Sunday, February 17, 2008

Holodeck Training

Holodeck Program
based on guidance from Alice Woodyard and Jody Baker

BEFORE OTHER TRAINERS ARRIVE

  • Let Lumi rest after airing her with Laddie.
  • To reduce likelihood of zoomies by giving Laddie some work, and to gauge Laddie's responsiveness, run him on two semicircle wagon wheels:
    • First drill: dummies thrown at 15 yards, birds thrown at 10 yards.
    • Second drill: dummies prepositioned at 50 yards, birds thrown at 15 yards.
    • In each drill, have Laddie pick up all the dummies first, then the birds.
    • Auto-whistle return on every pick-up.
    • If Laddie swerves to any bird, whistle sit and cast away.
    • If Laddie is not 100% responsive on every whistle sit, every cast, and every recall whistle, do not run him with the group.
  • White jacket.
  • Put on Laddie's collar.
  • Load pockets: pistol, ammo, ear protectors, slip cord for Laddie, radio.
BEFORE FIRST DOG
  • No birds for Laddie until his resource guarding has completely stopped in private training, and until he's had some good group sessions with dummies; therefore, make sure the gunners have dummies with them when they go out.
  • Arrange to run Laddie first.
RUNNING THE DOGS
  • No multiples for either dog until that dog has had some good series running singles.
  • No blinds with the group, since that would let the dogs continue to practice hunting for the article rather than responding to a whistle.
  • Run shortest mark first, longest last.
  • With Laddie, use a slip cord to prevent a break.
  • With Lumi, cue "mark" before each throw, then send as quickly as possible.
  • If Lumi looks away from the fall before I send her, radio to the thrower to pick up and throw again.
  • Auto-whistle recall on the first two marks of each training day. Based on how the dogs do, consider switching to contingent whistle for the remaining marks of the day.
  • After pick-up, anything but coming right back is bad. Attempt a whistle sit followed by a whistle recall, and if necessary, a verbal recall. If I cannot gain control quickly, walk out, take away the article, put dog on lead, and put dog in van for remainder of training day.
RECORD KEEPING PER MARK
  • Attempted break?
  • Head swinging, before or after throws? Which throws?
  • Did dog return uncued? Auto-whistle? Contingent whistle? Voice? Walk out?
  • If the dog did not come straight back, why (for example, RG, parading, Super D, zoomies, diversion)?
Conditions. Temps in mid 40's, overcast, wind 5-10 MPH from variable directions so it felt colder. Ground damp but no standing water where the group was training.

Before Group Training. As planned, Lumi rested in the van while I ran wagon wheels with Laddie. We ran more than I had originally planned. Laddie did well at lining to the article sent to, and on the few occasions where he veered to a duck, he was 100% responsive to whistle sits and casts. He apparently found the field highly distracting, and while he was eventually 100% responsive to recalls, it not always on the first whistle or verbal cue.

Group Training. The group ran a single triple with two blinds, one slightly to the left of the leftmost mark, the other down the center of the course taking the dog right next to the middle Gun, which had a holding blind so the thrower could retire for those trainers who wanted it.

Lumi ran the three marks as singles, shortest to longest. Laddie ran only two marks, also shortest to longest. I didn't have Laddie run the right mark because that's where flyers were being thrown and I didn't want him to be distracted by the cage of pigeons there. It didn't work, since he was still distracted by the pigeons on the other two marks, but that was the idea.

Neither dog is running blinds with the group right now.

I used no slip cord for Lumi. I used a slip cord for Laddie on the first mark, and when he made no attempt to break, I didn't use a slip cord on the second mark. Neither dog attempted a break the whole day.

I auto-whistled every pick-up.

The course was an indent configuration as follows:
  1. 70 yards into high grass
  2. 140 yards into high grass (this was the flyer station and the mark that Laddie didn't run)
  3. 220 yards into an open area
The course was narrow:
  • #2 was 45° to the right of #1.
  • #3 was slightly to the left of #1.
Here's how the dogs did:

LADDIE

Laddie ran as the first dog and retrieved white dummies. I walked him to the start line and back on a lead.

He pinned #1 and picked the dummy up immediately, but then started to run toward the flyer station. When I called "here", he veered back around toward me and delivered the dummy.

He also pinned #3 and ran the line exceptionally well. The line between the start line and the fall for #3 included a mulch mound, and Laddie was the only dog the whole day who ran over the top of the mound to get to the mark. Again, he picked the dummy up immediately and then ran toward the flyer station. But in this case, he did not respond to whistle sit, whistle recall, or verbal recall, and when he arrived at the pigeon cage, he dropped the dummy and began playing at the cage, being fended off by the gunners there. Meanwhile, I had begun running out in that direction, calling "here" several times. When I had gone out 50 yards, Laddie finally picked up the dummy and came running toward me. I turned and ran back to the start line, and Laddie completed his delivery there.

The following week, I asked one of the gunners at the flyer station for details on what had taken place. The guy said that when Laddie arrived, he dropped his dummy near the cage and sniffed the cage. The birds fluttered and Laddie jumped back. The gunner threw a dead pigeon, and Laddie ran to it and picked it up. Laddie then spotted the dummy, dropped the bird, picked up the dummy, and ran to the handler (me) with it.

LUMI

Lumi ran as the third dog and retrieved dead ducks. I walked her off-lead, except that as we were returning to the van and the gunners were firing guns for the next dog, I noted Lumi's body language and put on her lead for fear she might run back into the field for those marks.

During the series as planned, I released Lumi quickly after each throw, and she didn't look away from any of the falls.

On #1 and #3, she returned immediately. On #2, I could not see her behind the high grass and received no signal from the gunners. After about five seconds of not seeing any movement, I decided that she must have the bird by now and called "here". She immediately came running out of the high cover with the bird. As far as I could tell, she completely ignored the cage of pigeons.

On #3, Lumi ran over one slope of the mulch mound, the only dog besides Laddie not to run around the mound all day.

Once again this week, Lumi seemed unable to see the long thrower, even with waving and calling hey-hey. As I attempted to line her up, she focused either on the middle thrower to the right or the stand of trees, backed by a barn, to the left. After several attempts to line her up on the long thrower, I finally called for the throw while she was looking at the middle Gun, which was not too far to the right of the long Gun. As soon as Lumi heard the gunshot, she shifted her focus to the correct thrower, kept her eyes on the fall, and once I sent her, raced straight out to the mark, detouring only slightly at the mulch mound.

I felt Lumi ran an excellent series. As we were walking back to the van, one of the other trainers, who had expressed concern about Lumi's recall in the past, smiled nicely and said, "Looks like she's got the concept!"

After Group Training. One of the other trainers offered to throw walking singles for Laddie. Using white dummies and throwing each dummy after a gunshot, he threw singles of 60, 100, and 140 yards. I did not use a slip cord, and Laddie was steady on all the marks. I auto-whistled recall on each pick-up.

In each case, Laddie pinned the mark, picked up the dummy uncued and without hesitation, and then, despite the recall whistle, instead of running back toward me, Laddie turned toward the right in the direction of the thrower. Each time it happened, I called "here" repeatedly until he began running toward me, but even on the return, he sometimes swerved to the right again. I'm not sure why he was doing it, but it was most severe on the first mark and least severe on the last.

Training Group Advice. In the course of conversations with other trainers at the group, I received several suggestions:
  1. One trainer repeated a suggestion he's made before, that I send my dogs on the long mark first, not last. Today, he related his own experience in this area. He told me that his first dog was trained initially for Hunt Tests, with much shorter marks, and that dog had great difficulty learning to find the long Gun when they began training for Field Trials. In contrast, his younger dog trained for Field Trials from the beginning, and this trainer had the younger dog run the long mark first when they were running singles. It turns out the younger dog doesn't have the problem that the older dog has in finding the long Gun. I didn't argue with him, but my thought is that it was the older dog's lack of experience with long marks, not the fact that the guy had the younger dog run long marks first, that accounts for the difference in their abilities to find the long Gun.
  2. When I asked a senior trainer about retiring all throwers except the one who was throwing while running my dogs on longer marks first, he said he did not think that was wise, that the dog needs to learn to run past other throwers. I didn't argue, but my thought is that that is an advanced skill and the dog doesn't need to learn the advanced version of the skill from the beginning. A way to ramp up to running longer marks first would be to run them with the other throwers retired the first few times. The dog would still be aware the stations were there, but they wouldn't be as distracting if the throwers were hidden. After the dogs were successful that way, they could be sent while the other throwers were visible.
  3. I asked one of the senior trainers whether he thought Laddie would get better or worse on his returns if I continued training with the group. He answered, "I think he'll get worse." But when I said, "So you think I should stop training him with the group," he said that that was not what he was saying. He then explained that the group session isn't for training, it's for testing to see whether the yard work you were doing during the week was having good results. If you come back the next week and the dog is performing better, you know you're addressing the problem. If not, you know you're not. That seemed like good advice, but I'm not sure it completely explains why he said that he thinks Laddie will get worse. Did he mean, If the only training we do is to come to group training, Laddie will get worse? Maybe that's what he meant. I hope so. Of course if it's not what he meant, I hope he's wrong.
Comments. In previous group training, we worked on particular skills — such as multiples and steadiness — while others subtly, and then precipitously, declined. When that happened, I took the dog out of group training to prevent permanent damage to the dog's development as a field competitor. As far as I can see, that was not happening in today's session.

Lumi's performance was solid, suggesting that she may be ready for another raise in criteria. Some of the areas where Lumi is not working at the same level as the advanced dogs are retrieving flyers, running blinds, running multiples, and running longer marks before shorter ones. The problems with having her start doing any of those are:
  • Retrieving flyers. The risk is that she will begin to resource guard again, and after that will not even be solid on dead birds or possibly even dummies. I think we should stick with dead birds for the time being.
  • Running blinds. The risk is that she will practice using sight, scent, and hunting skills, rather than cues coming from me, to find the blinds. Because finding the blind is reinforcing, running blinds that way reinforces incorrect concepts of how to do it. I think we should stick to private drills that require responding to my cues to meet with success.
  • Running multiples. The risk is that Lumi's recent tendency to look away from the fall before being sent on singles will become an entrenched head-swinging behavior, eventually detracting from her ability to perform on marked retrieves. I think we should stick to singles at group training until we have a well-established ability to maintain focus until specifically cued to turn away from the fall.
  • Running longer marks first. The risk is that Lumi will run to the short thrower, then swerve to the long thrower (possibly requiring help), and finally succeed with the retrieve. It's possible that the success might train her to run such marks that way. However, it seems more likely that this mistake, if it occurs, will not be self-reinforcing since it inserts a delay in getting to the bird, and that therefore Lumi will do it less and less with experience. I also think that I could improve her chances of learning the skill correctly if I arrange for the shorter Guns to hide while running the longer marks. In fact, I've purchased three camo umbrellas that could be used for that purpose. The only problem is that some of the other trainers do not agree with that procedure and I'll have to ask them to help me train in a way that they don't agree with. Perhaps if I emphasize that it will only be for the first few times, that will resolve their concerns.
In conclusion, pending guidance from Alice and Jody, I believe that the next raise in criteria we should try is running longer marks first. I'm not sure how soon we should try that.

As for Laddie, he showed solid progress from the last time we trained with a group. Today, he displayed no resource guarding, no parading, no stalling, no rolling in the grass, and responsive to all recalls except trying to call him away from a cage of pigeons from 140 yards.

The question for me is whether the times he detoured on his return were self-reinforcing. If they were, then we must once again suspend group training until we can solve the problem. If they were not, then it is only a matter of training the correct behavior during the week and testing it at group training as suggested by one of the group trainers, because even if it occasionally occurs, if it is not self-reinforcing, and the correct behavior is self-reinforcing, then the incorrect behavior will ultimately extinguish.

Given the importance of group training, and the possibility that the group will be working on increasingly difficult set-ups in the coming weeks with the risk of leaving Laddie too far behind to join this year if he does not start soon, I am leaning toward the decision that we should continue group training, at least for one more week.

During the next few days, we will focus our private training on a single task: distraction-proofing the retrieve, so that Laddie learns to return immediately with the article no matter what diversions might be present in the environment. Another way of saying that, I guess, is to further strengthen Laddie's recall.

If we are successful, next Sunday Laddie should show distinct improvement, hopefully returning immediately on every retrieve. If, in spite of an intense distraction-proofing effort, Laddie again fails to return, that may indicate that not returning is so self-reinforcing that our private training can't overcome it as long as he continues to rehearse it with the group on Sundays. If that turns out to be the case, we'll know that it's time to suspend group training again.

I hope that's not what happens.

No comments:

[Note that entries are displayed from newest to oldest.]