Do you reward too much, too little or…both?
Chances are, you are doing both.
If you are feeding your dog a steady stream of rewards, always the same type and always the same quantity regardless of the difficulty of the behaviour, the stage in training and the context of the environment, then you are doing both. Rewarding known behaviours constantly is too much. Rewarding real achievements with one treat is too little.
Rewarding too much
When we start out training a new behaviour, we reward every repetition. That’s as it should be – it is necessary to build value for a behaviour, especially one that is not inherently rewarding for a dog such as sit/stay or walking politely beside you in a straight line. However, as your training progresses, that reward for every repetition needs to be faded out, and here’s where a lot of people get stuck. Every time Fido sits, he gets a treat. Every time she looks at you, Lady gets a treat. When walking, Fido is fed a constant stream of treats for taking two steps on a loose leash. This is rewarding too much.
Rewarding too little
We are often much too fast to say our dog “knows” a behaviour when in fact the Fido’s understanding is quite minimal. Until the behaviour is cemented in many different environments and has value built into it, the rate of reward needs to be kept relatively high. By fading out rewards too quickly, you may not be rewarding Fido enough at that point in his learning.
The other way we sometimes get the balance wrong is using the same reward for every behaviour, no matter how easy or difficult the behaviour is for the dog. If you feed Lady one low value piece of kibble to reward her for coming to you in your kitchen, and then try using that same reward when you are outside training to call her off a distraction, you are probably not going to have a lot of success recall in the real world. You are rewarding Lady enough in your kitchen, but not enough outside.
Getting the balance right
- The stage in your dog’s training is important. Reward frequently when your dog is learning a new behaviour and then fade out the rewards by asking for more duration before rewarding and not rewarding every repetition.
- Factor in the level of difficulty. Sitting is a natural behaviour, so is quite easy to teach. Walking beside you and not sniffing at all those tempting smells is a lot harder for many dogs because it is not natural for them, so the reward needs to be higher value than the reward for a sit.
- Think about the environment you are working in. When you are teaching your puppy to come to you in your house, you can use a lower value reward such as a piece of kibble. If you are working on the same behaviour outdoors where there are a lot of distractions, you need to go to something much higher value.
- Remember that there are many things your dog may value. We tend to use food as a reward because it is effective and most dogs like it. However, for some dogs, food is not actually their highest value reward. If you have a dog that is toy/play motivated more than food motivated, your reward for a sit might be a piece of food, but training a recall in a distracting environment might be more successful if the reward is a great game of tug.
Rewarding your dog is a balancing act influenced by many factors, but once you think about these factors, it starts to make sense that you should be using different rewards for different things, and that the rate of reward will vary as your training progresses. Make a list of what your dog values – food, play, affection, chasing a ball, sniffing – and think about how your dog ranks these. Have some fun experimenting with incorporating them into your training to achieve success and build a fabulous relationship with your dog.