. . . . . . . .

Shaping is defined as the teaching by differential reinforcement of new behaviours by systematically reinforcing successive approximations toward a final finished behaviour.  Basically you are splitting the behaviour down in to easy to manage sub-steps, you move through the steps by gradually increasing the criteria for a click and treat. The final finished behaviour is the behaviour you want to train, and you reinforce this with a “jackpot” of several treats delivered one at a time.

To shape a behaviour it is a good idea to clearly map out the required steps toward the final behaviour, and have these pictured in your head before you begin. By doing this you can observe your dog to see what behaviour he/she is currently offering, and think about how you might use this behaviour as a starting point for shaping a new behaviour. What you need to think is what does he/she already do that can be used as a starting point toward the final behaviour?

Let’s take a look at a very common training problem many dog owners have, pulling on leash. At the beginning of October I met a new fur client called Barney, a handsome, bouncy chocolate Labrador.  Barney is a great Labrador, however his owner was somewhat anxious that perhaps he was trying to turn her in to Elastigirl! Barney’s owner had tried a headcollar, making like a tree, and changing direction, but did not have much success with any of these.

The real problem was twofold, Barney had been inadvertently taught to pull on leash from a very early age; dogs simply do what works, he pulled this worked, the behaviour continued. Secondly, although his owner had tried several interventions to stop the behaviour, a few vital parts of the equation were missing, no effective communication, timely positive reinforcement, or valuable reinforcement to guide Barney and keep his interest so that he could learn a new desired behaviour.

My human client was not only keen to have her arms back to their normal length, but she was also eager to learn clicker training. So after some initial coaching, and going through the fundamentals of clicker mechanics, we set about a plan for shaping Barney’s loose leash walking.

When shaping a new behaviour a high rate of reinforcement (ROR) is important, with this in mind I decided to set Barney up for success by initially omitting the leash. Instead Barney wore his collar as usual and I introduced a front attachment harness, and would add the leash when Barney was ready.

Planning Criteria

With shaping it’s always a good idea to plan your criteria, for beginners it really helps to write down each sub-step toward the final behaviour so that you are clear about the criteria you are going to reinforce at each step.  Also if you have the steps written down it’s very easy to go back a few steps if you hit a sticky patch. So plan your criteria:

• Identify what the final finished behaviour looks like: Seems obvious enough, walking with a loose leash, meaning the leash forms a J shape (once added), but we also identified a final location to the left of the handler, (withers/neck in line with handler’s hip), which we called the “jackpot” location, as well as regular eye contact from Barney.
• Identify any current behaviour your learner is already offering that can be used as a starting point: We had fun with this one and got a little creative - keeping in my mind that we wanted Barney to succeed. The handler visualised a large circle surrounding her, and any walking behaviour in the circle to the left of handler got a click and treat (C/T). To start with it didn’t matter where on the left Barney was walking, as long as he was on the inside of the circle.
• List the sub-steps (criteria), that will take the learner through to the final finished behaviour: This was actually quite simple, to increase the criteria for each sub-step the circle would get smaller, until there was just the right amount of space for Barney to be loose leash walking with his withers/neck in line with the handler’s left hip.

So for the following weeks the handler committed to this plan. She started off at home indoors, then progressed to the garden, then to a nearby quiet open field, which she drove to.  All of this training was done in short sessions 5 to 10 minutes three times a day, with no leash, just collar and harness.

At each new location the handler went back to the first step and shaped the loose leash walking by systematically reinforcing successive approximations toward a final finished behaviour. Her timing was good, the ROR was kept high, and diced chicken breast was used for reinforcement, Barney was in Labrador heaven!

The entire process is now being repeated with a 6ft leash attached to the front of the harness. Barney is now in week two with harness and leash, and he’s doing great. My human client has been through a steep learning curve, her arms are back to normal, and she’s feeling ready to shape Barney all the way through the Good Citizen Dog Scheme.