Train Brainy - Basics of Training For Command Response

"Training brainy" allows you to establish trust and stimulate thinking.

1. Give command & hand signal.
2. Get the dog to do what you asked.
3. Give release word.
4. Reward.

The command & hand signal can be thought of as a "question". You are presenting the dog with a challenge. An opportunity. An assignment.

When the dog gets the answer right, you give the release word. Not only does the release tell the dog when he is done, but it also becomes a reward. The release word becomes what behaviorists call a "secondary reinforcer".

If you are using a clicker, it is important to understand the functional role your clicker plays. The clicker can be used in a similar capacity as the release word (secondary reinforcer). It can also be used as feedback (tertiary reinforcer). It can also be used as both. It can be used as a command, but not if it is being used in a reinforcement (primary or secondary) capacity. Clicker is often attempted in a "distractor" capacity. I don't find this works out well. More on this in another post!

The release word (secondary reinforcer) predicts the appearance of what the dog REALLY wants, which is food, play, etc. What the dog REALLY wants is called the "primary reinforcer". Primary reinforcers are motivators like food, sex, distance from danger, stress relief, social status, etc. Temple Grandin says that behavior comes from drive to stabilize core emotional states. These would also be considered "primary reinforcers" or "motivators".

There ARE dogs for whom getting the right answer, pleasing the handler, or receiving affection is a primary reinforcer. I find MOST dogs don't fall into this category.

Beginners, use food -- not just because dogs really like food, especially if hungry, but also for ease of controlled access. Non-beginners, play is a bit harder, but worth a try. It's addictive for both parties!

There are cognitive rewards (like stress relief or social status) which we must acknowledge exist as motivators, but which we have no vehicle with which to control the brain's access to these motivators.

So, great. Now we know how to train! All we do is follow the 4 steps, everything's fine. Right?

Chances are, the dog will make a mistake. You can think of your starting probability as 50/50 that the dog will get the answer right/wrong. When you see a mistake, do you teach through or do you regroup yourselves for another (hopefully better) trial?

• If you choose to teach through, you will need to make use of feedback to communicate precisely what is leading to the reward and what is making the reward farther away. Teaching use of feedback (tertiary reinforcers) will be covered in another post. Suffice it to say that feedback is a game of "hot & cold".
  

• If you are a beginner, or if you are not able to tell at if it is your mistake or the dog's mistake, I recommend simply telling the dog to "try again" (literally say those words to the dog) and start again from the very beginning of the exercise.

**Starting from the very beginning of the exercise is critical. If you have a dog that jumps on people as they enter the door, the beginning of the exercise is where the person is outside the door, not just where you told the dog "OFF".

It is also important to recognize that if the dog gets the answer right, the probability of the dog getting the next one right is high, but the probability of getting the next 20 right is much lower. Always quit when you have had success. 2-3 answers right is about quitting time on any new exercise. If you get 5 right answers in a row, without having to start from the beginning, definitely quit!