L3 Homework: Why Won’t My Dog Do What I Ask?

I promised I’d eventually catch you up on where Barnum and I are in on training. It won’t be chronological, so consider this part of an intimate peek into the brain of someone with fractured thinking. (You’re welcome!)

Sometime soon, I hope to post:

• All the socialization experiences Barnum had when he was ages 9 to 16 weeks;
• The videos and/or descriptions of the Level Two tests we passed;
• The videos and/or descriptions of the Level Three behaviors he has already passed;
• Various other training tidbits and milestones.

Today, let’s just pretend you already are caught up on the fact that we passed L2 several months ago, and are now itching with a deep and abiding curiosity to know my homework for L3.

The homework for Level Three of Sue Ailsby’s Training Levels is this:

Handler lists, in writing, ten reasons why a dog might not perform a required behaviour.

This question asks, in essence, “Why might a dog not perform a behavior that the handler/trainer believes the dog knows?”

Reasons do not include, “He’s stubborn,” or “She’s stupid.”

Here is my list, which I actually wrote a few months ago, when we were first embarking on L3.

1. He might be distracted.

This is a big one for us. I can ask Barnum to sit, and as he starts lowering his butt to the floor, any of the following can literally stop him in his tracks: a gust of wind (which carries scents to be sniffed or blows leaves to chase), a person speaking (even whispering or just saying one word), something moves (a bird flies past the window, someone else walks by), he hears/sees/smells another dog, etcetera. If he can see, hear, smell, or otherwise sense it, it can be distracting.

2. He might hear the cue wrong (have trouble distinguishing the sounds of two cues).

For example, I discovered that “Watch” (my former cue for eye contact) and “Platz” (my cue for down), sound very similar to Barnum, and he sometimes got them confused. For this reason, I changed the cue for eye contact to “Watch me!” and made the inflection very different, both of which seem to help.

Similarly, he might see the cue wrong — have trouble differentiating between two hand signals, which has also happened to me, and I’ve had to alter one a bit to make it clearer. This has happened with us with the cues for sit and stand, and with the cues for nose-targeting (“Touch!) and down.

3. He might be following a pattern that you have unknowingly taught, so that he is not following the cue you think you’re giving, but following the pattern he anticipates.

I remember reading in Karen Pryor’s Don’t Shoot the Dog! about how she had to inform another dolphin trainer that her dolphins were not jumping based on the hand signal she was using as a cue, but were instead jumping at even intervals that the trainer had unknowingly trained.

For this reason, I try to be aware of patterns of timing, or behavior chains, but sometimes I unintentionally train a pattern or chain anyway. For example, at some point, I must have done a pattern of sit, down, stand, down, with Barnum a few times, even though I know I didn’t do it consistently in any given session or in every session.

Nevertheless, I’ve done it often enough that Barnum, after being clicked and treated for a sit and a stand, would automatically offer down, no matter what I asked for. I’ve been working on scrupulously shaking it up.

4. He might be following environmental cues you aren’t aware of.

This “Oops” moment was featured in my video of Barnum passing L2 stand-stay.

At the beginning of this taped training session, I tried to work on eye contact, and Barnum — who had never done this before — barked each time I said “Watch.” I realized that, yes, “bark” and “watch” sound a little similar, but I don’t think that was all that caused the confusion.

First of all, at that point, Barnum was pretty good at recognizing the hand signal for “Bark!” but not the spoken cue; and so far, I have not taught him any hand signal for “Watch me.” For another, we had mostly worked on “Bark!” and “Quiet,” at night, and in the kitchen/living room. On the other hand, we primarily worked on eye contact in my bedroom, usually during daylight. It seemed to me that he was offering barking because that’s what the environment suggested I would ask for.

5. The dog might respond to unintentional cues given by the trainer/handler.

Speaking of hand cues, Barnum is a master at interpreting body language. Very often, he cues off of the position or movement of my eyes, hands, arms, head, and probably other things I’m not aware of, when I give a command.

Then, if my head, eyes, hands, or arms do that (whatever that is) while I’m asking for a different behavior, he performs the behavior that he’s come to associate with my body movement or position.

For example, without my being aware of it, we apparently did a lot of training of the “down” position when my arms were hanging down or palm-up in my lap. I was more likely to have my hands in fists or resting palm-downward on my lap for “sit.” None of this was conscious on my part.

I became aware of these tendencies when I started to give Barnum verbal cues only for “sit” or “down,” and noticed that he first looked at my hand position before deciding which behavior to offer. What I said was much less meaningful to him than my unconscious hand-shapes.

For instance, if I said “Sit,” and my arm was hanging down or palm-up in my lap, he would down instead. I had to retrain these cues by moving my hands a little bit with each repetition from the positions he associated with the behavior to completely different postures, and then randomizing them.

Even putting your hands behind your back — which many savvy clicker trainers do — does not always work to prevent unintentional cueing. For example, I discovered that I unconsciously tipped my chin up and looked at Barnum’s rump when I cued him to sit.

I learned that I could cue a sit simply by moving my eyes (to look at his hindquarters) or very slightly raising my eyebrows. Below is a video of me cueing Barnum to sit or down without using voice or hand signals. Or watch the captioned version of the video here.

Read the transcript of the video here.

6. You might be using a poisoned cue.

A poisoned cue is one that has come to either have a negative or potentially negative association for the dog, or one that doesn’t “work” anymore because it has become meaningless. Examples of the former include when the handler intentionally does something that the dog finds aversive, such as punishing an incorrect response, or when “shit happens,” such as when you are just introducing your cue, “Up the stairs!” and the dog’s metal dinner bowl comes clanging down the stairs at that exact moment as he is climbing them. (This is not a theoretical example, I’m sorry to say!)

An example of a meaningless cue as a poisoned cue is if you call your dog to come to you, and she ignores you and continues to sniff the exciting smells she’s enjoying, or playing with another dog, or chasing a squirrel, or whatever is more rewarding than coming to you. After you have done this a few times, you’ve taught your dog that, “Fido, come!” has no connection to Fido stopping what he’s doing to run over to you.

It is possible to have a poisoned cue and not realize it.

7. The dog may not like to do the behavior or feel comfortable with it in certain circumstances.

For example, if you ask your dog to down when there is another dog nearby that your dog feels may be a threat, your dog might be apprehensive about taking this submissive posture around the “scary” dog. Some dogs, such as Jersey, really hate lying down on a wet surface, such as a sidewalk that is damp.

8. He might have a brain fart.

Yes, that’s the technical term.

Seriously, I’ve seen this happen with all my dogs at one time or another. For example, occasionally I will give Barnum a cue he pretty much knows, and he clearly knows he’s supposed to do something, but it’s just escaped him. He’ll look at me like, “What? I’m sorry. Can you repeat that?” And if I do, he does the behavior just fine.

9. There might be a physical or medical reason the dog doesn’t want to do the behavior.

A few months ago, Barnum mysteriously began ducking his head when I tried to pet him on the head. In the past, I had worked a lot with him on first tolerating, and then enjoying, being petted on the head.

“What’s gone wrong?” I thought.

I had all sorts of ideas, relating to fear, linking head petting to other things we were training, etc. I kept trying to retrain it and getting nowhere. It turned out that he had an infected cyst over one of his eyes, and it was painful. He didn’t want to be touched anywhere in the vicinity. After one day of antibiotics, he was perfectly happy to have his head petted again.

10. You have (intentionally or unintentionally) raised the criteria for the behavior, and it’s too big a jump for the dog to understand what’s being asked of him.

This is where the famous “Four Ds” come in: duration, distance, difficulty, and distraction.

Friend of the blog and terrific Training Levels trainer, Eileen has given me permission to post her excellent videos on this subject (Eileenanddogs on youtube). Updated note! These videos are all either captioned now or on their way to being captioned. Definitely the first one (“The Missed Cue”) is already captioned, and likely, the others, as well.

She started with “The Missed Cue.” This video shows both her dogs responding correctly to a cue (“go to mat”) many times until suddenly, neither of them understands what to do anymore. A difference of a few inches is the culprit!

Eileen continues the series with, “Missed Cue: Paw Touch.” This video shows how previous reinforcement history can affect the response to a cue the dog otherwise seems to know. (This is not exactly the same as #2 in my list, above, but it is similar.)

In “The Missed Cue: Generalization,” one of Eileen’s dogs does not responding correctly to a cue because she hasn’t generalized to a change in the environment. (A different environmental change than I mentioned in #4, in my list above, but similar, because the object in this case is acting as a sort of cue.)

Onward! Upward! Outward! But first, to sleep.

-Sharon, Gadget (the Great Generalizer), and Barnum (SDiT, still young, but with an expanding mind)