Over the past few years, I’ve experienced a significant shift in my approach to training:
- I’ve become extremely aware of the need to improve my communication skills (clean handler mechanics, anyone?), and have been working hard to make that happen.
- I observe my dog’s emotional state and work with that first and foremost. Emotions drive behaviours, and so it seems a no brainer that we need must begin at that level.
- I do everything I can to give my dog choice and say in our training, play, and time together. To ask, and respect their answer when safely possible.
- To give you an example of my shift, let’s revisit the It’s Yer Choice (IYC) game. Made famous by Susan Garrett (I have no idea if she in fact invented the game, or just made it popular), sport dog trainers the world over – including myself – use this game as one of the foundation pieces they teach their new dogs and puppies.
Except I have been questioning this game for a while now. In fact, I question this game so much, I don’t think I will include it in my training anymore.
Really, you ask? You’re tossing IYC?
Yep. Out the window. The cornerstone of my training for probably a decade.
(Apologies to all who have trained with me and taught their own dogs this game with my encouragement! But… the never ending learning is what makes dog training so addictive and worthwhile, is it not? So hopefully my suggested reframe will be welcome!)
Very briefly, IYC is a simple game in which the dog/puppy figures out that if they want the cookie in our hand, they have to move away from the cookie. In other words, they must figure out to do the opposite of their instincts, in order to earn reinforcement (said cookie).
I love the idea of this game, the goal of which is to teach self-control using “choice-based” training to build value for work. This works as follows:
When our dogs (or us humans, for that matter) make a choice, and receive reinforcement for that choice, our brains enhance the neuropathways that led to that choice, increasing the chances of that choice being made again. As such, if we can set our dogs up to make choices we like, we dramatically increase the probability that they will repeat those choices.
[The same is also true if they make a choice we don’t like and receive reinforcement – like breaking a sit to chase a squirrel – which is why it is so critical to avoid letting our dogs make choices and be reinforced for behaviours we do not want!]
Neuroplasticity: Neurons growing in density
While initially seeming like an effective little game, upon careful reflection, there are three fundamental problems with traditional IYC:
1) It’s a misnomer.
The dog actually has no choice. It’s our way or the highway (so much of purported “positive” training is like this, a realizing making me increasingly uncomfortable and seeking alternative means of teaching). This leaves me questioning just how effective this game really is in achieving the aforementioned goal of value building through choice!
2) The game uses negative punishment (-P).
Ok, coupled with positive reinforcement (+R), but… negative punishment, people! As the first thing we teach our puppies? What kind of precedent is that setting in our relationship (recall that my dog’s emotional state is now my #1 priority)?
As a trainer seeking to use positive reinforcement only, whenever possible, I find this realization problematic.
[By negative punishment I mean: When the puppy approaches the hand full of treats, we close the hand, or take the treats away (negative). The result is a decrease in the behaviour of mugging the hand (behaviour has been punished) = -P]
3) IYC does not tell the dog what TO DO. Only what NOT TO DO.
I think this third point is the most perplexing, as the lack of clarity for the dog (and the handler) makes criteria difficult to evaluate, and thus, reinforce. I know many a dog who has shut down during this training, often then labeled as being “not operant”. Confused may be closer to reality.
As +R trainers, our goal is to tell our dogs what TO DO, to have clear criteria, to help them achieve it, and to reinforce our dogs for meeting criteria. Yet in IYC, we don’t care what the dog does, as long as the dog does NOT mug the hand.
In sum, if we want to train using +R methods as much as possible, starting with a +R/-P game focused on telling the dog what not to do seems counterproductive at best.
One last point, which is more practical than philosophical: Many dogs default to a sit or a down in this game. As a trainer of performance dogs, I do not want a static sit or a down behaviour to be my dog’s default offer! This is just my personal preference; there is nothing wrong with a down or sit as a default behaviour. But for a sports dog, other options might be preferable.
So, if not traditional IYC, what can we do instead?
Let’s not toss the baby with the bathwater. I still like the idea of having a dog who can work in presence of food and toys, and who won’t steal them or break their work/training to self-reinforce. So I do want a dog who understands the results of the traditional IYC game, I just want to find a different way of getting there. An approach that will use exclusively positive reinforcement to build the behaviour, that offers clear criteria about what the dog is TO DO, and that builds the dog’s operant behaviour.