It Takes Two To Tango and Mirror Train

Mirror training, or perhaps a variation of the Model/Rival method is a kind of training that's somewhat different from classical conditioning, but at the same time, needs it to function. I've often wondered how sheepdogs were trained to herd sheep with such precise movements and gusto that I've concluded that it's more nature than nurture, that they were just bred to be that way, hence they have natural instincts to herd. 

I was wrong, but breeding traits do help quite a lot. What they do, is either get a dog that's already trained or train one dog via operant conditioning to become a sheepdog. What they do next is to let the other dog or dogs observe how the trained dog does the task, then they reward the observer when they also complete the task. I suppose in a way that's already positive reinforcement training but with two technical trainers, you and the trained dog. 

Just like us humans, dogs are social animals. One very common method of social learning is mimicry of behavior or simply mirroring. You will see this behavior a lot with the chatty breeds like huskies who have a knack for "talking" and copying how their owners act. They are better at copying their kind, so when they see that another dog is doing something that gets them food, they are more likely to copy it. I don't know how fast it works, but if it's used to train teams of dogs to herd sheep for both work and competition, it ought to be pretty good. 

I'm talking about this because Matty's owner suggested I bring Maui and train him in front of Matty. Not only will it take advantage of this Model-Rival training style, but also let Matty play with another dog. I told her about the potential problems of dogs fighting each other, but she retorts that thinking and acting as if dogs will fight when they meet, increases the chances they will fight because they can "sense" our defensive actions and think that the other dog is a threat. 

I think this is coming from the raw experience of handling a large group of four and five-year-olds. She does have a strong point though. Being friendly about the encounters may actually lessen the chance of aggressive behavior developing unless you actually have a truly aggressive dog. 

Maui enjoyed the trip to the house because it was quite a walk to new places. When we reached the house's gate, Maui was ecstatic. He likely smells Matty, which he could smell off me whenever I come home training him. 

Even though we agreed to be friendly about the encounter, I still held Maui's leash tightly as we went to the backyard where Matty was waiting. He's not leashed though so Matty just rushed to Maui as soon as we turned the corner. Their noses met, and their body language already tells me they won't be fighting anytime soon. Their tails are both wagging and don't have any defensive signs like a lowered body, or growling. Maui kept looking back at me and then Matty as if he was expecting something, and was doing little pitter-patters with his front paws. It took me a few minutes to figure out that he wanted me to take the leash off, or at least that was the case because when I did, he ran off to the yard along with Matty and they played with each other. It was absolutely adorable. Seeing this, it will take a while before we can train because these two boys need to get the play-energy out of their systems first. 

While they were being the dogs they are, we had a discussion about Matty's training program and the specific training process. I explained to her the nuances of positive and negative reinforcement, and she shared a few things she learned that I'm not aware of. It's definitely something to digest, but it's been half an hour and It's training time. 

According to her, I just need first to train Maui normally, so that Matty could see how Maui reacts to the simple commands. This is a good time to let Maui refresh some of his obedience skills. 

First is the good'ol sit and stay. I called Maui's name and told the command. I had to say it twice before he did it, but when he did, I clicked, then tossed a treat that he caught mid-air. Matty was beside and watching Maui. It was interesting how he stared at Maui, looking up and down at him. I repeated the command and Maui followed it on the first try. I did this two more times and to my surprise, when I said the command, Matty did the same. 

I asked her owner if she taught Matty to sit prior to this and she did, but rather unsuccessfully, as even with the shock collar, Matty wouldn't sit still for more than two seconds. Now, Matty is sitting for more than four seconds. I clicked and gave both a treat. I tossed one to Matty first and I was glad that Maui didn't try to intercept it when Matty failed to catch it. Maui got his treat after and I did the sit-and-stay a few more times before taking a break. 

I needed to do some thinking. If I could make Matty mirror Maui, that means we can speed through a lot of the teaching processes, and since Matty is a retriever, he might be better at picking stuff up and placing it on my hand. If I do this well, it might improve Maui's fetch game. I'll bring the shoe next time. For now, I'll explore this new kind of training. 

I know Maui can stay still for 10 seconds before he needs to be reminded. I'm going to see if I can make Matty stay for the same duration. We resumed training and I switch between calling Maui and Matty, and sometimes, I called Matty when I meant to call Maui and vice versa. Didn't matter. Both of them followed the commands well, but for some reason, I could not make Matty stay for more than five seconds. All he wants to do after staying is to nudge Maui for not moving. 

This was amazing. First was "Shaping" which I think I'll be sticking with because it's quite efficient and I can teach Maui complex stuff fairly quickly. It's also partially relationship-based training because I've owned, trained, fed, and played with Maui for quite a while now, so I hope he regards me as an authority of some kind. It's definitely an upgrade from traditional dog training. 

Personally, I think some of the training methods are all the same, still under the classical conditioning because it's still reward-based training. They still consist of "marking" positive behaviors and rewarding them. I still think electronic training is good, but not on its own. I think it greatly compliments positive reinforcement training. It's also a very touchy subject because when you look at anything shock collar, you'll find all kinds of opinions about it. As is with all kinds of aversive training methods like prong collars and even spray collars. 

I'm growing more curious about these new training techniques, and if I have the freedom to train Matty, I might combine all these somehow and further speed up both Maui and Matty's training. I wish I had the guidance of an actual trainer, but for now, I'll learn through trial and error because my belief is that the learning from the small mistakes you make along the road of life contributes immensely to your success.