Transcript of Episode 261: Common Misconceptions Around Shaping: Why You May Find Dog Training Frustrating

Speaker Key

SG Susan Garrett

Transcript

00:00:00

In Shaped by Dog episode 259, I talked all about shaping and I said, if you wanted me to do a troubleshooting episode, just leave me some questions. And I got a lot of questions. So, all of that and more. And from those questions, I realized I don't need to do another shaping episode here on Shaped by Dog. I need to do several. And here we go with the first one. Your questions answered all about shaping.

00:00:37

Hi, I'm Susan Garrett. Welcome to Shaped by Dog. And wow, did we get a lot of really, really great questions. And I realized that there's a misconception about shaping.

And I think it's because most podcasters, most blog posts, most people, and probably including myself at one point, they describe shaping with the phrase successive approximations, meaning you reward the dog for doing one little behavior and that leads you to another little behavior and then another little behavior and another. 

00:01:08

So, the dog will eventually do what you want them to do. But from where I see things as a trainer, that can potentially lead to a ton of frustration for the dog because we're always expecting more. And that stresses and creates a lot of anxiety for dogs and a lot of pressure.

The other problem with that is we get a lot of what's called cheap behaviors built into that a little bit more and a little bit more and a little bit more when what we really, really want isn't super clear, both in our minds and in the dog's mind. So, shaping kind of gets a little bit screwy. 

00:01:43

I'm going to give you an example with my own dogs. Back in the early nineties, I was shaping both my Jack Russell Terrier and my Border Collie to do the exact same behavior. And my goal was to teach them to do a go out for obedience, a send away as my European friends might call it.

So, I started shaping them linearly using successive approximations with a clicker and a handful of cookies, me standing very, very close to the wall, clicking them for looking at it, for looking at it longer, for maybe getting closer to the wall. Eventually, I got them to put their paws on the wall. It was slow. 

00:02:18

And both those dogs were driven to work. So, I got a lot of anxiousness in their behaviors, a lot of franticness, but I got the job done. I got both of them to hit their two paws on the wall. But the next step in my training plan was I wanted what was referred to as a twofer, meaning I want the dog to show me they understand the behavior by I'm not going to click the first one and I'll click the second one.

And so, my Jack Russell, I did first, Twister. Now, when I was adding a twofer, I was maybe a stride away from the dog and they were leaving me and going and touching the wall. There was no cue given. It was still early in the process. So, Twister left me, went up, hit the wall, turned around to come back to get the cookie, and it was almost as if I could see the wheels turning in her head. And she went, “Wait, I didn't get the click. Did she not see me?”

00:03:08

And she turned around and went like higher up on the wall with more force as if to say “Look, I did it.” And of course, I fell over laughing, clicked her and gave her reinforcement. So, from then on, if I didn't click her, she wouldn't leave the wall.

She would just keep pounding on the wall, I mean, that was exactly what I was looking for. And so, my Border Collie who was brilliant at shaping anything, I was going to try her with a twofer next. And here's what she did, is she touched the wall, and I didn't click. But she didn't do like Twister did. 

00:03:43

She didn't come off and start walking towards me. She stayed there with her paws on the wall and she kind of looked over her shoulder at me like, “I'm doing what you asked. Do you not see what you've done here? Like, when are you going to click that? Do you want me to hold longer?”

And she just waited me out and I thought, “Okay, okay, okay. I see what I've done.” I clicked at the exact same spot every time, and she thinks the game is being in contact with the wall. So, I'm thinking through this all in my brain and I said, what I'm going to do now is I'm going to click her when she comes off the wall. So, she understands, ‘no, no, no, it's touch and come off.’ 

00:04:18

So, I'm waiting, I'm ready. But what she did is she looked up the wall further. And then she moved so that she could put her elbows and her paws on the wall and looked over her shoulder at me thinking, “Well, if my paws aren't enough, maybe you want more of my body on the wall.” And I'm like, “Okay, okay. She can offer something else.”

This is what successive approximations are about. You wait until they offer something that you can click. And so, what she did then was she looked down and she scooted her rear paws as close to the wall as she could. So, she got her entire body laying on the wall as if she was in the freeze position. At which point I fell over laughing because it was hysterical. 

00:05:01

I remember it like it was yesterday and it was probably somewhere in, I don't know, 1993 that that happened. And so, with successive approximations, there's just like a gamut of things that could go wrong. And if you have a dog who's driven to work, they may not give up, but they may get a little bit frustrated.

And with that frustration with the higher drive dogs, you might get vocalization. You might get them showing anxiety in other ways. Like you might get a spin and then a behavior, or you might get a little nip and then a behavior. Now, not all dogs are like that, are they? So, with dogs that aren't driven like that, they're going to go, “Oh, I don't get it.” 

00:05:42

And they might just go and lay down or they might like scratch their ear or stare at you. And so, many people think of shaping a success of approximations and that is something I want you to release. Send that out into the interwebs and let it be free to roam.

Because what I want you to think about it, what I described in the last episode is outside the box shaping. It's shaping by using behavioral blocks that your dog knows. Now, every single one of you listening to this, I know there are behaviors that you have reinforced in your dog that you have either given them a cookie, you've given them a permission to do something, you've given them a toy.

00:06:23

So, I'd like you, if you're driving in your car, just think about this. But when you get a chance, write down what are your dog's behavioral blocks, things that they have earned reinforcement for. It could be something like a sit, a down, a stand. I'm sure many dogs have received reinforcement for those three things.

00:06:41

Those are behavior blocks that could come in handy. For example, if you were trying to teach a dog to crawl on their belly, yet they'd never ever received a cookie for lying down, can you see how difficult it would be to get that dog to offer a down position when it's never been a previously reinforced position. So, list all of the behavioral blocks in a journal somewhere. So, you'll know what you have to work with when you're trying to shape a behavior. Okay, I'm going to get your questions.

00:07:09

I'm going to first remind you of some of the things we talked about in the last episode, how to have a successful shaping session. And I'm going to flesh that out a little bit more. So, super important that you have a hierarchy of reinforcement, meaning food that your dog goes, Cuckoo for Cocoa Puffs about.

Do you have reinforcement where the dog goes, “Oh my gosh, yes!” And do you have reinforcements when I say a hierarchy, if you have a dog like for example Tater Salad, if I use his number one reinforcement, he might get too crazy. 

00:07:38

Which is kind of funny because he was a pretty laid-back dog who really didn't want to work when he arrived here as a 15-month-old rescue dog. But he gets so jacked up. So, I would use like lesser value reinforcements for him when I'm shaping a more precision behavior. So, hierarchy of reinforcement.

And for those of you who say, “My dog really doesn't want to work.” That is where you're starting. You're creating a hierarchy of reinforcement so that your dog will take the food. And I spoke about that in episode number 259. Second, you're going to have those location specific reinforcement markers. 

00:08:12

‘Cook’, ‘search’, and I really encourage you to use ‘chow’ as well, which is ‘there's a cookie in the bowl. You can take the cookie out of the bowl.’ Or for those of us who are raw feeders, I put a spoonful of the food in the bowl.

Now, if I throw a cookie beside the bowl and I say “search”, they aren't to take their food from the bowl. Only take the food from the bowl if they hear “chow” or whatever location specific reinforcement marker you're going to use. 

00:08:40

So, cook, come into your mouth, search, look for it on the floor. And search is a great cue when we're working to create a targeting. Remember in episode number 259, I had you shape the dog onto a blanket. We made that blanket smaller and smaller. Now we have a paw target.

Well, the way we're going to build in more and more reinforcement for that dog, finding value in putting their paws on a target, is we're going to use what's called a “reset cookie.” So, that's where your cue, you would say “search”, which tells the dog you can get off and look for a cookie on the floor.

00:09:13

And I'll throw that cookie a little bit behind my dog so that they come back up and they find that target super easy. Now for a dog that's more experienced, I really want to challenge them, I might throw the cookie behind me. So, then they have to figure out how to come around and face me again on that target.

Now your dog in between, you're going to throw somewhere in between those two spots. So, the reset cookie is only possible if you have a dog who understands that location specific reinforcement marker ‘search.’ Super important. 

00:09:43

ItsYerChoice. If you are trying to shape a dog who doesn't understand ItsYerChoice, the dog is going to be so obsessed with the food in your hand, they're not going to be able to think about what to offer. Also, if your dog doesn't have really good ItsYerChoice, you can't use the location specific reinforcement marker ‘chow.’

Because if I say, want my dog to back up in a straight line, one of the things I might do is put a bowl with a cookie in it somewhere behind so that as I get a couple steps, one place I might reinforce that dog is by rolling a cookie between their front legs. But I also might just say “chow”, so they turn around and get the cookie right behind them. 

00:10:23

That might help them to not go in a crooked line when they're backing up. So, the use of “chow”, I use it pretty much every day when I'm training my puppy right now, I will use “chow” somewhere in our training. And that's just not possible without ItsYerChoice. Alright, Crate Games.

Crate Games and Hot Zone. Now I said you could do one or the other, but I got to tell you, if you work through all the stages of Crate Games, you have so many more of those behavioral blocks. Meaning you have a dog who will run to their crate when they see the crate or when you give them the cue to go in their kennel. 

00:10:57

And that is a behavioral block that means your dog has been reinforced for leaving you and traveling a great distance to do something that will come in handy when we want to do other things at a distance. So, Crate Games and Hot Zone. I would really focus on Crate Games, but you can do them both for sure because Hot Zone allows you to train more than one dog at one time.

Short sessions. Okay, here's what I'd like you to do after today. I want you all to commit to doing five sessions that are one minute or less. Video those sessions and come back and tell me what you learned. And remember, that's not the only piece of homework I gave you. I also asked you to write down all the behavioral blocks you know your dog has right now. 

00:11:39

And number six is related to number five. Please video your shaping and review the shaping. And there's two things I want you to look for. Number one, you're going to look at what did you expect and what did the dog do?

Now you're going to go back and you're going to ignore the dog when you look at the video the next time. What did you do? Because you're going to see a connection in between what the dog did and how you delivered the reinforcement or how you were standing or where you were looking.

00:12:07

So, really be critical of your mechanics because that my friend has the biggest impact on whether the dog has success or not. How did you arrange your antecedents? Spoke about this in the last episode, but it's just so important. Now, if I was teaching the dog to back up, I would probably start with myself kneeling on the ground.

If you're getting a dog to back up and you were standing up, then the dog might be thinking of things up here. When you're on the ground, the dog is going to be thinking a little bit lower. Plus, your placement of reinforcement can be more exact right between your dog's legs. So, that's encouraging them to back up more. 

00:12:43

So, the antecedent arrangements, it's not just what are the other distractions around you. It's your placement of reinforcement or how you're holding your tools or how you are sitting, standing, or kneeling yourself, how is that impacting that dog's ability to grasp the concept of what behavioral building block you are looking for them to offer you right now.

Because if you do this right, the correct response should be so obvious to the dog. And rather than waiting for the dog to offer something and the dog like trying to grasp something from outer space, like successive approximations sometimes look, both you and the dog are engaged in the process that you're a part of because it's happening fast. 

00:13:29

Now there may be some lulls in your training, but they're very, very few and far between. Also, if you're training, if you're shaping looks chaotic and frantic and all of a sudden, the dog gets a click and a cookie, then that also is not taking you in the path in the right direction. Because you will be building in all that frantic anxiety into the behavior that you're building.

And who wants that? None of us wants frantic or anxious in our dogs. Number eight, resist the urge to talk to your dog or help the dog. If the dog stalls out, give them a moment to process. And if it doesn't, maybe after like 30 seconds or a minute, the dog doesn't look like they're moving towards anything, then just break it off. 

00:14:11

Have them hop it up in the Hot Zone and rearrange your antecedents. Because if the dog stalls out and you help them by giving them verbal prompts or prompting them with your body or giving them a finger point or giving them a formal command, then you're not really shaping, you're telling.

Now, also along with that, I want you to be really conscious of your emotions. Don't sigh. “*Sigh*, like you ever going to get—.”, don't groan. You can be happy when you are marking and reinforcing the dog. Sure, show that kind of emotion but don't show disappointment because your dogs are going to feel that. 

00:14:48

You're going to be neutral. And yes, I can't help but get excited when I see my dogs doing what I expected. So, it's okay to celebrate with the dog as you're giving them the cookie but go back into the place of neutrality as you are just an observer of behavior looking for something to mark and reinforce.

Number nine, remember that evaluation minute. Every single session has to start with an evaluation session that's one minute or less.

00:15:14

And finally, I'm going to remind you one more time. Shaping shouldn't be about successive approximations. It should be about arranging behavioral blocks so that the dog can successfully move through things that they've previously been reinforced before.

Now, you might want to teach your dog to close the cupboard door with their nose. “But oh, Susan, I've never reinforced my dog for closing the cupboard with her nose before so that means I can't shape it.” No. Have you taught your dog to nose target? Well, that's behavior block one. What if we put a piece of tape on your hand now? Ah, behavior block two, he can target tape. 

00:15:49

Now let's put that tape say on a fly swatter or something that you can get away from your hand. Can they target then? Yeah, they can. Now let's put that on a cupboard door and extend the fly swatter to get your hand out of there. Can they target it now? Boom. We've got behavioral blocks that in a very speedy way has led our dog to offer the behavior we were wanting.

Okay and finally, yes, I'm getting to your questions. So, question number one. Should I be using a non reward marker when I'm shaping? No. Your antecedent arrangements are arranged in a way that the response you're looking for is the obvious response. So, no help from you. It is all on your planning, your setting up of the dog and the dog's offered previously reinforced responses. 

00:16:36

What is a good exercise to help a dog to learn to be okay with offering responses? Well, something as simple as the location specific reinforcement marker of search and the blanket as I spoke about in episode number 259. That exercise will get every dog motivated to offer paw targeting. And if it doesn't, chances are your reinforcement isn't high enough value to the dog. 

I've heard the comment, click for action and reward for position. Do you agree? You know, that is a great little rule of thumb to help differentiate between, are you shaping a stationary behavior or are you reinforcing a behavior of motion? 

00:17:14

So, so often people want their dogs to run away from them and then they click and reward them back at them. But I would click and reward the dog by throwing something and I wouldn't even click, honestly, I would just use a verbal marker like ‘good’ and throw their reinforcement out there to them. Now ‘click for action, reward for position’ isn't always 100 percent what we do because I've already mentioned reset cookies, right? By saying “search”, we aren't really reinforcing for position, but we're intentionally reinforcing to create a reset that allows a dog to do what we're looking for.

00:17:47

I try to shape, but my mechanics suck and my dog and I get very frustrated. So, I've heard shaping isn't for novices. Is this true? So, shaping is for everybody in my opinion because the more you do it, the better you get at it. If you and your dog are getting frustrated, that comes back to your antecedent arrangements. And I would go back to the paw targeting, start with something simple and then grow from that. Again, successive approximations are probably going to frustrate you and your dog more than shaping with behavioral blocks. And as I mentioned earlier, that when the dog gets frustrated, you'll get cheap behaviors like barking, whining, pawing at you, things you don't want are going to get built into that behavior. And that's going to be even more frustrating for you.

00:18:33

Susan, do you cue shaping sessions? No, because to me, they're just dog training sessions. And so, you know, how I've arranged the antecedents, or my environment is a pretty big cue to my dogs that we're about to learn something new or work on something that we've been working on in the past. 

What do you do if the dog keeps getting it wrong? I would end the session. Have them jump in the Hot Zone, give them a reinforcement for that. And then I would go to my video and evaluate what part of the antecedent arrangements were in opposition to what I really wanted my dog to do. 

00:19:06 

Now, if you have a dog that's been shaped a certain trick, and they just keep offering that trick over and over and over again, you can interrupt it by maybe doing a collar grab and moving them. But again, I really like the dog to figure things out for themselves, but in a way that doesn't frustrate anybody. And so, if I can rearrange the antecedents and create an environment where the correct is super obvious, that would be my first choice.

00:19:31 

Does shaping work with all behaviors and tricks that can be taught or shaping only for specific things? I use shaping for everything. So, I can't think of something that it can't be taught with. Some things I just can't fathom how to shape. Okay. Leave me a comment. Let me know what it was. 

Does this work with all breeds, even unintelligent breeds? Yikes. I personally don't think there are unintelligent breeds. I believe that there are breeds that are better suited for some skills than others. 

And yes, shaping works for all, not just dogs, but parrots and hamsters and rats and your backyard crows and squirrels. I mean, there's so many things. I don't want to encourage you to feed wildlife, but all animals learn by shaping even yes, us people. 

00:20:15 

Okay, a lot of things to take on board here. I want you to jump over to YouTube and leave me a comment. Let me know, what are your dog's behavioral building blocks, the behavioral units that your dog has been previously reinforced for that you can use in shaping another behavior.

And if there's a specific behavior, you'd like me to walk you through what that looks like, then I'm happy to do that. And in an upcoming episode, I'm going to talk about what we can do for those dogs that are just so frantic, it's hard to shape them. 

00:20:44 

But I'm pretty sure that if you re listen to this podcast and go back and listen to podcast episode number 184, where I talked about Jean Donaldson's ‘Push, Stick, Drop’, that you will have all the tools to absolutely fix what's been going on in your shaping session. But jump over to my YouTube channel. And while you're over there, please give this episode a thumbs up. But leave me a comment, let me know what more you want to know about shaping because I have two more episodes planned for you. I'll see you next time right here on Shaped by Dog.