The idea was to not let them pass the hide they were on (1, 2, 3 or 4) more than about a foot. Once they finished with a hide, they were to be restricted to half the distance between it and the next and no more than a foot past the next hide (essentially 3 feet).
What Joyce wanted us to do, but didn't explain in advance was to repeat the nose touch and multiple rewards at each one, just like we'd done last week. So the first few of us found out we were doing it wrong in the course of doing it and being criticized. I told everyone who followed Gimme and me what was expected, so they had that advantage. In Joyce's defense, her 92 year old mother has congestive heart failure and recently has declined substantially, so she was half expecting "the call".
Gimme did not respond well to this exercise. She did well for hide 1 and pretty good for hide 2. But, after that, the restriction of only having three feet to move in just really annoyed and frustrated her, so she stopped searching and resorted to "lying" to me instead. That means she would pick something and paw at it, trying to convince me there was odor there. Fortunately I've seen this before and so I understood what was going on - the key element is that she stops sniffing and just starts using her feet. I don't think Gimme got anything from this exercise, but we'll be trying it again. The other dogs did better with it, since they are all slower moving to begin with, so the leash restriction didn't really bother them.
The second search was a very large exterior under cover, with three hides and one food distraction hidden in a cinder block. We knew where all the hides were, but were told to act as if these were blind hides. Gimme did very well at this... she just went from one to the next to the next and never went anywhere near the food distraction. I was very pleased with her efforts and she was certainly happy to get to search more normally.
Susan (with Tucker) and Mary (with Grafton) and I have committed to get together once a week to practice nosework. For instance we are meeting after hours this Friday where Susan works, using the building for a bunch of threshold hides and then a container drill in the parking lot. Next time after that, I think we'll meet at this same closed store location. There is a 200 foot long rockery wall, so we can repeat this exercise, spacing it out. With that much room we can probably set up a dozen hides along the wall and fifteen feet apart. Some other time we can meet at the same location and do vehicle searches on the six semi-trailers parked end to end.
On Sunday we met in the afternoon at this same location and did a number of container drills with the food distraction. Tucker gives whole new meaning to distracted - he tries to pick up the container and drag it off. I stood on one end of the most delicious bag a few times; we finally resorted to putting a cinder block in that bag. After several container searches, we did one exterior search in some cabinets sitting outside.
Susan has her work cut out for her because of the trial. Purposeful food/toy distractions at trials are supposed to be inaccessible, but Tucker was #28 out of 34... Somewhere along the way the lid came off the plastic container holding the muffin, inside the distraction bag. He was able to get his nose in the bag and swallowed the muffin whole. Needless to say, Susan was quite upset and also needless to say, we have a real challenge now that Tucker believes food distractions can be had with the right amount of effort. He's so "vigorous" that my usual snap top disposable plastic dishes won't hold up - so I'm going to be looking for some sturdier stuff. He's entered in a trial next month... so we really have to get him over this before then. So no matter what else we practice - we will do container drills every time.
By the way... last night was agility class again and Gimme did well. I got stuck in traffic and didn't get there in time to walk the first course, so we were not smooth. Of course, lack of adequate direction is not Gimme's fault. But we got through the course and she had some fun. I've mentioned it before, but it bears repeating - she is so fast and fun. She manages to make challenging weave entries with no help from me - she totally gets that part of her weave job. If the entry was hard she often can't hold onto the weaves and pops out around the 5th pole, so we have more work to do. Still, she gets the whole package on the second or third try.
The other challenge we had was that she again decided she couldn't take the first jump at an angle. She doesn't have any problem when she approaches jumps on a slice mid-course. We ran into this once before... mid-winter and had to break it down for her then. We haven't had an angled start like that since, so I'm sure that one experience wasn't enough to cement the lesson. The really nice part was that once we broke that down... she was doing a fast two jumps into weaves... where she nailed the entry and held on for the whole 12 poles.
We spent our turn just doing those three obstacles, so Blynn had me find something else fun to do while the bars were being reset.
So I selected a straight tunnel to L-shaped tunnel where I used the body of the L-shaped tunnel as a layering exercise. The first time I sent her into the blue tunnel and ran ahead to position H1, cuing her to "tunnel" as she came out of the first tunnel. I wasn't surprised that she took the path shown by the dotted line. So I repeated it, this time running to the position shown as H2 and naturally she got it. The third time I went back to the H1 position and she totally understood. Since I'd never even thought of doing that with her... I was very pleased that she totally got it after just being shown what I wanted one time.
She's a smart cookie, but then we all knew that...
4 comments:
Have you thought of using non containers to teach distractions? Reyna will tear apart a suitcase to get to her ball, so we trained distractions in cabinets and drawers and then moved them to containers. Underneath doors in halls also works well if you are okay with some damage to the bottom part of the door for aggressive pawing
We have and I've done it with Gimme. In fact, the setup last night with the food in the cinder block on an exterior search was just that.
However, for Tucker, the issue we face is that he's learned (one trial learning) that he can get food from containers - so that is what we need to address.
I'm going to suggest less powerful distractors AND more powerful rewards. Problem is that his owner has unintentionally poisoned food rewards... so if she ramps it up too much, he'll get suspicious and it won't act as a reward.
Okay I have to ask...how do you poison a food reward? I have just never heard of that before.
Good question...
Generally speaking, when your dog does something you like, you reward them. So in short order, the behavior such as sit, which was of no importance to the dog, takes on the value of the reward. The dog thinks, "hmmm, sitting earns me filet mignon, I loooooove sitting."
We do this a lot when a dog is afraid or anxious about something... we wait until the dog sees the bad thing, then we start stuffing food in them. Thus the dog learns that seeing the bad thing predicts getting a good thing, treats. In time the bad thing is counter-conditioned to become a good thing (or at the least neutral).
There's an important thing about this... the dog has to see the bad thing and THEN get the good thing, so that bad thing predicts good thing. If you reverse that order consistently where dog gets cookies and then bad things appear... the cookies predict the arrival of baddies. Thus the cookies take on the negative association of the bad thing - becoming "poisoned". Thereafter, every time the dog gets cookies, they start looking around for the shit to hit the fan. This is usually the reason for the failure of classical counter-conditioning efforts.
You can do the same thing, even if you get the order of presentation right, by using low value treats in work to overcome something the dog dislikes. Or you can poison an environment.
Both of which have happened with this dog. He is somewhat touch averse and very much hates having his feet fiddled with. She did too much work too fast on this, with treats that weren't strong enough to overcome his emotional objection. Therefore... any big deal about food treats makes him suspicious. Likewise she cannot train him at all around their home where she did much of the earlier work or he'll slink off and avoid her for half a day or more.
And this was done with strictly reward based training that was done just a little wrong! Can you imagine what this dog would be like had she used aversives?
Anyway, we are working to find some super reward that he's never had before to use in nosework and nosework only.
Post a Comment