Atticus and I have managed to get through the “worst” of the rehab of his laminitic episode back in April. He has lost 200 lbs. and our training time is not focused so much on just exercise. We can now train for useful behaviors.
Since we’ve been walking so much and I would like to get around to riding him again, I’ve decided to focus on putting walking, backing, and whoa on cue.
There have been a lot of variations of what we’ve done since my last blog on walking. We’ve also worked on recall.
What I noticed and want provide food for thought is that when we train, we create magnets in our training area that we have to work around or work with, depending on the situation.
What I mean by this is look at the photo I inserted at the beginning. I’ve marked the current magnets in my paddock. The PVC pipe ring is no longer there and Atty and I didn’t work with it much before he had his laminitic episode.
The barn itself is a magnet since it provides shelter, it has been a location (or several inside) for training, it has hay, and the boys are fed in their stalls.
There is also a grooming stall or wash rack just under the roof in the upper right corner that is another location of reinforcement history (RH).
There’s a food pan in each corner and each has an RH and, depending on which corner, the corners themselves have RH.
Then there are the barrels (not labeled, but they’re white blobs) also have some RH and there are the teeter with RH and the target in the tree that’s now been moved to gate that has RH.
Imagine for a moment or two that each of these places is a giant magnet. Bob Bailey and Parvene Farhoody liked to talk about building mass around a behavior as a description of reinforcement history. Another analogy is like making a deposit in a bank account. But I like the mass idea for a couple of reasons and one is that it makes it easier to think about the behavior locus or site for training that behavior also has mass – or magnetism or gravitational pull – built around it. The location itself becomes the cue for the behavior as much as seeing the food pan or the target.
The environment is part of the cue. This is why we try to take care to set up “antecedent arrangements” or the environment to help the animal guess what we want. Or limit the possible answers.
But my main point is that the specific location as well as the item become triggers for certain behaviors. As such they also act as magnets – they draw the animal to that location to do the behavior associated with that location.
The animal wants to get reinforced and it will often offer the behaviors that have been reinforced a lot so she will audition those behaviors to see if they pay today.
Then if the animal is within the gravitational pull of a particular magnetic location, he will go there to offer the behavior that was taught there.
This is both good and “bad”. One way we can take advantage of this is to train different behaviors in different locations, especially if the behaviors are similar enough that the animal might get confused. Or even if they’re totally different but the animal wants to do Behavior A that was taught in Location A rather than try to learn the new Behavior B in Location A. It would be easier for both of you to pick Location B to train Behavior B in.
Now what happens if you have a relatively small space and you’ve taught a bunch of behaviors all over the place? You have to be very aware of those locations and the borders of the magnetic bubbles, that’s what!
I can use the barn and the feed pan near the barn to my advantage by getting more walking steps away from me because of the gravitational pull of those things will work to my advantage and get more steps in that direction before I need to click and reinforce.
However, going the other direction, it’s going to be harder for Atty to leave the barn and I need to be more generous by lowering the criteria to only two or three steps before clicking and feeding.
I also have to be aware of the pull of the hay box. He may decide that my Rate of Reinforcement (RoR) is too low and quit me to go eat hay. Can I get him back with my recall or a hand target? How close do I have to be for those to work? Is this really a problem or is is just a training challenge that helps me proof my cues?
And then halfway down the fence line, I had hung the target on the gate to also work on targeting. Yes, it could be a useful magnet, but I don’t want to reinforce there too much, if at all, because I want him to continue walking. It may come in handy later if I want to string cues together, such as “walk on” and “target”. But not right now. I have removed the target from the gate but that spot is still magnetic.
What do I do then? I have to be aware that that spot is coming up and prepare for it by clicking and feeding sooner and running up there to block the target with my body so that the target itself and that specific location is not the reason for the click and treat.
If I walk around the end of the paddock, I may get an offer to walk onto or back onto the teeter. If I don’t want that, what am I going to do? Be aware of it’s proximity, raise my RoR around it, and use a hand target if necessary to keep his attention with me and what we’re doing or he may go over there and audition the teeter behavior.
If he gets away from me because I wasn’t thinking ahead, what am I going to do when he gets on the teeter? Ignore it or reinforce it? At this stage, I think I can safely ignore it and call him or target him back to me.
If he’s new to clicker training and all I’ve done so far is teach him to target an object, what should I think about if I put something new in the paddock? Most likely my horse, even though he’s not new to CT, will go over, investigate, and target the new object with his nose. This may not be a good thing if the new item is my camera! Or the vet’s pricey diagnostic equipment!
Be aware that your horse may now decide that everything is a target that requires a nose touch! Or that the game you played yesterday was more fun and wouldn’t you really rather play that again than what you’re trying to train now?
In this video, Atty decides that the tree where the target used to hang is a possibility, so he heads for the tree. But I manage to “call him off” by using a hand target.
In this clip, Atty decides that the game we’re playing is walking on without me and tries to head for the barn and the feed pan there.
Now think about rubber bands. Each location and item has its magnetism or gravitational pull that draws the horse in and we have to be aware of the boundaries of each. As I said above, the direction I’m going relative to the barn, for example, I can either get lots of walking steps or very few. The magnetism pulls the horse toward the barn and I can get lots of walking steps to build duration of walking. But going away from the barn the magnetism pulls Atty back and I have to raise my RoR and lower my criteria.
My job is still to build duration of walking but it is also to stretch the rubber band of the magnetism so that I won’t always have to have a high RoR to get him away from the barn. We have to build this slowly over time. I can move the pan away from the barn so that that magnetic spot isn’t quite so strong. I can stop and turn around sooner before we’re drawn into the “black hole” of that magnet. I can stop just dropping feed into that pan to weaken that pull. I can work in other areas. I can put the behavior on cue and work on proofing the cue.
Be aware of the magnets and their pull around your training area.
Be aware also that the barn might not be where they want to go back to when you’re done training because maybe the horse isn’t ready to be done yet and going back to the barn is not a pull but a push. In that case, you will have to lower your criteria and raise your RoR to get her back in the barn so you can fill the hay box.
Let me know what you think of these musings of mine regarding magnets and rubber bands.