Wednesday, August 08, 2007

It's Really Almost Never the Dog's Fault, Is It?

I came home last week frustrated beyond belief. Laev had several weeks before inexplicably stopped going around the empty blind when sent, choosing instead to remain in heel position, go out a few feet and turn to bark at me, or to generally screw up. This was a skill she'd had for months, nearly a full year, and I couldn't understand why it had fallen apart.

We changed things. We put a helper behind me and shortened the distance to almost nothing, so that she had to be sent around the blind for an immediate bite. We added runaways to motivate additional speed around the blind. I double-commanded.

I video-taped, looking for something that was different. Was I moving differently when I sent her successfully? It didn't seem so.

Finally I resorted to negative punishment, taking her off the field when she failed to go out as sent and, to really drive the point home, letting her see another dog get the bite she missed.

Well, we got frustration, all right. She came back on that field ready to work, but she still didn't send reliably.

It made no sense. Why would Laev refuse to start the behavior which earns her the ultimate reward? Despite what was suggested -- "she doesn't think she has to" -- I knew she had no reason to spite herself and me. Something had to be confusing her.

Okay, I finally got it last Saturday after several days of training; we'd increased distance too dramatically. That didn't explain the initial loss, but that was why she was not progressing reliably. So we got ground paint and marked 5' intervals across the field so that we could bounce distance fairly.

I went home after training on Saturday feeling frustrated. My training director would like me to get our BH and trial for our 1 this fall, and I have backslid almost a year in this exercise. Well, darn it, this dog is going to learn to send to the blind again!

(You know where this is going....)

So I set up for dinner one night, putting a chair across the living room from Laev and myself. "Mark," I said, pointing my left arm at the chair alongside her face. She looked forward at the chair. "Revier!" I pushed my left arm forward, and Laev shot around the chair and returned to me.

I handed her dinner and then stared at my left arm. It had moved forward when I sent her, just as it had when I first taught the send long ago. Just as it had all winter and this spring. Just as it had NOT for the last two weeks.

My arm never moved in that video. Somehow, I had dropped a very salient part of the cue, and my dog was simply missing a chunk of information.

I sent an email to my club berating my foolishness and, last night, we set up again. "Mark -- revier!" and I pushed my left hand forward. Perfect blind search. Repeat. Repeat. Repeat.

Laev made one real error last night, in which she went out, hesitated, and turned to bark at me. I chalk that up to frustration with weeks of changing things trying to fix the problem that was mine. The few other mistakes in our two sessions were honest ones, as she tries to stay focused on the task before her with a helper behind her and I try to time my cues appropriately until she learns to ignore him.

I have no idea why my signal changed, but it's very obvious that the motion was more salient than my verbal cue. /sigh/ Now we'll work through the frustration we built in the last couple of weeks, and then we'll catch up to where we were.

At least, while I feel relatively stupid, I don't have any guilt for having unfairly used P+ or R- to "fix" the dog when it was my mistake. Yes, we had frustration, but I did not apply aversives. I'm glad of that.

It's a good thing our dogs don't write tell-all books...!