Friday, May 15, 2009

Those on the Ground Have No Fear of Falling.

So... yeah.

A couple of weeks ago, I realized that Laev had a gunfire problem. A real gunfire problem, not just the minor twitch I thought we had. It's tough with Laev; she doesn't show stress very obviously, and it's not until extremes that she starts to look as bad as she is. Where other dogs will be whining, panting, showing the whites of the eyes, or raising a paw, Laev flattens her ears and pulls back the corner of her mouth only. It's easy to miss. Then something goes wrong, and I wonder, "Where did that come from?"

There's also a consistent delay, because Laev really does have a nice work ethic, and she will try to continue for a while before things collapse. This is why Laev twitched at the gunfire in December but didn't break the heel until about 20 seconds later. This, I have gradually realized, is a regular pattern. Okay, I'm slow.

So I felt really stupid when I finally realized that her two major stressors in a trial were the reporting in (her first experience involved a dogfight) and the gunfire. Those can come in quick succession in a trial. She never used to have a gunfire issue, which is probably why I wasn't paying close attention, but we have a neighbor at home who likes target practice and I think he's done some sessions while Laev was trapped in her kennel, unable to escape the hours of ceaseless, intermittent shots, and she's now sensitized.

No problem, I thought naively. I borrowed a starter pistol and enlisted an assistant. I'd just have him fire a number of shots as we heeled and I'd reinforce heeling with me, and then we'd be fine!

I tried to split, I really did. But Laev does not react to recorded gunfire, only the real thing -- and it looks like distance is not necessarily a factor once she is sensitized. All I did was poison my cue, confirming for her that heeling with me predicted gunfire; she very quickly became so reactive that she'd jump just at the sound of a box clicker.

I left town for four days, which gave her time to flush the stress chemical cocktail from her system (takes about three days to purge so that the dog is starting with a clean slate). When I came back, I was able to heel her at club training with two gunshots, and she didn't have a panic reaction. She did, however, show me subtle signs of stress. "I didn't see anything, she looked fine," said most of the club, but I knew that if I'd pushed, I would have lost her again.

So today I went out for one final practice, after working a while without gunfire. Our trial judge, already in town, came to watch and offered his opinion: I was not firm enough with the dog. She broke not because she was afraid of the gunfire -- "look, her tail is wagging, she's fine" -- but because I had not taught her to "down or die."

I broke down, to my complete horror, and there is NOTHING more shameful than crying in front of a training director and a German judge. It's like crying in boot camp. Seriously, I should turn in my gear now. (The only thing worse is crying with sinus issues, which I had, making it all even more sniffly and gruesome.) I explained that though I wasn't going to sound rational now, I had rationally thought about this, and a week before the dog was shaking and crying during gunfire, so I knew it was a gunfire stress issue, I suspected where it had come from (the judge agreed with me there) and that I knew my dog well enough to know that she was stressed, though she didn't look typical. (Really, a wagging tail can certainly be a stress indicator as much as a happiness indicator! Look how many people get bitten and protest, "but his tail was wagging"!) He conceded that perhaps I did know my dog to know that she was upset, but that the issue was not the reactivity, but her lack of respect for me as a handler. It's good to love my dog, he said, but I had to force control.

My club friend had more immediate advice. "Don't take it so personally! It's a frickin' dog." He grinned.

"It's my frickin' dog!" I answered, but I took his point -- I shouldn't take it personally that Laev is sensitive. I blame myself for utterly destroying her training and I do get upset that she can't just trust me for five minutes, but I shouldn't think of it that way; I should think of it as a chance to improve her. Right after I get done destroying her training.

I know these people are experienced and are offering advice that has worked for them and others, but Laev and I don't work like that. He said that if Laev breaks the down to come to me during gunfire, I should require her to heel to punish her for the error. Heeling should be something she doesn't like to do. But I think heeling should be something Laev wants to do! and it should be a reward, not a punishment. And it used to be something she liked, before I destroyed years of training in just a few sessions by linking heeling and her trigger. /facepalm/

"There is no other way," he told me. And he believes it. So do my club friends, who all mean the best.

I believe there is another way. I have been told so many times, by so many people, that I will never get X without force. I've heard that it is impossible to train something the way I say I did. Yes, it may take me longer sometimes, because I don't really know what I'm doing 'til I've done it, but not having a map doesn't mean it's impossible to get there.

I hate being in disagreement with people who are trying to help me. I'm not trying to be unreasonable, and I'm not trying to be rude; I'm trying to do something I want to do. I know it's different than what they want, and sometimes I think that others think I'm judging them because I'm doing something different. But the truth is, I'm trying something here. I've never said I'm an expert -- heck, I just said I don't have the map! But that doesn't mean I can't try, right?

More, I have an ideal. I refuse to be the lesser of two evils -- if I have to force my dog to work with me, then it's no longer a game I'm interested in. Laev used to prance along with me to the field, even volunteering heeling en route -- today she was reluctant to work with me at all. That's not right; I miss being her first choice. If Laev heels or downs because it's that "or die," then I've lost sight of the reason I got a dog in the first place. We're a team; we'll get through this together.

Somehow. Honest. I'm pretty sure. I'd like to think so.

Anyway, I've changed her entry. We're going to just do a track only tomorrow morning. And it probably won't go well -- stressed dog, mega-stressed handler, thunderstorms all night and through the morning, and I don't know if the tornado watch will still be in effect during tracking -- but who cares? Those on the ground are not afraid of falling, and we can't get much lower. I'm going to support my club trial and support my trialing friends, and then I'm going to step back and do some serious evaluation.

I just wish I didn't feel like I had let down my friends, my dog, my training colleagues, and everyone who had wished us luck for this weekend. I'm supposed to know better, I'm supposed to get results. I hate having expectations.


Deborah Leão said...

I know this may sound simplistic, but why don't you just try classical conditioning, instead of operant? You know, pairing the gunfire noise with the reward, instead of asking for a behavior - and poisoning the cue?

You could maybe even ask your shooting neighbor to help, and reinforce Laev heavily during his target practice.

Many dogs are scared of the clicker sound at first, and learn to love it through consistent pairing with reinforcement. I don't see why the same could not be tried with other sounds.

You're a great trainer, Laura, and I truly admire your work and your persistence. Don't look down upon yourself. Lower your own criteria.

This is the first time you're preparing a dog in a sport you were not perfectly familiar with for a start, and you're trying to do it using methods that do not fit the traditional standards. Well, I guess that can be called lumping...

It's hard being judged by people who take such a different outlook on dog training - in a smaller scale, I've been there, and the only thing you can do is trust yourself. Don't be so harsh.

I love reading your blog, I have subscribed to your RSS feed, and I've been following you from far away for a couple years. Cut yourself some slack, you deserve it.

Becky said...

Who says you are suppose to know better? You are in unknown territory here, and I for one and grateful you are making the trek. I have been inspired by this blog and you and Laev. So much so that I just got a puppy to raise and train in schutzhund. It is a game, and it is supposed to be fun. Listen to Debbie and cut yourself some slack. Eat some chocolate and start again. Good luck with the tracking.

Dawn said...

Schutzhund shouldn't be treated as a game, nor should its training. I would much rather be neck pinched for a millisecond, over “being trapped in her kennel, unable to escape the hours of ceaseless, intermittent shots…” I digress.

But that’s just me, a person who has a ‘REAL’ relationship with my dogs. When we’re having fun, we’re having fun.

When it's time to obey a cue/command (semantics interfere with truth), my dogs are rewarded. When my dogs are confused, we retrain. But when my dogs willfully disobey, they are corrected.

All in truth, a 'REAL' relationship.

Shawndra said...

Laura, I am proud of you for sticking to your principles. Keep Laev happy...

Deborah Leão said...

Dawn, I'm sorry to disagree, but why shouldn't Schutzhund be treated as a game? Also, why is treating something as a game a bad thing?

From your point of view, I can understand that "real" means that you are allowed to use whatever means you deem necessary in order to obtain the desired results.

If you consider it necessary to use physical punishment to train your dog, that's fine. It's your opinion, and although I disagree, I truly respect it, and I would never consider the possibility of forcing you to change it.

However, saying that something is "real" and "not a game" is not enough to justify your choice of methods.

Have you ever heard of Bob Bailey? This guy trained dogs, pigeons and cats for war. He trained wild dolphins to perform behaviors in open water, 8 miles away from any human being. And he specifically states that he only used aversives half a dozen times in training - and not because he felt it was necessary, but because those who hired him demanded that he did so.

You don't need aversives for "real" training, and much less for a sport - which Schutzhund is, no matter what its original function migh have been.

So: if you think it's OK to use aversives, that's really your choice. But saying it is the only possible choice and justifying that by saying that Schutzhund is "real" and "not a game" is bad reasoning.

P.S.: I'm sorry for any mistakes I might have made, English is not my native language.

Dawn said...


I’m not aware of Mr. Bailey’s accomplishments in Schutzhund, but I will surely look into it. I do know of his and Mr. Spector’s accomplishments in developing a reliable service dog training program though.

Are you aware of Joel Silverman ?

I suppose we’ll have to agree to disagree on Schutzhund being a game or not.

If you didn’t tell me, I wouldn’t have known that English wasn’t your first language.

P.S. - I didn't say aversives is the only possible choice. I said that I use aversions to correct my dog’s willful disobedience. If you don't want to use them, then don't. And if I had a cookie for every time an all positive trainer insinuated that any use of an aversion would destroy the relationship between person and dog, I would be humongous. I'm sorry if I offended anyone with my reverse opinion of Clickerdom dogma.

Laura said...

Sorry, all -- I had an email issue which prevented me getting alerts on comments....

@ Dawn, who wrote: >>I would much rather be neck pinched for a millisecond, over “being trapped in her kennel, unable to escape the hours of ceaseless, intermittent shots…” I digress. << Trust me, Laev hearing the shots in her kennel is not my choice, nor my ideal. When I'm home during the neighbor's target practice, I bring her inside, but when I'm not home, there's little I can do. I do not understand your statement that a pinch is better, however, as I'm not exposing her to shots in lieu of a pinch collar...?

@ Deborah -- I did use CC in the beginning, but because we'd never had a gunfire issue before and I hadn't realized that it was a real problem now, I crammed. Badly. :) I do have a new gameplan now....

Ailigh said...

How does one define "willful disobedience"? Actions should be looked at (the ABC approach) not the unknown emotional reasoning behind a response.

Roger said...

I have to chime in. All dogs trained in positive only methods don't have "willful disobedience", one first needs to be obedient to be disobedient. If your dog only consistently performs its' cues when the external reinforcements on hand are higher in value than say a barbecued squirrel in the mouth of a ferrel cat running across the field, your dog isn't trained well. And unless you always have a higher value than that on hand, good luck training your dog outside of your fenced yard with positive only methods.

Laura said...

Okay, guys, I like discussion, but I'm closing this for comments now. It's no longer relevant commentary but now unfounded generalization, which I don't like on principle.

Unfounded? Yes. While I've not seen a barbecued squirrel in the mouth of a ferrel [sic] cat running across my acreage, I have downed my dog successfully while wrestling off a larger dog who attacked us from behind, no clicker or treats in sight. Roger's implication that "all dogs trained in positive-only methods" cannot be obedient is simply false, and this discussion has left usefulness for any party.

Thanks for reading, and relevant commentary -- even that not 100% in line with my thoughts -- is always welcome.