In case you missed them, here are my notes from the Friday sessions.
Look Away From the Dog, Theresa McKeon
This session was intended for professional trainers who wanted to learn more about human body language and how to be a better instructor. Even though I don’t fall into that category I’m glad I went! Apparently dogs and humans have very similar body language signals for stress. Who knew?! We both yawn, make ourselves smaller (crossed arms), appear menacing (crossed arms), humans tuck their thumbs while dogs tuck their ears, flick tongues, smile, cough/sneeze, face smash, blink excessively, and shake off the body.
What a Cue Can Do, Kathy Sdao; lab with Sarah Owings
Pretend your learner is a walrus; you can’t compel a walrus!
A really common practice is to use the cue during praise, such as “good Sit!” A cue signals that “reinforcement is available if you perform this behavior right now.” When we use the cue in praise we think the cue is a label, but dogs don’t use labels. For them it signals movement, and they can’t move into the required position if they’re already in it. If you use the cue during praise it reduces the effectiveness of the cue and confuses the dog.
Something I hadn’t even thought about is that learning a cue is a behavior! It requires practice for understanding. This goes for both the teacher and the learner; with enough time both will understand the process with fluency. I needed to hear this since I’ve been bad about putting behaviors on cue and now I’ll have more patience with myself and the dogs.
In the lab Sarah emphasized several things that will clean up our training and encourage clean loops. First, reward in the same location to reduce the search for food. This way the dog returns more quickly to the game instead of snuffling about. Second, reward away from you, you’re hard to leave! This way the dog will naturally return to the object (we were shaping) as they turn back toward you. Third, ritualize your training so your dog can predict what’s going to happen. Fourth, after you have a clean loop work on stimulus control with this strategy: feed the dog’s face for a few seconds, cue the behavior just before they do it, click/treat when they do it correctly, then feed their face again. This encourages the dog to pause in between behaviors so you don’t create a behavior chain.
Loopy Training, Alexandra Kurland
People at Expo threw around the word “loopy” like it was common knowledge. I have to admit I’d never heard of the concept until Expo. Loopy training basically means that you have a very consistent and fluent “loop” of behavior-click-reward-behavior-click-reward. You can see this in the Kay Laurence video below; this is what loopy looks like.
Related to shaping, I really was struck by this: don’t wait for a behavior that doesn’t exist. Wait until the behavior you want is already occurring due to variation before making it a clickable criterion. Shape the criterion when the behavior you want is already happening.
Game On! Train or Be Trained – Part 1, Jesus Rosales-Ruiz & Mary Hunter
This was a lab in which we played PORTL, a shaping game. We paired up and one person was learned and one was trainer; the trainer had to shape the learner to do a behavior. I chose to be the learner; I wanted to experience what my dog’s experience!
The original behavior was for me to push the bell. This led to a really critical question: is the behavior truly learned? Did the learner learn the behavior that you think you trained? I thought the behavior was moving the bell. I pushed the bell a few times so my trainer thought I had learned the behavior. This would lead to intense frustration if a cue was added!
Since it took less than 10 reps for me to push the bell, and we thought I had learned the behavior, we decided to try something more complicated. This highlighted how critical the order of operations is when creating a shaping plan. Granted, my trainer didn’t have much time to think through the plan since we did this spontaneously!
The criteria must be in a sequence that is logical to the learner based on their learning experience and motor patterns. I was supposed to flip over a hollow seahorse toy and place the bell in side of it. Our shaping plan went like this: touch bell, move bell, pick up bell, place bell on seahorse, touch seahorse, pick up seahorse with bell on it….confusion. I had no idea I was supposed to flip over the seahorse. We decided that an ideal plan would have been for me to flip over the seahorse first, and then reinforce me for touching the bell. This relates to a conversation I had with Jesus where he said, “train action before discrimination.” So “pick it up” before “the red giraffe”.
The ultimate goal is errorless teaching, which you can see in the video below.