. Ultimately I put all this together, and view and use greeting occasions as opportunities for proofing. The shock collar would be used only according to the principles I have already explained in previous posts;the command involved would "sit".
Brierley, with regard to your post #327, inducing a behavior with the use of a lure is eliciting. Food (or whatever you use as a motivator) has three functions to me: I use it to focus attention, to lure the dog into position (elicit the response), and as a primary reinforcer (paired with a secondary reinforcer, which is the word "good" spoken in a tone that is distinct from the command tone).
I don't reward approximations, that is not what I want the dog to learn. I ignore "wrong" responses and if the pup is focussed and doesn't get it before he starts getting frustrated or loses interest, I will physically help him out (often have to do this with clients' dogs for the the down). I want to get a rewardable response quickly. I am teaching single behaviors with short response times at this stage, not chained ones.
I do fade the reward to random schedule, and I do teach hand signals such as those used for handling to a blind retrieve, but only after the dog is solid on the verbals.
And I do indeed "punish" myself for bonehead mistakes, altho kicking myself is not an option at my age. I once punished my mistake by smashing my fist into an ash tree as hard as I could; it worked, I never made that particular error again, but the ER doctor said that I was darn lucky I merely cracked two bones, because from the appearance of the hand he thought I had crushed the whole thing.