Saturday, March 17, 2007

Faster Sits, No Barking, and #$%@&* Cats!

Okay, to fixing our sit-for-the-bite latency problem. Quick review: Laev had chained four barks before the sit, making her response a standard of 2 seconds, where I wanted a sit within 1 second, with or without barking.

The advice I received was to make a sharp collar correction to prompt a sit. While it is true that this would probably fix the behavior within a very few reps, and while it is true that Laev would probably not in the grand scheme of things really care much, I felt strongly that I should rise to the challenge and find a non-compulsive way to reduce our latency that final second.

This week I played with Laev, whipping her into a frenzy to chase a toy skimming along my ribs for speed and then freezing and cuing, "sit!" We had 1-second sits with no problem whatsoever. She's done this since baby puppyhood, and it has never involved much barking because it is very predatory in nature and low in conflict/aggression.

I went to training today and played with the toy instead of orienting Laev toward the helper. She did well, producing fast, silent sits again. For her second session, I started her chasing and sitting for the toy, and then I passed the toy smoothly to the helper as we spun and cued Laev to sit. She sat, and I released her to bite the toy from him instead of me.

From that, we developed a pattern of me revving Laev into a whirling frenzy, her teeth just skimming the toy as I whipped it away for several spins, then slipping my hand into her thick padded collar as I passed the toy to the helper just outside Laev's orbit and lifting her slightly as I cued a sit. (The lifting, as in the other obedience for the bite exercises, is to interrupt her lunging and allow her a chance to land in the sit position. It's not a trained cue.) She sat practically instantly, as was her habit for toy play, and the helper rewarded with a bite on the toy. No barking, no delay.

This is not quite as easy as it sounds for the handler! It induces vertigo, among other things, and I can say that my anaerobic capacity is embarrassingly low. But it got the correct behavior in an in-between situation which will likely carry over to our problem scenario.

Our third session was supposed to graduate from sitting for the toy and the helper to sitting for the sleeve and the helper, but we passed a cat on the way in. Laev's brain melted. We went inside without much effort, but all she could think of was the cat outside, not me and the toy. She wouldn't fire up even for the helper -- which stunned me, as I would have sworn if asked that nothing on God's green earth was more valuable than a helper. She simply lost it.

"You've got to fix this. I've never seen a cat problem this bad," commented my helper. That means a lot, coming from someone who works almost exclusively with dogs bred specifically for prey drive.

We swapped modes immediately. Someone brought my treat bag from the car, and I began reinforcing for attention as we circled. My helper traded the bite roll for a cat and the distraction work began in earnest. If Laev recalled to front when asked, she earned food; if she did not, I circled and guided with the collar and leash (mild R-, redirection) until she gave me eye contact.

The decoy cat was very malleable, and our helper bounced him on his lap, tossed him gently in the air, and swung widely him from side to side as he (the helper!) meowed at the dog. Laev figured out she could handle all of that, able to refocus with some visible effort for treats, until he placed the cat on the ground. Then she lost all ability to concentrate, and we did lots of circles! Laev hit a high point of arousal, nipping my fingers and drawing blood as she took treats, and whining frequently along with her lunges toward the cat.

Distance made no difference; her success rate didn't seem to change whether I was near or far. The cat she'd alerted on originally was at a good distance, anyway. Laev has always had incredible prey drive (I dislike that term, but it's a convenient shorthand); it cost us her first AKC leg at 11 months old, when she broke a group stay after a surprisingly good individual performance to pursue a bird which had landed nearly 200' away at the other end of the building. Yeah. So this is a tough project.

What I've found of late is that if I have Laev engaged when the distraction appears, I can usually keep her -- but if she locks onto the cat or squirrel before committing to an obedience task, it's all over. Today when she saw the cat while not under other orders, so to speak, her brain was free to melt away.

We made progress today, in that she was turning away from the cat to front to me, but it's a long way from finished reliability around a cat.


Shawndra said...

Laura, I just love reading about all your creative training efforts with Laev. I'm glad you don't cave to the conventional "yank on the collar" advice. The whole decoy cat scene - I could just see it.

Curious why you don't like the "prey drive" term?

Laura said...

I was having trouble cuing Laev because I was trying so hard not to laugh aloud at the swinging cat! He's such a nice cat, totally fine with that sort of thing....

"Prey drive" is just so darned un-scientific. :-) The term came originally from Hull's theory of drives, which was junked decades ago in psychology, and the remaining use of "prey drive," "pack drive," "fight drive," etc. allow different interpretations by different people and no clear *real* definition with a single, across-the-board meaning.

Consequently, we have a single term which is used to describe both a temperament ("that dog has a lot of prey drive") and a temporary state or "mode" of thinking ("the dog is not barking properly because he's in prey drive instead of fight drive"). Confusing and inaccurate, I think.

I recognize the need for a simpler way to describe "a dog motivated by a high natural predatory instinct and motions or actions relating thereunto." I'm just not sure that I like the term "prey drive." :-)

It'll have to do, though, until we find something better.