In this episode we discuss:
- The first step to dealing with a problem behavior is to identify the function.
- Figure out a plan to control access if possible and practical.
- If that reinforcer is not practical to work with, identify a substitute reinforcer that is as close as possible to the original.
- Choose your alternate behavior.
- Doesn’t need to be incompatible but could be.
- Should be cheaper and more efficient.
- There could be shaping involved, doesn’t need to be finished.
- Ideally, it’s something the dog is already doing.
- If you are teaching the alternate behavior from scratch:
- Teach it out of context first.
- Generalize with “different but not harder” before going straight to final application.
- Fade in components of the antecedent picture.
- Transfer to the “real” reinforcer.
- Decide if you’ll be using differential reinforcement.
- Ignoring can be hard!
- Extinction bursts are terrible for everyone involved and it’s often easier and kinder to avoid them entirely.
- If you are not planning to use differential reinforcement, plan to set up the environment for (mostly) errorless learning.
This podcast is supported by: MET Conference 2023
The Modern and Ethical Training Conference is coming up on March 4-5, 2023! MET is a virtual meeting of the minds between dog lovers and trainers embracing modern and ethical training practices. Speakers include trainers from all over the world who will share their knowledge during more than 18 hours of presentations. Click the link below to get your ticket.
Hannah: [00:00:00] You’re thinking, “Well I know my client wants this fixed yesterday.” I know that and I agree because I also want my dog’s annoying behaviors fixed yesterday. So I’m gonna be completely honest: every time I hear somebody say, “Oh, dog owners, they don’t wanna do the work these days,” I also don’t want to do the work. Most of the time I want things to just be– I mean, once in a while, it would be okay if it would just be easy, right?
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’ve 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 [00:01:00] clickertraining.com.
And this is your reminder: Don’t forget the next round of Zero to CD, my online mentorship program, will open for enrollment on Monday, April 24th. Oh my gosh. That’s getting closer every day. With orientation for the new cohort on the following Monday, May 1st. This program has always filled really quickly, so make sure to get on the email list to be notified. You can find out more about the program and get on that list by going to zerotocd.com.
So today, we’re talking about working with alternate behaviors as a strategy for dealing with problem behaviors or behaviors you would really rather your dog didn’t do.
This episode is supported by the MET Conference and patrons like Becky B, Catherine S, Veronica L, and Deb M, who are awesome. I love every single one of y’all. If you also would like to support the podcast, get your questions answered, and get access to our [00:02:00] super secret extra podcast episodes, you can go to patreon.com/DFTT.
So I’ve been considering for a little while now doing a series of episodes on dealing with problem behaviors within a positive reinforcement paradigm, because I feel like this whole category of topics is really not well understood by a lot of folks and frankly can be hard to wrap your head around even as a committed positive reinforcement trainer.
Realistically, a lot of us, like our starting point in dog training– a lot of people’s starting point in dog training is you have a dog that’s doing stuff that’s a problem and the whole reason you’re looking up training isn’t because, “Oh, I really wanna teach my dog cool stuff.” It’s first “I really want him to stop doing this annoying or potentially dangerous stuff.” And that’s the gateway, that’s the entry point for a lot of training.
It’s easy to think about reinforcing stuff we do like our dogs to do. It’s [00:03:00] a lot harder to figure out how to use the things that we do know and turn them into stuff that I want my dog to stop doing.
And I have dogs and I’m a real person and my dogs are real dogs. And I live a somedays very real life. And my dogs do things that I don’t like. And sometimes I have a hard time thinking about “how am I gonna deal with this behavior.” So far, so good, but tomorrow’s a new day. So I think it’s worth putting some time and conversation into.
There’s a lot of complexity and nuance here. And talking about alternate behaviors is just one facet of how we can deal with problem behaviors while still, again, operating inside that positive reinforcement paradigm. And we do think about things differently. Alternate behaviors is just part of it. But if I have to pick a place to start, this seems like as good a place to start as any.
I was inspired in [00:04:00] part to start here by a recent conversation with my friend and former (and hopefully soon? future? ongoing? repeated?) podcast guest, Kiki Yablon, who has also written a really good blog article about alternate behaviors, which I’m gonna link in the show notes just in case you haven’t already got her whole website bookmarked, which you probably should do.
But also because, when I’m in conversations with folks – and I’m thinking about conversations that I’ve had as a baby dog trainer – now I consider myself an adolescent dog trainer, but when I was a baby dog trainer, you would hear the oversimplified version. And I’ve said this myself. It’s very oversimplified of how to deal with problem behavior. It’s often stated as some variation of, “Well, instead of teaching your dog what not to do, teach him what to do,” and then you’re supposed to put like “TO DO” in all caps or bold – which is true. Or you may hear something like, [00:05:00] “Well, ignore what you don’t want and reinforce what you do want,” which sounds fantastic.
But I think, again, in real life, we know that it’s not always so easy, it’s not always so simple, and it’s not always fantastic.
Okay, so: let’s dig into exactly what it looks like to be successful working with alternate behaviors. And again, I just wanna emphasize that this is just one piece. Your training plan to address whatever problem behavior you or your client is dealing with probably will include an alternate behavior, but it may just be a part of it. But I’m gonna get more into that. I don’t know. I get excited. Okay. There are more episodes to come, so stay tuned.
Alright, so the very first step, the thing that I think is the most important to do when you’re addressing any problem behavior at all, but particularly when it comes to an alternate behavior, is identify the function of the behavior. This is not [00:06:00] always easy. Sometimes it’s straightforward. Sometimes like I don’t have to use a formal assessment or test to figure it out. A lot of times I just wing it. Sometimes it’s not easy to figure out though and you’re gonna have a lot more success if you can correctly identify what the function of the problem behavior is.
So of course, all behavior has a function, no matter how annoying that is. Behavior is designed to affect the environment, to have an outcome. It’s designed to get access to some reinforcer in the environment. Could be to avoid something, could be to access something, but all behavior has a function and the closer you can nail the specific function of your problem behavior, the more successful any training plan is going to be in terms of addressing it.
Again, there’s so much value in taking the formal approach in identifying the function. I do often admittedly wing it and one of the ways that I [00:07:00] can kind of test it is I try out my guess and if my training plan is working, then great.
But there are also times when really, truly knowing is important for safety, right? So one particular group of situational problem behaviors that show up is how dogs react to either other dogs or even more importantly people, because it can sometimes get kind of fuzzy.
Here’s an example: jumping up on people. It’s very commonly described behavior that’s perceived as a problem by a lot of humans and the function of that behavior isn’t always obvious. It can look kind of the same. Sometimes jumping up on people is a behavior that’s designed to solicit attention. “I’m trying to jump up because I want to greet you. I want interaction with you. I want petting and touching and [00:08:00] looking and talking and all the things that comes with that social interaction.” And sometimes jumping up is not. Sometimes jumping up is the nervous laughter of a dog who doesn’t really know how to interact and really kind of doesn’t want to, but doesn’t know what else to do.
And if we take a dog whose jumping-up behavior is not intended to access attention and we treat it as if it is, we can get in trouble. There’s other behaviors that can go in this category as well, but being able to identify at the very least “Is the function of this behavior to increase distance or to decrease distance?”
If the function of the jumping up was really to increase distance, to get outta that interaction, and we try to use distance to reinforce it, then best case scenario it’s just not gonna be effective. Worst case scenario we can get in trouble. We can push the dog to do other distance-increasing behaviors, which may be even less socially acceptable, like [00:09:00] growling, snapping, biting, et cetera.
So I do think that when you’re dealing with behaviors directed towards other people, other dogs, it’s a good idea to take a good hard look and make sure you’re not making assumptions. I say that as someone who has made assumptions. It took me a minute to figure out a dog that I was working with, that the jumping up was not attention-seeking. It wasn’t being maintained by my attention. It was actually a dog that was conflicted and, and concerned and really kind of wanted out of the situation. I don’t think that was unique. I’ve since run into other dogs with similar situations.
So being able to figure out “What is this behavior designed to access?” Are they barking because they want the UPS guy to go away? Are they barking because they’re very excited and they want you to come closer faster? “You’re too slow; you’re walking too slow!” Then we wanna know that. I’m trying to think of other problem behaviors. I have a bunch of examples on my online here that [00:10:00] we’ll get into as we go. But being able to figure that out and “what are we trying to get at” is first thing. Within that, having or designing a plan to control access to that reinforcer in some way. Could be stimulus control, could be opening or closing a door, could be controlling distance by moving the cue/stimulus further away, like having the person move away or have them move closer, moving the dog away or having them move closer, baby gates…
Is there a way that you can control access to that reinforcer so that you can have some leverage with it? Sometimes this is a very easy thing to do. Sometimes it’s really hard and sometimes it’s impractical and we may have other things we need to put into place around that as well. But is there a way to control access to that reinforcer?
I mean that’s the ideal situation, right? Like the ideal situation is, “Oh, I know exactly what the reinforcer that’s maintaining this problem behavior is.” I’ll just use barking, just because using the same problem behavior over and over again is boring. [00:11:00]
Let’s say barking. I know exactly the reinforcer that’s maintaining the barking and I’m gonna control access to that so that barking is never reinforced ever again with that reinforcer. And then I can use that reinforcer to reinforce anything that I want. Okay. That doesn’t always happen. And I mean, I’m saying that and I’m thinking, “Well, that rarely happens unless I have complete control over that reinforcer.” But if I can have any kind of way to control access, that’s super handy, at least temporarily so I can get something started.
Sometimes the reinforcer itself is a problem and in that case, I may need to have a plan to provide what I think of as like a “substitute reinforcer.”
So here’s an example: one of the problem behaviors in my life is Figment’s car-chasing behavior. Now it’s not really a problem anymore because I’ve done some training and we’ve addressed it and through a lot of experimentation, discovered some things [00:12:00] that are useful to get functional behavior from him so that I can enjoy being with him the way that I want to and he can enjoy life and we’re both getting our needs met because the human and the dog needs both matter.
But in the case of car chasing, the function of his behavior around cars is that he would very much like to actually chase the cars, and I believe – just having watched him perform similar behaviors in other situations – it would be to get in front of the car and stop the car from moving. He’s a border collie.
Obviously that’s not going to work. I can’t ever let him chase a car and get in front of the car and stop the car from moving. I can’t let him get in front of the car and stop the car from moving. I suppose if I were very creative, I could figure out a way for his behavior to stop car movement. But it’s not realistic, it’s not safe. Nothing about his behavior towards moving vehicles is safe. And so that’s not [00:13:00] a reinforcer that I’m going to use to shape his behavior.
So instead, I wanna provide some kind of substitute reinforcement. In my experience, I’m gonna have the best luck– It’s not luck. There’s no luck involved. I’m gonna have the best chance at a successful outcome if I can figure out a way to provide a reinforcer that’s very similar, very close to the one that the dog is after. So there’s also something to be said for providing for – we would say “providing for that dog’s needs,” but that’s almost a cliche phrase and I think we often aren’t specific enough about it. Because yeah, feeding him and making sure that he has food and water and safety, like those are really important. I definitely wanna make sure that he has that, making sure that he has sufficient enrichment – all of those things are not directly going to affect his car behavior. [00:14:00]
But, if I think about, “Okay, he wants to do this very, very herding dog kind of behavior and the stopping-of-motion, the getting to move and stop movement, all those things that are involved in that.” Well, I can provide him with opportunities in other contexts to exhibit those behaviors.
Okay. So that’s good. I still don’t really expect it to eliminate his car-chasing behavior, especially since it’s already occurred a couple of times. (Okay, so it’s already occurred more than “a couple of times”). So those are good. I really am a big fan, supporter of giving your dog lots of opportunities to express appropriate species- and breed-natural behaviors. I think that’s important for their quality of life. But I know that it can look like a lot of things.
I’m gonna have a much easier time if I can use a reinforcer that’s closer to that chasing behavior, [00:15:00] like running versus not.
So let’s just look at food. If I were to use food, I could room service, feed him like a baby bird, pop the food in his mouth and say, “Oh, here comes the car. Well, I want him to be calm, so I’m gonna hand deliver food into his mouth.” Now I know with this dog, he’s not gonna even take the food in that situation. Getting him to eat food in the face of moving cars: very hard. Now that’s a separate training challenge. And I mean, that’s a whole nother episode; that’s like a series of episodes.
But he’s much more likely to engage in a chasing behavior that’s just not directed towards the car. So chasing a thrown treat is gonna be easier to work with than trying to hand-deliver a treat. Chasing a thrown toy is going to be even easier, even if I keep it in my hand and let him chase a tug toy, ideally in the opposite direction of the car. But he still gets to move and chase and stop motion, which is what he likes to do.
That’s one of those fascinating [00:16:00] things. I love watching dogs, observing them and observing their behavior and seeing what emerges. And as an example, I talk about his Jolly Ball behavior. All of my dogs except for Rugby have loved playing kickball with a soccer ball or a basketball – I don’t really kick a basketball, well actually I have – or Jolly Ball. And with the belgians, they love to chase after it, jump on top of it and bite it, obviously, because what they really wanna do is get something in their mouth as quickly as possible. And with Fig, when I would kick a Jolly Ball, he really wants to go around, get in front of it and stop it from moving. That’s genetics, right? Like how cool is that?
Anyways, so knowing this dog, still it’s a chase behavior and movement is involved. If I can get a chase-y kind of reinforcer in play, that’s gonna be a lot easier to work with. So the closer I can get, the better [00:17:00] if I’m not able to give that reinforcer.
There’s other strategies that we can use when you can’t use that same reinforcer. Again, other episodes. There’s so much to this. Oh my goodness.
But let’s move on from the function and the reinforcer to choosing an alternate behavior.
Okay. Because one of the big obstacles I’ve discovered when I have effectively failed to deal the problem behavior effectively, even if temporarily, right? Like “This is my first rough draft of how I’m gonna deal with that. Oh, that didn’t work at all. Performance, feedback, revision. Come back and try something again.”
Choosing the right alternate behavior is huge. Even if I have the function, I know exactly what’s reinforcing the problem behavior, and then I try to replace it with an alternate behavior that’s not a good option, I’m gonna get real frustrated. My dogs can get real frustrated.
Here’s the classic example: sitting for petting. [00:18:00] Sitting sounds like such a great idea. I mean, how powerful do you feel if you can take your Larry the Labrador who’s throwing himself all over people, he’s so excited, he wants to greet people. We are a hundred percent confident that what Larry wants, the function of Larry’s behavior, is to get social attention from everyone and anyone. He wants all of the petting all of the time and he’s flinging himself around. He’s knocking over children. He’s a terror. And we teach him to sit quite calmly and silly and accept petting in such a dignified manner. What a power rush that would be!
That’s usually real hard. If you’ve met Larry, you know what I’m talking about here. That’s gonna take months and months of work. And if it’s your dog and you don’t have anything else going on, great. That’s never described me, because I always have a lot of things going on and it usually doesn’t describe my clients who are trying to get things done within the scope of a [00:19:00] behavior package. So that’s usually real hard behavior because even though the reinforcer is the same, it’s the complete opposite behavior. It’s very hard behavior. The dog’s not doing it to begin with and trying to go from bouncing and leaping and tongue swiping and wiggling and tail-wagging and all of the things to holding a sit stay. Ooh, my goodness. Okay. So that’s gonna be real, real hard.
But we could pick a really different behavior. Usually the best thing, if we care about efficiency, is to pick something that the dog is already doing and ideally already doing in the exact same context. So if in that picture, Larry is doing a lot of different things: one of them is jumping up, one of them is licking people, but he’s also already turning his body sideways and wagging and wiggling, we could run with that. [00:20:00]
And that’s actually my first choice of alternate behaviors when it comes to greeting people for most dogs because a very naturally-occurring behavior that a lot of dogs already do is that turning sideways, cashew movement with their tails wagging, their butt is wagging, and when they’re turning sideways and presenting their side to you and looking back over their shoulder, they can accept all of the petting and you’re much less likely to get busted in the nose. Right?
So I love that, especially if the dog is already doing it and we can just select for that thing that’s already happening in that like soup of other behaviors that are occurring. We can grab that one. Perfect.
But if it’s not occurring in that scenario, but it is occurring elsewhere that dog already has it, the behavior’s already fluent. Maybe he already knows how to target. Maybe he already knows– I was gonna say sit, but we’re looking for alternate-alternate behaviors to sit. A behavior that he’s already doing somewhere else that we could bring [00:21:00] into this situation saves us a lot of time if I don’t have to teach something from scratch.
So here’s another example: We’ll keep talking about Larry. Larry at the Labrador jumps all over people, especially at the front door and another behavior that he does all the time, especially when he’s excited, is he grabs things and puts them in his mouth. So a strategy that I’ve used very successfully a lot of the times with dogs that already very frequently – I might even say “they love to” – pick stuff up in their mouths, is use that as my alternate behavior to replace the jumping up behavior, the other behaviors.
So when someone comes over, Larry’s presented with a toy, he puts that in his mouth and then receives the petting and attention that he wants. And that’s really pretty fun. And it can work so quickly for dogs that already pick up stuff a lot, like that’s a very high frequency behavior. They do it all the time anyways. Just say, “Hey buddy, you know what? This [00:22:00] thing that you also do all the time, you really like doing, let’s also do that here when people come over.” And that can be a lot faster strategy, a lot faster solution, to get a usable functional behavior for Larry’s family so that they can enjoy having him and also have people to come over.
So something they’re already doing is really, really helpful. Save us a lot of time.
The behavior doesn’t necessarily need to be strictly incompatible. When I was first learning about training with alternate behaviors, I understood “alternate” and “incompatible” to be like almost interchangeable. Like any alternate behavior should also be incompatible with the problem behavior that I’m trying to control, to eliminate or to address. And sometimes you may want an alternate behavior to be incompatible, but it doesn’t necessarily have to be.
Here’s what’s so cool that it took me a long time to [00:23:00] figure out. (And then of course I’ve read about it since and it turns out that, like everything that I think I’ve discovered, other people have discovered it and are doing it better way before me). But anyways, what I’ve “discovered” is that something like Larry grabbing the toy in his mouth – he can definitely still jump up on people with that toy in his mouth. He definitely can. It doesn’t need to be– like the appeal of the sit is that you can’t both be sitting and jumping. Right? It’s one or the other. But you can have a toy in your mouth and jump (or sit for that matter). And so I would think, “Oh, well I can teach him to pick up a toy, but it’s not necessarily gonna do anything about the jumping.”
But here’s what’s cool: If the new behavior, the alternate behavior that’s replacing the undesired behavior, is cheaper and more efficient at accessing the same reinforcer, the dog will tend to do that [00:24:00] anyways, and the jumping up is gonna happen a lot less. And I just think that’s so interesting.
And so that’s one of the principles that I would apply here, right? Like, it’s not about putting a toy in your mouth – or rather, it can be about a lot of things. But a very functional, productive way to think about it, or a principle to apply here, is that whatever behavior you choose to be the alternate behavior that you’re gonna be working towards, working with, you want it to be cheaper and more efficient than the problem behavior. So it should be easier for the dog to do and more efficient at getting access to that reinforcer than the problem behavior. And as long as it’s cheaper and more efficient, the dog is gonna effectively prefer this desired behavior and you’re not gonna see the undesired behavior whether or not the alternate behavior’s incompatible. And I just think that’s so cool. Like it just makes me so excited.
And even if you’re reinforcing the new behavior, alternate behavior, [00:25:00] more often, even if the undesired behavior occasionally gets reinforced by accident or just because life is hard and and complicated sometimes, you’re still gonna see less of it. It may not entirely go away right away. That’s okay, because most of the time just seeing a lot less of the problem behavior goes a long way to helping everybody be happy and have a good relationship.
You know I am a fan of lifelong learning and I’m always looking for new people to learn from because hopefully one day I will have learned enough that I won’t feel insecure anymore.
So just in case you also have this problem, I wanna tell you about the Modern Ethical Training or MET Conference, which is coming up in March. It is virtual, and yes, it’s recorded. This is a relatively new conference that has some familiar names who have heavily influenced my own training, like Kay Lawrence and Bob Bailey, as well as friend of the pod, Leslie McDevitt (who I’m not mad at). But also some speakers that were less familiar to [00:26:00] me, like Lenka Blachová who will be speaking about canine greetings and Eliška Pavetta Šuhaj speaking on enrichment. And one that I’m particularly interested to hear, ethologist Ádám Miklósi, who is the co-founder of the Family Dog Project in Hungary, which I realized when I was reading his bio I saw featured on the Dog’s Decoded documentary from Nova. So he’ll be presenting on social hierarchy and dogs, and I’m very curious to hear what he has to say about current research on this old and often controversial topic.
This conference is approved for CEUs for CPDTs and you can find out more and register by going to metconference.com or click the link in the show notes.
But let’s say that there isn’t anything that’s practical, that’s already working, that either the dog’s already doing it in context or they’re doing it in a different context, and you really need to install a [00:27:00] new behavior. And there’s a lot of reasons that this might be the case. Something like go to a station, go to mat, is a really handy alternate behavior. Probably your dog’s not already doing it on their own. I mean, they may already be doing it because you’ve taught them to do that already, but they’re probably not already doing it in this context, so you may need to teach it.
And so if you’re teaching the behavior from scratch, you’re probably gonna wanna teach it outta context first, get it fluent first. And this is another piece that I didn’t really learn about how to deal with these kinds of behaviors in everyday life until I learned these strategies and principles honestly in competition training. And then I turned around and applied them. Oh my gosh, my lifestyle pet dog skills got a lot better, a lot faster.
So teaching the behavior out of context first and teaching it to fluency.
So what I used to try to do– I’ll use go to mat as an example because I do love go to mat or go to station as an alternate behavior in a lot of [00:28:00] scenarios. If I don’t know what to do, I’m not really sure what’s going on with this, again, I just need something to slide into that place at least for now, go to mat can solve a lot of problems. So I love that. It’s pretty quick to teach. It’s pretty easy for most people to teach and most dogs to learn. So that’s one of my go-tos. I’ve talked about that a lot. It works really well in so many situations.
So most of the time I’m gonna use a different reinforcer, because most of the time, I’m gonna go ahead and teach that with food, because food is the most convenient reinforcer for me to work with and for clients to work with and other trainers to work with. So I’m gonna teach it originally with food. That’s a different reinforcer. So because I’m using a different reinforcer, I’m going to need to have a plan to eventually transition it to the functional reinforcer, that reinforcer that was maintaining the problem behavior, whatever that was: attention, the mailman going away, whatever.
So I’m going to need a plan, but I’m gonna initially teach it with a different reinforcer. I’m gonna initially teach it where it’s very easy for the dog to learn the [00:29:00] thing, so not in the midst of wherever that problem behavior was occurring. In fact, I’m probably gonna change as much of the cue picture as I can while I’m teaching this brand new behavior from scratch.
Then the way that I thought I had to do it – or I don’t know, I don’t know what I thought – but this is what I thought it was supposed to look like. I thought I was supposed to come over to your house and I’m gonna help you teach your dog to go to station, in your living room or in your kitchen or somewhere that’s very easy when there’s nothing else going on, I’m gonna use food to teach it. Then once I have the dog going to mat and taking food, and going to mat and taking food, and going to mat and taking food, then we’re gonna put the mat in the hallway and I’m gonna start ringing your doorbell.
And I’m gonna ring the doorbell and you’re gonna tell your dog to get on the mat and you’re gonna try to get ’em to take food.
And sometimes that worked great. In fact, I would say my behavior was put under one of those intermittent reinforcement schedules where it worked great just often enough to keep me trying it. And I didn’t [00:30:00] know what else to do cause I didn’t have an alternate behavior. So meta here!
Here’s what I’ve learned through competition training that I have since applied to lifestyle training. What I would actually do is teach that go to mat behavior and effectively generalize it, teach it to fluency, and introduce a lot of other contexts that were different from the original. So like not just in your kitchen with food, but I might then also take it to the front yard, so different but not harder. (This is something we talk about in Zero to CD all the time: Different but not harder.)
Can we do it in the yard? Can we do it in the living room? Can we do it in the kitchen? Can we do it in the backyard? Can we do it in the driveway? Can we do it in the bathroom? Like lots of different places. Maybe we take it on the road depending on what’s relevant for this behavior.
But I would actually – and this is a cool part. This is not like a six month process. This is like, I’m gonna do one session in this room. Five minutes here and five minutes there. Adding. [00:31:00] irrelevant distractions that aren’t directly related to getting the dog to go to station when someone rings the doorbell. Again, I don’t necessarily choose this one as much as I used to because of the hardness aspect of it, but I might.
But I would get that really, really strong in lots of different, well-generalized situations and then start ringing the doorbell, but only after I’ve worked a whole bunch of lateral stuff.
And you’re thinking, “Well I know my client wants this fixed yesterday.” And I know that. I agree because I also want my dog’s annoying behaviors fixed yesterday. And I don’t wanna put as much effort into it as I feel like I really ought to. So I’m gonna be completely honest: Every time I hear somebody say, “Oh, you know, dog owners, they don’t wanna do the work these days,” I don’t want to do the work. Most of the time I want things to just– I mean, once in a while, it would be okay if it would just be easy, right?
Okay. Well, maybe one day. But anyways, I digress. [00:32:00]
It can be pretty fun. This is where you can kind of make stuff a game. There’s so much motivation that can be found in playing these games of, “Oh, well, can you do it while someone flashes the lights on and off? Can you do it with playing musical chairs and I’m the instructor here and you are my learner and so I’m gonna come over and I’m gonna play music on my speaker. When the music stops, you take your dog to station and you feed them, and then we go back to the musical stations again.”
There’s lots of ways that you could have fun with this and then put it in the hallway and add it to the doorbell and just like with handling and medical stuff, if you have already done a lot of weird stuff with this station training behavior, or whatever your alternate behavior is, then ringing the doorbell is just one more weird thing and you’re gonna be so much more successful and it’s really actually going to go faster overall. You’ll spend less time dealing with the doorbell because you’ll have fewer errors because you worked through [00:33:00] all of those other seemingly unrelated distractions before you added the hard one.
I just find that so fascinating and so cool. There’s so much room for creativity and who wants to do things the same all the time? I can’t; I can’t follow a recipe to save my life.
So thinking about it out of context. Then we need a plan to put it in context. That’s part of my plan is to do it in lots of different ways that aren’t directly related to the final hard way. If I can, I love to fade in aspects of that harder final aspect. So maybe I’m putting it in the hallway and I’m opening and closing the door, or I’m putting it in the hallway and I’ve rigged up a remote control doorbell, or I’ve hired someone to ring the doorbell, but nobody comes over, just the doorbell rings and it’s just random and it’s not connected to any particular event. Sometimes people come in without ringing the doorbell. We might have other people over if that works out and then bring the dog out second so they’re already in the living room sitting down and we bring the dog out and we do station stuff in the hallway, and [00:34:00] then finally it works it all up to we’re in the hallway, the doorbell rings, the homeowner opens the door, people come in, doggie is on the mat, et cetera, and oh, it’s magical and wonderful, and everything is a perfect utopia of dog behavior.
So we do need that plan to put it in context. And then the final piece, of course is going to be, we need a way to tap into that original reinforcer. So if attention has been determined that it’s maintaining the problem behavior, the jumping up, the jumping around behavior, we’ve got the dog on mat, then calling the dog for attention off the mat is gonna be what maintains that behavior. It can’t always be food – or I mean, it could always be food, but again, that’s where I think people say, “Oh, you cookie pushers, that’s never gonna work in real life because what if the dog would rather greet the person than take your treat?” That’s not usually the example that’s given, but I’m using that here because it fits my [00:35:00] story. That’s true. Because what’s gonna maintain that station behavior isn’t the food. That’s just the temporary reinforcer that you’re using to get that behavior in repertoire in the first place, because it’s more convenient. But really quickly you’re gonna want to add getting to greet people to that behavior.
So the way that that might play out is: you have the station set up in the hallway and the doorbell rings and you cue the dog to station and you answer the door, you let the people in and then maybe you take them away from the hallway. Maybe the living room is a more convenient place to let your dog greet and then you call the dog over to greet (now that would need to be taught separately, so stimulus control over greeting would be a necessary part of your training plan), but you call ’em over to greet, they get petted and then maybe you send them back to the mat so that you can call them back to get petted so that you can send them back to the mat so you can call them back to get petted, and then eventually it just becomes, “Oh, the doorbell rings. I walk up to the door, I [00:36:00] answer it. Dog’s on mat. People come in, I call a dog off the mat, they come over, get petting, and then they go about their business. Maybe they get a chewy.” There’s a lot of ways that that could possibly play out, infinity options.
But we do need a plan to bring the context together, both the antecedent aspect and then also the consequence, the reinforcer. That’s what’s gonna maintain the behavior.
And then the other thing I wanted to mention in terms of selecting and working with the alternate behaviors – I mean, I’m not done yet by any means. But the other thing I wanted to bring up under this bullet point is that it doesn’t need to be finished. You can actually do some shaping inside of this alternate behavior.
So maybe the dog isn’t fully laying down on the mat, but if they’ll even just tag up, if they’ll target the mat, you could call ’em over to be petted. If they’ll get all four feet on the mat, call ’em over to be petted. Think of the same shaping strategy or shaping plan that you use to teach the dog to do that behavior with food – you can use that exact same [00:37:00] plan, or very close to it, with the new reinforcer to reshape that behavior in context.
And that’s very clever. And also, again, goes really quickly, especially if the behavior is already in the dog’s repertoire. Again, they are because you have taught it. If they’re already doing that behavior in other contexts, it’s more likely to show up in this one too than if you’re starting completely from scratch.
So having shaped it once, you can shape it again with a new reinforcer really pretty quickly. So don’t worry too much if you can’t get the full behavior right away. That is okay. It’s not a fail. It’s just a shaping step.
Okay, so, the next thing that I wanna talk about in terms of working with alternate behaviors is that we need to decide if we wanna use differential reinforcement.
So a lot of our differential reinforcement plans, whether it’s DRA (Differential Reinforcement of [00:38:00] Alternate), DRI (Differential Reinforcement of Incompatible), there’s lots of different differential reinforcements, DR all kinds of things. And I’ll talk about some of those in other episodes, the ones that I think come up the most often for us.
But I wanna talk about differential reinforcement in general because I used to use a lot more differential reinforcement period than I do now, and I’ve kind of changed how I think about it.
So with differential reinforcement, you have these two behaviors, we’ll say jumping up and go to mat, to give us concrete things to talk about. You could, using differential reinforcement, reinforce the one (go to mat) and not reinforce the other one, not reinforce it differently, so that effectively puts the jumping up into extinction. It’s been reinforced in the past. It’s no longer being reinforced, so that meets our definition for extinction, and then only reinforce when the dog goes to mat.
So for example, you ignore [00:39:00] him while he jumps up. He jumps up. He jumps up. Nothing happens. Nothing happens. That behavior goes into extinction. He goes on to his mat and now you call him over to get petting.
Okay. I mean, that can sometimes work. Sometimes. Intermittently.
But sometimes it doesn’t work or sometimes it’s not practical. I would say probably it’s more that it’s an issue of being not practical, because for a lot of behaviors, being able to apply differential reinforcement can be really limiting.
One, it requires us to have really good control over the access to the reinforcer. Sometimes we do, sometimes we don’t. But if you do, okay, we could ignore it and let extinction do the job for us in terms of eliminating the problem behavior and then only reinforcing the good behavior with that same reinforcer.
But we don’t always have easy control. Even when we do, like even when the function of the behavior is getting your [00:40:00] attention – it is, as we would label it, “attention seeking” or if we wanna be very judgmental, “attention demanding.” I don’t like that phrase anymore, because what’s wrong with wanting your attention? Nothing. Like, that’s why we have dogs. Anyways.
Ignoring behavior is really hard, especially behaviors that were designed to get your attention. Like the reason that behavior is there in the first place is because it’s really good at getting your attention. Annoying things are really good at getting your attention. Car alarms, the battery in your smoke detector going out, that’s irritating and so gets your attention. So actually ignoring it is hard and often impractical. Especially if the behavior in question is something like barking and you live in an apartment and waiting out the barking is gonna get you evicted. Nah, that’s not okay. Waiting out, jumping up if you’re working with a big dog, that’s not good. It’s not gonna be practical. And being consistent with it is gonna make it really hard. And if that behavior is getting reinforced [00:41:00] intermittently, so you ignore it until you can’t anymore, and then you reward it, even just by accident, ugh, it’s gonna stick around for forever. So it makes it worse.
And the other thing is that the extinction process itself can be terrible for everyone involved. If you’ve experienced extinction, an extinction burst and the frustration that comes with that, it’s really freaking aversive. It’s very unpleasant. I don’t love it for the dogs. And if you’re the victim of the extinction burst, that’s also really unpleasant.
So with when a behavior goes in extinction, we typically see – that’s what we’re talking about with the extinction burst. We see this behavior all of a sudden go through this huge, like massive period of intensity and variability where they try a bunch of different things. They’re throwing themselves at the behavior. And if you’re the thing, then they’re throwing themselves at you. So if we let Larry’s jumping-up behavior go into extinction, he may start throwing himself at you. He can get a bunch of clawing you, grabbing at your [00:42:00] clothes and that can be painful and unpleasant. And it sucks for Larry too, because he’s like, “Why are you not looking at me?” And that can be really upsetting for everybody.
So I don’t love ignoring as a first line of defense most of the time. I don’t love using extinction to eliminate a behavior or to rule out a behavior. I would really prefer to work more of an errorless learning kind of strategy where I’m gonna set up the environment so that the jumping up on you is way less likely to occur. And so I’m not dependent on extinguishing the jumping up so that I can reinforce the go to mat. I wanna set it up so that the go to mat is very, very likely to happen and the jumping up is less likely to be expressed in the first place, so that the dog never experiences extinction.
So that might mean that I put up a barrier like a baby gate and I put the mat right smack in front of that baby gate so that Larry can’t help but step on it if he wants to approach you at all. And [00:43:00] the baby gate prevents him from getting to you to jump up on you. And “Oh my gosh, he stepped on the mat. Fabulous. I’m gonna give you petting Larry. And he stepped on the mat.” (And again, we’ve taught this separately.) And he steps on the mat and this time maybe he sniffs the mat. “Oh, great. Larry, I’m gonna call you over to get petting.” Maybe you even have a way to remote control open the baby gate? I don’t know there. But you could approach the baby gate, right? You can get creative. You know how to do this. But at any rate, I can set up the environment so that Larry can’t help but be on the mat and he gets reinforced so heavily for being on the mat, it makes it more efficient, it makes it cheaper and easier, and he just doesn’t do the jumping up. So there’s no extinction, no extinction burst. Larry’s very successful. He doesn’t feel frustrated. His needs are getting met through this other behavior. And you don’t have to suffer through the outcome of that extinction burst. And it’s more likely to be effective, because again, reinforcement drives behavior, so no amount of extinguishing the jumping up is going to build [00:44:00] the mat behavior unless mat behavior is being reinforced. Right.
So that’s the approach that I’m more likely to do.
And someone’s listening and they’re thinking, “Well, you can’t have a baby gate there forever, Hannah.” Well, of course not, but reinforcement history is such a powerful thing. I mean, you know this. If you’re the trainer who’s working with Larry’s family, you know that because Larry’s been jumping up for 18 months, this is a harder problem to deal with than if they had addressed it when Larry was 10 weeks old. But they didn’t because they read a book and they thought they had this. And so they’re calling you now that he’s 18 months old and he’s been jumping up for a year and a half, and now he’s 80 pounds, and obviously the elderly grandmother in the family also lives with them and she’s just had her hip replaced. There’s all kinds of complicated aspects of the situation that we have to work with.
But anyway, because he has been practicing this for a year and a half, you know that reinforcement history is very, very powerful. Well, it works the other way [00:45:00] too. So if you can put a ton of reinforcement history for this easier, cheaper behavior, then the baby gate’s not gonna be relevant. It’s not gonna matter anymore. He’s so much more likely to step on that mat than he is to jump up because it’s been working, it’s been working really well. It’s cheaper, it’s easier, it’s faster, it’s more efficient, and it’s very effective at getting access to the reinforcer that he really wants.
Okay, so to sum up: The first step in working with an alternate behavior– Well, the first to working with any problem behavior, particularly when an alternate behavior is part of your plan – is to identify the function of that behavior. The best alternate behaviors are going to be just more socially acceptable ways for the dog to access the exact same reinforcer. It’s so much less work and it’s so much easier to maintain. We talked about this in that in the “maintenance mode” episode that I did. So identifying the function, choosing the right alternate behavior, something that’s easy for the dog to be doing, ideally they’re already doing it, it may even be very similar to the [00:46:00] behavior that you’re trying to change, just a less offensive version of it would be lovely. You’ll need a plan to either teach it out of context and then bring it into context, both the the A and the C, the antecedent and the consequence.
Plan to utilize shaping steps as part of that strategy. Decide if you’re gonna use differential reinforcement and if so, what’s ignoring the behavior gonna look like? How are you going to control access to that reinforcer in a way that’s going to be practical for whoever is gonna be in charge of this (you or if you’re working with a client). How’s that gonna be managed and what’s that gonna look like? And if not, how could you set up the environment so that the learning is more errorless and you’re not depending on extinction to get the job done? Because that is [00:47:00] effectively what errorless learning means. Doesn’t mean that they don’t ever make the mistake – it just means that you’re not depending on extinction. You’re using shaping of the environment to make the desired behavior, your new alternate behavior, a lot easier to do.
So really the way that I think about it is, alternate behaviors are just kind of a fancy way of teaching your dog to ask nicely for the things that he already needs and wants. And it doesn’t mean you’re constantly managing him, you’re preventing him from getting access to the stuff that he needs, whether it’s access to the people or distance from the people. There’s always so many ways to get that dog’s needs met and we can teach him better ways to ask for them.
So I’m planning to continue this series talking about dealing with problem behaviors or getting rid of behaviors with positive reinforcement (which is such a weird thing).[00:48:00] And I wanna talk more about that.
But I would love to know, first, what are your thoughts about working with alternate behaviors? It’s not so simple. I’d love to hear some examples of stuff that’s worked well for you. I’d love to know what questions you have, either about alternate behaviors or questions that you have in working with problem behaviors or eliminating, stopping behavior using positive reinforcement. What are the things that come up for you? What are the questions that you personally have or what are the questions that you hear that you’d love to hear me talk about? If I don’t know the answers, I’ll either try to find someone who does or I’ll try to figure it out or I’ll tell you that I don’t know the answer. I’m not afraid to not know stuff (fortunately, because there’s so many things I don’t know).
I really do wanna hear from you. If you have a question or a thought or an example you’d love to share, you can definitely email me. The contact form on my website is a great way to do that. You can find the post for this episode on social [00:49:00] media. Facebook or Instagram usually is the best and leave it as a comment under there. You can try to message me on social media. I will have to admit that I get a pretty high volume of private messages on those platforms and so sometimes I am able to keep up with them and sometimes I am not. Email is a guaranteed way that your question will get in front of me, so you can always email me, leave it as comment or take your chances with private messaging. It never bothers me. You’re not going to offend me. So you don’t even need to say, “I’m sorry, I know you’re busy.” I mean, yes, but don’t say you’re sorry because I’m not sorry. You’re not gonna bother me at all. I’m gonna do the best that I can. So if you’re cool with that, I’m cool with that. And then, then we’re good.
But anyways, I wanna hear from you. So let me me know!
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’re already subscribed, well, thank [00:50:00] you. I really appreciate it! And there are still some ways that you 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. 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!