The Art of Generalization – Does Your Dog Really know a Sit?

“Brown Bear, Brown Bear, What Do You See?”  So begins the infamous household staple for children by Eric Carle.  My 2 year old son loves this story.  He correctly associates every animal and its specific color on each page, with 100% reliability; white dog, black sheep, blue horse, green frog, and so on.  But does he really know all of these animals and colors?  In real terms, he is fairly successful at identifying cats and dogs, maybe horses; the colors are very unreliable, still.  And here we thought we had a genius on our hands.

So, what does this have to do with training dogs?  Of course, everything!  My son memorized the animals and colors in the book, because we have been reading him this story since he was 3 days old. He learned the specific sequence of animal and color occurrence, as well as the context in which these animals and colors are presented to him.  And dogs are the same way. 

We may not be aware until we get video-taped by a friend that we developed a certain sequence when we work with our dogs, over and over.  Or we employ specific subtle body motions when we use a verbal cue to get our dogs to do something.  Often, a “Down” is accompanied by a subtle head nod that many are not aware of.  Maybe we use a specific pitch in our voice when we deliver the cue.  Probably, the dog is always positioned with the same angle to us when we ask the behavior.  We call it backyard obedience when we practice lots in our own environment and the dog looks super flashy but seems to know very little once taken out of that familiar context.   

A simple change such as asking a dog to go into a “Down” from standing next to us, versus standing frontally and looking at us, can throw the dog off.  Not providing the head nod is another simple change that confused the dogs.  Have you tried hiding and giving your dog a command?  Have you ever asked your dog a behavior by sitting on the floor?  Use the same body cues and voice intonation as you would use for a “Sit” but say “Banana” instead.  I bet many dogs would actually sit for “Banana”.  Be variable with your sequence.  Do you always ask for a “Down” after a “Sit”?  How about a “Down” first, then a “Stand”, and then a “Sit”.

Generalizing behaviors and testing (or proofing) them is crucial for the working dog – a mobility service dog, for example, or an autism support dog.  The same holds true for a competition dog who is expected to perform for a national rank.  In order to generalize behaviors, thus making them reliable under all circumstances, it is a good start to be honest with ourselves.  Keep a training log, or ask a friend to video tape you occasionally, so you can see yourself from the “outside”.  Take the behaviors to different places and environments.  Ask the behavior with a different voice, throw in “banana” and “frying pan” to check whether the dog actually listens.  See if you can get the behaviors from multiple angles, or with you out of sight.  The more broadly generalized the behaviors are, the more reliable and confident your dog will be, under all circumstances.  This is not only nice to have in a pet dog, but an absolute must for the service- or working dog.  Besides, you have your friends impressed with a dog that looks flashy and knows its stuff!

Generalizing is an art.  Feel your dog, and learn to read your dog’s processing ability and never let go of creativity in the journey.  Generalization takes time.  Dogs are creatures of habit.  But eventually, you will witness your dog getting the behaviors in various places.  This is an exciting and empowering feeling for both the owner/trainer and the dog and certainly worth aiming for!

“Brown Bear, Brown Bear, What Do You See?”  I see it all, reliably, and everywhere!

The Power of Involuntary Responses

Heed the power of the involuntary response! What is an involuntary response? It is based on the wiring of our dogs, from the “way they come”. Dogs are compelled/can’t help but to behave “that way”. The involuntary response also tends to be the one that frustrates us, gets our dogs in trouble, seems impossible to fix, and has tenacious qualities.

We are reminded how deeply engrained these involuntary responses are in our dogs when we consider chasing behavior. The dog may chase an object (squirrel, car, etc) for its entire life without ever becoming successful at catching the object (no reward), yet never loses its intensity and purpose in chasing that object. ….to never lose intensity without ever being rewarded… that is a very compelling response, time over time over time! We use classical conditioning to modify this wiring, mindset, and attitude in the dog to make behaviors manageable over time.

In stark contrast, behaviors we teach to our dogs through operant conditioning, such as a Sit or Down, must be reinforced occasionally or the behavior will become less reliable, less intense, and eventually fade. Needless to say, a behavior based on the wiring of the dog always has the potential to override a behavior taught through operant conditioning.

An example of the power of the involuntary response: My dear Sydney, who since passed, was a car chaser. He was very good at it; up and down the fence line he went, all the time. I never modified this and chose to do damage control instead to keep him safe.

In the last few weeks of his life, lymphoma had grossly overtaken his body, and there was not much of Sydney’s signature moving going on. BUT, he still chased the cars. As a fan of dog training and psychology, I watched on with amazement and found new respect for this involuntary response business. Sydney’s mind told him, as it did for 11 years, “must….chase…..cars….”. He was compelled to chase the cars, you could see it in the very little movements his body attempted. His facial expression even revealed a sense of confusion as his body simply could not deliver what his wiring told him to do. Behaviors so deeply engrained that they disregard the broken and dying body must be very powerful and strong.

Involuntary responses can be modified. However, it takes time to mold something as strong as the involuntary response into something manageable to us and there are often limitations. We need to be gentle and patient with our dogs and ourselves when attempting to change the wiring of the dog through classical conditioning. Achieving proximity to cars without the dogs lunging after them may take an inch at a time through methodical and effective repetitions of successive approximations, but it can and does happen, and therein lies the rewarding journey.

Let’s celebrate our dogs no matter how they come, ourselves for recognizing our dogs’ wiring, and realizing that there are ways to groom all behaviors. Some are just a bit more tenacious in their modification.

Happy Conditioning!

Should We Reward Effort?

Effort gets an A+! Or does it? Dog Training is rarely black and white. In the dog’s learning process, there are the desirable moments when the dog is correct, and we reward. There are other (common and normal) moments in the dog’s learning process where the dog is incorrect and we either withhold the reward, or go a step further and correct our dog, depending on how fair it would be to correct. But what about that fabric that fuels learning, that intangible concept of e f f o r t.

Do we communicate anything to our dogs about effort? If so, how important is that to the dog’s learning process. If we don’t, should we? How does it make us humans feel when we attempt a task, trying hard, giving everything, sometimes with success and sometimes without, and are acknowledged for trying. Do we care if nothing is said to us?

Innocent mistakes due to the dog’s overzealous desire to please us fall into the same category, in my opinion. How do we communicate those mistakes? Is it appropriate to withhold the reward? Is it fair to correct a dog, who knows the behavior, for an excitatory mistake?

This post is to challenge our approach and viewpoints when we train and communicate with our dogs and to be mindful of the fact that learning does not only entail marking either a clean cut ‘Sit’ or not, but attitude, spirit, and effort also matter towards payment. A dog that tries hard deserves to be acknowledged, in my opinion. A dog that makes a mistake because it wants to please us deserves to be acknowledged as well.

There are certainly exceptions and circumstances where we simply need to “get with the program” in order to achieve desired results and thus, recognition of mistakes and effort no longer applies. However, for the most part, I tend to err on the side of the dog to harness that spirit, to make learning a fun experience, and to avoid squishing the dog’s desire to work for me. Back to the human example, being acknowledged for trying hard makes me want to try equally hard or harder the next time.

“Thanks for trying, Woofi”!

Physical Praise Counts!

Physical Praise...the reinforcer we tend to forget. We all pet our dogs plenty but we may not realize that physical praise is actually a PRIMARY reinforcer which means it is a very strong (innate) stimulus to the dogs, one that did not need to be paired with a function (like secondary reinforces). Toys and food are the other two very common primary reinforcers and we use those a lot more.

It is common that dogs would prefer a toy or food instead of physical praise in a momentous time where a reward would be given, however, physical praise counts for the powers it holds!

How many of the ambitious pet owners amongst us have had mediocre success in teaching a new behavior in a training session but then find the dog offering the behavior perfectly on the casual stroll around the neighborhood? And we get upset because we were not prepared, did not bring food or toys along on the walk, and feel we miss an opportunity to reward. At all times, under all circumstances, you still have physical praise to give! Pair that with some serious verbal praise and more dogs than you might think are impressionably reinforced with that, it's marvelous!