Dealing with a behavior you'd really like to change? From leash reactivity to noise sensitivity, Sharon and I talk about it in this week's episode of the podcast.
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'll be talking to Sharon Carroll.
Sharon started competing with dogs over 30 years ago. Then she made the change from competing with dogs to competing with horses.
For the next few decades Sharon had a successful career riding and coaching through to the international levels of both eventing and dressage. She has been an Australian representative rider, and in 2013 acquired her EA Level 3 dressage specialist coaching certificate, the highest equestrian coaching qualification attainable.
She holds a Bachelor of Applied Science, a Graduate Diploma in Captive Vertebrate Management, and a Master of Animal Science, and is currently completing a Ph.D.
Animal behaviour, training, species-specific cognition, and welfare are key areas of Sharon's focus.
Sharon assists owners with behavioural issues in both horses and dogs, and is a fully certified behaviour consultant with the International Association of Animal Behavior Consultants (IAABC) in both species. Much of Sharon's routine work is with anxiety-based issues, and aggressive behaviours in dogs.
Just over three years ago Sharon made the transition back from competing with horses to competing with dogs, and now she is looking forward to progressing in a range of dog sports.
Hi Sharon, welcome back to the podcast!
Sharon Carroll: Hi Melissa. It's great to be here.
Melissa Breau: Excited to talk again. To refresh everybody's memory, can you share a little bit about who the dogs are that you live with and what you're working on with them?
Sharon Carroll: Absolutely. I have Dodge, who is my nearly 14-year-old Papillion. He does a lot of sleeping, a lot of eating treats. Fortunately he's not showing much signs of age, so that's really good. He still goes on big walks with the big dogs and spins around at my feet and jumps up and down for any opportunity for food.
Then I have Jericho. He's my 3-year-old Standard Poodle. He's got his CDX and his Rally Masters and his Expert Trick Title, and he's got a TEAM title. Lately I'm trying to get more TEAM training done, it's just trying to fit it in. He's currently doing a bit of scent training. I've done quite a lot of scent detection type training myself, but I've never really got into the competition side of scentwork at all, so I've been hoping to get lessons with Julie Symons and hopefully Jericho will enter the nosework competition scene one day. It won't be anytime soon, but one day.
And I've got Vincent. He's my baby boy. He's 20 months now. He's another Standard Poodle. He gets to work on everything. He's nearly ready to have a go in the ring in obedience. He's got a couple of TEAM titles, he's got his championship trick dog title, and he's doing a bit of nosework training and I started some tracking training with him recently with Lucy Newton. We've gotten through the tracking course and that's a new sport to me, so that's really exciting. I've been enjoying that. Vincent is a lot of fun. He's a great dog to work with.
Both my poodles are a bit low energy, and I have to plan my training sessions super carefully so I can fit in all the training without exhausting them or reducing their enthusiasm for training.
I have no doubt there will be a new puppy joining us this year, which is pretty exciting. I'm trying to make that super-tough decision whether to change breeds or not. Poodles here are not recognized as retrievers or gun dogs or working dogs, so they're really not bred for that, and you don't see many in the obedience ring.
Most of the breeders here focus on conformation and also breeding for therapy work and assistance dog work, and I'd really love to have a dog with a bit more energy to push me to work a bit. I used to compete with working line Border Collies, so it's a big jump from those to where I am now with Poodles, but I do love the Poodles, and it looks like Vincent's little sister might have a litter this year, and that might be too tempting, so I probably will have another Poodle next time you talk to me.
Melissa Breau: That's exciting. Puppies are exciting.
Sharon Carroll: Yes.
Melissa Breau: The plan is to talk B-Mod today, so I thought it would make sense to start with a rough definition. B-Mod is short for behavior modification, so it makes sense to start there. What is behavior modification and how does it factor into dog training as a whole?
Sharon Carroll: There's actually no definitive line between dog training and behavior modification. There's that bit of a blanket idea that people have that in behavior modification we're modifying the existing behaviors, whereas in dog training we're creating new behaviors.
But you can see that gets a bit hazy pretty quickly because certainly in dog training, and in competition training, sometimes we have a dog that's maybe coming in crooked in its fronts, and we strategize to fix that and have the dog come in straight, and we're modifying behavior. Conversely, of course, quite often in behavior modification, what we're actually doing is creating new behaviors to replace those existing behaviors that we don't want.
I can say, from a practical perspective, if a potential client rings up and says, "I have a dog that I'd like to be able to do xyz behavior," then I will consider that a dog training situation. Whereas if I have a potential client ring up and say, "My dog is already doing xyz and I really need it to stop," and usually they're going to be things like digging, barking, biting, escape behaviors, leash reactivity, and so forth, usually there's going to be a significant emotional component on the part of the dog. Usually the behavior is going to be driven by some sort of need within the dog.
So when I put all that together, that's what I would tend to refer to as something that's a behavior modification case.
Melissa Breau: I think when B-mod comes up, so do a lot of other phrases that are, for lack of a better term, we hear them all the time but maybe don't necessarily know exactly what they mean — things like desensitization, habituation, counter-conditioning, Premack, those kind of things. Can you give us a little insight into what some of those things have to do with behavior modification or how they compare? Can you talk about that a little bit?
Sharon Carroll: To look at the technical definitions of the terms, habituation refers to that process of response reactions, so when the dog or individual is actually responding less and less progressively to the same stimulus. Desensitization techniques are the methods we use to achieve habituation. If we're to look at two examples of desensitization techniques, they may be systematic desensitization and counter-conditioning.
But if we wanted to look at it more practically and have a look at those three, most of the time, when people talk habituation as a behavior modification strategy, really what they're referring to is encouraging that natural process of habituation to occur. We know, our dog, for example, can walk past our dining room table and it just doesn't respond. It's not finding it particularly exciting, it's not finding it particularly fear inducing, it just has no change in arousal and no change in behavior associated with that because it's habituated to it.
We know for us, as humans, we can walk down the street, we can walk past other humans, we can walk past vehicles going by, and if something particularly catches our eye for a very specific reason, we are just going to walk past it. We are not going to have an emotional response. We're not going to have a change in arousal or a change in behavior because we've habituated to that.
In behavior modification what we tend to look at for habituation is how can we use this concept to prevent behavior issues from occurring. If we want a dog that we picture as an adult dog is going to be able to walk down the street or walk through the park, and it's going to be able to walk past other dogs and walk past unknown people and not have a big change in arousal or a big change in behavior, then we really need to set that situation up. That dog needs lots and lots of practice and simply walking past dogs and people and nothing fear-inducing happening and nothing exciting happening, and the dog will habituate to going past them and be able to walk past them on a nice loose leash and have no change in arousal or change in behavior.
Where it tends to go wrong is a lot of people are so enthusiastic about giving their dogs all these positive experiences that they encourage their dog to go up to every dog or every stranger. The problem then is what we're doing is drawing attention to those dogs, for example, and so our dog now starts predicting dogs as exciting things to interact with. So it warms up to one and has a fun encounter, and warms up to the next one, and next thing it's actually pulling us toward dogs because it just sees them as very fun and exciting things. And so instead of reacting in the presence of dogs, now it's reacting several meters from dogs, and now it's looking for dogs so that it can start to get excited.
At that point we see a dog that's potentially leaping up and down, lunging, barking, pulling towards other dogs or people, and that gives us our own set of problems. But then beyond that, what happens when we block the dog from performing that behavior, potentially we get frustration from the dog, so that adds to our existing arousal.
Then we have a risk for the dog to tip over into aggression. They can turn around and start attacking the leash, they can attack the handler, or start to bite the handler when they try to restrain them. And they certainly can snap at the dog, if they make finally contact with the dog, purely because they've become so aroused that they tip over into aggression.
So when we talk about habituation and behavior modification, people are usually referring to how can we use that basic concept to prevent behavioral issues from occurring.
But if we look at systematic desensitization, that refers to that gradual habituation to an arousal stimulus, so we commonly use that in behavior modification when the problem is caused by an inappropriate arousal. It really doesn't matter the valence of that arousal, so the dog can be super excited, or it can be super fearful, it really doesn't matter. What we're trying to do is reduce that arousal response, and we do that by gradually exposing the dog to low levels of that stimulus that it can cope with, and then steadily decreasing that distance only when the animal reliably fails to react at the previous level.
So we want to get to a zone where the dog kind of notices the other dog is there, but we don't get a big change in arousal, we don't get a big change in behavior, the dog can still listen to us, it can still take treats. We hang out in that zone until the dog is so comfortable that it really hasn't even noticed that the other dog is there, and then we get a little closer. And over time we get a little closer, a little closer, eventually we're in the immediate presence of the original stimulus. I think this way the technique aims to raise the threshold for the response.
Then, if you look at counter-conditioning, that's when we're exposing individuals to repeated stimulus pairings of the opposite balance, so negative and positive. The practical use of this in training is that we might have a dog with a negative balance emotion like fear, and instead of that dog seeing that stimulus and just feeling fear, what we're going to do is pair that situation, or pair that stimulus, with us perhaps giving food, something positive.
So now, after several repetitions where the dog is given food in the presence of the stimulus, then the stimulus itself starts to predict that good things are going to happen for the dog. So often that systematic desensitization and counter-conditioning most commonly are great together in some form and work as one basic principal.
I think the other one you mentioned was the Premack principal. This one is thrown around a lot when it comes to behavior modification. The basic principal is that it's the relative theory of reinforcement. It really suggests that we can use behaviors to reinforce other behaviors. What it states is that the more probable behaviors will reinforce the less probable behaviors.
A really simple example or way to explain this is if you have a child that loves ice cream but doesn't really like vegetables, and you put down a bowl of vegetables and a bowl of ice cream, by free choice, giving them complete free choice, what's the more probable behavior and what's the less probable behavior? The more probable behavior is eating ice cream. But if we instead made eating the ice cream contingent upon eating the vegetables, if we said we're going to make the opportunity to perform that more probable behavior contingent upon doing that lower probability behavior, so you have to eat your vegetables before you can eat your ice cream, then we're more likely to see the lower probability behavior occur.
And that's exactly what happens with our dogs. For example, a dog that loves working with sheep, its higher probability behavior would be interacting with those sheep, but we might say to the dog, "Can you perform some position changes," or something, and then we're going to reinforce that behavior you've given us, that attention you've given us, that ability to perform those behaviors. We're going to reinforce that by allowing you to run over there and interact with the sheep. So we're using that more probable behavior to reinforce the less probable behavior.
The only thing I would say that I often see happen when people talk about using the Premack principal in their training is that you have to remember that you are encouraging that high probability behavior. Every time you use it as a reinforcer, you are encouraging that high probability behavior. So just make sure that that high probability behavior is a behavior you actually want.
You don't just go, "My dog loves doing xyz, or "My dog loves chasing birds," or "My dog loves chasing bicycles," so I have to put that into a Premack principal. No, you don't. You don't have to. You can use other strategies to create impulse control around those things. So always make sure that the high probability behavior is a behavior you actually do want to encourage.
The other thing I would say is make sure that your high probability behavior supports your lower probability behavior, and by that I'm talking arousal levels. If I want my dog to remain calm and lay on its mat while bicycles are going by and things like that, that might be a low probability behavior for some dogs, so I need to reinforce that with another low arousal behavior. My high probability behavior in that case might be eating, so I might reinforce that low probability behavior of remaining calm in that setting by giving the dog a treat.
Whereas what I'm not going to do is say, "You've remained calm for so long, I'm going to jump up and play tug with you," because now you've got your high probability behavior not supporting your low probability. After many successions of that, the dog won't actually be lying calmly on the mat. It might be lying on the mat, but it's going to be lying there with its tongue hanging out and its eyes popping out of its head because it's anticipating that really exciting high probability behavior, that reinforcement of tugging. So just make sure that your high probability behavior, arousal-wise, supports your lower probability behavior.
Melissa Breau: I threw some phrases around there and asked you to talk about them, but I know you're doing a webinar on this stuff next week. What protocols or processes are you talking about in the webinar? Can you give us the 101 version of that right now?
Sharon Carroll: Absolutely. We're talking about five protocols. We're talking about offered durational engagement, which is essentially my version of putting together some systematic desensitization, a counter-conditioning component, and also an operant training component, so a component where we are reinforcing certain behaviors individually as well. So we have offered durational engagement, we have distraction, which is active management, what to do when things go wrong. If we notice that our dog is starting to increase in arousal, or starting to change in behavior, we might want to jump in with some distraction. Or if we see that a situation is about to happen that the dog is not yet ready to cope with, we might jump in with some distraction.
We talk about stationing, which is stay behavior. It usually includes staying on a mat or a platform or a bed. It might even be stationed with the handler. We talk about reinforcement of an incompatible behavior. For example, if we have a dog that's jumping up on us, it can't be jumping up on us and have four feet on the floor. So if we reinforce the incompatible behavior of four feet on the floor, we end up with a dog that's now got four feet on the floor in our presence instead of jumping up on us.
The last one I talk about is management. Management is where you just do something to stop the behavior or avoid the behavior from occurring entirely. If you have a dog that triggers when it sees other dogs, and you avoid triggers, you just avoid going places where there are dogs, that's a management strategy. Or if you know your dog is going to jump up on guests at the door and get really excited, and you just decide to put the dog in another room, that would be a management strategy.
Management strategies aren't aimed at changing future behavior as such. They're not a training strategy. But they are really useful as part of a behavior modification program because we stop the practicing of that unwanted behavior. And I would also say that it's important to note that sometimes it's okay to just use management long term.
People often feel like they're some sort of failed dog trainer because they decide to use management in certain situations, but really, especially with something that comes about very rarely, if you have your grandkids over visiting not very often, and only a few times of the year do they actually use the swimming pool, but when they are in the swimming pool, they're running around and giggling and splashing, your dog is going to run the pool fence, barking at them. But if that's the only time you find it does it, then take the dog, put him in a nice, quiet room, give him something to chew on, problem solved. Use management. Don't put all your effort into retraining that if you don't need to.
Melissa Breau: If those are the solutions, you've given me a little of this already, but what problems are we really looking at that these types of things can help us solve? What behavior issues are you talking through in the course of the webinar?
Sharon Carroll: I think if you have those five protocols in your toolbox, if you have your offered duration engagement, your distraction, your stationing, your reinforcement of an incompatible behavior, and your management, if you know when and how to use those, you're probably going to be able to deal with almost all behavior issues. But in this webinar we certainly do cover a lot of different issues. When people were asked for what issues they wanted covered, we ended up with a giant list, so I think we've got something like seventeen behaviors to look at.
We're going to include things like leash reactivity, off leash dog-to-dog behavior issues, transport issues, both the loading issues and the traveling issues themselves, separation-related problems, sound sensitivity, fence running and barking, hyper-vigilance in the home, dogs who are uncomfortable around strangers, over-excitement when you get the leash or the collar or the harness out, counter surfing, jumping up on guests, attention seeking during meals … I think that must be about it.
Oh no, actually we also look at dogs that reject treats. The reason I wanted to look at that is that's a really big stumbling block for a lot of people. They really do want to get involved in dog training, or they want to modify some existing behavior, but they think they can't because their dog won't take treats. And so we look at the issues that cause that to happen, to help those people out, hopefully. We also look at the opposite problem: the dogs that over-arouse to treats or training. They see the training situation setting up and they get really excited or certainly over-aroused for one reason or another.
We also talk about strategies for building confidence in those super-fearful dogs, those dogs that genuinely almost can't leave the home or can't come out from under the bed. So there's lots of things we can use those protocols for.
Melissa Breau: That's quite a list. Would you be willing to talk us through one of those as an example? How you would approach changing one of those using the protocols you mentioned?
Sharon Carroll: Sure. Let's look at sound sensitivity, because it's probably one that is not covered as often as maybe leash reactivity or something. Sound sensitivity can be fear based, and most commonly is going to be fear based, but you can get dogs that are just overaroused or stimulated by a certain sound. It really doesn't matter what balance that emotion is, whether the dog is fearful or whether it's excited. What we go ahead and do is use that offered durational engagement, so that element of systematic desensitization, that element of counter-conditioning.
When we talk about the systematic desensitization with these guys, with these sound-sensitive dogs, of course we're not talking about decreasing distance the same way we were for those visual triggers. We're talking about reducing volume; reducing the intensity means reducing volume. We can lower the volume directly, if the device that's causing the noise can just have the volume lowered on it. We can start with performing that offered durational engagement in the presence of incredibly low volume and then slowly build that intensity up. If it's a device that we can't lower the volume on directly, we can try increasing distance from the object that's emitting the sound, because obviously that inherently lowers the volume.
What we can do as well is we can record that noise, maybe on our phone or something, and then replay it back on our phone. That way we can control the volume. Say it's a microwave or a blender or a washing machine, something that's making noise and we can't control the volume. We can record that sound and use the recording of that sound in our desensitization program and build that volume up towards the real-life volume.
Another thing we can do with that recording-type thing is that we can change the environment as well. Let's say we have a dog that every time the doorbell rings, the dog just straight away goes to frustration. It's out of control, there's barking, there's running, there's excitement, and the person says to me, "I don't know where to start. I can't even get in there." You can record it, you can play it quieter, and straight away the dog still runs to the door. Well, let's take that recording to a park, or let's take it somewhere else that's a different setting, and let's start our desensitization there, because that's going to straight away interrupt that pattern that the dog has. He's going to hear the doorbell, but he can't run to the door because there's no door. So you're going to have time to get in there and get some training done and then shift back to doing it in a home setting.
Melissa Breau: I love that idea of taking it to the park. That's fantastic. You teased in there leash reactivity is more common, maybe gets talked about a little bit, but I'd still love to talk about it, if you're up for it. How do you approach barking and lunging at people, at dogs, that kind of thing?
Sharon Carroll: I think leash reactivity is seriously the big one. I think a lot of behavior modification work for behavior consultants works around leash reactivity, partly because sometimes it's been inadvertently created, but not always. Certainly there are some other reasons why it can happen.
When we're talking about it, that sort of thing, we're talking about a dog that's lunging and whining, crying, barking, potentially growling, so there can be all sorts of vocalization. It may be directed at another dog, it might be a person, a car, an animal, a bicycle, a skateboard. There are a variety of underlying emotions and motivations. It's important at some point to work out what is driving this behavior, because it certainly can affect how we're going to address it from a behavior modification point of view.
Some common examples of what might drive this behavior will be something that's fear or anxiety based, something that's frustration based, something that's stimulated by just a fast movement or a sound. It can be predatory behavior. It can be purely high arousal or excitement within the dog itself before the trigger even presents. So it can be state driven. It can have owner-guarding behaviors. It can have genetically based dog aggression issues. There can be lots of reasons why a dog might decide to lunge and bark.
We start with that offered durational engagement, so using desensitization and counter-conditioning and instilling some new behaviors in the dog. But, for me, I don't usually start with a station behavior of any sort, unless I absolutely know where this is coming from, because if it's a fear based behavior, the last thing I want to do is be asking the dog to sit and stay sitting in front of me while I'm working with the dog, because I want that dog to feel free to avoid or escape or give me information. So I don't start with a station behavior.
However, once the dog is mostly not reacting to the previously triggering stimulus, then I'm going to consider adding a station behavior. The reason I'm going to do this is because if the behavior is purely fear based, if the dog is just performing the barking lunging behavior because it was scared of that stimulus, then once we go through our desensitization and counter-conditioning, the dog's not fearful anymore, guess what, the behavior completely disappears and we don't need to do any more on that.
But, if there's a desire component involved, if the dog is actually chasing the bike because it likes chasing bikes, or because it's got an owner-guarding behavior and people come near the owner and it's barking and lunging because of that, so there's a component that's not fear based, but it's actually driven by desire on the part of the dog, then just desensitizing and counter-conditioning won't be enough.
It won't help, because what will happen is, yes, it's really good at helping with impulse control, it's really good at getting the dog literally desensitized to that stimulus being present. However, it's still going to leap out and perform the behavior when the object gets close, because you haven't told it not to, so it's still going to perform that behavior because it wasn't triggered by fear.
At that point we often have to add a heavily reinforced station behavior. That might be sitting in the heel position. It might be sitting in the straddle position. We're saying to the dog, now we've put that work in to help you get impulse control, we've desensitized you to the presence of a stimulus that you used to find incredibly stimulating and arousing. We've put that work in, but on top of that, guess what. I need you to perform this behavior in the presence of these things, and I'm going to heavily reinforce your ability to do that. So we do have to a little bit at some point often tease out why the behavior is occurring, just so we can get the exact right behavior modification approach.
Melissa Breau: I know we were talking about that in some of our back and forth before this, before actually doing the call and hitting record. I know another component that you mentioned was this idea that clients will have a dog that barks and lunges once, and then they see what they think is the same trigger again and they get a different response, less or none or something else. You said you have clients ask about that all the time. Do you want to talk a little bit about that too?
Sharon Carroll: Of course. It is a common question. It's a super-common question, but I think this is because as humans we try to work out why the dog is doing it. So we go, our dog's lunging at bicycles, okay, but why does it sometimes lunge at bicycles and not at other times. Our dog is lunging at other dogs and we're looking for a specific thing. Is it big dogs, is it small dogs, is it fluffy dogs? We're trying to work out is there an obvious visual trigger that we can predict, and often there is, but there are lots of times when there isn't.
It can be that the dog is perceiving something that is definitely there that we don't perceive because we're humans. Maybe there is a certain scent that triggers the dog, or a certain sound, or maybe there's something visual that we're just not seeing. But also it often is that there is a different energy level or behavior of that stimulus.
For example, maybe an old Labrador flopping on by, it's nice, low arousal, not really that interested, it wanders by and it doesn't trigger our dog. But a little Toy Poodle goes past and it's quite active and trotting along and it's a bit more energized, maybe it's having a bark and a lip in the air and that might trigger our dog. People, at that point, if it's the first time they see it, they go, "Oh, it's small dogs." No, not necessarily, because the next time a really nice, slow dog, a small dog with low arousal, might wander past and you don't get that behavior.
Equally, sometimes some people approach the dog in a nice, calm, relaxed way and they're casual and they don't make a big approach on the dog, and you don't get a response. But then another person might be all excited and they approach the dog quicker, or they approach with a higher energy, or they talk in a high-pitched voice, and the dog does trigger. And so it can be the different energy level or the different behavior of the stimulus itself.
It can also be trigger stacking, which a lot of people are familiar with. For example, we head out the door, our dog sees a bicycle, it doesn't react, but then a few houses later some dogs run at the fence line and bark at it, and then a little while later some small children go running past, then we get to the next corner and a bicycle goes past, and our dog does lunge at the bicycle. So it can just be that trigger-stacking effect, which is essentially increasing arousal. Arousal is very accumulating in dogs, whether it's positive or negative, it all just adds into the one pot.
Another thing to note is it can be completely state driven by the dog, meaning that your own dog, when it's at low arousal, may not react, and when it's at a higher arousal it might react. If, for example, we go for a nice, relaxed, sniffy walk and the dog is at a very low arousal level, and a dog goes trotting on by, it may not react or it might, so you look up and increase its arousal a little bit, nothing big happens. But then the next time, we maybe have just finished playing tug with our dog, or playing fetch with our dog, or we've been running, and the dog is a bit more aroused, and then a dog goes by and they'll turn around and snap at that dog. So it may be that the dog's own arousal state is what's driving the difference.
Melissa Breau: I think it's also worth just talking briefly about the elephant in the room anytime behavior modification comes up: this idea of intentional fallout and why positive reinforcement is so powerful when we're talking about behavior modification. Can you address that?
Sharon Carroll: Sure. That's a tough question, Melissa, because I know I'm going to have to use the word "punishment" in there, and people hate that word, and I don't want to be linked with that word. But I do agree with you.
I'm guessing what you're referring to here is the fairly frequent use of punishment-based approaches to address behavior issues and behavior modification. It's certainly not the modern approach, but it is still common in part of society. And certainly most clients, when I talk to them, they've tried some sort of punishment-based approach to fix the problem, or they've at least had people tell them how to fix the dog's issues with a punishment-based approach.
I guess it's still there, because it is the traditional approach, and often information like dog training is passed on generationally within a home. Certainly for me, my parents are in their 80s now, so it's a long while ago that I was a small child. But certainly in the home that I grew up in, dogs were absolutely chastised. If they toileted in the house, someone would chase them around with a rolled-up newspaper or rub their nose in it. If they put their feet on the kitchen bench, someone would bang the kitchen bench loudly or yell at them. So for a lot of people, that's the exposure they've had growing up, and then they got a dog, and then they used those techniques, and then their kids got dogs and they used those techniques.
I think for us in the dog training community, or in the dog world, we think everything has progressed so much, but it's important to note that not a lot of pet owners feel the need to go out and seek out new information. The information they have has served them well, usually, and so they continue along that path, not really meaning to, but maybe not realizing there is new behavior information out there.
I guess the other problem with punishment-based protocols is that they do work, unfortunately, in lots of cases, or they at least appear to work. They do stop the behavior initially. That punishment is not contingent, obviously. It's never going to work. Rubbing our dog's nose in its toileted area is never going to work. If we come home and there's rubbish spread everywhere from the trashcan and it's hours later, and we try to explain to the dog what it did, certainly that's never going to work. So when punishment is not contingent, it's never going to work.
But if the punishment is contingent on the behavior, so every time the dog puts its front feet up on the kitchen bench, someone bangs the bench really loudly and frightens the dog, then certainly that is going to essentially work. After a few repetitions, the dog is going to stop putting its feet on the bench.
What I would say is when we say punishment based training works, we have to define "works." We have to define what are we really talking about when we say it works. It might stop that behavior then and there, but what's the fallout? For me, I think we need to factor in the dog's mental health. I think we need to factor in the impact of those techniques on the relationship between us and the dog, the impact of those techniques on the future of learning of the dog. We know that punishment based training can affect future learning or how they learn.
Also I think when people are saying that punishment based training works, they're probably not factoring in the potential fallout, so the fact that we can actually get new problems created by the punishment. And for me, I'd say why do it? Why would we do it? Positive punishment by definition means to be aversive. It has to be aversive enough to stop the behavior, so it needs to be really unpleasant, usually fear inducing and intimidating or painful. Why would we want to do that to our dogs, if we can achieve the same or better results without making our dogs feel uncomfortable or fearful or confused or conflicted?
And of course some of these techniques don't work, and that's probably what keeps me in the job is when people have tried these techniques and they haven't worked for them. Sometimes people do manage to fix one of the unwanted behaviors, but just because they targeted the actual behavior instead of the reason for the behavior, they just often get a new unwanted behavior.
Or due to the punishment-based approach, the dog's arousal increases. Then we might get some redirecting onto the owner and we might end up with a situation where the dog is biting the owner now because of the punishment based approach. That's because if the dog is performing the behavior due to fear, and then we just add more fear and intimidation and pain on the dog, we're going to end up with a dog that basically is going to lash out because at that point it's trapped and it doesn't see an alternative.
Another time I can only see things go wrong is when a fearful dog is being asked to station and it just can't cope with being pinned to the spot. Certainly we know ourselves, I see this happen frequently, these are what people would term the "without warning" attacks when warning behaviors have been extinguished.
If you were to say to me, "I'm really scared of spiders," and I would say, "Well, I'm going to put a spider on your arm," you'd probably say, "No, you're not. I'm going to leave." But if I were to say to you, "I'm going to hit you with an electric cattle prod if you leave," well there's a fairly good incentive for you to stay still while I put that spider on your arm.
Now if you just don't like spiders, you're probably going to cope. You're not going to enjoy it, you're going to forever feel horrible about the experience, but you are going to cope. But if you really, really are scared of spiders, there's going to be a point that you're going to break, and it might be as I place it on your arm, or it might be when that spider starts crawling, or it might be when I put in on your face and not your arm. There's going to be a personal threshold that you can't tolerate, and then, because you've tried to suck it up for so long, you're in such a high arousal state that when you react, it's really ugly. You're going to throw that spider across the room, you're going to scream, you don't care about the cattle prod anymore.
And that's exactly what happens with these dogs that have been asked to sit there. They're scared of children approaching, maybe, they've been told to sit, they try to move, they've been corrected for that, they curl their lip, they get corrected for that, they growl, they get corrected for that. So in the end they're just sitting there, sucking it up, and if they can tolerate it, that dog forever may never lash out at a child, but it's forever going to be very uncomfortable.
But if that dog does have a breaking point, maybe the child touches the dog, maybe the dog can cope with that, but it can't cope with being hugged around the neck, or can't cope with being stared at, or maybe two children come running at once. There's going to be a personal breaking point for some dogs, and when they do go at that point, it's really ugly, just like it is with us and the spider. So at that point they just lash out. They don't care about corrections. They just react, and it can get pretty ugly, and that's where those big "without warning" attacks come from.
When we're training new behaviors, it always seems so straightforward and obvious to use rewards based training, especially for the people that are in the dog-training world. But when trying to get rid of a behavior, it's a bit of a human thing to get tunnel vision and only see that we're trying to get rid of a behavior, so we start looking at ways to stop that behavior.
And punishment is intuitive for humans in that situation. Even if we took a person who had never owned a dog before, and we told them that the goal was to get the dog to stay out of the kitchen, the first thing they'd do is yell at that dog or chase it when it came into the kitchen. If we said to that person, I want you to teach this dog to not jump up," the first thing they're going to do is just yell at the dog and push it away. So punishment based training is intuitive for humans. We know how to stop something from happening.
But if I was to give that same person a bag of treats and then say to that person, "Now I want you to stop the dog from jumping up on you, but you can't use an aversive; you can only use these treats," they would not know what to do, because using the words "based training" requires knowledge and understanding and thought that just isn't required if we're prepared to use punishment.
So punishment is the easier option because you don't need that same knowledge base, and unfortunately, behavior modification is where the aversives still come out, even from the more modern dog trainers, because many people just can't get past the fact that we are usually trying to stop a behavior in behavior modification, and by definition that's what the punishment quadrants are all about. They're about reducing the occurrence of a behavior.
Whereas behavior specialists, instead, we're going to focus on the conditions that lead to the undesirable behavior so the behavior is less likely to occur, or we're going to replace the current behavior or response with a different response and heavily reinforce that.
It's hard, though, to see so many people still reach for punishment based strategies when attempting to modify behaviors. It really just isn't necessary. I honestly believe there are effective rewards-based approaches for every situation. But yeah, tough question, Melissa, so I hope I answered that question in a useful way.
Melissa Breau: Yeah, I totally think so. I have one last question for you, Sharon, the one I always end on. What's something that you've learned or been reminded of recently when it comes to training?
Sharon Carroll: I just got reminded of this very recently, and that is how often the dog is learning something other than what we think we're teaching them. Often we have this giant plan, we can picture exactly what the end goal is, we can picture the cues we're expecting the dog to notice that are happening. And so we think we are teaching something, but somewhere in amongst it all, we thought we had made specifically relevant cues, but the dog saw other cues we hadn't even thought about, and now we've got a dog that is not performing the behavior for the reasons we think the dog is performing the behavior for so often.
I'm so aware of this because I see it happen time and time again with dogs, that we think we're teaching a certain behavior and they're learning a completely different thing. In my case, it was my older dog with nosework. I just spent a little bit too long on boxes and containers. I thought I was just teaching the dog to be fantastic at finding odor. My experience in scentwork has always been on container top searches, on scent detection in that sort of setting, as opposed to wandering through the environment looking for scent. So I'd spent way too long on boxes and containers.
My dog is phenomenally good at locating boxes. You can put boxes out and he'll find the right box. But I thought I was teaching him how to locate scent. He thought he was learning how to decide which box had the scent in it. So when I put the scent somewhere else, I put it on a chair in the middle of the patio area and asked him to search, he looked at me like I was crazy because there were no boxes present, and I suddenly realized what was happening. I thought, he doesn't know what this cue is at all. He doesn't know I'm asking him to look for scent.
What I'll do is try some empty boxes out there, he'll realize from that cue, that context cue, what I'm asking, and once he gets in amongst the boxes, he'll use his nose and then he'll realize that the hide is on the chair. But no. He went out and he searched all those boxes, and he looked around and went "No, no odor in any of these boxes," and I went, "How can you be standing inches away from the odor and not put your nose on the hide?" Purely because he thought the game was check the boxes for odor and didn't realize the game was search for odor, wherever that might be. So as soon as I recognized that, then of course I just changed strategies and he's fine now.
Just recently someone said to me they were teaching their dog a go out, and in Australia a go out is they have to run 25 meters to a defined area on the ground. There's four pieces of PVC piping making a box on the ground, and the dog runs in a straight line and goes into that box and sits. The person said to me, "I built this distance up, the dog is phenomenal, it goes really fast, it goes on cue every time, goes the whole 25 meters, gets in the box, everything's great. But now she's just blowing me off. I give her the cue and she just doesn't do it. She just wanders off and starts sniffing." I said, "That doesn't sound quite right. Let's talk about what's happened."
When I talked to the lady, what had happened was she decided to teach this by putting a bowl in the box, a perfectly fine way of doing it, there's lots of different strategies. She had decided to put a bowl in the box with a bit of food in it, and she had been sending the dog out there, and it had been going to the bowl and eating the food. And then she'd moved on. She had taken the food out of the bowl, and the dog would run out to the box and she would go up and give her the food.
But sure enough, what did the dog learn? The dog had learned that the cue meant run to a bowl. It hadn't realized the box was even relevant. It just thought the bowl was the relevant cue. So when she took the bowl out, the dog looked at her like she was mad because, "How can I run to a bowl when there's no bowl to run to?" She thought the really obvious cue was the box on the ground, and the dog didn't even realize that was cue. It just thought the bowl was the really obvious cue.
So sure enough, when she tried to send it out, she showed me these videos, she's trying to send this dog out, and it's looking at her like, "How can I go to a bowl when there's no bowl there? You're telling me to run, but there's nothing to go to." It was just wandering off and sniffing, and she took it as the dog didn't know the exercise and it was wandering off. I went, "No, absolutely, this dog does not." So that's fine. We went back and put a bowl in there and did a bit of reduction of the size of the bowl and then faded that prop out, and the dog was absolutely fine. But it's so easy for us to think we're teaching one thing and they're learning something else.
Melissa Breau: I think that's true of people too. Sometimes we think we're teaching another person something …
Sharon Carroll: And that's not what they're learning.
Melissa Breau: No. Awesome. Thank you so much for coming on the podcast Sharon! I appreciate it.
Sharon Carroll: It's been great!
Melissa Breau: And thank you to our listeners for tuning in! We will be back next week with Nancy, Megan, and Amy to talk about whether size matters — and what the differences are when training a little dog versus a big dog.
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!
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!