Today Lightning turned 7mo. For our training, I took him and Laddie to our closest training property, about a three hour round trip. I had hoped to train with some others, but we ended up working alone.

As I’ve described in previous posts, Lightning is now in PRT Stage 2, corresponding to Mike Lardy’s TRT Basics, which means we’re working on yard work and field work in parallel. At the training property, he and I did some of both.

We began with some location-proofing for the Fetch Game that Lightning has been learning in yard work the last couple of days. We had trained it indoors and had also practiced it in our yard, and now we were able to practice it in a field training location with the associated environmental distractions, but at this stage without the distractions of a training group nearby. Lightning has developed a reflexive, instant, lunging response to the Fetch cue, picking a bumper up off the ground and delivering it to me. He showed it in today’s training at the new location when he was right in front of me as well as when he was some distance away. 

After the yard work, we drove around and settled on two locations for field work. At each location, I ran Laddie on some blinds with diversions and keyholes, and then created a setup suitable for Lightning’s current field work objectives for the circumstance where we’re training alone. It’s been too cold lately for swimming, so we were only able to work on land.

My two field setups for Lightning were mirror images of one another. In each case I set up two stickmen, one at around 20y, the other at around 35y. I used a mat for our start line. I would have liked to use a holding blind behind the mat, but the combination of a stiff wind and rock-hard ground made that impractical.

At both locations, I ran Lightning on an angle away from the wind. Judges don’t usually set up marks for the dog to run into the wind because they’d be able to follow the scent to the bird.

I used 3″ white bumpers with streamers, and pistol shots, for all marks. These were poorman marks. For the single marks, I left Lightning in a sit at the start line, walked out, fired the pistol, walked back to him, and sent him on his name. For the doubles, I left him at the start line, walked to the first station, threw the first mark, walked to the second station, threw the second mark, walked back, sent him to the go-bird (the second mark I’d thrown), and when he delivered that, sent him to the memory bird (the first mark I’d thrown).

For the first setup, I ran Lightning on four poorman singles and one poorman double, as follows:

1. Single to short station, thrown outside the angle made by the two stickmen.
2. Single to long station, thrown in the opposite direction, so again outside the angle.
3. Repeat of #1.
4. Repeat of #2.
5. Double made up of the same two marks, the long one thrown first and retrieved last.

For the second setup, I ran Lightning on two singles, a double, and a final single, again all poorman marks, as follows:

1. Single to short station, thrown outside the angle made by the two stickmen.
2. Single to long station, thrown in the same direction, so this time inside the angle. 
3. Double made up of the same two marks, the long one thrown first and retrieved last.
4. 60y single, the only mark of the day not thrown from a stickman but instead thrown from a position in the angle between them, and on an angle back to a fall also in that angle. The bumper was visible in flight but past the crest of a rolling hill on the ground, so Lightning couldn’t see it till he was close to it.

Lightning ran straight to every mark, including the memory birds for the doubles. Not all of his returns were great, but they got increasingly better. For both marks of the last double and for the long final single, he picked up the bumper, came straight back to me, and held the bumper till I took it. I’m not ready to require delivery to hand yet, much less bringing him to heel to deliver, but he even did that for delivering the first mark of the double, positioning himself facing the field and ready to launch toward the memory bird when sent.

Though I didn’t have an assistant to act as gunner today, I felt Lightning was nonetheless making progress on our two current field work objectives with today’s session: running singles with multiple guns out, and lead steadiness. He was dealing with stickmen rather than human gunners, but hopefully it was similar. And he wasn’t practicing lead steadiness because he had to be steady while sitting by himself at the start line, but hopefully his steadiness training was still benefiting.

So that’s where Lightning is at this stage in his training.

Hi, everyone, Lightning here. As a follow-up to the awful photos Daddy posted earlier, here’s a video of Ryley and me romping in the field, with guest appearances by Mommy and Laddie.

 

 

I thought it might be interesting to post some photos I took in the bedroom this afternoon to show how small Lightning still is.

At just shy of 7mo, Lightning is just over 40lbs. For contrast, I’m showing him next to Laddie, my 8yo 68lb field Golden. Ryley, DW Renée’s 9mo confirmation-bred Golden, not shown, is up to 72lbs at last weighing and hasn’t grown into his paws yet. I don’t think Lightning is ever going to catch up to Ryley in size.

I think Laddie and Lightning are both gorgeous dogs in beautiful condition. Sorry my attempt at photography doesn’t show it.

 

 

Well, my estimate for how long it would take Lightning to learn the Hold behavior chain I described in my previous post was far too pessimistic. For the indoor training, it actually took six short sessions, five yesterday and one this morning. For this last session, I broke a piece of Muenster into eight pieces but we only needed six trials, alternating left and right heel and in various indoor locations, to establish that that chain is now solid indoors. I don’t think it will take long to proof outdoors.

But that probably won’t happen today. After temps in the low 80s at times last week, this morning we had wind chills in the teens with winds gusting to 45mph.

In the meantime, we’ll start the next step of our yard work. In the TRT program’s section on Force Fetch, Mike Lardy calls this step “Ear Pinch”. In our PRT program’s section on Formal Fetch, we’ll call it “the Fetch Game”.

After again reviewing Mike’s video presentation on Ear Pinch, giving a sense of our performance objective, we’ll again depart entirely from the training method we’ll use to achieve it.

The behavior sequence for the Fetch Game in PRT is as follows:

1. Leave the dog in a sit with the bumper somewhere near him/her. Don’t let the dog pick up the bumper yet. 
2. Walk a short distance from the dog and turn to face him/her.
3. Cue Fetch. While you remain still, the dog is to grab the bumper and start running toward you.
4. Run away shouting and laughing, darting left and right, possibly hiding behind trees or other objects.
5. When the dog catches you, bring him/her to a sit in heel position, grasp the bumper, and cue Drop.
6. Put the bumper down again somewhere in the dog’s vicinity, and walk away again for the next trial. For the hardest version, put the bumper behind the dog.

The theory behind the Fetch Game is that for Lightning, and I think for most retrievers and other predators as well, one of the best parts of a retrieve is the chase. The trouble with the normal retrieve pattern is that the chase comes first, making the return and delivery something less to look forward to, that is, less reinforcing.

The Fetch Game reverses that pattern. Instead of a chase to the prey, the chase comes after the pick up, providing one of the strongest intrinsic reinforcers in the dog’s psyche. That’s why it’s important not to start running away with all that excitement until the dog has grabbed the bumper. The grab itself is the behavior we want to reinforce.

Although our behavior chain is different from Mike’s, we’ll follow the same model he does of shaping the behavior incrementally. For example, we’ll place the bumper between the dog and us at first, and we won’t require the full delivery sequence at first. If the dog doesn’t run to us as soon as we say Fetch, we’ll say Here a moment later until that’s no longer needed. If the dog doesn’t pick up the bumper on the way, we’ll do a mini-retrieve at first, tossing the bumper a short distance, cueing Fetch, waiting a moment while the dog runs to the bumper, and running away excitedly at the instant the dog grabs the bumper. After two or three trials, we’ll experiment with fading out the toss and just placing the bumper on the ground until the dog has the idea. If the dog simply doesn’t seem to understand that Fetch means to grab the bumper, we’ll start with a session or more of incrementally shaping that behavior, much as Mike does in the video but using treats for reinforcement rather than ear pinches. Then we can build the really high reinforcement history for Fetch using the Fetch Game.

I’ll leave it to your imagination to come up with your own solutions to building the Fetch Game behavior chain incrementally, so that you can get to the point of being able to play the final version. If you’re lucky, the dog will respond correctly from the first trial, as Lightning did.

You could start this game indoors, as I did because of the unpleasant weather, or in the yard from the beginning. Unlike the comparatively difficult to train Hold cue, your dog, like Lightning, may pretty much understand Fetch immediately given his/her months of previous work in the PRT program.

Once learned, proof the Fetch Game for a wide variety of locations and distractions. For example, a cool version would be to place a thawed bird in some shallow water, cue Fetch, and have the dog grab the bird, chase you, and deliver the bird after getting to you thru some reeds or swampy terrain. 

Even after Lightning is fluent with the Fetch Game, I plan to continue practicing it occasionally for some time. I believe that the more I practice reinforcing him with a chase after he has picked up the article on the Fetch cue, the greater will be his reinforcement history for grabbing the article on cue and delivering it. But I’ll be sure also to practice a Fetch, Heel, and delivery right near me, since that’s a common situation. Then, his Fetch response will be available that day when he drops a bird on the return during a competition and I need him to pick it up and complete the delivery to avoid disqualification. The judges won’t be thrilled with the dropped bird, but hopefully they’ll be pleased with Lightning’s well-trained Fetch response.

Referring to the TRT Flow Chart’s “Basics” section and the right column called “Field Work”, I think Lightning is a bit ahead of the game on lead steady. But I want to continue working with him on singles with guns out till his returns with multiple guns out (or stickmen out) are satisfactory.

Meanwhile, I’m satisfied with his performance on formal obedience for the “Yard Work” column, especially because we’ve now integrated those skills into every retrieve in the field, so we’ll continue to work on them continuously anyway.

Therefore, we’ll stay with the first two items for field work, but we’ll go onto the next item for yard work. In TRT Basics it’s called “Force Fetch”, but in PRT Stage 2 we’ll call it “Formal Fetch”. I’ve also heard it called “Trained Retrieve”. Whatever the name, it consists of training three cues: Fetch, Hold, and Drop. (I’ve always used “Out” but we’ll switch to “Drop” for consistency with TRT.)

You can learn a positive approach to training a formal retrieve from several online websites and videos. Generally they use an approach called back-chaining, which trains the cues in the reverse order they’ll occur. So first you train Drop, and then Hold, and finally Fetch. As the dog learns Hold and Fetch, he/she already knows what comes next, so the whole chain accumulates reinforcement history.

For PRT, however, we’ve already trained a version of Fetch we called “Take it” and a version of Drop we called “Out”. Lightning has been pretty fluent with both for a couple of months. So other than the names, only Hold is new to us at this point.

Coincidentally, Hold is where Mike begins training Force Fetch. That’s what we’ll work on now for Formal Fetch as well.

But at this point, I’ll depart entirely from Mike’s approach. Rather than copying Mike’s style of teaching Hold, let’s define a specific behavior chain and then shape it, that is, incrementally improve and reinforce each version of each step until the dog can perform the entire chain fluently.

Our definition for Hold will be that the dog be able to:

1. Stay in a remote sit holding a 2″ bumper while we walk some distance away.
2. On verbal Here or whistle Come-in, come to front position and, with a verbal or whistle Sit cue if needed, sit in front position while still holding the bumper.
3. On Heel and a hand gesture, stand up, swing to heel position on either side, and, with a verbal or whistle Sit if needed, sit while still holding the bumper.
4. Continue to hold the bumper even if we place our hand near or on the dog’s head or chest, or touch the bumper’s rope, or tap the bumper in any direction, without saying Drop.
5. When we finally grasp the bumper and say Drop, the dog should instantly open his/her mouth and pull his/her mouth away from the bumper.

Because of Lightning’s earlier training as described in previous posts, and no doubt thanks to his instincts as well, sometimes, in a non-distracting room of the house, he is already able to do an approximation of that entire behavior chain. He even seems to find it self-reinforcing, so on those occasions I don’t seem to need any extrinsic reinforcement such as treats, tug, or a thrown ball.

But other times, we need a traditional shaping session. The first one, for example, looked like this:

1. I tried sliced hot dogs but Lightning wasn’t interested enough. So I switched to a small bowl of cream cheese, which I dipped my finger in to give him a taste when I wanted to reinforce a correct response.
2. I tried working from a standing position but leaning over was too hard on my back so I switched to sitting down on the lower steps of our stairs.
3. We did about thirty trials in rapid succession.
4. For most trials I started with a verbal Sit cue, since Lightning generally popped up after each trial. Sometimes I just let him keep standing.
5. I presented the horizontal bumper in front of his mouth. Most of the time he just grabbed it. Occasionally he didn’t and I said “Take it,” and then he’d grab it.
6. This next point might have been the most difficult for an observer to see, but it was also the most critical. Lightning would let go of the bumper after a split second. I needed to use the cream cheese to reinforce the Hold moment, not the let-go moment. Yet I also had two other imperatives: First, I needed to maintain a high Rate of Reinforcement (ROR) to keep him fully engaged, having as many successful trials per minute as possible. And second, I needed to gradually increase the amount of time he held the bumper before trying to drop it.
7. Because my timing was reasonably good, it gradually dawned on Lightning that he would not get the reward if he opened his mouth before I said Good (I also could have used a clicker), but he would get the treat if I said Good (or clicked) while he still had a tight grip. Once I saw that dawning comprehension, after eight or ten trials, I began saying Hold the instant he closed his mouth on the bumper.
8. Then, gradually, he held on a tiny fraction of a second longer after each few trials, though with plenty of backsliding.
9. When he was clearly, intentionally, holding on for nearly a full second, we did one more trial, the best of the session. I said Good and took the bumper, and I presented the bowl for him to lick the rest of the cream cheese.

So as you can see, in a span of a few hours, we trained the behavior chain as if we were close to completing the indoor phase of this training, and later had to do a beginner’s session just trying to get Lightning to hold the bumper for a full second. For a third session later that day, with a slice of Muenster cheese folded into 16 pieces, we still couldn’t get a good version of the behavior chain, but Lightning had jumped between sessions to being able to hold the bumper in a sit for several seconds. For the final few trials, i tossed his piece of cheese behind him so he jumped up and ran to get it, then came back to me. While he was standing in front of me, I presented the bumper to him and cued Sit. The first time I did that, he dropped the bumper, and of course no treat. The next several times he sat down without dropping the bumper. Yay! Hold is a difficult skill to train.

In fact, it’s worth pointing out that as with many natural retrievers, after months of the dog retrieving  marks, it would be reasonable to think, Oh, my dog already knows how to retrieve. That’s where the word “formal” comes in. Up till now, Lightning’s been doing an informal Fetch approximation. It won’t stand up in competition. It’s the formal training we’re doing now that will prepare him for the rigors and distractions he’ll someday face in the exhilaration and demands of an event.

On a theoretical note, I believe that traditional trainers view this training as forcing the dog to perform the various skills of Force Fetch. I see it differently. What I see happening is that faced repeatedly with a particular scenario, the dog, having experienced the outcome of various choices, learns to choose the behavior that produces the most desirable outcome. In other words, what feels best. From that perspective, Mike and I are doing much the same thing. Mike’s dogs feel best when their chins don’t get cuffed. My dogs feel best when they get a reward. The results are the same but I get to avoid using a physical aversive.

Over the next few days and weeks, using a large number of short sessions, we’ll continue shaping Lightning’s indoor performance until he’s solid on every step of the behavior chain, as well of course as being able to hold like the dog in Mike’s video. 

When we finally have the behavior chain trained indoors, we’ll take it to the yard outside the house and re-train it there, and then again at various training locations.

When we’re done, this behavior chain, eventually minus the extra Sit in the middle, will be our delivery to hand and we’ll use it for all future retrieves. (We’ll need to train Hold for coming out of water eventually, too, but we’ll discuss that later.)

The Hold behavior chain won’t finish our yard work for the Formal Fetch, but it will satisfy our requirements for the first step, Hold, and as it happens, it will also satisfy our requirements for delivery to hand.

How long will training Hold take? That depends on the dog. I think about three weeks for Lightning, but again, Hold is difficult, and it could take longer. Nonetheless, it’s worth every second of the investment to have a solid Hold and delivery to hand.