Teaching Self-Control

What dogs need self-control training? All dogs, but especially those that pull on leash, won t sit still for more than a minute, are hard to control, and any dog who goes nuts to be petted by people.

Dogs don t have self-control naturally, because they don t need it in the wild. When they are hungry, they look for and chase food. When they are tired, they sleep. When they feel rowdy, they play.

Most people don't like living with dogs that act wildly. So it is up to you to teach your dog, that exercising self-control is a good thing.

Keep in mind that dogs don't care what you want; they are only interested in what brings good things to them. So we make it our mission to teach them that exercising self-control is to their benefit. This is the key! It is a known law of learning and if you can use it, your dog s learning will progress much faster.

Another law of learning is that dogs abandon behaviors that don't benefit them in any way. So we also need to be sure the dogs aren't getting any benefit from unwanted behaviors. If the dog is getting any type of reward or benefit from it s behavior though, it will not stop, even if there is also punishment involved.

So, lets look at one behavior that dogs do, that makes us feel they are out of control:

Teaching a well behaved greeting

Why do dogs jump up on people they are meeting? Because the greeting rituals for dogs that are greeting pack members who have been away, includes: buzzing around them excitedly, licking the returning dog s face and sniffing. Licking at the dog s face is a submissive gesture and it s exaggerated in dogs that live with humans because the social group is continually being fractured and then re-united.

Dogs jump on people because we are vertical and they are trying to get to our face. We also tend to let puppies get away with it, then change the rules as the dog commits the crime of growing up. Also, no one has made it beneficial for them to do otherwise.

Punishments like kneeing, pinching toes or hanging with a collar are ineffective, laden with side effects and inhumane. Imagine yourself being punched in the stomach for smiling and extending your hand in friendship. It s not the dog s fault that their greeting ritual is at odds with our preferences.

The key to training a dog not to jump up is to strongly reward an alternate behavior that doesn t allow the dog to jump. It s called counter-conditioning. The dog can t jump and sit at the same time. He also can t chew on your carpets while working on getting the contents out of a Kong toy, or chase squirrels while maintaining eye contact with you.

The behavior you choose as your counter-conditioning behavior must be one that physically cannot occur at the same time. For instance, don t teach sit to counter condition barking. These behaviors can occur simultaneously (and will). Reward silence instead and be sure the reward you use is better than the benefit the dog gets by barking or if possible, be sure barking isn t being rewarded in any way.

You have to practice the new behavior till it's at a very reliable standard, before you can use it in a real-life context. In the case of jumping up, the dog should be rewarded for a sit and stay with distractions, like people nearby, before using this to counter condition such a strong, hard-wired impulse like jumping up for a greeting. Especially practice the sit-stays at the front door and on the street (or anywhere the dog is likely to meet people).

Secondly, the dog needs to understand a no reward mark (NRM). Which is just like it sounds. It is a sound that MARKS when the dog gets NO REWARD for a behavior. This signal is just one more bit of information and feedback you can give the dog about his behavior during training. Praise tells him he s on the right track, a click tells him he s correct and a no reward mark tells him that the behavior he just offered is not going to be rewarded. Common NRM s are Too Bad, Try Again, Oops and AHH. It is best to keep this feed back neutral, so Ahh is probably the worst choice, because we tend to be really negative with it. The NRM is not meant to be a punishment, just feedback. The dog made an incorrect choice, and you are telling him. I use ahh, but that s a kickback to my days as a traditional trainer. I also try to use Nope, try again as much as possible, to get myself away from using ahh.

Before you start training, be sure the behavior is one you don t want at all. It is much easier for the dog to understand that no one gets to be jumped up on, rather than trying to teach him, you and grandma don t like it, but dad and Billy don t mind, as long as dad s not wearing a suit.

Start by setting the dog up. Open your arms wide and invite the dog to jump on you. Really ask for it. As soon as his front feet leave the floor, reject him totally with a no reward mark and turn away. This tells him he guessed wrong about what you wanted and he has zero chance of being rewarded. Your timing is critical. Giving the signal as his feet start to leave the ground is worlds away from giving it after he has planted his feet on your chest. Mark the behavior as soon as you see it begin. With some dogs, you can tell when they are thinking about jumping up, that s when you mark it. It is helpful to tether the dog or have someone hold the leash so you can get out of the dog s range until he sits.

After you mark a couple of jump attempts, he will probably try something else when you invite him to jump up. A lot of dogs will simply sit. If he doesn t, you may suggest it. As soon as his butt hits the floor, click, praise enthusiastically, and give him a treat from your pocket. If he jumps up when you click or praise, don t give the treat until his feet are all on the floor. Be sure to keep the treat low so you don't encourage him to jump for it.

Practice until you can t get the dog to jump on you no matter what you do. At this point, the jump up invitation is becoming an alternative sit cue. Not only does it drive the point home, but it prepares the dog for those unavoidable idiots who will sabotage your training by asking for your dog to jump on them. If one person in 10 encourages and praises the dog for jumping up, the behavior will continue. You should not need to do the thinking for the dog by continually reminding him to sit with the verbal sit command. The training will have a much greater impact if the dog figures out what makes you click on his own.

Dogs can learn rather quickly in the first session not to jump on you. What they don t learn, is to not jump on you tomorrow either, or when you are upstairs, or when it s another person. This is because dogs don t generalize well. It s normal for dogs, plan on it and structure your training to help the dog overcome it.

Help the dog generalize, by repeating the training session in many different places, then try it with many different people, until the dog understands it with different people. However, be sure each person understands what to do  meaning they only give the dog attention if he s sitting;

if he jumps, they walk away. You can give the no reward mark, so that you are sure it is given as early as possible. You are also the one who clicks and gives treats to the dog. The helper gives praise and attention to the dog while the dog is sitting.

Once the dog understands the behavior, meaning he s doing it reliably in many different locations with different people, you are ready to try it at the most difficult level (at home at the front door). It helps to have some understanding friends at this point. Be sure they all know the drill before you start. You can also practice this each time you come home or first greet the dog after being away.

The first person rings the doorbell and enters. If the dog jumps up, they give a no reward mark and walk out. It is likely that this will happen, because of the dog's failure to generalize, but the former training will come back quickly. If the dog remains sitting, you click and treat and the dog gets some petting from the person. Do this repeatedly for 10 entries (even if the dog is getting it right after two or three entries). This may be because the newness of this person is wearing off, but the repeated entries are chances to practice correct behavior and to build a reward history for it. After the first person has done 10 entries, go on the next person and do the same thing for 10 more entries.

Now, repeat this exercise the next day, or within the week. It will be like the dog forgot all his training when the first person walks in, but it will come back faster and faster each time you practice it. Until the dog realizes that jumping up gets him nothing, EVER. And sitting gets him treats and praise and attention. The dog also has to learn that these rules hold true when you are out for a walk, in the pet store, and at the front door and that it doesn't matter what human it is. If you have multiple dogs, you will need to teach the behaviors to each one separately until they fully understand them.

This may seem like a lot of work and the main problem with this type of training is that people don't persevere long enough for the dog to make the blanket generalizations. Then they end up saying, Oh I tried that, and it didn't work . This technique works every time, if you don't quit too soon. Then you have a dog that is good for a lifetime! With only the occasional refresher or treat. You can do dedicated training sessions that are repeated often or you can work on it as the situations arise but the more practice sessions the dog gets, the faster he will learn.

It is a great feeling of accomplishment to do the work yourself, to get through the this isn't working feelings and to see the success because you persevered. You will also become a better trainer and your dog will be smarter and appreciated by more people. So ask yourself, is a month or two of repeated training sessions worth it to have a dog that won't jump on anyone the rest of his life?

www.DogScouts.com
Copyright Dog Scouts of America
5068 Nestel Road St. Helen, MI 48656 (989) 389-2000

 Top of Page