23 JulEpisode 17 | The Dog Training Triad Part 2: Rewards
Once you have set the stage to up the chances that you'll get lots of behaviors you want (see The Dog Training Triad Part 1: Management), you'll need to figure out how to reinforce those behaviors. In this episode, Annie discusses, among other things:
-The difference between reinforcers and rewards
-Discerning whether something is a reinforcer
-Figuring out your dog's individual reward cornucopia
-Understanding rewards as currency
-The changing value and appropriateness of rewards
-"Real life" rewards
-The importance of treat size when using food
Subscribe: iTunes
Transcript:
Annie:
Hi, there. This is the second episode in a three-part series on the training triad. The training triad is made up of three things: management, reward and timing. And, in my opinion, these are the three big things that you need to consider when you're creating any kind of positive reinforcement based training plan.
So in the last episode, we talked about management and management is just setting the stage for your learner to succeed, really creating a path where you are definitely going to get lots of behaviors you like, I like to call it the yellow brick road. So it’s a path that has boundaries which make it impossible to get a lot of behaviors that you wouldn't want. But within those boundaries your dog is going to have a lot of opportunity to do all the things we want him or her to do. And we talked about different ways you can think about management: you're going to physically manage their space, manage their time and you're going to manage their energy. So definitely go back and listen to that episode.
But now we are going to be moving onto the next step. So you have this well thought out path where your dog is going to be engaging in lots of appropriate behaviors, things you want him/her to do. Now what? We need to communicate to the dog that we like all these behaviors that are going on. We need to encourage these behaviors and how are we going to do that? Well, we're going to reward them.
I’m using the word reward here, rather than reinforcer, but truth is to use these words pretty interchangeably, I find most dog trainers do which is really fine but I just wanted to explain the difference even though, like I said, we often just use one word instead of the other. But the truth is that not all rewards are reinforcers and not all reinforcers are rewards.
Technically speaking, reinforcer is anything that encourages the likelihood that a behavior is going to happen again and usually those are rewards. Right? Anything that your dog likes is gonna be reward. And if your dog engages in a behavior you like and you reward appropriately that behavior should be reinforced. The thing is there's another kind of reinforcer which is called a negative reinforcer so the reinforcement that we usually think of we think of that reward is now good stuff right: money, love, affection and attention, but a reinforcer is anything that encourages a behavior, that’s the definition of what a reinforcer is. Those good things, those things that we like that tend to reinforce behavior, those are called positive reinforcers but there are also negative reinforcers and negative reinforcers are things that are going to reinforce a behavior because you basically want them to go away. A good example of something that is negatively reinforcing is that annoying beeping sound that you hear in a car when you don't have your seatbelt on. Right? The behavior putting on your seatbelt is encouraged or reinforced because it makes the annoying sound go away.
An example of negative reinforcement in the dog training world would be like pinching a dog’s ear until the dog sits and once the dog sits, the ear pinching stops. The behavior of sitting has been encouraged and reinforced because it's made the ear pinching go away. So these are not usually things we think about when we think about reinforcers but those are reinforcers. And when I said a reward is not necessarily a reinforcer, what I mean by that is there are lots of times that as a teacher, trainer and manager, you might give what you think that as a reward to your learner that just because you think of that as a reward doesn't mean that it's necessarily positively reinforcing to your dog or to a person as the case might be.
Anyway, this brings us to the essence of what a reward is which is it should be something that your dog enjoys and I think part of the fun of starting training or getting to know a dog that's new in your life is figuring out what they're really into, what did they enjoy? Now, of course, a lot of the rewards we use in training are going to be food-related. Your dog has to eat, we’re the ones usually giving them access to food. Most dogs enjoy eating, so we might as well use food to our advantage. And what you can start doing with food is really creating, kind of like, a hierarchy or like a currency list of what your dog is most interested in food-wise. Right? I think about it like down at the bottom around like a quarter or maybe a dollar would be a piece of kibble, up at the top, say hundred dollar bill, would be a piece of bacon or roast beef or turkey or chicken and then you have to fill in that rest of the chart.
And when it comes to using food in training, I suggest using things that are going to break up really easily. At School for the Dogs we use a lot of lamb lung, which is great because it breaks up into very, very small pieces and it doesn't get your hands too gross, it isn't slimy or sticky. But the most important thing I consider is that it breaks up into really small pieces and this is because we’re going to be getting a lot of rewards in training especially in the beginning and we want the pieces to be small because you want your dog to swallow the treat right away, you don't want them to waste a whole lifetime chewing. Right? The way I think of it is like learning is happening all the time and in your training session you want as much of the time to be about learning whatever it is you're trying to train and not them learning how to choose something.
So you can use the smallest piece that you can get away with it. I would say something the size of the TicTac or even ideally half the size of the TicTac is a good size guide. I also like using treats that are like pencil eraser size. Zukes makes treats like this. Tricky Trainers by Cloud Star is another line of treats that I like. These are a really nice because they break up really easily with your thumbnail and one of these pencil eraser size treats can be broken up into like six or seven pieces without getting too crumbly. And that's a big thing, you know, one time we got sent a product, some sort of like, treat-dispenser product that came with a milk bone in it at School for the Dogs and we all kind of looked at each other. And it wasn't even like there's anything that bad quality-wise about a milk bone we all kind of like obviously this was not a product made by a trainer because a trainer wouldn’t use something like a milk bone because you just can't break a milk bone into little pieces without it getting super, super crumbly.
When it comes to treats I also really like using things that a dog can lick. That's another way for them to eat something very quickly. We use what we call Liquid Treat Dispensers– I’ll link to these in the show notes. You can find them at storeforthedogs.com. We fill these with peanut butter, cream cheese, baby food, liverwurst, any number of things. I also like using Kalles which is like a coe roe fish paste from Sweden that we also carry. There is also a brand of liquid treat pouches called Bark Pouches which are really great. And there are these hard sticks, kind of like deodorant sticks, called LeanLix which I also like to use.
Anyway when you're figuring out your dog’s treat currency you want to experiment with lots of different things on commercial treats, lickable treats. Also go to the deli section of the supermarket and try getting some sliced meats, try getting some cheese. One thing that we also do a lot at School for the Dogs, we’ll often get string cheese or hot dogs- both of which are great because you can quarter them lengthwise and then chop them into lots of little bits so that you get, like, you can probably easily get like 50 or hundred pieces per thing of string cheese or hot dog. And then throw them in the freezer so they’re going to freeze into hard little bits and when you training with them, when you’re ready to train with them, you take them out of the freezer, they’re not going to get all gross in your hands.
Other human things that I like to use, human foods that I like to use for training include Cheerios which she can break up into small pieces, peanuts you can break up. And we’ve been talking a lot like things in the meat and dairy category but some dogs really like fruits and vegetables. You can use peas, you can use blueberries. My dog loves blueberries. You can use a banana chip.
So once you’ve selected five or six treats that you think your dog likes, what you can do is give him or her a taste at each one and then step out these five or six treats in like a semicircle around your dog, hold your dog back and let him or her go and see which treat he or she goes to first. And then you can figure out where that goes on your currency list then repeat with another five or six treats.
Now food is a great tool in training but there are lots of other kinds of rewards that you’re gonna use in training. I think, after food, attention for most dogs is probably the biggest reward that you can use. And of course play, whether that's a game of chase or fetch or tug or whatever it is your dog is into maybe it's just rolling on the bed with your dog and being super silly. An opportunity to play with another dog could be a reward or an invitation to come up on the couch or to jump up if that something your dog is not generally allowed to do. Anything that your dog enjoys, anything that your dog is into and I think what’s cool, like I said, about getting to know your dog in terms of figuring out what he or she is into, is that it's like you're figuring out the fingerprint of who they are, right? It’s similar with people, no two people find the exact same cornucopia of things rewarding and that's part of what makes each of us individuals and, I think, it's also part of what makes us good friends or parents or lovers are bosses or whatever is knowing what your subject is into .Right?
In the book, Don't Shoot the Dog, which is an excellent read by Karen Pryor, it's kind of about how to use positive reinforcement in life in general. She talks about how when her sons were little she felt it was really important that they learn to be good gift givers so when it was her daughter's birthday she would take her sons shopping to pick out a dress for her daughter. And I love that anecdote because I think it really is important to teach children to be thoughtful and in that way. We underestimate the value of being a good gift giver but when you give someone, like I said, doesn't necessarily need to be a sibling but whoever it is that you are trying to care about and show that you care about when you give that person a gift that really speaks to that person it’s meaningful, right? And of course this is again a way in which humans are vastly more complicated I think then then dogs. There's a lot more variety in the things that we find rewarding in life than for dogs, I think. For dogs we can get really far simply with food, play and attention-humans are generally a little bit more complex. But again you know what makes you think about what is a reward because for some people you know being hit is rewarding, for some people loud screeching music is rewarding and these are things that for someone else would be punishing.
I have a relative who once told me he sometimes daydreamed about going to jail because they give you free food and you don't have to pay rent and you have a place to sleep so you never know what someone is going to find rewarding.
Of course, you can pinpoint with great precision what your dog might find rewarding or person or whoever, if it makes behavior more likely to occur. If your dog is doing something over and over and over, you have to think about what the consequence is to the behavior that he or she is engaging in. And that consequence is reinforcing the behavior, that consequence is a reward. If your dog is doing something you don't like over and over and over and every time your dog does this thing you yell at him or her, chances are your yelling is actually only being perceived as a reward because it is part of what is positively reinforcing that behavior. How do you know the behavior is being positively reinforced? Well, most most likely it is and that's why it's happening again and again. And to some extent it’s the same with dogs, in that, for some dogs leaving the dog park is going to be a much bigger reward than going to the park.
Another thing to think about with rewards is that the value of a reward can change from one moment to another or from one situation to another. An example might be seeing a leash, right? When your dog sees the leash in your apartment he or she is super psyched because it means there's about to be a walk, but when you show the leash in the dog park it means something else.
My dog, a good example would be a tennis ball. Most of the time he really doesn’t care very much if I show him a tennis ball but if we are near a pool or any other kind of body of water he will go absolutely crazy for the tennis ball. Same tennis ball but it has totally different meaning to him if he's in a place where he knows I’m going to throw it and he can run and jump into a body of water and fetch it, which is pretty much his favorite thing to do.
So actually, interestingly, the leash I was just talking about in that scenario and also the tennis ball, these are rewards that are different than a reward like affection or food. These are secondary rewards or conditioned rewards-we tend to call them secondary or conditioned reinforcers. And a conditioned reinforcer or a secondary reinforcer is just anything that your dog had to learn to like. It’s something that your dog was not born knowing about. Right? Your dog was not born knowing about what a leash is and that the presence of the leash predicts going out. Your dog was not born knowing what a tennis ball was but your dog was born knowing that food is good or that play is fun.
Conditioned reinforcers play a huge role in our lives as humans. Money is conditioned reinforcer. You did not come out of your mother's womb knowing what a dollar bill was but, over time, you learned the value of that money because it could buy you lots of the primary reinforcers that you need and want to live. Words are simply secondary or conditioned reinforcers. Right? A word has no inherent meaning except for the meaning that you learned to give it. So I can reward you by saying “good job” but that's only because you've learned the meaning of those words.
Secondary reinforcers are extremely powerful. Just think about how money runs the world when, in and of itself, a dollar bill or check or a credit card or the numbers that you see on the screen when you check your bank account, they only have meaning because of the meaning that we give them. But those kinds of conditioned reinforcers like money wield a lot of power because we give them so much meaning.
There are a lot of things in your dog's life that are secondary reinforcers that you maybe haven’t even tuned in to yet. But once you start thinking about it, you'll notice, you know, maybe the sound of the refrigerator opening is a secondary reinforcer that your dog has picked up on because every time you open the fridge or oftentimes when you open the fridge, he or she gets something good. These kind of secondary reinforcers or conditioned reinforcers, and again you know the word conditioning just means learned, can help a dog really, like, predict what's going to happen.
In the dog training 101 class that I teach sometimes at School for the Dogs, my dog Amos is usually the demo dog and it's really the only class at School for the Dogs where he gets to participate, he gets a lot of attention and lots of treats and if he is at school with me and I start putting the chairs in the circle, he starts getting super excited. For him chairs in a circle has become a predictor of something good to come. A circle of chairs is a secondary reinforcer.
Again to reiterate, primary reinforcers are all those things that no one had to teach you to like, right like, food, sleep, play, attention, sex is a primary reinforcer. Actually from what I understand sometimes dolphins in training are rewarded by getting access to dolphins of the opposite sex. And secondary reinforcers are basically everything else and they’re secondary not in that they’re any lesser value, they’re secondary in that they need to be learned. But a secondary reinforcer can certainly be just as powerful or even more powerful, sometimes, than a primary reinforcer.
My mom's dog loves being outside, it’s her favorite thing in the world. She has literally jumped through a second-story screened in window in order to be able to be outside and it seems that the only thing she likes better than being outside is taking a ride in the car. So my mom has figured out that one way to get her to come when she's outside is to open the car door. The sound of the car door has become a conditioned reinforcer or secondary reinforcer that, to her, is even more rewarding than being outside.
And actually when you think about it, you could add a tertiary reinforcer, or a kind of third step, by proceeding opening the door in that situation with some kind of cue, say ringing a bell. With enough paired associations, the bell could be a learned reinforcer that's associated with opening the car door which is associated with going on a trip and then you have a cue in that you can ring the bell and have the dog come to you.
But, like I said, there are a million ways in which the value of a given reward might change and, with the dog, we may only ever be able to perceive some of the factors that go into the value of a reward at any given moment. But I think of it, kind of like, your dog is always putting whatever you're offering him or her on a scale against whatever he or she is already enjoying. So, in the case of, the dog who doesn't want to come inside being outside in that moment is far more rewarding and has a higher currency amount perhaps than anything else you might be able to offer, but, in another situation, because maybe there's fewer scents outside or your dog is more tired or more hungry, offering your dog bacon as a reward might be more valuable than that joy is being outside.
And just like I was saying, I think, part of the fun of getting to know your dog is figuring out what his currency is, you also have to figure out when certain rewards might work better then other rewards and often I think that has to do with figuring out what you can offer your dog as a reward versus whatever reward your dog is already enjoying in the moment be it playing or sniffing or eating where running around outside.
Certain rewards can also be inappropriate in certain situations. For example, you probably aren't going to want to reword your dog for behaving well at the vet with a tennis ball. You probably won't tip your waiter with a gift card. Sometimes a certain reward might not necessarily be inappropriate in a specific situation but it seems that way to the person giving the reward. I know a woman named Leslie Hock who runs a nonprofit in Romania where she works to try and get parents to bring their children to school and she's had a lot of success and, I believe, that the main way she's done this is by paying parents to bring their kids to school. And I'm sure that that's an idea for a method that would seem in bad taste to a lot of people but the fact is she's had a lot of success. Now it's true that most people I know send their kids to school either because it seems like an inherently reinforcing thing to do, it just feels like the right thing to do, or because the truancy laws-it's negatively reinforcing. They do it so that they don't get into trouble. But in Romania, I guess, negative reinforcement wasn't working and the inherent value of school wasn’t obvious but the value of money was obvious to these parents and, in the end, she's getting the behavior that she wanted, which is encouraging the behavior of parents bringing their kids to school.
An interesting thing about negative reinforcement is that it always produces behavior that's really just enough to get by. Negative reinforcement, again, is doing something to avoid something bad. Like I said, you put on your seatbelt to make the annoying beeping sound in the car go away but nobody's putting on their seatbelt with great flourish. Or taxes: we pay taxes because it makes the fear of being hounded by the IRS go away but I don't know anyone who pays more than they need to or who pays their taxes very early or with great excitement.
I have a cousin who’s son is in high school and he has figured out exactly what he needs to get in every single class in order to pass, like down to what grade he needs on every test. I think he has an app that helps him do this, so if he knows that he needs to only get a “C” certain test in order to pass the class, he puts in just enough effort to get that “C”. What’s interesting is that his father is actually a really excellent positive reinforcement dog trainer and he said to me once, “you know I know that the laws of learning are the same with dogs and humans but I just cannot figure out how to train my son to get better grades. I cannot figure out how to motivate him because nothing that I've offered seems to be rewarding enough.” And I said, “Oh gosh, I know it's rewarding to him.” And the reason I know is when he had his Bar Mitzvah I said to his parents what can I get him and they said that money- anything having to do with money, he's just really into money. And, of course, I think there are many parents out there who do pay their children for getting good grades but my cousins thought that this was in poor taste and brushed off my suggestion but I'm pretty sure it would've been effective reinforcer for this kid. And it really made me think about how kind of arbitrary it is that some rewards are deemed fine in some scenarios and not in others.
**Music**
In the next episode of this series, I am going to talk about timing, which is the third part of this training triad that we’re talking about. And in talking about timing, I am going to talk about how we use a clicker. But I just want to mention while we are on the subject of conditioned reinforcers that the clicker is a training tool that is a conditioned reinforcer. The clicker is a small handheld device that you operate with your thumb. It makes a sharp noise and that noise, after repeatedly pairing with a reward-treat or play or whatever, the dog ends up learning the meaning of the click and that click can be a powerful tool for training. And I don't actually always use a clicker, I often use a word, just like we learn the meanings of words, like I said, “good job” has meaning because of what you've associated with those words and I can use the words “good job” with you as a secondary reinforcer, a conditioned reinforcer. Same thing with the clicker but I often do use words, usually I use the word “yes pairing that always with some kind of primary reinforcer: food, play, affection, etc. And just like money is so powerful in getting what we want from human beings, this kind of conditioned reinforcer, be it a click or a word like “yes,” I find is hugely effective when training dogs especially when we are thinking about this training triad, creating a really well-managed environment, choosing our rewards and then delivering them with really good timing which is where the clicker is going to come in.
So just in summary on the topic of rewards: figure out what it is your dog likes and figure out how you can use those things that your dog likes to reward behaviors that you like. If you’re going to use food rewards, think about food rewards that can be delivered quickly, easily and efficiently and that can be swallowed quickly by your dog. One reason that food can sometimes be better in training than play or attention or affection or anything else that your dog likes is it just faster to deliver it, it’s faster to get a lot of reps if you're using food then you have to throw a ball every time your dog does whatever it is you want him/her to do. Although, often, I find it can be really effective to switch back and forth between different kinds of rewards, use that food rewards during some training, switch to play etc. etc.
Remember that every dog is different and that part of your job as a trainer is figuring out what's rewarding to your dog and also figuring out what's rewarding during which situations. Some things might be rewarding in one situation but not in another. You know also you’re thinking about creating that kind of currency or figuring out what your dog's currency is, there may be times when you really need those hundred dollar bill treats, right? When you’re asking your dog to do something complicated or if your dog is stressed out or you're working in a new environment, you're setting new criteria, whatever the reason is, there are times when you're really going to need those big guns. Other times, you can use those dollar bills or 50 cent treats.
Also start thinking about what real life rewards you can use with your dog. Like I said, maybe an invitation to jump up on the bed or a chance to run down the block with you. What are non- food rewards that you can give to your dog that can reinforce behaviors you like.
Lastly, start keeping an eye out for what conditioned reinforcers your dog enjoys. What things has your dog learned to feel good about, like a said, maybe it's the leash appearing out of nowhere or maybe it's you opening up your sock drawer. My dog gets excited quite often when I open up my sock drawer because I think he knows that's often a predictor of me putting on my shoes which is often a predictor of us going out for a walk.
Fun dog fact of the day: I mentioned milk bones before. Did you know that milk bones used to be called Maltoids and that they were originally made on the Lower East Side in Manhattan and now they are made in Buffalo New York but I believe in the last 110 years, they have been produced in one way or another in New York State. It's true that I don’t use milk bones very often in training; I don't think you'll find many a good dog trainer who does. Like I said, simply for the reason that they don't break up very well, but I do like a good milk bone reference. Here's one from my favorite shows from my childhood, Cheers:
“What’s happening Norm?”
“It’s a dog-eat-dog world, Sammy, and I’m wearing Milk Bone underwear.”
**Music**
Our Woof Shout Out today goes to a human, Milla Chappell. If you have been on our Instagram lately, you’ll see she has been taking amazing photos of our classes, some of our students. Milla is a photojournalist who has turned her attention to dogs and the relationships between dogs and their people. I am a big fan of her work. If you would like to book Milla for a private session, mention that you heard about her in this podcast and you will get a mounted 11×14 inch print when you sign up for a shoot with her. Which you can do by emailing milla@realhappydogs.com.
I’m also collecting questions for an upcoming Q&A episode. If you have a question that you would like answered, please go ahead and either send a voicemail through the Anchor app or you can call and leave a voicemail at 917-414-2625 or just email podcast@schoolforthedogs.com
I have gotten a few questions about details regarding running a small business. As you all know, School for the Dogs is in New York City coming for private lessons, playtimes, classes, you name it and I’m always excited to learn about tools that other small businesses use to run things smoothly and I am also happy to share some of the tools that we use. One tool that has been indispensable for us has been Pike13. Pike13 is the platform through which all of our scheduling is done, all of our sales are run through Pike13 and they also built our scheduling app which is in the App store, its SFTDNYC. As a business owner, I love it because I find it really easy to use, it's intuitive, I have all my own reports in one place, they are simple to run. It’s just snag free. But I also love it because we are always hearing from our clients how user-friendly our website is, how easy it is to book things, people love our app so that makes me feel really good because anything we can do to make the entire dog training experience fun, easy, stress-free for our clients, for me, that is a big win, because I think our human clients are just as important as our dog clients.
So if you’re a dog trainer or a gym owner or a yoga studio owner or anyone who is trying to manage a whole bunch of clients and a space and staff etc etc, definitely check out Pike13. You can learn more at schoolforthedogs.com/Pike
**Music**
Special thanks to Nicole Toombs for her really wonderful ukulele rendition of the Harry McClintock song, Big Rock Candy Mountain. You can find Nicole on Youtube, Nicole Toombs.
LINKS:
