Learn more about the proper use of food in training in this factsheet (used with permission from the Pet Professional Guild)
Why should I use food when training my dog?
Surely if I ask my dog to do something, he should just do it? Dogs are thinking, feeling, and intelligent creatures. Though they are not little people in fur suits, like us they will do things that are fun and rewarding and avoid things that are not. Asking your dog to do something ‘just because you said so’ is like your boss asking you to work for no pay. How motivated would you be to do it? You could force your dog to comply, but what would that do to your relationship? How would you feel about somebody who forced you to work? On the other hand, if you set up the game that your dog understands as "If I do what you want, you'll give me a treat," it's a relationship-building win-win situation. The dog is motivated to learn, and both of you can enjoy the training session. You love your dog, so why wouldn't you want to use a training "tool" that gets him excited and makes him happy?
Why does my dog only listen to me when I have a piece of hot-dog in my hand?
Dogs are so intelligent that they don't only learn what we WANT to teach them - they also read between the lines and learn things we don't mean to teach them! So if you start off with that piece of hotdog waving under his nose as you train him, he will expect to see and/or smell that piece of hotdog every time. This is a common complaint in food training, and there's an easy fix. Simply make sure that you use the following steps:
- Cue the behaviour,
- MARK the behaviour (with a clicker or word, to let the dog know "that is the right answer and a reward is coming"), and only then
- Reach for the treat. If the treat always appears after the behaviour, the dog will learn that he doesn't need to see or smell it ahead of time to earn it.
How do I ensure that I don't have to bribe my dog forever when I would like him to do something?
First of all, there is a difference between a bribe and a reward. The bribe comes first, before the behaviour; the reward comes after the behaviour. It's an important distinction, because "bribing" is not an effective way to train, but rewarding/reinforcing is.
If you bribe, your dog will only perform the behaviour if there's a piece of food in front of him first. If you reward, then your dog will perform the behaviour in hopes of earning that piece of food. And after the learning phase, when that behaviour is fluent you can start to phase out the food rewards.
For example, if you are teaching your dog to sit on cue, you would start off by reinforcing every sit with a piece of food. At some point that behaviour will become so familiar and easy to your dog that when you say "sit," he will do it without thinking. When he is that good at sitting on cue, you can reward him with food sometimes but not every time. Then you may be able to substitute praise (good dog!) or what is called a life reward. If he wants you to open the door so he can go out into a fenced yard, he can "earn" that by sitting at the door. Or if you ask him to sit before you put his food bowl down, that is also a life reward.
Basically the food is a training tool, and will not be required forever. But also keep in mind that nobody goes to one math class and comes out able to solve algebraic equations! Give your dog a lot of practice in a lot of different contexts before deciding that he "knows" how to do something.
If I train a lot, won't my dog get fat?
Whatever amount you put into your dog for training during the day, you must deduct from his dish at dinner time. If you stuff your dog full of treats all day long, and then feed him his regular meals, he just may get fat! Treats are wonderful, and powerful treats (like hotdogs, cheese and peanut butter) may be necessary when you're training a difficult behaviour or in a difficult environment with a lot of distractions. But other than that, you have a great resource for training in that bowl you fill up twice a day. Instead of just putting the bowl down and letting him eat it all at once, use a portion of it for training. Some trainers give their dogs their entire daily ration in training sessions throughout the day.
Also note that dogs don't really work for the taste of food, they work for the smell. If you use very tiny pieces of food, they will be just as happy to work for them as if you offer a big chunk. Treats can be broken into pieces smaller than a pea, and maybe 1/4 of a pea for tiny dogs. When I use string cheese, I smear just a tiny bit on my finger and offer a lick.
If my dog will not work for food, then how can a positive reinforcement trainer help me train my dog?
Positive reinforcement training isn't about food, it's about whatever your dog likes well enough to want to work to earn it. The reason we use food most often is because (1) most dogs love to eat, and (2) food is easy to work with. But there are many dogs who work for a tossed ball or a tug on a rope.
At what point do I stop using the food to train?
When you are comfortable that your dog will perform the behaviour anywhere, under any circumstances, then you do not need to use food to train that behaviour any longer. Remember, if you were using a training method such as collar corrections, this question would read "at what point do I stop correcting my dog," and the answer would be the same. Both methods use motivation. You choose if you want to motivate your dog with something he likes and will work to attain, or something he dislikes and will work to avoid.
My dog will do sit and down in my house, but when I take him outside to the park I have to bribe him again. Why does this happen?
Again, you don't take a beginner's math class and solve algebraic problems. You don't take skiing lessons on bunny slopes and then tackle the Swiss Alps. Learning comes in steps. Your house is a Nursery School environment for your dog. It's the easiest learning environment, with the least amount of distractions.
Out in the back yard may be a Kindergarten level environment, because now there are butterflies to chase and the sounds of neighbours. In the front yard, you may be moving into Middle School, because now your dog can see people, other dogs, and moving cars. The park is going to be Graduate School, one of the most difficult environments in which to learn because there is so much going on that is competing for his attention.
One of the big differences between dogs and humans and the way we learn is the ability to generalize. If your father teaches you how to change a tire in your garage, you can probably figure out how to change a tire in any other environment as well. But if a dog could be taught to change a tire, he would be more likely to get confused if the car was anywhere but in the garage where he learned the skill. We have to purposely work at helping the dog learn to generalize each new behaviour. So when you teach him to sit, you need to teach him to sit in the house, in the yard, at the park, when a plane is traveling overhead, etc. Each new set of circumstances is a new level of difficulty for your dog.
What is positive reinforcement? Does it just mean that I give my dog treats?
Positive reinforcement is one of the four "quadrants" of behaviour modification. In the simplest of terms, positive means to add something and reinforcement means to encourage or strengthen a behaviour. So when training your dog you want to build new and appropriate behaviours. The most efficient and effective way to do that is to provide something pleasant to your dog when he exhibits a behaviour you like. This reinforcement can be anything your dog wants access to, a walk, its food, a toy or your attention.