Different people have different opinions on when to attach a cue to a behavior. My own opinion, which works for me, is to wait until the behavior looks the way I want it to look BEFORE I attach the cue.

When I’m working on a given behavior, my dogs will keep repeating what it is they believe is going to be successful. Part of the shaping process is picking out the performances that are a tiny hair’s breadth closer to what I want to see in the end and reinforcing THAT. Or, you can look at it a different way: since I try to reinforce 80-90% of all trials, I omit reinforcing the worst 10-20% of trials.

When the dog is predictably repeating the behavior up to my minimum Mental standard each time, then I start to attach the cue. I start off by saying the cue SILENTLY to myself and watching to see if I predicted accurately when the dog was about to perform the behavior. When I’ve gotten to the point where I can predict it accurately to myself, I start saying the cue out loud. ET voila! It usually doesn’t take very long before the dog is responding to the cue.

What if the dog doesn’t do what I want when I cue it? Well, gee, I hold all the aces in THAT deck! The dog is trying behaviors and repetitions of behaviors in an attempt to get me to click because I hold something that dog wants (badly). All I have to do is withhold the click and primary reinforcer. The message is “that’s not going to get you what you wanted.” If the dog is happy to forgo my reinforcer, then it’s time for me to examine what is happening–is my reinforcer good enough? Is the alternative behavior a self-reinforcing one (for instance, sniffing bitches is a self-reinforcing behavior for most intact males)? Did I make my increments too high, so that the dog didn’t have a 90% chance to succeed? Am I trying to click for more than one criteria at a time (which is babbling with the clicker–like a baby, pleasant to listen to but meaningless)?

My own personal experience with my own dogs, who live with me and are wonderfully generous about covering up for my trainer’s deficiencies, is that they apparently enjoy training, in and of itself. I can use training treats of just plain dry Cheerios and they will work forever for them! Believe me, plain dry Cheerios are NOT a preferred food with my pack. But in a training situation, it seems to me that the marker of success is more important to my dogs than the actual substance. It’s like that tacky plastic prize I won at a birthday party when I was 11 years old–I could have gone into anystore and bought one, but it wouldn’t have the same meaning as the one I got for winning the Scrabble tournament. And yeah, when I’m feeling discouraged about my ability with words, I dig that silly thing out and look at it.

In a given training session, I might work on three or four different behaviors, none of them yet on cue. I just take a short break between behaviors–maybe play with the dog a little, maybe just stand there and consult the ceiling as to what to do next, maybe just move to a different part of

the room. The dog usually throws the last behavior that was reinforced at me and I just ignore it. Then they go on to try something different and out of that something different, I can pick out the tiniest, most remote approximation of what I want and reinforce that. Also, I often use props in training– retrieve articles, a target stick, jumps, the light switch, the dryer, the laundry basket, etc. When I put that prop away, the dog knows that something different is afoot.

Dogs actually pick out MANY cues from the environment as to what to do. It’s the trainer’s task to prune out the superstitious cues (those things that are not really cues but the dog thinks they are) and the extra cues that will not be wanted in the future.

There are many ways to train something the dog would never offer to do spontaneously. First of all, mentally break your task down into the tiniest possible increments. If you look at each increment, you will be able to see various ways of free shaping, luring, or capturing the behavior.

For instance, I had been working on light switches with Bretta, my Belgian Tervuren bitch. I’d taught her targeting as a puppy, and I’d also taught her to touch what I touch with the target stick. So it was easy to get her bouncing up and down by the light switch and touching it. Then I started to withhold clicks for light touches, only clicking heavier touches. At one point, she accidentally moved the switch itself and I jackpotted that. That’s as far as I’ve gone with it–about one try out of ten, she moves the switch. Since I was doing this just out of curiosity to see how long it would take (about 10-15 minutes, actually), I haven’t bothered refining it. Now I know the how to do it part, the rest would just be refining what she has already.

As I became more serious about training Bretta to flip light switches, I’d got a friend to construct me a light switch on a heavy piece of masonite, connected to a light bulb (a small 25 watt bulb). That way, I could teach her that the light bulb changing (on or off) was the marker for success. –a prop, which serves as a cue for the desired behavior!

Leave a Comment

Shopping Cart
Scroll to Top