I recently read a blog about “choice” that got me thinking. If you’re interested you can see it here: http://eileenanddogs.com/2015/06/19/giving-dogs-choices/
If I’m shaping something with a simple yes/no system, then a correct choice earns a cookie and an incorrect choice gets nothing – being wrong is “neutral”. The dog may not actually perceive nothing as “neutral” but that is a story for another day.
But most of my training is not shaping simple behaviors – it’s building those behaviors into solid chains that can hold up under challenging conditions. That is why I put so much energy into building relationship and love of work; because over the long run teaching behaviors is the easy part. Getting those behaviors to hold up in the ring is where the challenge lies.
Many dogs fail in competition because they are disengaged; they do not enjoy working, and being with you isn’t all that valuable to them without a classic reward. Of course, backchaining to a jackpot reward outside the ring is an option, but dogs with lower interest levels in food and toys aren’t very good backchaining candidates because of the length of the chains required for competition.
Since “choice” makes training more fun for my dog and since “fun” is my ultimate weapon in developing ring engagement, I want my dogs to have as many choices as possible in training, and I want the options to be attractive; greater than neutral. I need training time to be consistently fun to offset whatever stress I inadvertently build into my program. To do this I pick and choose from my available reinforcers and reinforcement packages so that I can “get my way” over time.
Here’s a video with Brito. Note that everything that he does with me is reinforcing; I simply reward some of his choices at a higher level. He can choose to self reinforce with his ball if he wants to; it is freely available to him. He can also wander off, which he does here a couple of times. (Wandering is the only choice that receives no reinforcement from me but it does have a value of its own)
Here are his options with approximate values as motivators on this day:
1. Wandering around: 2-3
1. Personal praise: 3 – 5
2. Chewing his ball: 3 – 7 (Depends how long we’ve been working)
3. Fetching his ball: 5 – 8 (The fresher he is the more value he finds in fetch.)
4. Cookies: 4 – 7 (thrown cookies have higher value – or at least bring out a better picture . Cookies are never freely available because he would simply sit down and eat with no thought for me or training.)
Here are the ways that he can get reinforced:
1. Looking at me/wagging his tail/staying close: I will ALWAYS reward the choice to engage with me because it is so fundamental to our basic relationship, even if he’s not actually doing any work. I want interaction with me to be a default. He earns personal praise and occasionally a cookie.
2. Picking up his ball: That is his choice; he can do it at anytime and I won’t stop him. In this case he is self rewarding with chewing.
3. “Outing” his ball on cue: I’ll give him a food reward or a ball toss. I have worked hard to get a clean release off a ball so I value this highly and will reinforce it even if it is not the focus of a session.
4. Placing his ball in my hand on cue: usually this earns a cookie. I reward this to offset his innate desire to sit by himself and chew his ball; he is interacting with me and I reward that. Which brings me to my next point:
5. Chewing his ball near me rather than alone: Usually personal praise; scratching his head and talking nicely to him, while he chews. As a result I find that he rarely chews by himself; he prefers to stay close to me. What the heck – he has his cake and eats it too.
6. Picking up his dumbbell; on cue or otherwise: This is my highest value option and what I am working towards, so this choice gets the highest value package; I give toys, food and praise. I try to time my cues to when I believe he will comply – with greater and lesser success. Next time I’ll probably leave off the cue and let him work it out for himself.
Can you see how his entire training session is rewarded? Over time, he’ll pick up his dumbbell more often and with a very positive attitude so that he can get the best package of rewards. I have structured this session so that the packages that include me are always the most desirable. I don’t have to punish him at all or provide unpleasant options for “failure.” Over time he will choose me.
If you want to try it, think carefully about what reinforcers you have access to and what their values might be to your dog. Think about what work you choose – more intrinsically motivating work is easier than less desirable work. Think about your package of rewards; what is it? Is it well developed? If your dog sees praise as a conditioned punisher because you only use it when you’re not handing over a cookie – you have bigger issues than the ones I’m talking about here. Fix that first. And if you pair a dumbbell retrieve against squirrels in the trees, well…. you’re probably doomed for a first attempt.
Give your dog some positive choices in training. As long as there is a difference between the choice that you want your dog to make and the alternatives, you will win over the long run. I would assume that it takes longer to train this way and I don’t normally do it when training simple behaviors, but when I’m moving on to more relationship based training then it has a place, especially with more fragile dogs or with dogs that have relatively little interest in classic reinforcement options. Pay attention to what options you have available and structure yourself to win, but at the same time give some choices. Real ones. Not a forced choice between what you want and an undesirable alternative.
If you want to read more about reinforcers, Deb Jones and I wrote an entire book on this topic. It will give you plenty to think about:
Here’s the video, unedited: