“But he only does it when I have treats!”
“I have to pretend to have a treat or she won’t listen!”
Does this sound like you? Did you find that your dog learned the basics very fast when you used treats as a reward…but then seemed to forget everything when asked to perform without them?
Some may tell you that this is an inherent problem with using rewards in training. However, the issue is more about how food is used than about using rewards in general. Follow these tips and you can get the benefits of reward based training without sacrificing reliability.
- Dogs are very perceptive to body cues, intentional or otherwise. When you are working with your dog, do you find yourself reaching for a treat before each cue? Or using a treat to lure your dog into position beyond the first several sessions? When treats are used in this manner, they become part of the cue to perform. The same thing often happens with people bending of kneeling before calling their dog then when they try standing straight, their dog no longer responds to the command. Same with bending over to get the dog to lie down, try standing up and the dog no longer responds. So don’t be predictable about the treats. Sometimes have one in your hand, sometimes leave it in your pocket, on a counter or in a hidden spot to surprise the dog. And avoid using a food lure to get your dog to perform beyond the first several sessions.
- Food isn’t the only way to reward a dog – mix it up! Use food, toys, play and “life rewards” so your dog doesn’t become too dependent on food being the only reward you provide. When I worked at a doggy daycare and Jagged was young, I taught him an extremely solid recall and stay using a release to play as the reward. In every day, life my young dogs practice sit-stays every time they are released from a crate or go in or out doors or vehicles. No treats involved, the reward is being released to do something they want (if they don’t hold their position, the door closes). I use chase-me games as a reward for my dogs pretty often too. Anything your dog wants is a potential reward and once you start using that in daily life, your dogs reliability will skyrocket.
- Be unpredictable. Always giving your dog a treat every time he sits may initially seem to produce reliability quickly…until you ask him to perform without a treat a few times. Consistently rewarding every correct response doesn’t teach your dog to work through stress and to keep trying. If your dog quits when things get even a bit stressful or because he doesn’t see an immediate possibility for reward, it will be very difficult to build reliability. After you have gotten the behavior you want on cue, start being unpredictable with the rewards. Ask for a sit, play with your dog, ask for another sit…then run away (if your dog enjoys playing and chasing). Or sometimes reward after the first sit, sometimes after the second, third or fourth. Remember to also use life rewards – if your dog wants something (say to be let outside) he must first sit and wait to be released to it. When rewards can come from anywhere and be in any form, your dog will work for the chance of a reward instead of the promise of one.
- Can your dog sit in a box? By a fox? In a boat? Near a moat? In the car? Will he sit near and far? Getting behavior in dog training is pretty easy, at least when it comes to basics like sit, down and stay. Most people can teach their dog to do these behaviors for a treat in just a few quick sessions in their living room or kitchen. So why are there countless dog training books, classes and professional trainers? Because getting the behavior is just the first step in developing reliability. And most people would agree that reliability is what makes a dog “trained. Having your dog sit for a treat in the kitchen is fairly useless if you can’t get him to perform the behavior when you really need him too. So you need to teach your dog to perform the behavior in a variety of settings, for a variety of rewards (treats, toys, play, affection, praise, release to something he wants, etc). Sometimes he may need to perform the behavior a few times before earning a reward. And sometimes he’ll get a reward on the first try. When you first introduce new scenarios, make it easier for the dog to be successful. High value treats and easy behaviors. Once he’s had a bit of practice, he’ll start to get better and better at generalization and you will see reliability developing faster and faster in new situations.
I’ve used sit as an example here but the tips work with any behavior you want to become reliable. Set your dog up to succeed, watch your body cues, use a variety of rewards instead of just food, be unpredictable and take a hint from Dr. Suess – teach your dog to do it in a box, near a fox, over there or anywhere!
Some places Savvy can sit…