Reward-based Dog Training

I think everyone agrees that reward-based dog training is a good thing. Rewards in dog training are the pay that the dog gets for performing the behaviors we want from them. High value rewards are invaluable in training. So, can reward-based dog training go wrong?


When rewards are a good thing

When we begin to train a dog, we definitely want to use rewards. The food is the payoff for the dog. He does something that, frankly, makes little sense to him so he gets that tasty treat! When he does it really fast or particularly well, he gets several treats! Wow! What a great thing it is for him to do what we train him to do!

At the start of training it’s crucial to use rewards and use really good ones! We want to motivate the dog and keep him engaged and happy to learn the behaviors. Reward-based dog training is what I’ve done for more than 40 years! I would never suggest not using high value food rewards at the beginning of training. But there comes a time that the rewards become more widely spaced and eventually almost faded away. Because we don’t want a dog that refuses to complete the commands unless he gets food.


When rewards start to go wrong

Once you’ve begun your training you need to carefully gauge when you should start to make the treats random. In other words, you need to give the dog his food rewards on a random basis. This time a treat, next time no treat, then perhaps several (for a job especially well done), then back to none, then one again. You get the idea.

The reason for this random reward system is to leave the dog guessing. It’s not that he never gets a treat. It’s that he never knows when he’ll get one or more. It keeps him from tuning out when he doesn’t get a food reward. He just never knows when the food will fall from your hands like manna. It keeps him engaged and working hard to unlock that treat bag!

Fading the rewards

Before you begin to fade the food rewards you should be sure the dog is fluent with the command. That means he truly understands that “Sit” means “put your little butt on the floor.” If he isn’t fluent and you stop rewarding him or even use random rewards, he will have no motivation to continue learning.

When the dog is fluent it’s time to fade rewards. But how do you know when the dog really knows the command? Jean Donaldson, a great trainer, uses what she calls the “Push, Stick, Drop” method. In a nutshell, the method is this:

Push, Stick, and Drop

  • If the dog performs the command correctly 4 out of 5 times in a set, PUSH to the next level of difficulty.
  • If the dog only gets 3 out of 5 correct, STICK to the level you are on and repeat the training.
  • If the dog gets 2 or fewer correct in a set, DROP to the previous level.

So, reward the dog more often when it is still learning the behavior (DROP) and use fewer food rewards as the dog becomes more and more fluent. You will always reward the dog randomly throughout its life because you always want the dog to work hard for that occasional reward.

Using Donaldson’s method helps you to be consistent in your progress and to advance the training at an appropriate speed. You do need to challenge the dog to perform without the reward (other than praise) so that you don’t make his performance based solely on the food reward.

Pushing the dog to perform with only random food rewards keeps his level of expectation up. He will work harder to get that reward! Be sure, though, not to stretch out the rewards too far, too fast. This will only build frustration in your dog and, consequently, in you. You want him engaged, happy, and willing to play the game of training! Reward-based dog training never completely ends.

What if the dog loses interest?

If your dog seems to have lost interest in training you must find the reason why. There are several things that can cause a dog to choose to not stay engaged.

  • Did you try to fade the rewards too quickly? If you went from rewarding each time to not even having rewards with you, the dog is going to lose interest!
  • Is the dog confused about what you want? If he’s not fluent in the command he can’t perform and if he can’t perform, he doesn’t get the reward and loses interest.
  • Did you reward him every time for too long? If he’s become dependent on the reward to perform the behavior you need to go back and then start fading the rewards slowly.
  • Have you become predictable? I always say dogs can’t measure but they can count. If you reward every third time the dog performs the behavior, he will learn that and a 4th request will likely be ignored. He’s got his pay. He’s done. Randomize the rewards.
  • Rewards need to be delicious and interesting! Give treats from both hands instead of always rewarding from the same hand. If you use a ball as a reward try doing something different before you throw it. Turn your back and throw it over your head. Your dog will be delighted at the changes. By keeping the dog guessing you’re making training more fun and this builds motivation, anticipation, and engagement!
  • Know when to quit. A tired dog cannot learn. It’s crucial that your dog end every training session with a win. Be sure to stop the session before he’s too tired to succeed.
  • Reward-based dog training and corrections

    Along with food rewards it is necessary to add corrections as your training progresses. A dog that is allowed to ignore commands it is fluent in is an untrained dog. Just as with a child, you need to set the rules and consequences for misbehavior.

    Instead of going into a long explanation of what corrections you should use and how to use them, I suggest you read ”What to do with a Bad Dog. The article will instruct you on how to handle a dog that is ignoring commands he knows in a far, humane way.


    Working together for success!

    By using a reward-based dog training method and incorporation corrections when it’s time you will have the happy, well-trained dog you dreamed about. And if you find that it’s too much to handle on your own, remember Joyful Dogs of Michigan is always here to assist!