I am a motivational, marker based, dog trainer. I don't use physical pain as corrections. But that doesn't mean that consequences for mistakes do not factor into my training. People can say, I never punish my dog for getting it wrong, I just don't reward the mistakes. But if you have ever told your child, you do not get dessert because you didn't eat your dinner, or you may not go out and play because you didn't clean your room, you have just withheld a reward for failure to perform. That is a correction in the sense that you are doing it in hopes of decreasing the likelihood that a behavior will continue.
If there are litterally no consequences for getting it wrong, then that means being wrong is just as rewarding as being right. Imagine trying to train a stay if she got the toy every time whether she got up or held her position....I'd rather not.
So the question is really, when there is an error made by the subject in the learning process, what are you going to do about it? If I have learned anything from my 3 years with Reyna it is that I better have a plan for when it goes wrong that is just, if not more, well thought out than when it all goes perfectly.
I think it's relatively easy as a motivational trainer to say...I am not using physical pain to correct issues so I'm all set. But take a look at the following video. Sam has a Level 1 foundation in Sue's training levels, and I needed to teach her to stay out of our playroom while she was staying with us for the weekend. I have layered in a high rate of reinforcement, but watch when she starts to make a mistake. Even a weight shift towards her is enough of a pressure correction to make her back up immediately. Sam is not nearly as sensitive as Reyna, and yet something as "gentle" as taking a step forward sends a very strong message of "that is the wrong choice, back up".
I think as positive trainers, we need to consider the effects that our corrections have on the learning process, just as we would hope a compulsion based trainer would critically look at their methods, to ensure that we are always being fair, taking the dogs body language and emotional state into consideration, and objectively looking at whether or not our methods are humane. And then being willing to change them, for that situation, that dog, or even overall in our training if there is a major problem.
I'm not saying never use pressure (obviously since I use it the video above), but I am saying we need to be deliberate and thoughtful in how we employ even physically pain free corrections. If you have seen Eileen's shock collar video where she displays how negative reinforcement is used to teach a dog to climb on a platform, I can tell you right now I could do the same thing to Reyna with pressure alone and it would be just as inhumane in my opinion.
Before Reyna I really was oblivious. I thought that since I used a clicker and treats and wouldn't go near a prong, choke, or shock collar I was acting in the best interest of the dog. But by focusing on the methods I was using and not the attitude of the dog sitting in front of me I created a dog who found very little joy in her work. Here are a couple examples of misused punishment/negative reinforcement and the fallout because of it:
Crate training:
We brought Reyna home at 9 weeks old and the first thing we wanted was for her to sleep through the night in her crate. So I slept next to her crate the first night and if she would whine and I knew she didn't need to potty I would say "hush" and give her a hard stare until she disengaged and laid back down. That is A LOT of pressure especially for a 9 week old puppy. It worked too. We did the same thing to make her stop whining when we were in the yard and she was on the porch. It made her so sensitive to eye contact that by 8 months old if she was dozing and I looked at her she would get up and pace and whine.
Loose leash walking:
When training Reyna to walk on a loose leash I put a leash on her and pulled on it just enough to make her uncomfortable (doesn't take much for a puppy). When she stepped toward me I rewarded. This might have worked fine if I had laid a foundation of reinforcement for walking next to me, and then walking while dragging the leash, and then some luring into position with the leash on to remind her what a great place being next to me was. I didn't. When I took her to a training group they asked if I ever played with her because she was so uncomfortable on leash.
Proofing behaviors:
I used to use a No Reward Marker pretty often. If I asked for a sit and she offered a down, or just didn't sit I would say "wrong" and turn my back for a few seconds. Again, maybe for a different team this would have worked just fine...My Greyhound in college caught on pretty quick, the boxers I had growing up did. It completely demoralized Reyna. She stopped being shape-able and anything that sounded like a new cue would make her shut down and walk off. I can't tell you how many times she just walked off in a training session.
It took a lot of rethinking, and changing the way I train, to turn our relationship around. Now I try to always have a plan for what will happen when things go well AND when they don't. Also I always consider how the consequences of her miss affect her attitude in training. Does it build drive to try again? and does she exhibit joy when she gets it right or just relief? My goal is always joy!
No comments:
Post a Comment