Dog Spelled Forward :: Dog training and behavior consulting services for Bergen County, New Jersey

Dog training is about communication.

As I said earlier, we don’t know what our dogs are thinking, or even how they think. Chances are though, it’s not in the sentences and paragraphs that we tend to use.

But we can communicate with them. If you have any doubts, check out the agility videos in my videos section and explain to me how that trainer achieved that level of performance without an almost constant stream of communication between himself and his dog.

The most basic form of communication in training is the positive reward marker. It is a signal (usually audible) indicating that the behavior being performed at that very moment is correct or desirable. Some trainers use a clicker (a small device that makes a clicking noise) while others use a word such as “Yes”. With marine animals some trainers use a whistle. The only real requirement is a signal that is distinct.

The marker is “loaded” or “charged” by creating an association between the marker and a reinforcer, usually food. This is done by giving the signal and following it with the reward. Some trainers will do this in conjunction with a basic behavior such as making eye contact, other will just pair the two stimuli until the relationship is apparent to the dog.

One important aspect of the positive reward marker is that the reinforcer does not have to immediately follow the signal - there can be a delay of a few seconds.

A clear and unambiguous means of communication has now been established: when this signal is sent a reward will follow. The implication being, “I like it when you do that.” Moreover, we can give this signal at the moment of the desired behavior and then reward our dogs a few seconds later: we not only don’t have to carry the food in hand, we can hide it, paving the way for removing constant rewards and also replacing food with other reinforcers such as play.

Being able to send this clear signal whenever we see what we like (and are prepared to reward it) is very powerful. We can communicate behaviors we want to see more of, even when we didn’t ask for them. For example, every time a jumpy puppy’s rear end touches the floor we can mark that moment. Over time, jumping may decrease. Every time a distracted dog looks at his human, the moment can be marked. Where will the distracted dog’s attention tend to wander to more often?

Training behaviors also becomes much easier. We can now communicate exactly what we are looking for, as opposed to communicating what we do not want, which may mean repeated punishments and maybe even training by process of elimination.

Here’s a chimpanzee learning to ride a Segway. Note how he corrects himself after a couple of crashes.

Dog training is about communication.

I want my dog to sit when I ask:

1) I lure her into a sit and reward her her for it.
2) I start to say “sit” and then lure her into it and reward her.
3) Lather, rinse, repeat.
4) After a few more trials she sits when I say sit.

I didn’t teach her how to sit. I taught her what the “sit” sound means.

In order to be successful, training needs to involve two way communication. While dogs don’t speak in sentences and it is literally impossible for us to get “inside their heads” they are constantly providing information to to us via “body language” and their behaviors.

There is one form of communication that we primates use that dogs can use in almost the same way: eye contact. Eye contact means “you have my attention.” When we call our dog’s name, what we are looking for is eye contact, very much the same way we expect it when we call a person’s name.

When we train behaviors such as wait, stay and heel, we “shape” our dogs into giving us eye contact. A good trainer never releases a dog from a wait or a stay without eye contact and most trainers are probably more concerned with where their dog is looking during a “heel” than anything else.

Last week I was out walking one of my dogs early in the morning. I was not feeling well and was really not very connected to Gage at the time. We walked across a long municipal parking lot that is always empty at that time of day to his favorite spot, just outside the entrance. This is a very quiet area and Gage is probably my best loose leash walker, so I can get away with not paying attention.

We got to the far end of the lot and went through the smaller part of the entrance gate for people. The larger part (for vehicles) was closed and locked. After we went through, we stopped. I stood there for a moment, probably still under the influence of Nyquil and definitely inadequately caffeinated. I looked down at Gage. His body was pointed across the entrance of the lot, directly at his favorite spot. He turned and looked me dead in the eye, as if to say “Can we go there? I really need to go.” We did, and he did.

We brought Gage into our home about 3 years ago. He was already 6 - 8 months old and terrified of everything, especially people. Getting any sort of eye contact from him was very difficult. He now uses it to tell me where to walk him.

Two-way communication indeed.

Dale Carnegie said “Remember that a man’s name is to him the sweetest and most important sound in any language.” But I think he missed the forest for the trees: what we really love is the sound of our own voices. We love to talk, and if we don’t get some sort of acknowledgment, we tend to repeat ourselves.

Meanwhile dogs don’t send or receive very much information vocally. Sure, there are plenty of dogs - entire breeds actually - that seem to love the sound of their own barking, but this is not a dog’s primary way of transmitting or and this is important, receiving information.

I’ll repeat that.

Vocal communication is not a dog’s primary way of transmitting or and this is important, receiving information.

When we train a dog to do something “on command”, such as sit when we say “sit” or lie down when say “down,” we are not teaching them what those words mean. We have assigned a sound (a cue) to an action. We could just as easily get them to sit when we ring a bell or lie down when we snap our fingers.

When we repeat our commands to a dog, one of two things happen. They either learn to not respond the first time: “sit” becomes “sit. sit. sit.” Or worse, they learn to ignore it completely. Think of what happens when the radio stations discover (or decide) that a song is popular. After the 1,000,000,000,000th time it comes on you don’t hear it (or are tempted to jump out of the car, but I digress.)

This effect is what trainers are trying to avoid when they tell you to not repeat your commands. It’s really tempting to repeat a cue when a dog doesn’t respond right away and even the “pros” do it at times, but there’s good reason to try to say things once and wait it out.

Some people have a real problem with the idea of using treats to train a dog. They think that it’s “bribery” or a “crutch.”

When food is used properly in training it is being used as a reinforcer. This type of training uses operant conditioning to teach (or perhaps un-teach) a dog behaviors. (Think of the word operant in terms of operating on the environment and being conditioned by the result.)

Food is used most often as a positive reinforcer: something that is presented when a dog performs a behavior that we want to see more frequently. The addition (that’s where positive comes from) of the food reinforces the behavior.

The science behind this influences everything from which park benches pigeons return to in city parks to whether or not you are going to order takeout from the same restaurant you did last Friday night. It’s not up for debate. It’s a natural phenomena.

The reinforcer doesn’t have to be food. It just has to be, well…reinforcing. Some dogs will “work” for a game of fetch while others will watch you throw the ball and just stand there. Some dogs will work for a game of tug, others a scratch behind the ears, and still others an opportunity to chew on an old shoe.

Trainers use food for a very simple reasons: most dogs will work for food. Some may require really interesting food, and others may even need to be fed a meal during training (I.E. not fed before class), but when it comes to reinforcers food is the lowest common denominator and works very well in a room with 6 or 7 other dogs.

A good trainer will introduce the idea of using other reinforcers early in your training and get the treat out of your hands very quickly, but there’s nothing wrong with using food in a classroom to introduce new behaviors. As a matter or fact, there’s some pretty basic science behind it.

Like many trainers, I don’t like to see retractable leashes in my classes and ask that handlers bring six foot long nylon, cotton or leather leashes. Many students are surprised when this comes up. Why don’t I like them?

First, it’s worth explaining what I think a leash is used for: keeping your dog from walking (or running) away from you. That’s it. Nothing more, nothing less. (It’s not a means of communication, but that’s a different discussion.)

So right off the bat, the retractable leash sends a mixed signal - sometimes your dog can walk away from you, and sometimes she can’t. How does she know? She pulls and sees if she can. If she can’t, she tries again. And again. And again. Already you can see how training a dog to walk nicely on a leash when pulling on the leash is how she figures out if it’s OK to walk away from you is nearly impossible.

As far as classes go, retractable leashes are almost impossible to use effectively. During a class you need to be able to keep your dog near (which is difficult enough if it is her first class,) have a hand free for treats, perhaps have another hand free for a clicker an oh yeah, learn something. The retractable leash makes this twice as difficult with its rather large handle and its tendency to slip into “unlocked mode.” I frequently recommend that students step on their leash while working on an exercise in order to free up a hand, but this doesn’t work with retractable leashes either because the fine nylons leads will slip under most shoes.

Last, and by no means not the least, is safety. After effectively training your dog to pull on leash, you are now betting her safety on two things: that you will have the leash “locked” when it needs to be and that the locking mechanism will hold at the critical moment when your dog lunges after a tennis ball on the street, a squirrel or another dog that doesn’t want to be lunged at.

If you want to give your dog some freedom, buy a long line and a “snap-back” and find a safe place to play. With some practice and some training, a long line can be a lot more fun anyway!

The AKC Agility Invitational is on Animal Planet Sunday. Here’s a link to the schedule, including all of the reruns.

Don’t know what agility is? To oversimplify, it’s essentially an obstacle course for dogs.

Take a look at this video (with a slightly odd ending):

and here’s the same team, on a different run from a different angle:

I’ve been playing around with Squidoo for the past week, and have made a few pages.

Playing tug - which is similar, but slightly different from this tug page.

Body Language - which is also similar to a page here.

and a new page about training with treats which will have a companion page here next week.

Check ‘em out and enjoy!

I don’t like dog parks. Most of them are poorly run, if “run” isn’t too generous a term. Today I came across a great example of what can go bad in a dog park.

Go ahead, read while I wait.

Done? Good. So let’s recap: A greyhound, which is bred to chase, capture and kill small game, attacked and severely injured a small fluffy animal. The small fluffy animal was a dog, which the owner (and apparently the reporter) think should make a difference.

It doesn’t. (That’s a PDF.)

Even without being present, I feel confident saying that the greyhound didn’t “turn mean.” He turned greyhound. Neither of these dogs should have been allowed to get close to each other at all off leash, and on leash they shouldn’t have been allowed to get close enough to each other to bite.

Obviously, the owner of the greyhound was irresponsible.

But, quoting the article:

“She acknowledged that Lafayette’s dog park has a separate enclosure for smaller dogs, but she said Shiloh was attacked on the way out of the park and that small dogs shouldn’t have to be hustled under lock and key out of the dog park for fear of being mauled.”

She wasn’t exactly acting responsibly either. A cynical person might say that she seems to think it’s everyone else’s job to be responsible so that her dog can run free without her worrying about it.

Either her dog was in the same area as the larger dogs, or if you read the comments where she sort of changes her story a little, her little fluffy dog was traveling through the large dog area. In coursing that’s called “the lure.”

My objective isn’t to trash this story, or diminish the seriousness of what happened to this poor dog. It’s to tell you this: stay away from dog parks, or at a minimum know exactly where you are taking your dog, who is going to be there, what to expect, and how to keep your dog safe and under control.

Never assume the other person is being more responsible than you. If you do, then you’re right.

This is pretty entertaining, and good example of a creative way to exercise your dog. Simon seems to be a Jack Russell Terrier, and probably has a lot energy and a desire to chase and bite things. If he doesn’t eat the balloons, letting him have some fun by giving him something he is permitted to chase and bite is not a bad idea.

I love when the guy calls off the time part way through…I swear it looks like Simon speeds up.