I just brought Phoebe along so she could say hello. Really. Kay Laurence, her Genabacab grandma, was leading a workshop one weekend in October, but we didn't plan to participate. Phoebe and I could only stay for the morning, in any case, as we were due at the annual Tidy Lawn Halloween Party for Dogs that afternoon.
I also knew that everyone registered for the workshop would be an advanced practitioner of the art of shaping and cueing. I'm a pretty sloppy clicker trainer, as Phoebe would be the first to tell you. From cooking to filing to dog training, precision is not a natural inclination. I, frankly, didn't feel qualified to join the gathered group of clicker wizards.
Yet I did want to get a look at Kay's micro-shaping techniques, which were the focus of the seminar. I know a click can mark any behavior, no matter how small. But can it mark a mere muscle movement, a motion so small the dog is unaware of what it's done? Taking notes off to the side, Phoebe asleep in her X-pen, I jumped when Kay looked around and said, "Let's see... Gale, get Phoebe and let's try this out."
Oh, rats. Couldn't we just hide over here and watch everyone else? Never mind, I've passed the CAP test; we could do this. Kay asked me to shape Phoebe to put her forepaws on a brick. Easy-peasy. After barking at me a bit (translation: "What? What! What is it you want me to do? Tell me clearly!"), she focused and settled to the task. Soon two paws were on the brick. Well, one paw was on the brick and the other slapped the brick before falling off the side. Phoebe's shaped behavior was as sloppy as my training. Not to worry, I thought, the behavior is rough, but it can be polished—shaped for quality—in subsequent sessions.
Then Kay took over. She placed the brick again in front of Phoebe. Phoebe approached the brick, about to repeat the behavior but got a click before she got there. Kay kept clicking on Phoebe's approach to the brick, rather than for the desired ultimate behavior. What was she shaping? The slant of her shoulders and shift of her weight. Kay was marking Phoebe's muscle movements so that when eventually she did place both paws on the brick she would already be balanced and in perfect position; Kay was polishing the behavior before it was established.
Building a behavior from the inside out, rather than getting a rough approximation and then smoothing it, is the essence of micro-shaping. It's tricky stuff. Not only does it require a reversal of our training plans, but a high degree of clicker precision. The handler's mechanical skills both in timing and in tossing the treat must be smooth and quick. Kay used the treat toss to reposition Phoebe's weight—back on her heels and ready to shift forward—in a way I could not manage (at least, not with an audience) without a lot more practice.
Micro-shaping also relies on the handler's observation skills and ability to anticipate the dog's tiniest muscle movement. You need to see what it is you want to reinforce, see what happens just before it, and see it every time in order to keep up a high rate of reinforcement—also a key to success.
Is the challenge of micro-shaping too high for a lackadaisical trainer like me? On the contrary, it may offer the solution to that roster of quickly, easily trained behaviors that never become truly polished. (Phoebe and I have a lot of those.) Like preparing to paint a room by first spackling and sanding the surface, rather than slapping the paint over all the flaws, I can save myself a lot of trouble later on by doing it correctly from the beginning.
And Phoebe? Makes no difference to her. Any clicker training session is puzzle-solving. Whether I'm clicking her for a shift in her weight, instead of standing and walking on her back legs, it's all good fun.
Wanted: A nice boy. Smart, quick, companionable, handsome, comes from a good family. Enjoys children and family life. Likes to herd sheep and jump through hoops. Phoebe will be 3 years old in February, and will be in season to boot. Time to think about having some Genabacab puppies on this side of the pond.