So we set up for an obedience for the bite session, with a new goal of a 1 second latency. Revved her up, lifted her collar to bring her out of a scrabbling lunge, cued sit.
Laev barked four times, settled into a sit. Two seconds. Not good enough, so we repeated.
Laev barked four times, settled into a sit. Two seconds.
That's okay, she's a bright dog, she'll think through the fact that she's not responding quickly enough. Rev and cue.
Laev barked four times, settled into a sit. Two seconds.
Okay, we have a pattern. We'd never paid attention to the number of barks as we were working her down from longer latency, we only counted seconds. But somewhere along the line, Laev got the idea that she needed four barks before her sit, and that was locked in. Because she took the time to bark, she couldn't get below a two second response time.
Well, we'd let frustration take care of it, we decided. So we did more reps. I don't know how many more reps we did, but it was something like 15-20 -- and she never got below two seconds. Nor was she reinforced for those slow reps, but she never got it "right," in our criteria; she *knew* she had to do those four barks, because obviously we'd reinforced for them before.
Okay, I'm not that smart, but I'm not that dumb, either. We switched to another behavior. I opted for a recall, just so that she could get some reinforcement, and because I want a strong recall, too.
Laev sees the helper, Laev revs. I let her out ahead of me, straining on leash. He fires her up. I call, "Come!" Laev turns, comes to me, sits. I click and tell her to go for her bite. Laev gave me a shocked glance -- "Really? For that?" and then spins and races for the helper. :-)
I was quite proud of her recalls for the bite (much better from the beginning than the sits), but we're going to have to do something about this sit problem! Our thought at the moment is to take it back to just playing with me, getting fast sits for the toy -- she's done that from puppyhood, but not lately, and while it's a different picture than sitting for the helper (a *huge* difference in control!), the skill should carry over.
Laev is so good at keeping me humble. /sigh/
"she can go do whatever she wants to do now" = R+!
If I end the limited hold by releasing the dog to "go do whatever she wants," won't she move forward and reinforce herself? I don't want to reinforce an incomplete behavior.
I have not applied an NRM to the barking, because the barking itself is not wrong. (She's barking already prior to the cue.)
Laura &
clarification
Laev does understand the concept of a limited hold, which is how we got down to two seconds from eight (it's hard to snap out of that kind of drive!). She just didn't know it was possible to do it without the extra barks and get inside the new limited hold span.
The problem was not that of a limited hold, but that I had inadvertently reinforced a short chain. (Bad Laura, no cookie.)
Laura &