The goat palace is almost finished. We were hoping to get it done yesterday afternoon, but we didn’t quite make it. The three yearlings are feeling very squashed in the stall by the oldest female, Thanzi. She is making it very clear that they are TO STAY IN YOUR CORNER. I am glad we decided in our construction to use the entire space the lean-to provided and didn’t just settle for making a small goat pen. They will have plenty of room to spread out.
So for this morning it is back to July and the Goat Diaries. I had gotten as far as mid-morning of E and P’s second day of clicker training.
Good training begins to have a rhythm to it, especially in these early stages where you are asking for simple behaviors, and you’re keeping the rates of reinforcement high. It’s get the behavior – click and feed, get the behavior – click and feed, – get the behavior, click and feed. It becomes a training loop. We’re looking for clean loops.
When a loop is clean you get to move on, and not only do you get to move on you should move on. That’s the mantra of loopy training. Often people change criteria too fast which ends up confusing the learners. Or they stay too long at one step so they build a glass ceiling into their training. To the learner backing up means three steps and only three steps. If the handler asks for four, there’s frustration. The learner knows the behavior. It’s three steps and three steps only!
The mantra of loopy training helps you to know when to move on. It also helps you to know when you should pause for a moment to let your learner show you what he has learned. Canine trainer, Kay Laurence refers to these pauses as puzzle moments.
In these early sessions with these goats I was beginning to establish some training loops. P in particular was such a fast learner, it was time to give him some puzzle moments to see what dots he was connecting. If you aren’t sure what a puzzle moment looks like, P is about to show you.
Session 3: 11 am
I started with P out in the pen. He was ready, eager to touch a target, but my attention was elsewhere. I was busy setting up the camera. I was very aware that I might be missing a window of opportunity. We began with a little targeting. He oriented to it, I clicked, fed, and then clicked and fed again while he was still out of my space. The jumping up on me to try to get the food that he had been doing in the previous session was almost completely gone. My active use of food delivery was paying off.
Click for targeting. Feed where the perfect goat would be. The perfect goat would have all four feet on the ground. He would be looking straight ahead, and he would be outside my personal space.
After I clicked, I fed P so he had to take a step or two back to get the food. My concern here was the food delivery caused him to curl his neck so his head was in the orientation it would be for butting with his horns. I didn’t want to trigger that behavior. But head butting is a forward moving behavior. Here he was moving back, so I hoped that his feet would keep his head from thinking he should be charging me.
Get them while they’re standing still.
I fed P so he had to back up a couple of steps to get to the treat in my hand. Before he could come forward again, click, I was giving him a treat – this time where he was standing. I wanted him to get the idea. Standing still, away from me, is a good thing. Click treat, click treat. I was tightening the training loop down to the tiny fraction of a second in which he was standing still looking straight ahead.
The neighbors were mowing the hill up above the barn. P kept turning his head to the side to check them out. His feet were still, but I didn’t want to make such a full head turn part of the behavior. I had to wait, hoping his feet would be still when he finally looked back in my direction. Click then treat.
When I clicked, I used my food delivery to move him back a couple of steps. I wanted to be able to click again while he was still standing back out of my space. I also wanted his head to be straight. If I clicked too many times when his head was turned, I was concerned that I would build that into the base behavior. So I had to wait to click until his feet were still AND he had his head straight. Asking for two criteria at once was pushing my luck. The first couple of times he was too quick for me. He straightened his head, but just as I began to click, he was shifting forward.
I moved him back again with the food delivery. He took his treat from my hand. Before I could click again, he had come forward into my space.
I work hard to avoid putting my learners into a macro extinction process. Here’s what that means: This behavior has been consistently working to get me to hand you treats. Only now suddenly, it’s not. You’re not going to be reinforced for this very successful behavior.
We all know how frustrating this can be. You put your money in the vending machine and nothing comes out. Time to shake the vending machine!
My training rhythm was broken and P didn’t yet have enough experience in the game to know what to do. His repertoire of behaviors was still too limited to offer me something I could reinforce. Instead he was trying to go directly to my pockets. I suspect by this point the small children he had grown up with would have dropped pretzels and peanuts all over the floor and everyone would be happy. The children would be giggling, and P would be gobbling up the goodies. Only this wasn’t how I played the game. How annoying!
P gave a little chuff of a sneeze. I had llamas years ago, so I recognized this sound as a sign of frustration. He tried both my pockets. Nothing. He gave a head toss which I dodged, and then I got lucky. He dropped his head away from me enough so that I could reinforce him. The food delivery moved him out of my space, and we were back on track building good behavior.
Training is not without moments of frustration. I was beginning to recognize what this looked like in a goat. A little tail wiggle, a snort, a head butting gesture – these all told me that P was struggling a bit to make sense of what was happening. Why wasn’t I just giving him treats! That’s what the children would have done. And if they didn’t give him treats, he’d just jump up on them, and that was sure to make them scatter their peanuts and pretzels on the ground!
But here this was different. He was clearly frustrated. Doing what had always worked in the past, namely crowding into me didn’t work. Looking away, taking a step back, produced treats! It made no sense to him, so while it produced treats it also produced a puzzled goat. And a puzzled goat can very quickly become a frustrated goat. Noted.
I was monitoring carefully. Always I am asking myself is this working? Is this the best strategy? How much frustration is too much? What should I change? Should I stop?
There is a time to be clicking, and a time to just wait it out and let your learner work out the puzzle. Through the food delivery, I had shown P the answer. Back away and you get treats. Would he put the pieces of the puzzle together? I waited. The skill here is to be quiet, to remain as non-reactive as you can be and let him figure out the answer. A puzzle you solve for yourself, is an answer you will own.
He could sniff at my pockets. I remained non-reactive. How frustrating! I was not playing the game fair. The children would have been flailing their arms about and pushing him away. Which meant they would also have been dropping treats. Push on the vending machine, and it scatters goodies over the ground, except not now.
His feet took him back a couple of steps. Click – treat. The next time the backing was even more definite.
He caught on fast and began to back away from me. When he came forward into my space, now I could wait. It was a puzzle moment. What would he do? I had shown him the answer through the food delivery. Would he find it now on his own?
The answer was yes! He backed up, not just a little, but multiple steps. And he backed with energy. Very neat!
P was definitely a quick study. He was beginning to understand that he could get the food by doing other things besides jumping up or bumping my pockets. It was a really fun session watching him catch on so fast. Though I got the impression that he was still very confused. Backing was clearly working, but it didn’t make sense to him. How could backing up get treats to appear? He was a very puzzled goat.
I sympathized. We’ve all been given sets of instructions that make no sense. Whatever is logical – do the opposite. How maddening is that! Especially when it works!
I would find out in the next session if P could reconcile himself to this new inside-out world order.
(Note: we had moved on in the treats. I was now using a mix of peanuts, peanut hulls, sunflower seeds and hay stretcher pellets as treats.)
Training time for this session: 6 minutes.
Video: Video: Goat Diaries Day 2: A Quick Study: Note you will need a password to watch this video: GoatDiariesDay 2 E Learns
“A puzzle solved is a behavior owned.” P showed me he was making the connections – fast!
Video: GOAT DIARIES/Day 2/Problem Solving: Note you will need a password to watch this video: GoatDiariesDay 2 E Learns
Coming next: Day 2 Continued – Two Different Learners
I am LOVING the Goat Diaries! you keep laying out rules/principles that I know internally but haven’t verbalized or written down. This is allowing me to examine ideas that were too nebulous to examine. I am one of those people who can show you how I do something but find it really hard to explain my thinking…this may be the missing link. Thank you!
On another note, I remember most vividly an owner with a poodle puppy who showed us how well her pup had learned “leave it” since the last week’s lesson. Her pup ran backwards a good 20 feet or more, and looked SO pleased with herself! “Typical standard poodle” I thought. Seems like your goat is part poodle!
Thank you for a great comment. These goats are indeed great teachers. And yes, part poodle is a good way to think of them.