Train Your Dog: Excitement Level and Value in Rewards
I discovered a need for an article that goes in-depth on the effects that different rewards can have on the goal of reinforcing of our dog’s behaviors. We are going to talk about different types of rewards in terms of value and consequent level of excitement.
The given value of any reward—be it food, play, praise, or a life reward—is entirely up to each individual dog. This being said, there are some general guidelines and things to consider when choosing a reward to reinforce behaviors in our dogs.
Food rewards are the best rewards. Why? Well, not only is food a primary reinforcer that relates directly to our dog’s biology, food is most practically and reliably manipulated and handled on a day to day basis. While praise and play may be handed out willy-nilly to our dog, food, in general, is not. In fact, mealtimes with kibble are a great opportunity for training sessions.
Food often has an array of value for each individual dog. For example, tripe (the edible lining of various farm animal’s stomachs) has, in general, a very high value for most of our canine companions. Other dogs go nuts for raw broccoli stems. I trained one dog once who, because of her issues with fear, required a very high value food item for her to accept anything as a reward. I found I had to reward her with a lick off "the peanut butter stick" (a honeycomb stick I carried around with a small jar of peanut butter in my training pouch).
Your dog gets kibble regularly at mealtimes, and this may make for enough of an enticing reward when inside the house during a calm, distraction-free training session. Kibble may not have the same desired effect of keeping your dog’s attention during an increase in stimulation during a walk or at the dog park.
Any reward that is too high in value may have the opposite effect of the desire for increased attention and willingness to learn. For some dogs, the attraction to the reward can throw them a bit too much into hindbrain mode (a state of mind more instinctual than governed by logical thought processes). The increase in level of excitement is to the detriment of the lesson, with the dog more focused on getting the reward no matter what, rather than figuring out the puzzle of what it is they need to do in order to actually get it.
On the whole, food of the appropriate value level for the sessions or behavior is a neutral reward that keeps the dog in place of the behavior and at the appropriate level of excitement.
Many dogs are more motivated by toys than by even high-value foods, such as tripe. My own dog turned from agility competitor to agility shining star once I figured out to use a game of tug at the end of every run in place of treats. Suddenly, my "interested, but not all that stoked" agility partner flew through the various obstacles in record speed. With such wild success, you might think I would have turned to a game of tug for all my behavior reinforcers.
What made this reinforcer so successful was that it heightened the level of excitement that helped my dog move faster and with more "hindbrain" than a slower, more logical thought-out approach would have. Imagine a scenario where you are trying to teach your dog how to "stay," remaining calm so as not to break position, and as a reward, you whipped out a toy for a game of tug. You will have only pulled your dog out of the stay position for the reward (what’s rewarding, staying in position or not?). Like with my dog, you will have brought his excitement up so drastically from the intended level of calm required to perform the behavior that he may have even forgot what you were doing in the first place.
In summary, toys for a toy-driven dog can fall into the trap of being too high value to be appropriate in most basic obedience training cases, keeping in mind that for any positional behavior (such as stay), it can only be rewarding in association with the behavior if it is given while remaining in the same position.
When we think of praise for our dogs, we may think of petting them, or some other form of physical interaction that comforts them. What comes as a surprise to many owners is that such physical "comforts," as we perceive them to be as humans, do not translate the same for dogs. In fact, more often than not, "comforts" translate as quite the opposite.
Dogs, unlike us, are not hardwired to relate physical interaction from humans to mean a comforting, reassuring act. More often than not, they don’t like it at all! Even eye contact in a dog’s perception can remain as intimidating from its owner as it would be perceived from another threatening dog. Many dogs learn to tolerate these cultural differences from their human counterparts, as they are often paired with other, more desirable things, such as treats, verbal cues, interaction after a long day at home, scratching that hard-to-reach itch, and other forms of pack approval desirable to the dog.
The value of these types of rewards vary greatly between dogs and can depend on each dog’s individual personality, and the associations built with the type of praise. Many dogs learn that the words "good dog" mean approval of their behavior, as it is also often paired with other things such as treats, so that the formally meaningless English word cue becomes a secondary reinforcer. A secondary reinforcer is a somewhat stepped down version of a reward but still enough of an effective, understandable form of communication.
None of the aforementioned reinforcers vary in value as greatly as the simple life reward: an environmental reward that has value to each dog as an individual. Too often owners turn a blind eye to what bountiful opportunities they have to use life rewards to their training advantage. In worst case scenarios, owners allow their dogs to self-reward for a number of undesired behaviors without even realizing it.
Take on-leash greetings for example: a dog pulls on the leash to greet another dog. Greeting another dog is an extremely valuable occurrence to a number of dogs, and should be treated as such. Too often owners allow such a valuable reward to work against them, allowing their dogs to pull on a leash (undesired behavior) while greeting another dog (highly valuable reward). Not only is this dangerous, it’s downright nonsensical!
The Premack Principle has very practical application when dealing with life rewards. Suitably dubbed, "Grandma’s Rule," the Premack Principle is known to be written in a "first/then" format: first, eat your vegetables, then you can go outside to play. It is a way of rewarding lower probability behaviors with behaviors that already have a high probability of occurring.
Owners and trainers alike can find many opportunities to work "free" rewards into their loose-leash walking training of their ever-smelling sniff machines. First, allow for a duration of acceptable walking with a loose leash, and then reward with periods of free reign to "smell the roses." Remember, too, that any type of structure in your dog’s life is what helps them to better understand the expectations put on them, and thrive in our often confusing human environment.