In this episode we discuss:
- Why we often consider shaping to be an advanced technique.
- Why I think it doesn’t have to/shouldn’t be that way.
- Why I think how many of us first learned about shaping is contributing to the problem.
- How the expectations we have around shaping (based on how we were taught) might be getting in the way of doing good training.
- The two expectations that can get in the way: that shaping is linear and that it should be spontaneous.
- Changing how we think about shaping to consider the whole ABC contingency can allow us to do better training without having to struggle so much.
- You can set yourself and your dog up for success and it is still shaping!
- There is no cheating in shaping as long as you are being honest with yourself (and your dog).
- Setting up for success means a lot more than just training in a low distraction environment.
This podcast is supported by: Zero to CD
Zero to CD is an online group mentorship program designed to provide support, structure, and accountability for people who are new to competition obedience and looking to earn their Novice level title AND to make competition obedience accessible to all dogs and handlers through force-free, positive reinforcement-based methods.
Hannah: [00:00:00] Now, I used to think that I need to put my dog in a very simple, distraction-free environment. We talk about, “Oh, go into your bathroom, close the door, and pull the curtains.” And that’s true; that reduces the distractions by a lot. But you know what? It changes the heck outta the context because the number of behaviors that I actually want my dog to perform in the bathroom are low.
Hey there, fellow training nerds! You’re listening to Drinking from the Toilet. If you like to geek out about combining the science of behavior with positive reinforcement philosophy in real life, you have come to the right place. And I’m your host, Hannah Branigan: teacher, trainer, podcaster and author of the book Awesome Obedience and its companion, Awesome Obedience: The Field Guide, which are both available from clickertraining.com.
I’m also the absent-minded genius behind my [00:01:00] online membership program, Zero to CD, which opens again for enrollment at the end of October, which is this month. This month is October. To be honest, I’m probably more absent-minded than genius, so I’ll have to work on that description a little bit more. But at any rate, make sure you are on my email list if you would like to get the announcement about exactly when the doors open for enrollment for that program. You can go to zerotoCD.com for more info and to get put on that list.
This episode we are talking about shaping and what makes it so hard – and some thoughts that I’ve had, or maybe complaints or concerns, nothing so structurally-sound as a soapbox, but definitely some big feelings that I have around how we think about and talk about and, to a certain extent, apply shaping concepts.
But before we get into that, I’ve gotta send a shout out to a couple of awesome folks for supporting this podcast on Patreon. So huge thanks to Jane L and Wendy S. I love y’all and really [00:02:00] appreciate your support. If you’d like to join them, support this podcast, get your questions answered, join our monthly Q&As and get access to some super-secret extra podcast episodes, you can go to patreon.com/dftt for all the important information about that.
So, let’s dig in.
So first, why is shaping so hard?
This is a question or topic that comes up a lot. I get a lot of emails about it and it’s definitely something that comes up at seminars and workshops.
And I think we have… I’m gonna go ahead and say it, “misconceptions.” Or just there’s this general– I do think it’s a misconception; that’s not a secret. But it’s definitely a general belief or attitude that shaping is only for advanced trainers. It’s super hard. It’s hard on the dogs. You have to have like all of these skills you have to have been training for I don’t even know how many years. And obviously you have to have like some really high-test trainer’s dog: border collie, malinois, something like that.
And I don’t believe any of that is [00:03:00] true. I think shaping is amazing. And if we’re coming to it with the belief that it’s hard, it absolutely is one of those things that becomes something of a self-fulfilling prophecy or completely self-fulfilling prophecy. But I don’t think it has to be as hard as it’s sometimes made out to be. I think we can do good training and take advantage of all of the wonderful things that shaping, AKA “training by successive approximation,” has to offer. And I don’t think you have to be some kind of savant, or just hugely experienced and advanced trainer, to do it.
So I was thinking about it and I really think the reason that shaping can seem hard comes down to two things. First is the expectation that it’s going to be a linear process and second that it’s gonna be spontaneous. And I think a lot of that comes from the [00:04:00] way we were originally taught or how shaping continues to be talked about.
I know for myself, the picture that I had of shaping “growing up” as a trainer involved a lot of sitting still, staring at the dog, clicking when the behavior happens and then tossing treats across the room. And I was given the impression– I’m not gonna say that people told me this or like that it was– Well, I’m not alone– but I definitely had the impression that doing anything to encourage the behavior or make the training itself more convenient was cheating.
And there’s not really cheating in dog training, right? I feel pretty strongly about this. I mean, there’s certainly cheating when it comes to rules in different sports and organizations. You can absolutely be creative and bend the rules there (which is not something I do because I’m a rule follower, except for when I’m not. Well, we don’t need to talk about that). But there’s not really cheating in dog training. The [00:05:00] principles of learning are the principles of learning and we can’t really get around those.
So I was doing some thinking about it, because again, I got a bunch of questions that have come to me in this general area, so I was thinking about it. And I said, “You know what I need to do? I need to do some serious research and I gotta figure out what’s going on in the industry.” So I sat down to do the research in the way that I currently do all of the important research for anything important in my life.
And so I went to TikTok and I checked a few different search terms in the TikToks and I spent like very serious committed time to watching the TikTok videos so that I could get a real feel for what is going on in the state of dog training.
It was interesting to me during this very valid and productive time that I was spending here on this, like totally billable hours– What was interesting to me was [00:06:00] that not a lot has changed, at least not in the first fifty or a hundred videos or so that I watched in this structured time, in how shaping is being presented as a technique from when I was first learning about it from the old email lists and Yahoo Groups.
And that was really interesting to me. I saw a lot of references that I recognized. A lot of trainers or instructors presented shaping as a game of playing Hot or Cold. Maybe you remember playing this game where you’re moving around and someone saying “Hotter, hotter, oh, you’re burning up, oh, you’re catching on fire!” or “cold, colder, freezing cold.” I dunno, that’s how I played it because I can’t just say hot or cold; not interesting enough; I needed more than that.
And definitely I have seen shaping described that way. So it’s just hot or cold and no other information is given. And again, a lot of the examples that were [00:07:00] presented under these hashtags were pictures where the trainer is sitting down motionless to some degree– actually, there’s a lot of different variations of it. But one of the things that I laughed about was when I would see a trainer that stood up or moved around, they would often say something like, “Well, I’m just gonna cheat a little bit here.” And again, it’s bit of a judge-y word and I think we would be better served to ditch it. There’s not cheating in this. There are a bunch of different choices we can make in how we set up our training and they are all of them gonna have trade-offs. Every single one of them is going to have trade-offs. If we come back to shaping, which is just building behavior through successive approximations, that means that you are starting out by reinforcing some rough version of the [00:08:00] behavior and then gradually incrementing towards a more and more refined version of the behavior.
And the more I thought about it, the more I realized that this is really one of the places where considering the whole ABC contingency (that three-term contingency: antecedent, behavior, consequence) as the smallest unit of behavior is really important. Because, again, behavior never occurs in a vacuum. So anything that you’re training – or well, any behavior, not just training. Whether you’re training it or not, any behavior is already attached to the condition of the environment in which the behavior is occurring and the consequence that occurs following that behavior.
So when we say successive approximation, you are not just limited to what you are seeing the dog do, as if they are just performing it against some green screen totally separate from the things before, the things after and what’s occurring in the environment at the same time. The whole successive approximation applies to the [00:09:00] entire ABC contingency: the antecedent, the behavior itself and the consequence.
Anything that we do in the antecedent part of it – in the setup of the training session that is not exactly the same as the cue, the antecedent conditions, the picture/context where your goal behavior is to be performed – that is part of the shaping. The behavior itself, the muscle movements that the dog makes, the actions that that they do, those are part of it. And then of course, the consequence as well. So I really think it’s beneficial to consider how you reinforce or how the behavior is being reinforced as part of those shaping conditions as well.
So now we have a whole bunch of stuff that we can manipulate other than just looking for some small muscle movement that the dog is doing and depending entirely on that.
And I really think that that is where a lot [00:10:00] of us get stuck. We are looking for a behavior to occur. It’s not already occurring and if we make any changes to the A or the C, that’s cheating. Again, there’s not cheating in dog training.
So I might– I’m almost certainly going to make changes to the A. I’m gonna create a setup that is beneficial for the behavior that I’m trying to train, that makes it very likely to occur. And anything that I do in that training setup is going to be further away, most of the time, from the conditions where I need that behavior to be performed, to achieve the goal for why I’m training this behavior in the first place. A lot of the time, but not always, the way that I reinforce is– Again, it’s a temporary thing. This is what I’m doing to make training the behavior convenient and part of the things that I’m gonna need to do in order to make it functional to meet my goal for why I’m training this behavior in the first place is probably gonna be to make some changes to how I reinforce, [00:11:00] including maybe even transferring that whole behavior onto some other reinforcer entirely.
So loose leash walking, for example. I often start training it with food. But in order for it to be functional for me in real life, I’m going to be very quickly transferring it to real life reinforcers in the environment. And because I’m using food to train that behavior, it’s a totally different function, totally different reinforcer, but it’s still shaping because I’m still reinforcing smaller versions, rougher versions, less accurate versions of the behavior initially.
And again, if I’m defining the whole behavior as the whole ABC, it looks different. But my goal is to incrementally move towards the end goal behavior that I have, the end ABC that I have, which in the case of loose leash walking is going to be that my dog moves at my side, stops when I stop, starts when I start, changes direction, matches my speed [00:12:00] and all of the things whether I have food or not, and is doing so in order to access the environmental reinforcers that are already present, like the opportunities to sniff, to pee on stuff, that kind of thing, and forward motion, which is a big part of what tends to reinforce the behavior that your dog does on leash, whether it’s loose leash walking or not. Forward motion tends to reinforce whatever behavior is occurring in that moment.
And again, those choices are always going to have trade-offs. No matter what you do, there’s going to be trade-offs. There are absolutely advantages to sitting in a chair and being totally motionless while you’re training a behavior. The primary advantage there is that you’re really eliminating any unintended prompts and that’s legit. I will often still sit– well, on the ground most of the time because chairs are mainly for storing laundry that you’re gonna fold at some point, but you can also sit on laundry….
But anyway, sitting. I will often sit when I wanna take my body [00:13:00] out of the picture or just outta the equation temporarily, still in the picture because again, behavior never occurs in a vacuum. Whatever the picture: how you are standing, where you’re standing, the room that you’re in, the time of day, the temperature, the smells, the footing, all of that is part of the environment. It’s part of the context, it’s part of the A, it’s part of the antecedent, it’s part of the cue, and that’s something we gotta be aware of.
But I definitely – for myself, and after some amount of time of instructing a lot of different teams in dog training – I know that some of the hardest pieces of that antecedent picture to eliminate are those unintended prompts, the body movements and positions that the trainer is doing that they don’t know about. We are the worst at evaluating our own behavior and it’s so easy to accidentally introduce a small amount of movement or a particular body position or [00:14:00] something you do with your hands that you don’t notice during the training process. And then because the behavior’s being reinforced in the presence of whatever you’re doing with your hands and your body, that’s part of the cue.
I can think of an example immediately. I’ve been working on Figment’s position changes. So going from sit to down to stand and sit to stand to down, and all of the different– there’s six different variations there. And I was doing it in such a way that the way that I was standing – and I was standing, not any way in particular on purpose, but just standing in a way that is comfortable for me so that I could access the treats that I was using to reward – and I discovered that the simple action of me dropping my hand to my side after delivering a treat and standing a certain way had become the cue for him to stand. So me standing completely still, giving him a treat and then just dropping my hand on my side, [00:15:00] which is about as neutral as it comes, had become the cue to stand because I had reinforced sit to stand in too many sessions too many times with that exact picture. And it wasn’t a cue that I was adding. It wasn’t a prompt I was adding to the session with any kind of of intention. That was just me existing and feeding him and then trying to return to home as quickly as I could without adding anything to it. And just through repetition, it became the cue.
So sitting a chair might be one of the ways that I remove that part of the cue. There’s a lot of things that I can do to vary what I do with my hands so that it isn’t part of the cue. And I didn’t realize that was happening until I tried to do something else with my hands and I discovered that the cue disappeared, or I tried to get some other behavior to occur and I discovered that, “Oh, this picture has become the cue to stand.”
Those are the hardest pieces of the picture to isolate, especially if you’re training by yourself, which I think most of us train mostly by [00:16:00] ourselves and that’s hard. It’s a challenge. There’s ways to solve it, but that’s not what this podcast episode is about. And anything that I do, if I’m standing up on my feet, I’m gonna be shifting my weight, if I’m delivering food – all of those little movements that I’m not considering as part of the session are becoming part of the session and they’re becoming part of the cue. So if I sit in a chair, it changes that picture, and at the very least, my feet aren’t doing anything. So I’m not moving towards the dog. My movements are less of the cue. But then of course sitting in a chair could become part of the cue.
So sometimes I will have teams, you know, “Let’s try putting you in a chair just to take your feet outta the picture so you have less to think about.” Fewer moving parts on the part of the human is usually a good way to bring focus to something else and there’s an advantage to that. Being able to toss food anywhere, across the room yes, has an advantage of having the dog move out of position so that you have an automatic reset built into your loop. So those things aren’t wrong or bad, but they’re also not the only way [00:17:00] to do shaping. If you were to be standing up, if you were to take a step and move to a different location, every repetition with every loop after every treat, that doesn’t make it any less shaping. It’s just a different trade-off. You’re just introducing a different variable into the picture, which is gonna have a different trade-off.
Luing doesn’t make it not shaping. I’ve talked about this one way at the beginning of the podcast. A couple of years ago, I did an episode (DFTT #2) on luring versus shaping and how they’re not opposites. You can still be reinforcing successive approximations of the behavior, of the whole ABC picture, and also be luring for part of it or all of it. The trade-off, of course, being that whatever you’re doing to lure the behavior is becoming part of the cue.
Sometimes that doesn’t matter. Sometimes I can always lure the behavior. I don’t care if I have to lure it because that becomes the cue to the behavior. And other times, that’s something I want to get rid of pretty [00:18:00] quickly. So whether or not that’s to my advantage depends on, to a certain extent, the goal. What am I training? What’s my goal for this behavior? How do I need it to be performed? What are the cues I need to be part of that picture? How much of a problem would it be if luring or the actions surrounding luring were a part of that? Or how quickly do I fade it? How quickly do I increment the A side of my shaping plan in order to get to a different looking A, a more accurate version that’s closer to where I do need the ABC to be for my goal?
So I really do think we need to rethink the picture of what we consider shaping to be. A lot of what we’re talking about here is– or what I’ve been talking about so far– a lot of what I’ve been ranting about so far is addressing that aspect of the behavior being [00:19:00] spontaneous. And that’s definitely an expectation that I’ve run into a lot with folks.
In fact, I remember way back, a long time ago, one of the trainers that I was working with in agility – and this was several generations of dogs ago – I remember me talking about how I’m very excited and very into clicker training and I’m really enjoying shaping. And she said she didn’t believe in shaping because her dogs were never going to spontaneously perform a set of weave poles. And that kind of caught me for a second because I was like, “What does she think shaping is?” And again, this was a long time ago. Well, time as a construct, but anyways, I have no idea if that person still has that belief or not, because I had a bunch of beliefs at that point that I absolutely have let go of and moved on from and I hope nobody is holding me to those. Please don’t. Please don’t hold me to beliefs I [00:20:00] held two years ago, honestly. Anyways. Reserve the right to change my mind at any time. It’s my prerogative.
But back to that, what really is shaping? It comes down to that successive approximation.
So what would not be shaping? Well, successive approximation is that incremental reinforcement of different stages of the behavior, which can look like a lot of things. The only thing it really doesn’t look like is reinforcing the whole behavior occurring spontaneously and entirely. That’s not shaping. That’s capturing, potentially. And I agree. I think even if you were to put however many monkeys in a room waiting for them to write Shakespeare, if you were to put that many dogs into a room and wait for them to perform 12 weave poles spontaneously, you’d be waiting very long time. And dogs don’t live long enough for that and frankly, neither do I.
But that’s not what shaping is! Shaping is “how can I manipulate the environment to make a piece, a slice, a component of the [00:21:00] goal behavior likely to occur, and how can I deliver reinforcement?” And then the “successive” part of that is going to be then what increments do I need to make to the A, to the B and to the C to get to my final goal that I’m training towards?
That’s the setting up for success that we do talk about. Now, I used to think when I heard– and this is where I’m kind of limited in how I was thinking about behavior– when I heard “set up for success,” what I envisioned was “that means that I need to put my dog in a very simple, distraction-free environment. So we talk about, “Oh, go into your bathroom, close the door, and pull the curtains.” And that’s true, that reduces the distractions by a lot. But you know what? It changes the heck outta the context because the number of behaviors that I actually want my dog to perform in the bathroom are low. Almost zero. Very close to zero. Almost all the behaviors that my dogs do in the bathroom, certainly when I’m in there, are ones that I would rather [00:22:00] they’d not do. In fact, when I am my very best self and I have very little other things going on in my life, I do like to train and reinforce my dogs to wait at the bathroom door and actually not enter the bathroom at all when I go into the bathroom because for whatever reason, they all need to come with me into the bathroom and then like, invade my personal space when I’m on the toilet and I don’t enjoy that.
I got sidetracked there, but my point being, setting up for success probably does include being thoughtful about what other stimuli are present in the environment, but does not necessarily mean going into your bathroom. That may be not at all beneficial for your training plan. That may be counterproductive for your training plan. It may be closer or further away. Again, this is where it’s about trade-offs. Like what makes sense for your goal and what makes sense for the increments, the [00:23:00] approximations for that behavior that you’re gonna be shaping, be reinforcing.
So that’s one of the things that’s struck me lately as I’m teaching other folks to do shaping, as I’m talking about shaping, as I’m answering questions about shaping, is totally unconsciously– but it’s very true that I find myself focusing a lot less on, I’m gonna use air quotes here, “criteria,” as defined by “what I see my dog doing, like the actions my dog is doing, the movements that he’s making.” When I’m making my shaping plan, I’m thinking a lot less about that and a lot more about how I’m setting up the session, that antecedent arrangement aspect of it. And then also, what can I do with the C. What kinds of reinforcement strategies, reinforcement procedures, benefit the behavior that I’m trying to train, maybe not necessarily– or I’m gonna keep in mind the goal for that behavior, but benefit the particular [00:24:00] slice in this setup, in this antecedent arrangement that I’m working in.
So again, behavior never occurs in a vacuum. So when we are setting up the training session, we’re arranging those antecedents: where I’m standing, how I’m standing, the position that I’m in, the movements that I’m making, the room that I’m in, what else is in the room, or maybe I’m outside, time of day, like everything that is part of the time space continuum that is where the behavior’s going to be occurring, all of it counts. And all of that plus your dog’s previous learning history, all of that determines what behaviors are likely to show up right now, and that includes – it is not limited to but it absolutely includes – your dog standing and staring at you.
The reason that I bring that up is because I think one of the [00:25:00] biggest problems that folks reach out with, and I personally certainly have seen (in other people’s dogs, obviously, and maybe what I’ve run into) is where you want to shape a behavior, and the behavior you have in mind is one that involves your dog moving in some way and the behavior that your dog is exhibiting in this moment is standing and staring at you.
And that’s a really common one. It’s a really common place to get stuck. And I suspect a big part of that is: one, the antecedent picture of a training session looks very similar and/or completely exactly identical the same. No matter what behavior you’re training, you’re wearing your treat pouch like so, you’ve got your training pajamas on, you’re training at the time of day that you have available before the kids wake up or whatever.
And you’re standing there with your clicker and treats, you’re ready to reward, or your ball, whatever, and your training vest. I don’t know how you train. But you have a way that you look and a way that you sit up when we’re gonna go [00:26:00] do some dog training. Maybe you’ve got your phone in a tripod. Maybe you’ve got it sitting in a coffee cup. Maybe you’ve got it wedged in a Kleenex box. I don’t know how you video your training. Maybe you don’t video your training at all, but you should. That’s another podcast, episode two.
But that whole picture is a training session and probably for a lot of us, one of the earliest behaviors that we have reinforced the most, so it has the most just mass, the most volume of reinforcers received in the picture of what looks like a Training Session, is standing and staring at you. It gets reinforced a lot and there’s nothing wrong with that. I like my dogs to look up at me because it feels good. It gives me some sense of self-worth, which I’m really desperate for. So I like my dogs to look at me, and it is something that I reward early and often. It also means that on the cue of capital T Training Session, that’s the first and most likely behavior that I’m gonna see unless I [00:27:00] deliberately and strategically reinforce other behaviors as well in that same umbrella antecedent picture.
And because I really do get a lot of self worth from my dogs looking at me, I don’t ever want my dogs to experience extinction for looking at me. Like, I don’t want the behavior of looking at me to be extinguished. I don’t want them to experience extinction conditions around that behavior at all, if at all possible, because I don’t want my dogs associating the training picture and looking at me with frustration. I don’t want that to be part of our training relationship in general and I certainly don’t want frustration-related behaviors – particularly things like barking, biting me, sniffing or wandering off – I don’t want any of that to be associated with looking at me.
So I wanna be really careful with that. I want to have a plan for how I start my shaping session, what that looks like, so that [00:28:00] staring at me is never extinguished for sure, but also doesn’t get any of the junk that comes along with that frustration. Frustration being a function of extinction. I don’t want any of that in those training sessions.
And I think that that’s one of the biggest problems that we run into with that old idea of shaping, which is focused entirely on criteria, is focused entirely on the behavior, and it’s somewhat rigid in the picture of “you have to be standing still and rewarding in as inconvenient a way as possible for the behavior that you’re trying to train.”
That picture also includes a lot of waiting. It includes a lot of the dog staring at you and getting increasingly more desperate and frustrated when you continue to appear to have frozen like a Zoom call with a bad internet connection. I maintain that there should be very little waiting that occurs in a training session, whether you’re going to categorize it as shaping or not.
[00:29:00] You know, here’s another thing. I think all good training sessions include a certain amount of shaping. They just do. Whether you’re calling it shaping or not, there should be some shaping involved. Again, unless you are capturing the whole behavior in the context where you want the behavior performed, using the reinforcer that you want the behavior to be performed for. Did that sentence make sense? The function of that behavior. Anything short of that’s going to be shaping. And so all good training sessions are going to have a shaping element to them. They may have other things going on as well, and all of that is okay, but also shaping.
So there should be very little waiting. There should be very little extinction. There should be as limited as possible your performances of your dog staring at you and starting to squeak and bark and whatever, and you sitting there doing nothing. A lot of bad stuff happens in that space and it doesn’t need to! We can do really good [00:30:00] training, we can do better training, without that. Right? By being mindful of the antecedent picture that we are creating, how we’re setting this up, from props to your position, all the A, all the C, how you reinforce, all of that is part of your training setup.
And a good trainers using a clever training setup and, asterisk, is aware of the trade offs. I think that’s the piece, to circle back to the cheating thing– When you’re not aware of the trade offs that you’re making, that’s when you get into trouble, and that’s when it can maybe feel like cheating or seem like cheating. No, not cheating. I’m mindfully making this training choice very aware that by bringing a prop into the picture, very aware that by standing on the other side of the mat or the other side of the platform, that that’s becoming part of the cue.
And if I need the– I’ll use that as an example. If I wanna train my dog to go to a platform – this [00:31:00] applies to any kind of target, but I’ll just say platform to keep it simple. So I want him to go to a platform and my goal behavior is that at the end of the day, I want to be able to send him to that platform from 50 feet away. So the end picture there is the dog is in the starting position– which is part of this. Starting position is part of your shaping setup. It’s part of the A, really; it’s part of the conditions. The dog is standing next to me. The platform is 50 feet away and I am sending the dog away from me to get on that platform.
Now, if I were to start there standing next to my dog, I have the treats, the dog– and I’m not gonna change the C here, I’m just going to have that be food reinforcement from my hand. But if I were to set that up so I’m 50 feet away from the platform, I’d be waiting a very, very long time. Right. So I could shape that behavior 50 feet away from the platform – and don’t get hung up on the number, distance away from platform doesn’t matter here – and the dog is near me and the food is on my person. [00:32:00] And I could start by clicking when the dog looked in the direction that the platform happened to be.
Now 50 feet away, my dog’s probably not aware the platform exists on the planet. Particularly Rugby. He’s aware of shadows and sniffs and all kinds of things, but not that. And then hope, by clicking a head turn in that direction and then rewarding from my hand, that he tried more head turns, and then the head turn became a step and became another step and became a third step and a fifth step until eventually the dog is moving the whole 50 feet, however many steps that would be for a dog – let’s say 50 steps (it’s gonna be more than that; depends on the side of the dog) to the platform for the treat that’s still in my hand.
That’s a long, tedious process. That’s hard. That would be making shaping really hard.
A different way, still shaping, is that I start with the platform very close. I put the platform right in front of me and my dog is right in front of me. So the platform is on the tips of my toes [00:33:00] and it’s on the tips of my dog’s toes. And he can’t get any closer to the food, even if I don’t move my hand at all. I just stand there and I have food and my dog has a learning history of taking food from my hand, which he probably does at this point. And if he’s gonna move any closer to me at all, he happens to bump against or step on that platform. Bam. I can click and I can reward from my hand. I’m still keeping reward for my hand – I’m pinning that one as our C here so I’m just changing the A. And maybe I toss a treat at that point and then he comes back towards me and I click and treat from my hand and maybe I toss another treat and he comes back towards me, he bumps the platform, click and treat from my hand.
Now, we could also vary the size of the platform. I could have maybe start with a full size Klimb platform with the tall legs on it and everything. For a big dog, that might not be a big deal, for border collie or an aussie or something like that. Might not be a big deal for a small dog. Might be a bigger deal for Mastiff. Might be a real big deal. It’s one of those weird paradoxes in dog training. I could take the legs off that platform and I could put it flat on the ground. [00:34:00] I could also start with a enormous platform, like barn door. And I don’t really mean a literal barn door, but it could be a literal barn door, but not at the price of lumber these days.
But it could be something really, really big and really, really inviting that your dog’s very likely to put his feet on, he can’t help but put his feet on, because it takes up so much of the training space that he’s gonna happen to step on it. I can’t think of– I don’t know, a gymnastics mat. But I could modify variables like that to make the stepping up on the thing– So it’s very low, it’s very inviting, and it’s so big he can’t help it. Bigger surfaces tend to be easier to step on than smaller surfaces, right? (Like except for contacts; those are very easy to step on no matter how big. And piles of dog poo. Doesn’t matter how big your yard is. If there’s one turd in there, I will find it with my foot.)
But anyways, those are things that you would modify in the A that are still part of shaping and they are just as valid as trying to shape your dog to move fully across the yard just by [00:35:00] reinforcing their actions. And which one’s going to go faster?
Well, I can tell you. Changing where you put the platform and where you stand, particularly with your dog’s reinforcement history of getting food in front of you from your hand, it’s gonna make contact with that platform a lot more likely. And then once you’ve reinforced that, maybe then you shrink it to a Klimb platform and then maybe you put the legs on it. Maybe you do short legs first and then taller legs. Then you could add distance to it. And because your dog has been reinforced multiple times, many times – hopefully at this point, they have a huge reinforcement history, tons of deposits into that bank account for putting their feet on and stepping up onto that platform – then moving it a little bit further away from the source of reinforcement isn’t that big a deal because now the behavior has context. That’s one of the advantages to back chaining as a strategy – both back chaining multiple behaviors, but also back chaining within a shaping session like this
That gets a lot easier. Now it’s, “Oh, the game is to put my feet on the thing” and then you can increment how far away from you. So [00:36:00] now changing the A. So now the dog is facing me, I’m facing the dog, but the platform is a few inches off center from being the space right between us and he has to take a little bit of a step onto it. And now it’s a foot away and now it’s two feet away and now it’s 10 feet away and now it’s 30 feet and 50 feet. With context, that often goes faster and you get more reinforcement history for that by starting with it a lot closer.
And this example also is part of why I love thinking about behaviors in terms of components. Whatever behavior you’re looking at, even if it’s just a sit, I guarantee you, I can find ways to divide that behavior into little component behaviors and then those behaviors can be divided, like, infinitely. This is very exciting to me. But certainly going to a platform– If you think about the function of getting on the platform is to get food from your hand and then the function of going to the platform is to get on the platform so that you can get food from your hand. Now you have kind of [00:37:00] this component aspect of it.
I can give context to the whole thing by creating reinforcement history. One of the ways to get access to the food that’s in my hand is to put your feet on this surface, this weird surface that the human is really interested in, for whatever reason. And then add a component of going to that platform to then get access to the food that’s in your hand. So much easier because you’re building the function onto the “going to” part of the behavior. Right? Like that’s the advantage here of breaking it down in this way. And you lose that if you’re trying to shape your dog to go across the yard. You’re also making it a lot harder because he’s leaving the food to get to the food. It’s just– yes, that’s doable, but that’s the hardest way to make that happen. There’s a lot of easier ways.
Now, yeah, you could change the C, right? You could leave the platform where it is and when your dog turns his head in the direction of the platform, you could mark and then throw food towards the platform. That would change the C part while leaving the A the same.
And you [00:38:00] could do both. You could change the A in all the ways that we just talked about and drop the food on the platform. That might also be part of it. And then as you increment the platform further away from you, you could do all of those things and toss the food towards the platform. Those would all still be shaping and that would all still be really valid and probably a lot better and a lot faster and a lot more effective than trying to keep the A and the C the same and just increment the behavior itself, because that’s gonna be really tedious and yeah, really hard.
All right, so we are over 30 minutes on this and we’ve just gotten into digging into the spontaneous piece of why shaping is so hard and that maybe if we think about it a little bit differently, it doesn’t have to be quite as spontaneous as you might think. And we still have other things to say regarding the linear aspect of it. And I have a lot of other thoughts as well. This is not even 10% of the ranting that I have available for you on this subject, [00:39:00] but I think we need to do a part two.
So I’m gonna leave you here and I’m very curious to hear, just for historical context, how did you learn about shaping? Like what does shaping mean to you or what did it mean to you a few years ago? What does it mean to you now and how do you think about that? Because I really am very interested in kind of the anthropological or is it sociological? I don’t know; neither of those are my area of expertise. The human experience in clicker training and dog training and particularly in this case shaping.
So hit me up. Find the post on Facebook or Instagram for this episode and let me know, or you can certainly send me an email. I would love to hear from you.
Thanks for listening. If you like this episode, well, you have good taste and I hope you’ll hit the subscribe button on your podcast app to make sure you don’t miss the next episode. It might be even better than this one. If you are already subscribed, well, thank you. I really appreciate it, and there are still some ways that you [00:40:00] could reinforce me if you were so inclined. You could always leave me a five star review on iTunes or Stitcher or wherever you happen to be listening to this podcast. And you can also check out and support the sponsors because they help make the podcast possible. You can find links and information about them and the other things that we’ve talked about in this episode by going to the show notes, which can be found at www.wonderpupstraining.com/podcast. And while you’re there, you could also pick up a free PDF training template to help you plan your training sessions. There’s also some other articles and previous podcasts and that sort of thing, which you could always find if you were interested. So until next time, happy training.