Reinforcers and rewards can be powerful tools, but get them wrong and you can get very tangled up. I see a lot of newer trainers want to use only one type of treat, or only bring a tiny amount of good stuff and expect their dog to work for the lowest value items. This is what I think of as the cardboard fallacy. Suppose for a minute that you could get paid for your job in one of two ways. Either we will give you your pay this week in gold, or in cardboard. As of this writing, baled cardboard goes for between $10 USD and $162 USD per ton. Gold on the other hand is about $40 USD per gram. Now, one gram of gold is going to occupy 0.05 cubic cm and one ton of cardboard is going to occupy 9 cubic yards. Setting aside that these items are sold in units that are on completely different measuring systems, I could pay you $700/week in gold that would sit nicely in the palm of your hand. Or I could fill your front yard several times over in cardboard to give you the same value payment. The biggest thing here is that if both a handful of gold and your entire front lawn full of cardboard are worth the same thing, why is it that you want one thing more than another? Especially if it is raining. There is a huge disadvantage to $700 worth of wet cardboard sitting on your lawn.
If you are going to give a gift, if you want it to be well received, you need to know your recipient well enough to figure out if they are going to appreciate $700 worth of cardboard, or if they need to get paid in gold. I look at rewards for dogs like gifts; if it is going to be a good reward, I need to know my recipient at least a little bit. To begin with, I need to know my recipient on the species level. This means that I need to know a little bit about what dogs like to do.
Off the top of my head I can honestly say that most dogs like food. There are a few rare exceptions, but for the most part, dogs like food, unless they are extremely distressed. As a group, they have preferences for meat, fish and other high protein sources, and for fats. This means that beef liver is likely going to be a high value item for them, and dried toast will still be valuable, but less valued than is beef liver. Orange slices MIGHT be of interest, but they are likely going to be less valued than is meat. There may be a few dogs who are exceptions to this, but most dogs are going to like beef liver better than dried bread and dried bread better than orange slices. Orange slices are the cardboard of the dog currency world. Yes, some folks would be delighted to have up to 630cubic yards of cardboard as a gift, but these folks are the exception not the rule. Most people are going to prefer a small amount of gold over the same amount of cardboard, and most dogs will prefer a tiny piece of liver over a large slice of orange.
The next level I need to know a little bit about is the breed level. Within the species “dog” there are many different types of dogs and they are generally grouped into 7 different groups; the sporting dogs, the hounds, the terriers, the herding dogs, the toy breeds, the working dogs and the miscellaneous group. Each of these groups have some commonalities. The sporting dogs are those dogs used for hunting, usually birds. Most of them really like to fetch, so offering them a game that involves bringing back a toy is a natural reward for them. Hounds on the other hand are used for hunting by sight or scent and most of those dogs don’t want to fetch but they DO like to search. Terriers love to dig and tear things up, and they like searching but many of them aren’t as interested in fetching. Herding dogs like chasing and gathering things up, but they are generally less interested in digging. The working breeds include the guarding dogs and the livestock guardians and they have some interesting traits. Guarding dogs love games like tug where they can pretend to take down the bad guy, but livestock guardians don’t like tug at all; they are often hard to motivate because they descended from dogs bred to follow the flocks and get between the flock and a predator. Toy breeds and dogs in the miscellaneous group like things related to the groups that the dogs originated in.
What do the groups have to do with what rewards your dog will like? Ask the next Basset Hound you meet how he would feel about herding sheep as a reward and you can see quickly that knowing something about the general preferences of the groups of dogs can really help to inform you about what to try first as a reward. If instead I offered my Chesapeake Bay Retriever the chance to swim if he did something I liked, I would be more on track than if I offered a Basset Hound the chance to herd sheep. Now you may be thinking to yourself that you have met oodles of Border Collies and German Shepherds who love to play frisbee or swim, and you would be right, but that brings us to two points; herding dogs tend to be pretty flexible in what they like to do, and the individual can give us a better picture of what they actually like to do as an individual.
When you get a puppy, you will find that he is fairly flexible in what he likes and doesn’t like to do. I have seen young retrievers who will spend a whole day fascinated by a herd of sheep that would not attract an adult retriever in the least. Herding dog puppies will often learn to chase and grab things easily and that is easy enough to translate into a number of interesting and rewarding behaviours such as tug, fetch or search. You can often teach a young terrier or guarding dog to fetch in a similar way. Doing this at a young age is quite straight forward, but trying to get it started with an older dog is much more difficult. This is where knowing your individual is important.
My dog Eco really, really likes biting bad guys. We did protection work when he was younger and he is really good at it. Eco’s pinnacle reward is biting bad guys. Bad guys are gold to Eco. It is not practical to carry around a guy in an agitator’s suit to reward him on a regular basis, but that is his top reward. We started when he was about 9 weeks old, teaching him to “worry” a rag and then to tug and then to tug big and then to get a good bite on a bad guy. The most valuable behaviour to me is self control. This means that I had a perfect situation to use in order to reward Eco for self control. As a youngster, the best way for him to get a chance to bite a bad guy was to show self control and follow my cues. If he didn’t show self control in training, he didn’t get his turn to bite the bad guy. In this situation, knowing my dog and knowing what I had taught him to do this worked brilliantly.
Friday could care less about biting bad guys. If I offered her the chance to do so, she would just stand back, looking confused because that isn’t what we taught her as a youngster. What she does love though is frisbee. John took the time to teach Friday to chase and fetch back a frisbee and when they are working on obedience behaviours, this is her gold reward. Heeling and coming when called are always faster and sharper if she thinks she is getting a frisbee at the end. Again, this is not a reward John can use all the time, but in a training set up, it works really, really well.
John made a mistake when Friday was a puppy. He used only her kibble for rewards for many months. At first, she liked her kibble well enough, but later, she got tired of it and eventually, it became a cardboard reinforcer. I remember one memorable class where she gently took the kibble out of John’s hand, and then panted with it stuck to her tongue! While this was funny, it did nicely to illustrate what happens when you use cardboard to reward a dog; she humoured him for a bit, but she really, really did not want kibble as a reward.
Our first Chesapeake Bay Retriever was named Bear and Bear had a few things he loved to do. One thing he really enjoyed was swimming. He loved to swim so much that we had to teach him to get out of the water when we told him to do so. What might have enticed him enough to come out of the water? The opportunity to swim again! Bear would get out of the water in order to get the chance to go back in again. Getting into the water was Bear’s gold reward. While it was really difficult to carry a river in our pocket, we did an awful lot of training around open water to take advantage of this great reward.
In class we can use some play and some life rewards (the rewards that your dog just loves to have and that just happen to happen along the way) but for the most part we are going to use food. There are a lot of reasons to use food. To begin with it is fast. I don’t have to take it away from the dog after he is done with it. I can use it in gradations to pay really great behaviours with really favourite foods, and really easy behaviours with so-so foods. This means though that you have to know a little bit about what your dog likes and doesn’t like.
There is an easy game to play to help dogs to tell us which foods are gold rewards and which foods are cardboard rewards. Take two different foods; say kibble and sausage. Give the dog a piece of kibble. Give the dog a piece of sausage. Now, offer kibble in one hand, sausage in the other. Which does he choose? Kibble? You would be surprised the number of dogs who choose kibble over sausage. Now take your kibble and cheese and repeat. Let’s say that the dog chose kibble over sausage, but cheese over kibble. Now you have a hierarchy of treats. You know more now about what your individual dog likes; this particular dog likes sausage, prefers kibble, but really gets excited about cheese.
Observe your dog carefully and be aware of what drives their behaviour. I have seen dogs who will do nearly anything to get to roll in mud. One of my dogs particularly likes to go out to the barn and smell the horses (they are less enthusiastic about the activity). I had one dog in class who would do nearly anything for the chance to heel on leash; he loved heeling! There are many dogs who love to scavenge and will do a lot in order to get the chance to sniff around for treats. Don’t dismiss these reinforcers just because they are infrequent or unusual or because you don’t like them; you don’t have to like the gift you are giving; what matters is if the recipient likes it.
Once you have a good long list of things your dog likes, you can pair them with what your dog is doing. If your dog is learning something really difficult, choose something he likes a lot but doesn’t lose his mind over. It is no good using a bad guy to teach Eco to fetch for instance because he would be so very excited that he would throw the item at me in his enthusiasm to bite bad guys. I might choose a lesser and more convenient reward for bringing a toy back to me such as the opportunity to chase and get another toy! If I am working on quiet work, such as holding an item while sitting, I might use a medium value food treat so that Eco wouldn’t get too excited, AND so that he would still know that he was on the right track. Once the behaviour is well established for a medium value reward, then I might set up a situation where he could have his gold reward, but he would need to be very well trained on a behaviour before I would reward it with the chance to bite a bad guy.
The end of this story is simply that good training often requires that you have a variety of reinforcers available to use under differing circumstances. There are thousands of great things your dog can have as reinforcement; food, toys, the chance to greet a friend, time off leash, time to himself, time with you. I even let one of my dogs fetch a lawn chair as a reinforcement once! Be creative. Don’t pay only in cardboard and know what is gold to your dog. And remember to bring at least three different types of reinforcers to class each time so that you have a good chance of reinforcing your dog if he changes his mind about what he likes today.