What and When Questions Are Smart Questions When It Comes to Training

A few weeks back I wrote about the “why” questions: People love “why” questions. Perhaps because we are curious, because we think “why” questions help us understand our own dog better (which sometimes they do, of course!) but most of the time it's because we'd much rather our dog not do something and we think that by “knowing” the why he/she will not continue to do so.

Why questions, as I stated, are not very useful when it comes to animal behavior and training. The reason behind this- as a form of recap- is that why questions, while being interesting, seldom help us move in the right direction with our dog’s behavior and training.

So then, what kinds of questions should one ask in order to make changes in our own dog’s behavior?

“What” questions are really great starting points. As in “What is your dog doing that concerns you? Irritates you? Etc."

Answers to what questions are observable, and as such, specific.  For example: What is your dog’s response when a stranger reaches out to pet its head? Answer: Lowers the head, emits a low growl and moves away.

If someone were to give me the response above to my question of what is your dog doing… I would most definitively get a clear picture of the interaction between dog and stranger for this very specific action- the stranger reaching towards the dog’s head with the hand.

I am also getting a clear concrete “picture” of how the dog is responding to it. I might not know what the dog is thinking because nobody knows, but I can infer by the dog’s body language, distance increasing and growl that he does not want to be reached out by a stranger.

“When” questions are also very helpful because they tell us similarly to “what” questions under which specific circumstances the behavior is taking place. As I have stated before, behavior never happens in a vacuum so knowing the precise conditions will help in coming up with a plan of action and will very likely set up the scenario so that the dog does not continue to engage in the dangerous, unwanted behavior.

Here is an example: When would you like your dog to walk on a loose leash? The answer to this question might vary. When we are taking a walk in town, when going to and from the car, when your dog see other dogs also on leash. When is it okay for your dog to pull or walk on a taught leash? When we are practicing scent discrimination on leash or when we are practicing skijoring. An activity in which a dog is attached by a long line and is pulling you on a bike, sled, etc.

Finally, "what" questions are also critical in reaching and understanding of what we want the dog to do instead of what the dog is doing at present.

If we do not have a clear picture of what we want there is really no way that we can reach our goal of communicating clearly to the dog in what we want him to do instead.

Good Timing

I am in the kitchen waiting for Deuce and Rio to finish their breakfast; this morning it was served in their bowls. Deuce appears at the kitchen while Rio bee lines to Deuce’s dish just in case he has left something behind. Once she realizes there is nothing left she comes into the kitchen. I give Rio her daily cartilage support supplement, which they eat as if it was a treat. She is now gone and looking out the window.

I call Deuce to give him his and as I am doing this I realize what’s going to happen next… Rio is now standing next to me and giving me a look as if saying: what did he get that I did not? Ah, of course! She has “forgotten” that she got the same thing just a few minutes past. I grab a piece of kibble that is on the kitchen counter and I give it to her - her consolation prize.

Timing is everything in animal training. Animals do not have the capacity of separating in time a behavior emitted with a reinforcer gained.

Typical case scenario: The dog that relieves himself inside because he has not learned where to go or because he has not been giving the opportunity to do so and cannot hold it any longer. The owner coming home finds the “accident” and admonishes the dog in a not so friendly tone of voice or perhaps something worse.

The person, of course, thinks that the dog realizes that he is being yelled at or punished physically because he eliminated inside. However, the sad truth is that the dog has no idea whatsoever why the owner is acting so threatening.

From the perspective of the dog, that owner is acting threatening “out of nowhere” or because the dog was laying somewhere, came to say “hi”, etc. Because that is what the dog was doing the minute the owner became ballistic.

Not only has the dog not learned where he should eliminate, but now is potentially afraid of the owner. So it is not that the dog is feeling “guilty” it “appears” guilty because it is “appeasing”. And it is demonstrating “appeasing” behaviors because it feels threatened/afraid and is trying to communicate to the person that he means no "harm” or threat.

The same is true with wonderful rewards for our dogs. Consequences,  good and bad, but I am hoping good, must be delivered immediately so that the dog has a chance of linking the antecedent - the behavior to the reinforcer.

Here is another interesting example: Someone is trying to teach the dog to sit instead of jumping on people.

They ask the dog to sit but they are slow in delivering the reinforcer (treat) so now the dog, which has very little impulse control, begins to jump again and the treat is delivered at this particular time. Ask yourself: which behavior was truly reinforced?

The sitting or the jumping? Indeed the jumping! ( &%*$*@). Or perhaps both. So here we have what is called a behavior chain.

This particular one we don't want. What we want is a dog that sits and not jump/sits. There is so little wiggle room when it comes to poor timing in communicating accurately with our dogs.
For those of us that are clicker trainers/owners the timing again can make our intentions crystal clear to the dog, which is what we want or we can again create confusion for the dog and reinforce a behavior that we don’t necessarily want.

Using a clicker is no saving grace! By using the clicker as an “event-marker” which means to the dog: for this behavior (the one clicked) you are getting a reinforcer we can not only speedy the learning process but have a dog that is really interested in working with us. But this is only true if the timing of the behavior and the click – followed by the reinforcer is done with exquisite timing.
For this reason, I teach my clients that when using a clicker they must really nail the timing of the click. And if needed, it is okay to delay for a few seconds the reinforcer.

It does not take the dog too long to realize that after a “click” comes the reinforcer, and as such they have learned that the movement from our hand to the treat bag means delivery will begin. This is important: Delivery will begin… The anticipation of the reinforcer will – if it is not too prolonged, add value to the reward.

The take - home message then, is to work hard in delivering the “click” with extreme good timing instead of worrying about clicking and delivering soooo fast that we find ourselves fumbling with the treat.

The same applies to “real life”. If we strive to improve the timing of the delivery of the reinforcer we will have not only a more responsive dog but the behavior we want to see more of will become the norm.