Friday, December 11, 2009

Training Paradigm of Choice

I started reading Control Unleashed today (stoooooooked). There's the requisite stories of how dogs were harshly treated in traditional classes, then their owners saw the +R light and now they have excellent dogs. In CU a lot of the focus was on owners and trainers that misread calming signals, and that got me thinking about something.

As humans, communicating with dogs, an entirely different species, we really have no idea what they're doing. I mean, we can guess, and we've made some really good guesses, but not always (Remember when people thought dogs were wolves, so they should act exactly like wolves, then we realized we don't even really know how wolves act?). This is why learning theory is THEORY and not learning FACT. We think it works, and it appears to work, except when it doesn't. Gatsby, for one, has clearly not read the book, or he'd be perfect by now.

So I can't really find it in me to be up in arms over traditional trainers. They are just going off of what had worked for them in the past (which is totally explainable with current learning theory, by the way,and is part of why I get so excited over operant conditioning-based training [I really love science]).

So why am I a pure positive trainer?

Fallout.

Frankly, my technique sucks. I soak up all the theory I can get my hands on and love fitting it all together (the critical thinking rant is a separate post), but gorramit, put a clicker in my hands and I might as well have paws. What's the worst that comes out of a poorly timed click?

Gatsby gets a piece of kibble. That's it. Maybe dried liver, or a piece of cheese. One point to the dog. We'll try again until I get it right. Then he gets lots of kibble.

I have no doubt that correction-based training works. No one can really argue that fact. People trained dogs before us and they'll train dogs after us. Obviously, I think positive reward based training is best (otherwise I wouldn't use it, only the best for my Gatz). But in the hands of a poorly trained trainer, the chance and magnitude of dog-shattering fallout is much, much higher.

I think pretty damn highly of myself, but even I don't want to take that chance.

1 comment:

  1. It sure is a fierce debate, and it sure is thought provoking.

    I'm glad you like CU. I read most of it, and enjoyed it, but I honestly didn't feel like it helped me much. I do love the Look at That game, though. Perhaps I should finally pick it up and finish it..

    ReplyDelete