10 DecEpisode 183 | The matrix of expectations & resilience, issues with the “Positive Reinforcement” label & more
A School For The Dogs Instagram Reel that was meant to help dog owners understand how to tell if a trainer is a "positive reinforcement" trainer or not resulted in a battle in the comments section about what kind of dog training methods are best. Annie considers how the very title of "positive reinforcement dog trainer" is problematic, and talks about how the conversation led her to consider the possibility that maybe the divide between dog trainers comes down to expectations about what we want dogs to do and how emotionally resilient we think they may be.
Subscribe: iTunes
Mentioned in this episode:
“Clues a dog trainer may not be positive-reinforcement based” Reel on Instagram
Related Episodes:
Episode 2 | What is “Good” dog training?
Episode 47 | A busy person's guide to operant conditioning
Episode 61 | The Greatest Animal Trainer On Earth: Ken Ramirez
Episode 120 | News flash: Positive reinforcement isn't about being kind!
Transcript:
[music and intro]
Annie:
Last week, we posted a Reels on the School For The Dogs Instagram, and the title of the reel was “Clues that a dog trainer might not be positive reinforcement based.” And, you know, reels can be kind of tricky to do because they're short and to the point and meant to be kind of flip and quick and ephemeral.
And, you know, it's not like we workshop them for weeks. We do them pretty quickly. And there have been a couple of times where I've sort of regretted ones we've put up, not because I didn't think they were good, but because they deal with a topic that in reality is quite nuanced and complicated. And when you reduce a topic like this down to something that is 20 seconds long or 10 seconds long and lip synced to music, it can be misleading and certainly reductive.
Last week when we posted this reel saying, Hey, here are some tips that or some clues that a dog trainer you're working with might not be positive reinforcement based. I wrote the text for this reel and I guess the avatar in my mind of who was reading this is someone who is like I was when I was when I first got a dog, and first found a dog trainer. I mean, I didn't research different kinds of training. I just went to the closest doggy daycare that was offering puppy kindergarten classes and had no idea about the language people use or methods people use.
And so, I guess, often in things I do when I'm thinking about clients, I'm thinking about the client I would've been, and if somebody's following us on social media, I assume that means that they're kind of into what we're doing. So, I thought of it as like, Hey, if you're into what we're doing, here is how to maybe try and figure out if a trainer you're working with or following is doing a similar thing. And the shorthand for the kind of training we do that's most well understood is positive reinforcement training.
Now, I refer to it as positive reinforcement training as a kind of shorthand all the time. But I think that there are other things we can call it, and positive reinforcement training can be kind of misleading. I talk about this in the second episode of the podcast, the episode called “What is good dog training?”, because I think it frankly is kind of a problem that there isn't a better way to refer to the kind of dog training training we do.
And I have chosen to generally call it good dog training, but you might also refer to it as science based dog training, or reward based dog training or clicker training, or positive reinforcement training. Each of those labels also presents some problems.
But in this reel, I wrote, you might be able to spot some clues that a dog trainer is not doing the kind of dog training we're doing. And for instance, the word “balanced” is a term that often comes up on the web pages or in conversations with dog trainers who I guess I would say are not doing what I define as good dog training.
And generally speaking, I think that that means setting unreasonable expectations and using aversive and coercive methods to get behavior without recognizing the associations a dog is making and the possibility that a dog might be making the wrong associations with things that are intended to punish behaviors.
The thing that I really don't understand about so-called balanced training – and it's interesting, right? Because balanced seems like such a harmless word. But it's a euphemism for a balance between reinforcement and punishment that, you know, I've heard self-titled balanced trainers talk about having to use all four quadrants, which I'll talk more about that in a second.
But the thing I always run up against, and I would be curious to hear a so-called balanced trainer talk about is how can you account for the way a dog is feeling when they're, say, being shocked with an electric collar? And how can you account for the associations a dog might be making that you are not aware of?
So, you know, the example of a dog who is wearing an electric collar attached to an invisible fence gets shocked even maybe just one time as he is running to greet the UPS man, and does not realize that the shock associated with the boundary, but perhaps now thinks the UPS man is bad news. I don't know how you prevent against that without being able to use language to explain, Hey dog, don't worry. It's not the UPS, man, it's the boundary. But of course you don't have that luxury with dogs.
Anyway, because of course this is social media, it wasn't only the avatar that I had for a client like myself 10 or 15 years ago. There were plenty of dog trainers who have methods that are different than ours, and I guess you could say ethics different than ours, who watched this reel and left comments.
And I swear to God, I do not wanna be posting things that are gonna be divisive. If there's ever anything I've put online that has been divisive, it is not because I have wanted to start arguments. It is because I have been thoughtless. Because, you know, when you're running a business, you basically want people to like you. And yes, we're trying to do something to change the world. I often say that we're a mission driven business in that way.
I don't think the answer is starting comment wars on social media. And I also don't think the answer is bashing other trainers or their methods on this podcast or elsewhere, with the possible exception of Cesar Millan only because I feel like he is the most famous dog trainer in the world and represents a lot of what I think is wrong about dogs and the way people think about dogs in popular culture. So if there's one area where I'm hypocrite, it's maybe when it comes to him.
But beyond that, some of the people who commented on this reel said basically just that. Like, you shouldn't go around bashing other people's methods. Why don't you just focus on doing good stuff yourself. And okay, point taken. However, to be fair, I don't think we were really trying to bash anyone else. We were just saying, Hey, here are some clues that someone might not be doing the kind of dog training we're doing. Which, you may choose to go to someone who's not doing the kind of dog training we're doing.
Which perhaps, though, goes back to the problem of calling it positive reinforcement dog training, because some of the people in the comments then said, Well I am a balanced trainer, but I use lots of positive reinforcement all the time. And, of course, when trainers talk about using all four quadrants, they're talking about operant conditioning, which is the consequences of a behavior.
And the consequence might either be adding something to encourage a behavior, encourage the likelihood a behavior’s gonna happen again, or taking something away to encourage likelihood of behaviors gonna happen again, or discouraging a behavior is gonna happen again by either adding something or taking something away. And those are the four quadrants, as they're called, of operating conditioning, as codified by BF Skinner.
Positive reinforcement being the quadrant where we are trying to give the animal, in this case, something the animal wants in order to encourage them to keep doing behaviors that we want. So I see it as kind of like the everybody wins quadrant.
However, you cannot go through life having your behaviors influenced or influencing the behaviors of anyone else purely using this one quadrant, I think. If you're using positive reinforcement, you're very likely also sometimes using negative punishment. The example being the elevator game, where you're bending over to give your dog food on the ground while their butt is on the ground, sitting on the ground. You bend over as your dog starts to stand up, perhaps. You then stand up taking the food away, removing the desired thing in order to discourage the behavior of standing. You have now used negative punishment, and it's not really that big of a deal.
You also will likely punish your dog without meaning to. The other day I stepped on Poppy’s foot while we were playing tug, she ran into her crate scared, and hasn't wanted to play with that tug toy since then. You could certainly say I punished the behavior of playing with that tug toy, although it was not my intention to do so.
So I would say positive reinforcement trainers are the ones saying, Hey, this is the quadrant we really want to try and exist in. Whereas balanced trainers might say Yes, but also we have to acknowledge that we're going to be using these other quadrants. And I don't quite understand in doing so how they can deal with the learning by association that happens when you use anything aversive or coercive with a dog on purpose or not on purpose.
But the conversation in this thread really started to make me think that maybe a big part of this divide that there seems to be between the so-called positive reinforcement dog trainers and those who do not refer to themselves as such, is about expectations and consideration of a particular animal's abilities. And what got me thinking about this was two things.
One, the TikTok that I talked about last week, the woman chasing her cat around her living room with her Christmas tree saying that the goal was to traumatize the cat with the Christmas tree so that the cat would become scared of the Christmas tree.
The other was another comment that was on our positive reinforcement reels. This person on Instagram, who comments frequently on our posts by the name of oc_dog_training, that's OC underscore dog underscore training, looks like – it looks like a man and it looks like he has a business called On Command dog training.
Based on comments he's made in the past and from his profile, I think he is a trainer who specializes in working with gun dogs, and I have no firsthand experience working with gun dogs, just to be clear about that. Anyway he wrote a response to someone else's comment, “come out to the field and show me how you could control a retriever at 300 meters without using aversive. I've been waiting to see this for two decades. I’ll continue.”
Now to be clear, pretty much all of this person's comments have been antagonistic and seem to suggest he thinks what we're doing is, I don't know, for sissies. And I have actually reached out to him and said, look, I don't know anything about gun dog training using aversives or otherwise. And you seem to be passionate about the methods you use for dog training in general. I would love to have you on the podcast to talk about this.
And he has said that he doesn't think that that would be constructive to which I replied saying, Look, if you would just like to record something that I could share on the podcast, ‘cause you maybe don't wanna actually have a conversation with me, I would be willing to share that. I have not gotten any kind of response to that.
And, it's curious to me. It's curious to me, first of all, why people write angry comments on the social media pages of people who are doing anything that they don't agree with. And I'm sure this applies to any field you're in, that there are people who do this, and there are people who feel like I do, which is that like, how do you have the time to be doing this?
I mean, I could spend all day finding Instagram accounts of people doing things that I think should not be done to dogs. And I guess I could be spending my time commenting about how awful I think it is. But instead of that, I am trying to walk the walk, talk the talk, do work that is constructive and set a good example. And once in a while, perhaps of course I might slip up and post something that seems offensive to certain kinds of people.
But anyway, it's strange to me, I guess the world of trolls in general, and dog training trolls in particular, because if you don't like the way we're doing things, don't follow us on social media. What else can I say? I don't think any problems are gonna be solved in the comment section.
Anyway. What he said, though, about how would you train a dog at 300 meters? Really got me thinking because I thought, it's not something I would ever expect a dog to do. Now I have to the guess part of this is just because I live in New York City and the dogs I've owned and the dogs I've worked with, no one needs dogs to be behaving at such great distances off leash. And were I asked how I would, say, keep a dog safe in a 300 meter wide field without a tether, I would say I would put up a fence.
Now can a dog be trained without aversives at such a distance? I would guess so. Again, it's not something I know very much about, but I do know that people have worked with herding dogs for, probably for centuries, if not longer at great distances, getting behaviors by whistling and using arms signals. That kind of training must predate the existence of shock collars.
And I also know that I know people who have trained animals at great distances without aversives to do some pretty incredible things. Some of these have been discussed with the trainers themselves on this podcast. I'm thinking about Ken Ramirez training butterflies to fly on cue, or Bob Bailey who taught dolphins to do search and detection work in the ocean without using aversives at distances much greater than 300 meters.
But again, not something I know very much about. And it got me thinking less about, well, how would I do that? And more about why someone would want to do that and what we expect of the dogs in our lives. And I think maybe it's that question of expectations that really separates these two, let's call it dog training camps. That, coupled with an understanding or an ability to guess at what a particular animal or cannot handle.
It's almost like you could plot these two things on an X and Y axis. You know, one axis being how emotionally resilient an animal is. And some dogs would be on the one side of the spectrum being quite emotionally resilient, and others not so much.
So on the one side of that axis, you have the cat or dog or whatever, that's just bomb proof, can handle anything without too much stress, whatever you are asking. My old dog Amos was very much like this. He took everything in stride. Whereas I would say my current dog Poppy is on the other side of the spectrum, where every little thing can be a potential indicator that the sky eye is falling.
And then on the other axis, you could be plotting what your expectations are of that animal. And I sort of think that how humane you are as a trainer or just an animal owner maybe has to do with where things fall on those axes. You know, if you're asking the animal to do something that is asking a lot, and you have a kind of bulletproof dog, and you need to be using aversives to get to that point, maybe it's not that big of a deal?
I think that this is kind of the argument that a lot of people make in favor of doing training with E collars. For example, I know people who have e-collar on a dog because they don't want to be living behind a fence, is what they say. And they want the dog to have free run of their property. And they feel that that dog can handle that kind of aversive, and the cost-benefit analysis of it, I guess, works out in favor of controlling the dog using a shock collar.
Of course, the argument I would make is that we don't know the emotional state of the dog. We don't know what associations the dog is making. And for those reasons why not use the least invasive possible method of controlling the animal by, for instance, putting up a fence, or letting the dog out only on a leash, and every animal is different. Every dog is going to experience those circumstances differently.
The best that we can hope is that whoever is training this dog knows that dog better than anyone else, and, I guess, can guess better than anyone else what may or may not be okay for that dog.
And this made me think about the woman with the Christmas tree and the cat who has been insisting on social media that her cat is fine. And I believe her cat is fine because she knows her cat better than anyone else knows that cat. And I suppose that cat in this particular instance was not so severely traumatized.
And if there were fallout behaviors either, they were not severe or they went unnoticed. Stress can lead to different behaviors that you might not like. And certainly the cat must have been traumatized in some way in order for the whole thing to have worked. Although, as I mentioned in the last episode, a tree being jabbed at you that your owner is holding and running around the living room with is very different from a tree standing in the corner. So she might have developed, a fear of the former, but not of the ladder.
Anyway, I guess we have to trust a pet's owner to know what is or isn't okay for that pet. But I think part of the positive reinforcement mindset is assuming that very little that is punishing or coercive is going to be okay with any animal, and sort of starting with that assumption. And also starting with the assumption that we're only going to ask things that are reasonable to achieve using smart management and force free non-coercive methods as much as possible.
And for me, a lot of it comes back to what's really necessary. I mean, is it really necessary to have a Christmas tree in your house? Is it really necessary to have a dog with whom you are working the fields as you go around shooting other animals? And maybe there are times when these things are necessary and using some degree of force or punishment or coercion makes sense.
The example I thought of when I thought of, when I was trying to think about like, well, what would be a human correlation to this is using a snot sucker on my daughters. Now I'm thinking about this because both of them had colds this week, and I had a cold.
And if you're not a parent, you might not know that there are these bulb shaped snot suckers that you have to use on babies because they can't blow their own noses and they get stuffed up. And they're generally not big fans of this. And my six month old has certainly cried the handful of times in her short life that I have had to use this on her.
Now with my older daughter, who's three, I also used this thing on her, I don't know, half a dozen times when she was very little, and she never liked it very much, but I never put in the effort to train her to feel okay about it, only because it was so rare that I had to use it. And again, there was a sort of like cost-benefit equation that we're all working through all the time.
And I don't know, it seemed kind of like no big deal. She could handle this unpleasant thing happening very rarely. And she grew older and is now able to blow her own nose. And if I ever do need to use the snot suction bulb, we talk about it and she's okay with it. The little baby on the other hand is not so crazy about it, but she recovers from the trauma very quickly. So I think this is an example of taking into consideration, you know, the importance of what needs to happen and how the animal is going to perceive it and deal with it and recover from it.
Of course, it's a little different because I am trying to clear my daughter's nostrils for their own benefit. Whereas the TikTok woman was not putting up a Christmas tree for the cat's benefit. The gun dog trainer is not training the dog for the dog's benefit, but that's how I see it.
Maybe you could argue differently, like it's better for the cat to get a little bit scared of the tree than to climb the tree and get hurt. Or you could say that the gun dog ultimately benefits from being a gun dog and that therefore warrants whatever training needs to be done to get there, no matter what the methods are.
But just by contrast my good friend, Ilana, who I had on as a guest a few months ago, she has a son who has seizures. He literally has seizures and stops breathing when he gets very upset by something. And Ilana has therefore had to go to great lengths to basically do what in animal training is called cooperative care, or to shape the behavior using positive reinforcement, shape this behavior of getting something stuck up his nose and being okay with that.
Now does that make her a better parent than me? Yes, [laughs] probably does, but I think it illustrates my point in that so much of any relationship you're in with any animal, if you are trying to influence that animal's behavior, I think that these are two things to consider. What are your expectations, and what can your animal reasonably handle?
And I think, as a so-called positive reinforcement dog trainer who feels that that is not the best title for this kind of training, maybe that would be a better way, for me at least, to define my point of view and my thoughts on how I want to train dogs. I want to assume that a dog can't handle anything and set my expectations very low.
This doesn't mean we can't accomplish amazing things with positive reinforcement, but it's the difference between like approaching a block of marble with is a pick and chipping away at it in order to reveal some predefined idea of what a sculpture should be, versus starting with the amount of clay and considering the clay that is in front of you and what it may be capable of and building up from there.
This way of think about dog training is largely informed by the fact that I have worked with many very fearful dogs living in New York City, which is a place that is not easy for any animal to live in. We are asking a lot by asking dogs to live in our studio apartments on the 25th floor in crowded cities.
And I think a truly good approach is to not ask more than is reasonable for your dog, making every attempt to understand your dog, reading your dog's body language, noticing your dog's behavior, but also understanding that you will never be a mind reader. You will never truly know how your dog or cat or Guinea pig is thinking and feeling, and that even using anything purposefully punishing or aversive could be understood or experienced by the animal in ways that you may never fully know.
[music and outro]
