[TracyJo]

Okay, today we started on working on more eye contact. I'm trying very hard to keep the leash as loose as possible (trying to mimic off lead ) and her attention on me with good eye contact. While we worked on easy stuff - heel, down, stay etc, I c/t'd every time I had eye contact. If it was prolonged, I pretty much just kept the bar open as long as I had the eye contact - when she'd look away, I would not verbally or physically correct her, I just kept going with different exercises and waited until she looked back to me. Is that right? While the bar is open and I'm pretty rapidly feeding her tiny bits, should I click before each bit?
She did pretty well, it's been quite a while since we've really worked on anything. I did notice that she's swinging her hiney out when we stop so she can face me - I did not correct or reward these sits, would just step up a few steps until she sat correctly (correct sits with eye contact earned a c/t).

 

[Zucker]

have you tried doing c/t only for "look at me"? Instead of making things more complicated for the dog and doing obedience routines and giving her the c/t for listening to the command AND eye contact, simplify it. You know, baby steps First teach her the concept (unless you've already done that) that eye contact gets treats. THEN integrate that into the obedience routines.

Just say "look at me" in a happy voice and the second you get eye contact c/t. Once she's got it, gradually faze out the actual command. You might get more consistent results that way.

I don't understand what you mean by the bar being open... and I think you should be clicking before each treat just to reinforce the c/t idea. Although honestly, I'm not too certain how that would apply to long term commands and if you really should be acting like a treat dispenser while she's doing a good long term command. The example of "stay" comes to my mind. You don't get the long stay by saying "stay" and then treating continuosly. You get the long stay by teaching a very short stay and c/t'ing at the end of the well performed exercise. Then you gradually increase the time of the stay as the dog gets better, and still c/t at the end of the exercise.

Just my thoughts. I'm not big on clickers so maybe some one more experienced with them can give you more advice. My understanding is that every things, just like with all training actually, but especially with clickers gets broken down into tiny steps. That way the clicker can mark a very specific action and over time you merge those learned behaviours into an acutal routine, even if that routine is only a single obedience command.

 

[TracyJo]

Yes, I've taught her to "look at me" she's very good at it actually when I ask her but now I want it to be more spontaneous... ie. when we're working, you should look at me without the "eyes" command. Bar open = treats are a comin'. My quasi-problem is that although she understands the click = treat and whatever she was doing when she heard that click is what earned her the treat, the click often times also translates into "a treat is comin' oh my God a treat is comin'!!!" and she breaks whatever she was doing in excitement over the treat... Open bar= reinforcement without the click and less distraction.. does that make sense?

 

[Zucker]

see, that's why i'm not crazy about clickers. They can be great for teaching a behaviour, but then in later training they don't offer you the ability to vary the reward's intensity.

So essentially what is happening (correct me if i'm wrong) you want her to sit and stay. You say sit, she does, you then click and she gets up all excited expecting a reward? hmmmm.... perhaps you could try fazing out the clicker in that instance? c/t at the end of the task, so "sit" she sits, then say "stay", that way you get the exact group of behaviours you want, and the c/t when you're ready to release.

 

[Sooz]

Tracy, I was at the local kennel club this week observing members proofing their dogs for upcoming events. All types of dogs were there, from Brittany Spaniels to Portuguese Water Dogs, ages 3-8 yrs, all extremely well mannered from tons of training. We talked about training to get "the look" from our dogs. While no one there was using a clicker, they all advised me to cut the treat into tiny bits and 1) elicit the look, 2) praise profusely and 3) reward with a tiny treat. When you get a better (longer) look, give praise but with extra (more or bigger) treats. A long look would get a whole treat, maybe 2 whole treats. Keeping this in mind, it seems to me that when using a clicker (and I am no expert), clicking marks the correct behavior and the treat is the reward...so, one loooong look would get ONE click (correct behavior) and one BIG reward. To continually click for the duration might be confusing...but bigger treats never are. Just my thoughts...

 

[doberkim]

well, clicking signals the end of the behavior - if you want to continue the behavior, you need to introduce a BRIDGE that will reinforce the behavior but NOT signal its end, and encourage the dog to keep going. this bridge is NOT the click - a click does and always means, GOOD JOB END OF BEHAVIOR REWARD COMING!!!!!

i disagree that size of treats is a motivating factor for most dogs too - because my dog will work just as hard for a huge steak as a little crumb of something - most dogs are most concerned with high value items and variety - i offer jackpots not because they are huge, but because it becomes a big party to signal the end of something REALLY GOOD.

i would introduce a bridge-
i personally dont use a word for attention/looking at me. i dont want it a command, i want it a default. if anything, i use the animals name - but for me, once heel position is a known thing, i expect attention in it when we are working, unless released.