In this episode Laura and I talk about how to build and then maintain behavior chains, and how to change those that may have been taught unintentionally.
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 I have Laura VanArendonk Baugh here with me to talk about behavior chains.
Hi Laura, welcome back to the podcast!
Laura VanArendonk Baugh: Thank you so much for having me.
Melissa Breau: I'm excited to chat. To start us out, do you want to just refresh everybody's memory a little bit, share a little about you, a little about your pups?
Laura VanArendonk Baugh: Yes. I am a behavior nerd, and that's probably what it should say on my business card. I think it says something like "trainer," but nerd would be a better description.
I have two dogs who live with me currently. Undomiel is my working lines Doberman and Penny is my service dog retiree. She was a service dog in training who now is not working. I'm not sure what that settles her as. She has important jobs now, like make sure the couch does not float up to the ceiling, check and see if the pond is wet — really big, important dog stuff. Undomiel is actually sitting right here next to me, supervising this talk, so I will have to make sure I say only nice things about Dobermans.
My background — I'm a crossover trainer. If we jump way back, I probably made my switch to clicker training, oh my gosh, roughly twenty years ago now. That sounds like a lot. But I really appreciate having had that other perspective because I feel like it forces me to ask a lot of questions, like, I don't do this because it's nice, I don't do this because it's fun, although both of those are true. But I also can say I'm doing this because it's more efficient and it's a better system and all of these sorts of things.
In my background I have done competitive obedience, I have done agility, I've done tracking, and some protection sport, and all the fun things. And right now I am primarily an instructor. I'm with the Karen Pryor Academy and I own my own business, Canines In Action. And I just love nerding about behavior things.
Melissa Breau: Do you want to mention those books in there anywhere?
Laura VanArendonk Baugh: Oh yeah, I guess. Thanks. I'm also from the Midwest, so self-promotion clearly does not come naturally to me. I have written Fired Up, Frantic, and Freaked Out, about working with reactive or fear-aggressive or such dogs. I've written Social, Civil and Savvy, about socialization for young dogs to prevent such issues. And I have written Dragons, Unicorns, Chimeras, and Clickers, which is a children's book about using clicker training with fantastic beasts like unicorns and dragons and werewolves and all the fun stuff. And apparently I am incapable of writing any book with an easy-to-pronounce title, but that's what I have.
Melissa Breau: I haven't read the children's book, so maybe I'll have to pick it up.
Laura VanArendonk Baugh: It's quick and easy.
Melissa Breau: I would hope, but still. So I wanted to talk today about behavior chains, like I mentioned in the intro. I think it makes sense to probably start with a general definition, so can you tell us what a behavior chain is?
Laura VanArendonk Baugh: I need to preface this with: this definition can change, depending on what building you're standing in on which college campus. The way we are going to talk about behavior chains today is a series or a sequence of behaviors linked together by cues, such that one cue provides the reinforcement for the previously performed behavior. You will also hear "sequences," and again, these terms can be used differently in different fields of behavior, science, neuroscience, etc. And we do not have time to get into that, and that's not really a productive argument for us to have today. Just if those terms do get used separately, just make sure we know what we're talking about today: we are talking about a series or sequence of behaviors, such that one cue is the reinforcement for the previously performed behavior.
Melissa Breau: Excellent. What are some examples of these behavior chains that we might actually want to train?
Laura VanArendonk Baugh: There's a lot, and my nerdy side would say that pretty much everything in life is actually a chain, but that's not really helpful to break things down for.
A really common example that people might think of would be a retrieve, whether that's a formal or an informal retrieve. My dog is going to run away from me, my dog is going to pick up the object, my dog is going to return to me, and then, in an ideal world, the dog will hand me that object. Depends on how formal or informal this retrieve may be.
But then, if we actually break those behaviors down, picking up the object is a lot of behaviors. Picking up the object and then turning and then coming back to me are all separate behaviors, so we can actually get really nitpicky about these. But that may just be a thing that nerds do over snacks and drinks.
But I would say that's probably the easiest one to reference for a lot of people is think of a retrieve, whether it's formal or informal, and you can see all those different parts that go into one thing that we call a retrieve.
Melissa Breau: What about unintentional behavior chains — things that we may not train on purpose or that we may even actively dislike? Do you want to talk a little bit about those?
Laura VanArendonk Baugh: Oh my gosh, we do that all the time. Again, one that I see a lot is we have a dog that's jumping on people, and so we teach our clients to teach the dog to sit to meet people.
This seems like a great idea, and it's not a bad idea. But if we're not thinking about behavior chains, what we frequently end up with is a dog who runs up, jumps onto a person so that we will tell them to sit, so the dog will sit so the dog can get reinforcement for sitting.
That cue for "sit" is reinforcing the jumping up, and because we weren't thinking about behavior chains while we were working, we have unintentionally created a very, very effective and really solid behavior chain, and we will have this dog who gleefully bounces off my chest in order to get to sit for a treat. So that can happen a lot.
Just talking about reactive dogs and things, I've also seen the dog hit the end of the leash and bark so he can be asked to come back and touch a target, or come into heel, or do something like that, again because we're not considering how behavior chains work while we're trying to deal with other unwanted behaviors.
Melissa Breau: I'm really glad you mentioned that example, because I think a lot of the time the term "behavior chain" has a really strong association in people's minds, specifically with sports, and we don't always think about some of the less sport applications or how some of this filters in.
Laura VanArendonk Baugh: Honestly, when we start talking about behavior chains, it's usually in a "How do I fix my obedience retrieve?" or competition heeling is one long behavior chain, and then all of these things. And then when you start thinking happy behavior chain thoughts everywhere you go, please send help. But then you start seeing where, like I said, everything is actually a behavior chain. And suddenly some of these behaviors that did not seem to make sense — "I always treat him for sitting; why does he keep jumping?" Because you keep treating for sitting. He has to jump in order to get the treat for sitting. It's such an easy slippery slope, if we're not aware of what we're doing unintentionally.
Melissa Breau: Yeah, and if we don't stop to think about it and really look at the behaviors. So for those that we do want, how do you think about the process of training and behavior chains? What's your, I guess, broad approach when you know there's something like that you want to teach?
Laura VanArendonk Baugh: My general approach is to be really, really annoying to my poor students. I emphasize so much the importance of building that foundation and spending more time on the foundation than you probably think you need, because we really need to build this fluency. And then, if you have those foundational pieces, the behavior chain will almost magically coalesce on its own.
This sounds a lot like wishful thinking, but go back to what is a behavior chain. It's a collection of fluent behaviors linked by really solid cues with good stimulus control. So if I have fluency and I have stimulus control, then I have a behavior chain.
Where I think people start getting into trouble is we have mostly fluent behaviors, and we have two out of four aspects of stimulus control or something, and then we start building our chains. That seems to work in the beginning, but it's not going to hold up long-term. And then we start thinking we have behavior chain problems. No, what we have is fluency problems, or what we have is stimulus control problems.
So my process is just spend more time on the beginning and you can spend less time on the end. That's not the fun part, nobody wants to do the sexy, sexy stimulus control practice, but that is what's going to hold your behavior chain together.
Melissa Breau: So basically teach each little piece before you try and knit anything together.
Laura VanArendonk Baugh: Oh, absolutely, yeah. Like, each individual behavior and all the parts of each individual behavior. Again, just because the dog does it on cue does not mean I have full stimulus control. And this is not a stimulus control talk, so insert link here to alternate stimulus control talk.
But that's the thing that people … your behavior chain depends on stimulus control. Stimulus control is the reinforcement for everything that's happening in your behavior chain. It's what's providing that positive reinforcement for the behaviors that are happening in the chain.
So if I don't have good stimulus control, guess what? I don't have control over what's getting reinforced. If I don't have control over what's getting reinforced, we have chaos. Everything about training is control freaks controlling what's being reinforced.
Melissa Breau: That feels fairly simple. That's not crazy-complicated. So then, where do we all muck it up? Why do we struggle with all this stuff? What goes wrong?
Laura VanArendonk Baugh: One of my favorite Bob Bailey quotes is, "It is simple, but it is not easy." I think behavior chains are a great example of that, because literally it is about fluent behaviors with good stimulus control, and that's it. That's really all I need. But then we do, as you say, muck it up quite a lot.
Frequently, one of the easiest mistakes to make is what I mentioned already: we have mostly fluent behaviors, mostly we have a good start on stimulus control, and so we assume that those are going to hold up where they might not be as thoroughly tested as maybe I should have.
The other thing, and this is one that I think is deceptively tricky, is we think the behavior that we see getting messed up is the behavior that is broken, and so we spend a lot of time working on that behavior. "Why does he not bring the object to me in the retrieve?" or whatever it might be. What's actually broken is a different behavior in the chain, because again, each of your behaviors is what's reinforcing the previous behavior, or each of the cues is serving as reinforcement. So if there's something later in the chain that's fuzzy, it's going to not be reinforcing something earlier in the chain. That's what we think is broken, But that's not the behavior that's actually broken.
So I think probably the trickiest part of behavior chains is identifying where the actual breakdown is, because sometimes it's the one that you see messed up, that can happen, but sometimes it's not. And then you have to actually think about fluency and stimulus control, which we should have been thinking about already. Hello, I was right. Why didn't you listen to me? No, that's not actually how I say that. But yeah.
Melissa Breau: Do you want to talk a little bit about when we've got it down, when we've got the behaviors happening correctly in the correct sequence, what do I have to do to keep it that way?
Laura VanArendonk Baugh: The good news is if I've built it really well, it's often self-maintaining. If I'm having to do a lot of heavy lifting just to keep it going, that's a sign that probably it shouldn't have been going yet. Hello, I need to go back to my foundations. I say this a lot. But if I've got a chain that is put together well and is solid, I just need to keep that reinforcement going.
And I need to make sure that my cues stay both … two things. One, my cues are still the cues that I think they are, that I'm not unintentionally replacing some cues or bringing other cues in, and then I need to make sure that my cues are still reinforcing, we are still moving toward some primary reinforcement, that each cue is serving as a conditioned reinforcer toward that primary reinforcer.
If I've got those two things, we'll be good. I said that very glibly, like, "Oh, I just need to make sure my cues are clean and my conditioned reinforcers are …" Yes, that is a very simple sentence to say, but as we already mentioned, sometimes we get distracted and we're not thinking about the way we're using our cues, and we can easily start turning something into something else.
Frequently I think a verbal cue is what's prompting the behavior, but it's actually a contextual cue. So when the environmental cue is present but I haven't said the verbal cue yet, "Oh, no, he's anticipating, oh, no, I don't have control of this behavior." And no; it's my cue is not what I thought it was.
Melissa Breau: As part of the maintaining process, do you need to continue to pull pieces apart and build fluency individually? Or once it's in the chain, can it stay in the chain?
Laura VanArendonk Baugh: I'm going to weasel right out of that and say it depends. There are definitely cases where … it's never a bad idea. It's never a bad idea to pull out pieces and work with them individually. All that's going to do is help build your reinforcement history. "Oh no, too much reinforcement" is not something I say very often.
But some chains honestly can be pretty self-maintaining, and those are usually simpler chains. Those are usually ones where we're not likely to have a lot of competing or conflicting cues or something. But something as simple as a recall or "Go to your mat and lie down," which is actually a very small behavior chain, "Go to your mat and lie down," I don't have to break apart "Go to your mat and lie down" terribly often to keep those fluent in a chain.
Those are going to be there because it's short, it's easy, the cues are very, very clear. I probably don't have a cue for "lie down," probably arrival at the mat is the cue to "down.' So it's going to be really hard for me, as a fallible human, to botch the timing of that cue and start confusing my poor dog and that sort of thing. As long as I'm reinforcing the end result of that chain, the dog is going to maintain that chain pretty well.
So there is light at the end of the tunnel. Not every chain needs very, very careful maintenance. But I need to keep an eye on things. And if I see something starting to wobble, fixing it sooner rather than later is probably a good idea.
Melissa Breau: To go back to those unintentional or undesirable behavior chains for a minute, if you're looking at something like the examples you mentioned earlier, how do we begin to break those down and rebuild them into more helpful behaviors, now that we've dug a hole for ourselves, so to speak?
Laura VanArendonk Baugh: I think one of the first things is just being aware that unintentional behavior chains happen, so that we are looking for behavior chains, as opposed to a single, really simple, unwanted behavior that's completely detached from the rest of the world — because those don't typically happen.
And then, once we're looking for what is cueing this, what is reinforcing this, being aware of what our cues are, and then being aware of what the timing is that those cues are being given. Any cue that we have positively trained is a conditioned reinforcer, so whatever behavior is occurring when I give that cue is on some level being clicked.
I think sometimes that can be almost a scary thing to say, because then people are like, "Oh no, my dog is running across the field, but if I call him, then I am reinforcing running across the field." No, no, call your dog, that's fine. But if you're noticing a pattern that you're continuously giving this cue during a behavior that you don't want, you might be aware that you're clicking this on the regular, and that's going to start leading to unwanted behavior chains.
For the example of the jumping, if my dog keeps jumping on me so that I can cue "sit" so that he can sit to greet someone, how many times does this have to happen and then I can go, "Oh, wait, there's a reliable pattern here. I should look at that." And now I can see the dog is jumping, that's when I say "sit."
What happens if I, one, don't say "sit" while he's jumping on me? Honestly, in an established thing, the dog's going to continue to jump, and harder, because he's exaggerating the behavior, you're going to get a little bit of an extinction burst, and "I am pounding my paws into your lungs; please say 'sit.'"
A better option might be to say "sit" sooner. "We're going to bypass. I don't need to jump on you to get the cue 'sit.' I can approach you to get the cue 'sit.' Cool. I get to save energy and greet people sooner. Life is great."
This is the … I think "system" is too grandiose a word, but these are the things to look for: What are the cues, and what's the timing of those cues being delivered? That will tell me what is being reinforced. And then I can look at do I want to reinforce that or not. And if I don't, great, let's not give the cue then. Let's give the cue at a different time before or after.
Melissa Breau: That makes sense. Part of the reason we're chatting at all today is because you're doing an awesome webinar for us on all this stuff on behavior chains on November 4. Do you want to talk a little bit about what else folks should know about the webinar, just to help them decide if they're interested in attending, maybe who should attend?
Laura VanArendonk Baugh: If you are a nerd with insomnia and you want to talk about behavior chains, this is definitely the place for you. I love behavior chains because, like I said, they really are everywhere. You can't actually get away from them, and it may start to interfere with your daily life, like, you're waiting at a stoplight for the light to turn green, and you're starting to analyze all the little behavioral chains that go into this chain of the streetlight, at a traffic light. I'm just speaking for a friend. Anyway …
Honestly, I just see knowing more about behavior chains, and being aware of how they work and how pervasive they are in life, so that understanding them better just means we're better at behavior in general, both wanted behavior and unwanted behavior. And we can build really polished sport behavior chains.
Like you mentioned, most people, when we start talking about behavior chains, we start thinking about it for the first time in a sport context. And then you start realizing this can actually be everything in my life. I walk into the kitchen, my dogs automatically go to their beds so that I'm not tripping on them in the kitchen, and then when I do this, they automatically move to this other space. These are all behavior chains, and it's great because it makes less effort for me and more reliable behavior and more reinforcement for them, so everybody wins.
That's not necessarily a checklist of people who should come. But I guess I would just say if you are interested in having a metric to more easily identify what cues can I shift, what timing should I shift, what specifically am I looking for to make these behaviors more or less reliable, depending on if they're wanted or unwanted behaviors, then that's what we're going to talk about.
Melissa Breau: Excellent. To round out our conversation today with one final question, if we were to drill all of this down into one key piece of information you really want people listening to take away or to understand, what would that be?
Laura VanArendonk Baugh: If you have fluency and if you have stimulus control, then you have a behavior chain. It really is that simple. That doesn't mean it's easy, but it really is that simple.
Melissa Breau: Awesome. Thank you so much for coming on the podcast, Laura.
Laura VanArendonk Baugh: This was a blast. Thanks so much for asking me.
Melissa Breau: Absolutely. And thank you to all of our listeners for tuning in. We'll be back next week with Kim Brophey to talk about the connection between ethology and behavior.
If you haven't already, subscribe to the 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!
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!