E81: Hannah Branigan - Behavior Chains & Competition

Summary:

Hannah Branigan, of Wonder Pups Training, is back on the podcast to talk about behavior chains, and a little bit about her new puppy (because he's adorable).

Hannah is the host of Drinking from the Toilet and blogs at wonderpupstraining.com.

Next Episode:

To be released 9/28/2018, an interview with Shade Whitesel about crucial concepts when training for competition.

TRANSCRIPTION:

Melissa Breau: This is Melissa Breau and you're listening to the Fenzi Dog Sports Podcast brought to you by the Fenzi Dog Sports Academy, an online school dedicated to providing high-quality instruction for competitive dog sports using only the most current and progressive training methods.

Today we have Hannah Branigan of Wonder Pups Training back on the podcast to talk about behavior chains, and a little bit about her new puppy because he's adorable.

Welcome back to the podcast Hannah.

Hannah Branigan: Thank you for having me.

Melissa Breau: Absolutely. To start us out, do you want to remind listeners who you are, share a little about the dogs you have, and a little about the new addition?

Hannah Branigan: Sure. I am Hannah Branigan, in case you've already forgotten. I have my own podcast, Drinking From The Toilet, as well as teaching for FDSA and playing in a lot of different dog sports. Obedience is my primary focus, that's where I spend most of my time, but I'm really interested in doing really good training, and diagnosing and breaking things down, and reverse-engineering cool behaviors and high-quality performances, and figuring out how we can systematize that and then how can we teach it.

For my own dogs, I currently have five dogs. I have three Belgian Shepherds, one Border Terrier, Rugby, and my new puppy, Figment, is a Border Collie.

Melissa Breau: Let's start by talking about Mr. Figment. With a new puppy, what have you chosen to focus on, and I do assume the goal is to eventually do obedience?

Hannah Branigan: Definitely we'll do some obedience. We'll probably do a lot of different things. I cross-train with all of my dogs. Again, obedience is my central focus, but we tinker in a lot of sports, so right now I'm not doing anything with him that you would consider sport-specific, nothing that is exclusive to obedience or exclusive to agility or exclusive to anything else.

My theme for him right now, my word for him, is really balance. So what I'm trying hard to do is looking at the puppy that I have today and building his skills in how he and I interact together, how he interacts with the world, and trying to develop balance, because he's definitely got … he has his preferences, I of course have my preferences because I'm old and set in my ways, but there are things that I want to build with him. But while I'm building these new behaviors, I want to make sure that I'm not creating a lot of reinforcement history for a particular picture that will make it harder for me to then get other behaviors to change that picture later on.

That was really vague, so let me give you an example. For example, I do want to do sports with him, and so we play a lot of motivational games, drive-building games, where he's moving and I'm moving and there's a lot of excitement and a lot of arousal. At the same time as I'm playing those games in a training context, I am juxtaposing them with less exciting, less arousing, more relaxing, more thinking sorts of games. So we might play with his ball for a few minutes and work on some toy skills, and then immediately in the same session we'll go to some mat, and shape relax on a mat with some food, and then we might go back to working on some toy skills, and go back to relax on a mat.

That switching between food and toys, and again being able to balance that, is something that I'm focusing on a lot in terms of his reinforcement, and also just in terms of all the behaviors that we're working on.

Melissa Breau: I've been doing quite a bit of thinking lately about how much every dog is really unique and their own individuals, and that means that any new dog is going to be totally different than any dog any of us has owned or trained before that one. How do you go about getting to know a new dog or puppy? For that matter, what do you want to know about them?

Hannah Branigan: I'm not doing anything particularly formal in that way. We mostly just go through life, and as we come across something like a cat, or an unfamiliar person, or whatever, I'll pause and I'll observe his response. Maybe it's like, "Oh, that's not a big deal," and the behaviors I see in this context are totally fine and appropriate, great. I'll just maintain that. On the other hand, sometimes we'll see a car moving, and I'll observe his behavior — and I put observe in air quotes because his behavior is quite obvious in that context — and I'll say, "Oh, that's not really what I was looking for, so maybe I need to do some training here to get a behavior that I can live with when we see cars going by."

Really I'm just living my life with my dogs the way that I would normally, and then, as we get to a new situation or some new picture or some new experience, I'll just keep an eye on him and ask him the question, "How's this for you?" and just be ready to look for his answer, and then adjust what we do at that time.

Melissa Breau: Is there anything about Figment that feels new for you in terms of training?

Hannah Branigan: I would say the biggest thing that's different with Figgy is that he's the first dog that I've ever had personally in my house that is easier to stimulate with toys than food. In fact, his food-eating behavior was not really present when he first came home to me, and it disappears rather easily if we go to a new situation. When we get out of the car at the park or at Panera or something, he may not take food right away. There are situations where I've accidentally over-fazed him and he's not taking food or toys, but there are a lot of situations where we get into a new environment, or there's people, or there's other dogs and they're moving, and he'll chase a toy but not take food.

That's a new thing for me, because for my other dogs, they would usually take food first, and then the toy play would be the more fragile behavior. With him, he may not take food at all, but he'd be willing to chase a toy. And there's a little bit of a trap there, because, one, having worked so hard to get Rugby to play with toys with me, it feels nice to have a dog that is so easy to stimulate with toys in that way. So it would be very easy for me to say, "Don't worry about the food. We've got toy play, and that's sexier anyway, so I'm just going to use the toy."

Having worked through this with students and clients, I know that that wouldn't benefit my long-term goals, that wouldn't fit that theme of balance I'm looking for with him, so I know that, "OK, you're not taking treats; I need to go back a step. Where were you taking treats last? Where can you eat food?" and work from there, rather than blowing past that threshold where he can't eat food anymore and then just lean on the toys to get through it, because that way is not going to get me what I want in the long run. It's going to be very limiting and I'll eventually get stuck.

Melissa Breau: I'm guessing at this point he's probably little enough that he hasn't been introduced to much in the way of actual behavior chains yet, right?

Hannah Branigan: Certainly not formally. Behavior chains are around us all the time, that's just part of how we naturally function in the environment, but we don't have really anything on cue. Everything that we're doing training-wise is just on building fluent reinforcement behaviors, and a few movement skills, and those foundational individual behaviors, and way down the road we'll work on turning those into more formal, finished behaviors. Once we have those really trained to fluency, then I'll worry about creating sequences or behavior chains from a performance standpoint.

Melissa Breau: Before we get into this stuff too much, can you explain a little bit what a behavior chain is?

Hannah Branigan: Sure. A behavior chain … and you will hear words like "behavior chain" and "behavior sequence" can be used, there's some debate in the behavior world as to what the definition between a chain versus a sequence is. There isn't a lot of agreement. So some folks have their definition, and other folks have a different definition. I'm going to go ahead and just use behavior chain and behavior sequence pretty much interchangeably and talk about the phenomenon as a whole, and we'll worry about splitting down the jargon later in another podcast episode.

Melissa Breau: Fair enough.

Hannah Branigan: A behavior chain or a behavior sequence is a sequence of behaviors that are held together with cues. Those cues can come from an outside source, like if we're talking about a dog, they might come from the trainer, the handler, but they can also come from the environment.

If you need to get in your car to drive somewhere, you'll perform a behavior chain. You'll walk out of your house, so that's one behavior. You'll see your car, and the sight of the car tells you, "Oh …" more specifically, the sight of your car with the door closed is a cue in the environment: If I'm going to drive my car, I need to open the door. So that's your cue to reach your hand out, take the door handle, pull the door open, which of course is its own little behavior chain. We can really zoom in quite a lot on these and go crazy.

Now you're standing in front of the car with the door open. That's your cue to get in the seat. Now you're sitting in the car. That's your cue to maybe close the door. You're sitting in the car, it's not running, that's your cue: "Oh, I need to put my key in the ignition," and then turn the ignition.

Each behavior that you perform creates some change in the state of the environment and your conditions around you that then signal you to do the next step, and the next step, to eventually reach your goal of backing out of your driveway and driving to the grocery store to get some more Oreos or whatever. So that's an example of a behavior chain where your actions changing the environment are functioning as your cue to do the next thing.

And of course if you didn't know how to do each of those individual behaviors, you'd be in trouble. If you were faced with — and this has actually happened to me in a rental car before — you get in the car, you close the door, and you cannot find where to put the key because it's one of those weird cars with the new …

Melissa Breau: With the start button.

Hannah Branigan: Yeah, and you just sit there for a while and you're like, "I guess I'm going to have to go ask someone for help." The chain broke because there was a behavior that was required in the middle there that was not yet fluent. Once you learn how to do that behavior, now you know how to start this new, weird, funky car, and you're able to complete your behavior chain.

Melissa Breau: For anybody wondering, you have to step on the brake and then push the button. Usually.

Hannah Branigan: Yeah. I had to go back in and ask for help.

Melissa Breau: I had to ask for help too. I could not figure that out. To get back to dogs for a minute, what are some examples of behavior chains that are helpful for competitive obedience?

Hannah Branigan: In competition the whole performance is a behavior chain, from the time you get your dog out of the crate to when you put your dog back in the crate, or back in the car, or whatever, after the performance. That whole sequence of events is a behavior chain, really. And it's one of those combination sequences where some of the cues are coming from the environment and some of the cues are coming from you as a trainer.

For example, we'll look at the retrieve. You throw the dumbbell, the dog is still sitting at your side, and then you're going to give a verbal cue to send the dog to go get the dumbbell. At that point, though, you're no longer giving any cues. The cues are all present in the environment.

So the dog leaves your side, runs out, the sight of the dumbbell tells the dog what to do with it. "Do I put my feet on it?" Hopefully not, because hopefully you've trained your pickup to fluency. So he sees the dumbbell, he scoops it up with his mouth on the bar of the dumbbell, now he's got the dumbbell in his mouth, that's his cue to return to you. He sees you standing there with your arms at your sides in that formal soldier kind of posture, that's his cue to come sit in front. At that point you cue the give and then you cue him to come to heel. So you've given two of those cues for those behaviors, and the environment has given the rest of them. So that's one exercise.

But then again, that whole performance is a behavior chain, and we have sequences of behaviors that we have to perform to move from one exercise to the next. So once you're done with the retrieve on a flat, now you need to cue a series of behaviors to move your dog into the retrieve over the high. Once you do that one, which is another little behavior chain, you have another behavior chain that links the retrieve over the high to moving over to the broad jump.

I think we forget that there are still behaviors happening between "Exercise finished" and "Are you ready?" So that's something that I have made a personal project to figure out.

Melissa Breau: I know your upcoming class gets into this, and specifically into the behavior chains, in a big way. Can you share a little bit about that?

Hannah Branigan: Yeah. It's called "Unchain Your Performance," I think, and it's kind of a joke because I can't do anything seriously. But the idea is that in order to unchain yourself from that feeling of having to have food in your pocket, having to have a treat in your hand, and get multiple behaviors, we can build behavior chains, build these sequences of behaviors, that end in the primary reinforcement, the food, the toy, whatever. That's the little plan, which I thought was hilarious but maybe not everybody gets.

But anyway, it is a common problem, in that it is the nature of dog sports that we have to leave all of our sources of primary reinforcement, all the stuff that we've been using for months or years to train this dog to do this particular set of behaviors, and we leave them all over here on our chair or in our crate, and we have to go all the way into this other space, which is usually clearly delineated by ring gates or hay bales or ropes or something, but it's quite clearly a different room in the environment, and there are no food or toys in there. And it does not take many repetitions for your dog to catch on to the idea that "She left all the food over there, but we're going into this room where there is no food." That's not the whole thing that makes performance hard, that makes competing hard, but it is a piece of it, and it is a piece that we want to account for.

So one of the things that we can do is we can teach the dog that by being very systematic, I can create this sequence of behaviors that does lead to access to the reinforcement, even when it's back at your chair. I can communicate that concept through successful approximations, which is shaping in a way, in that we're gradually building these increments of performance by holding these sequences together in a continuous stream of cues and response and cue-response and cue-response that leads into the ring, we do the performance, and then out of the ring, and bam — that's when you get access to your reinforcer. All of that is connected instead of having a big gap where we've left the food and then nothing happens, and then I come ask you to do this whole bunch of things, and then there's this gap, and then magically food appears.

So it doesn't work to just feed at the end of the performance. There has to be that connection, and we need to condition the dog, we need to teach them that, "The reinforcement you're getting right now, here at this chair, is related to the behavior you just performed."

Melissa Breau: You mentioned in there this idea that it does give people freedom from treats and toys, at least on their body, in order to get into the ring. But when you're talking about clicker training and things, folks often say not to bother loading the clicker anymore because dogs figure it out. If we're consistent about rewarding when we leave the ring, would a dog just figure it out that after they compete they get good stuff; it's just in the next little bit of the room?

Hannah Branigan: Well, yes and no. If you were consistent with how you performed that whole sequence of events, yes, it would probably totally work. That's not usually the case in real life.

We certainly have examples of dogs being able to catch on to sequences of events as cues that reinforcement is coming. For example, if you feed your dog at 5:30 every night, those behaviors that you're doing that are leading up to dinnertime, your dog is totally aware of, and when you close your laptop and stand up and start walking toward the kitchen, your dog's already responding to that. That consistent flow of events that you're creating with your behavior there effectively becomes the conditioned reinforcer aspect would walk forward in time from scooping the food into the bowl to walking to the kitchen … to closing the laptop and then walking into the kitchen.

So yes, that effect is there. However, it does depend on how good of a predictor your behaviors are, which requires consistency. When we're talking about something in the real world, like a performance, there's a lot of stuff that happens in between. If we're talking about a 2-, 5-, maybe 8-minute long obedience run, that's a lot of stuff to depend on just repetition.

There's a lot left to chance. Could the dog figure that out? Yes, and probably some dogs do. But we can do better than that by breaking it down and being systematic and deliberately building that from the ground up, or from the chair back, or however you want to think about it, so that we know for sure that there's a connection, instead of just hoping that if I do this a hundred times, eventually it will come out right.

Your dog can be learning so many things, and if you're inconsistent in that space, like sometimes you come out of the ring and you continue straight to your crate and you give them the meatball, but sometimes you come out and somebody says something to you and you look away from your dog, and then you forget where your chair is, or you get there and … stuff happens. The bigger that space is, the more likely that there's some external variable that's going to come in and mess up your party. It's easier, it's more effective, to be more deliberate about how you build that.

Melissa Breau: It feels like, looking at this, that there are two approaches to this whole deal of helping your dog understand that the ring has value. So the idea of building reinforcers you can bring INTO the ring with you, like personal play, which I know is a big topic at FDSA, and then this concept of teaching the dog that the good stuff will come when they're all done and they get back to their chair or their crate or whatever. Are some dogs better suited for one approach or the other?

Hannah Branigan: I don't think it's an either/or kind of thing at all. It's just incorporating … however, again, the whole performance is a sequence of behaviors, so part of that sequence can and should, in my opinion, include a series of, we'll say, very easy behaviors, preferred behaviors, like play. Ideally, play is a very high-frequency, it's a very easy behavior, it's very cheap for your dog because it's so fun. So there's some behaviors that are easier as part of your ring performance. Some of the behaviors you're going to be asking of your dog are a lot harder, like finding the correct scent article under pressure while there's some strange man staring at him, holding a clipboard, and being very judgy.

So it is, I think, very helpful to apply a little bit of the easy-hard-easy pattern that we use in training a lot, where you ask for something a little bit harder and then you ask for something easier and you reinforce it, and then you ask for something a little bit harder and you ask for something easier. We can play that pattern out in the ring by "I'm going to ask you to find the correct scent article, return with that, take it, call to heel, exercise finished, now I'm going to cue an easy behavior, which is playful interaction with me," whatever that looks like for you and your dog.

Those are behaviors that need to be trained and conditioned separately, out of context, and then you can work them into your performance. But they still have to be something you have to condition the dog to expect. We have to build it. It can be systematic, be deliberate about building those easy behaviors into the performance so the dog knows what to expect. That's part of what makes it easy.

Easy things are things we expect. He knows that "Oh, and then we'll play, and she'll pet me, and at the end of that play session there's another call to heel, and we go do another thing, and then there'll be another play, and she'll pet me, and then she'll call me to heel, and do another thing, and then after that we get my leash on, and we go to my chair, and I eat my cookies."

So they can have all of those things, and it's not one or the other. It's looking at the whole picture and being thoughtful about how do you want this to happen, what are the actions you're doing, the cues that you're giving to your dog that tell the dog what to expect, what's next, what behavior should he perform, and then being ready to give those cues so that you're maintaining that continuous interaction that we would call it connection in the ring.

There is a common phenomenon that I experienced with my own dogs, I didn't really know what was going on, and I've since seen in a lot of student dogs and go, "This is something we need to train for." It's that we are often very good at incorporating play behaviors at the end of some training exercise that we do. We call the dog on a recall, the dog comes to front, finished, exercise finished, "Yay, good boy! You're amazing and you're beautiful!" We pet and we play and we do all this beautiful play, and then that's the end. But we have to have a very clean way to go from "I'm petting and playing with you" to calling to heel.

What I noticed with my own dog is when I would stop petting him, and I'd pull my hands back up and I'm ready to move on to the next thing, there was this withdrawal kind of effect where he would stand and just look at me and be a little bit flat. There was a question mark on his face, and I didn't realize it in real time, but I saw it on the video. So many things are clearer on video after the fact.

I was like, let's try replicating it. Maybe that was just in the ring because he was stressed. So I tried it in my front yard, and damned if he didn't stand and stare at me in the front yard when I tried to move from play back to call a heel. It was like, oh, this is a chaining problem. I need to be able to time that call.

One, I need him to know to expect that this play session will end with another opportunity to work, and hopefully that opportunity to do those behaviors is also positively trained and so isn't perceived as aversive. That's part of my job. And then I need to time that cue for while we're still engaged in the interaction, because if I am petting him and we're engaged, and then I just stop playing and take a moment to collect myself, he's just standing there looking at me, and then I say, "Heel," and he's like, "What?"

There are two things. One, that ending of the play is potentially aversive. You're taking the toy back. Whether there's a real toy there or an imaginary toy, there has to be something to fill that space. I realized I was having gaps in my performance, and it was making the setups at the beginning of each exercise harder than they needed to be, because there ended up being a disconnection in-between the exercises.

Even when I thought I was doing a good job at maintaining the connection, there was still a little bit of a question mark on his face, and subsequently when I would ask him to come to heel to set up for the next exercise, he would be a little slow, a little bit panty, and then I would fall into some horrible habits of patting my leg and shuffling my feet and repeating cues, which I never do in training, and only add to the weirdness of the ring performance.

Melissa Breau: Most people probably wouldn't think about that. You're giving your dog a reinforcer, "OK, you did a good job," now I can get back to work. From the human's perspective, that makes sense, but from the dog's perspective, I can certainly see how that might feel a little bit jarring to be in the middle of this interaction and have somebody be like, "OK, time to go."

Hannah Branigan: Especially if they didn't even say it, if they just stopped, like you're eating lunch with a friend and you're having a conversation, and she just gets up and leaves. And you're like, "What?"

But we tend to rush dogs with reinforcement anyway. I think it's just more obvious when there's not a physical thing involved.

Food is relatively easy because the food is swallowed and then it's gone, and then that is a very clear signal that that reinforcement is and both of you are aware of that, so there doesn't have to be an in-between. But we still tend to rush dogs: "Hey, come back, come back, let's go, let's go." Whereas we could let them swallow, let them re-engage, and then give the next cue.

With toys, it's a little bit harder because the toy is still there and you have to take the toy back. We still have a tendency, we get in a hurry as trainers and we want to rush the dog back so we can do the next thing. So "Give me the ball back," and then we reach toward the ball, and the dog isn't aware that the reinforcer is over just yet: "I'm still pulling on your toy." And then, when there's not a toy there at all, I think it's even more abstract. So we really have to be thoughtful about how are we signaling that we're now going to the next thing, and make sure that that transition from this easy behavior to this new next thing, this new harder behavior, is not a surprise, that it's expected, and that it's a positive, that the dog has good feelings about that transition.

Melissa Breau: To shift gears a little bit, we've talked a lot now about behavior chains and the intentional, careful planning, what it takes, think through, "OK, these are the pieces, and this is the order I want them in, and these are the bits in-between, and I want to think about that." What about unintentional behavior chains? Let's start with the obvious: Is it possible?

Hannah Branigan: Of course it's possible, and it happens all the time. Behavior chains — that's how we interact with our environment. There are many, many things we do where we have to do a long sequence of different actions to get whatever outcome it is that we're after.

We can certainly set up what happens to people, but we'll look at dogs because that's what we're interested in right now. It certainly happens in the dog world. Many of us have experienced the undesired behavior chain of the puppy that jumps up at you and then sits. I'm trying to teach you a polite greet, and the greeting that we end up living with or settling for in the short term is the puppy jumps up and then sits. "Did you see me? I sat! Did you miss it? Let me jump up and then sit again." These sorts of things can happen in lots of places where there's some behavior occurs on the way to the behavior that you want to reinforce, and then it gets built into that sequence of behaviors.

Of course it's harder to do that if it's a behavior chain you want to occur. That's like the Murphy's Law of dog training, kind of. But if you want the behavior sequence together, it will take months of careful planning and successful approximation to create that structure. But if it's a behavior that you don't want because it's annoying, like vocalizing, like squeaking in your crate, or something like that, or jumping up and then sitting or …

Melissa Breau: Or ping-ponging to the end of your leash before returning.

Hannah Branigan: Yes, ping-ponging to the end of your leash and then coming back for a treat, and then end of your leash and then coming back for a treat. Those things, of course, you only have to reinforce them, like, one time and you've bought it for life.

So the trick for those unintended behavior chains, undesired behavior chains, the ones you don't want, is to reinforce before the undesired behavior happens. So click before the puppy jumps up, which does mean you're clicking before they sit, but you're clicking before they jump up, and it interrupts the undesired behavior chain. It's the universal trick for fixing that in all cases.

Melissa Breau: Before the puppy squeaks, or before they hit the end of the leash.

Hannah Branigan: Right. Exactly, yes. Click before the thing you don't want to happen, happens, and it interrupts that chain, and you can get the reinforcement that was going to come at the end of the chain is now coming, you're short-circuiting, you're making a little shortcut to that reinforcer. "You don't have to sit. You don't have to jump up. Great. Now you're stopping 2 feet away from me." Or in the case that you're hitting the leash, maybe you're just dipping your head and looking right back, bam, I'll reinforce that, because that, for me, is a lot better than having my shoulder snapped every three steps.

Melissa Breau: Depending on the size of the dog, but yeah. All right, I've got one last question for you, Hannah. It's the new last question I've been asking everyone when they come back on: What's a lesson you've learned or been reminded of recently when it comes to dog training?

Hannah Branigan: I would say something that Figment has reminded me of frequently right now is the importance of setting up the environment to make the behavior that you want very, very easy to happen, and the behavior you don't want to see very, very unlikely to happen. So that he's always rehearsing the things that you like, so that you have opportunities to reinforce them, and rarely — and hopefully never, but probably because we're all human, rarely — practicing the behaviors that you don't like.

Almost all of the places through the course of the day where I start finding myself feeling frustrated and maybe tempted to fall back on some old training habits that are not tools that I want to use anymore, it's almost always that I have failed to set the environment up so that he's set up for success. And it means that I may need to put a leash on him temporarily while we're walking through the house to go out the back door, if there's going to be a cat in the kitchen or other predictable features of the environment that I need to change his behavior around.

So I would say that's something that I am having to think about. You don't think about it as much with your adult dogs, with your old dogs, because they have their habits like we have our routines, and so those types of things are now very much unconscious. But with the new puppy, he doesn't have those routines yet. He doesn't have a lot of habits yet.

I do have habits, and some of my habits are not setting him up for success. So I need to change my behavior so that he's always rehearsing the things that I do want. If I catch myself starting to feel frustrated because he's practicing behavior that I don't like because it's annoying or frustrating or dangerous, I need to change my behavior to change his environment so that I have what I want so that I can reinforce it.

Melissa Breau: Thank you so much for coming back on the podcast Hannah. This has been great.

Hannah Branigan: Awesome! Thanks for having me.

Melissa Breau: And thanks to our listeners for tuning in. We'll be back next week with Shade Whitesel. We'll be talking about crucial concepts for competition. Don't miss it!

If you haven't already, subscribe to our podcast in iTunes or the podcast app of your choice to have our next episode automatically downloaded to your phone as soon as it becomes available.

CREDITS:

Today's show is brought to you by the Fenzi Dog Sports Academy. Special thanks to Denise Fenzi for supporting this podcast. Music provided royalty-free by BenSound.com; the track featured here is called "Buddy." Audio editing provided by Chris Lang.

Thanks again for tuning in -- and happy training!