And one of the main criticisms of using treats in training, is that without them, the dog is not controllable.
So is this true?
Does the use of food in training produce dogs that are only obedient in the presence of food?
Training versus bribery
Training and bribery are two very different concepts. Effective training creates an automatic and reliable response to a command or cue.
Or you blow a ‘sit’ whistle, and the dog ‘sits’.
Bribery on the other hand, creates a response to the confident anticipation of a reward.
But if you are training with treats, how can you possibly avoid bribery taking place?
To understand we need to look closer at what a ‘trained response’ actually is.
The trained response
We can confidently say that we have generated a trained response when we can give a specific signal (like a recall whistle) in a range of different conditions and get a reliable response several times in a row, without offering or giving any reward to the dog.
But, and this is a big but, if we persistently give a signal and then fail to reward the response, over time, that response will die out. This is always the case unless, we punish all and every possible alternative that the dog might dream up.
This phenomenon whereby a behaviour disappears in the absence of any benefit to the dog, is called extinction. And it has been widely observed and studied in mammals for many years.
So whilst we can define a trained response as one that is carried out without the promise of any reward, we do need to bear in mind that in the long run, rewards must be given with sufficient frequency to perpetuate a trained response.
A bribed response
A response generated by the certainty of a reward (because you are holding out food in your hand, or because you always and predictably reward the dog with food for a particular action), is a bribed response. The dog is not being trained, he is simply performing in the expectation of a reward. Be it food, or any other form of reward. If the dog perceives that the reward might not be forthcoming, or if he believes he can get a greater reward elsewhere, your apparent ‘training’ will break down.
Creating a trained response
To create a trained response rather than a bribed response, we need to understand some of the theory behind how dogs learn, and we need to use rewards in a particular way. You cannot do this without information.
You will find this information in good dog training books, and in some of the training articles on this website.
For a detailed explanation of how dogs learn and how you can create a trained response in your dog using food, you might also like to check out How To Win At Dog Training.