My Education Continues - Primary Reinforcers
I have been taught that dogs do what is immediately reinforcing, and I have been taught that food is a primary reinforcer. This means that it has innate value to a dog. A secondary reinforcer, or conditioned reinforcer, is something the dog learns to value. Well, OK, so far.
I learned, through my Karen Pryor Academy education, how to use food as a primary reinforcer to increase the likelihood of behaviors being repeated and to use food to teach new behaviors. OK.
KPA offers a course about non-food reinforcers. I looked into this course because I would like to know more about training without using treats. I found that it is a course about how to teach a dog to be reinforced by something he has learned to value by using food to teach it in the first place. So, it is a course about teaching conditioned reinforcers; the non-food reinforcers are not primary. Stay with me here, I am actually going someplace.
Common sense tells me I do not have the whole story, and this has been wiggling at me for a long time. If food is THE primary reinforcer, and dogs do what is immediately reinforcing, and dogs actually have the experience of eating only about 2% of their 24-hour day, what else is reinforcing their behavior? Is it all about food? Does it all come back to food? Or is all else that is reinforcing to them secondary, learned reinforcers? I just don’t buy it, and so my education continues.
When I attended the training course at NEI (see the blog post about that below), I was one of three dog trainers in attendance, and no one I talked with had any idea of what an LGD is. In my introduction to the other students, I rattled off one of my standard descriptions of LGDs to people new to them; I said that they own real estate. I remember seeing Steve, the owner and head trainer of NEI, smile just slightly, patiently, and I remember feeling that I had stumbled into something. Indeed.
Being Trained by Rufio – My Natural Encounters Adventure https://www.bensonmaremmas.com/post/being-trained-by-rufio-my-natural-encounters-adventure
My LGD Podcast with Natural Encounters, Inc.! https://www.bensonmaremmas.com/post/my-lgd-podcast-with-natural-encounters-inc
We watched video after video after video of training being done in zoological situations; the “real estate” of those animals was a major consideration in making training decisions. I watched this, and I thought about my LGDs, because if real estate is important to tigers – and I know it is, that seems like common knowledge – why isn’t it common knowledge about dogs? Is real estate more important to animals that are confined, I wondered? I tucked all this under my hat and came home to ponder this and ponder this. It has really been on my mind for the six weeks following the course.
I felt that what Steve was trying to say is that real estate is a primary reinforcer, not a secondary reinforcer; that the physical autonomy of an animal is of innate value to that animal. I created the graphic at the top of the page a couple of weeks ago to help me visualize this, but I wasn’t sure enough about the conclusion I had reached to write about it yet. Today, I began an eight-week course taught by my favorite scientist, Susan Friedman. At the end of her three-hour live lecture in the class, we were given the opportunity to ask questions; yep, I was ready with this one! When she confirmed for me that my conclusion was correct, I actually cried. It means so much to me to know why. It isn’t enough to know how, how to get an animal to do something; I want to understand why the animal is willing to make that choice.
This is the course I am currently taking:
As a trainer, if I understand why something works, I can make better training choices, but for me, there is also an ethical question. In a previous course with Susan, she suggested that it is appropriate to consider what an animal has to give up in order to accommodate a trainer’s request for a behavior, given that before the trainer’s request, the animal was already doing something of value to him. So, just because I can teach an animal something, do I have the right to ask for it?
Susan created a graphic that presents her hierarchy of behavior change procedures. She advocates using the least intrusive training methods possible to achieve the desired training result. This hierarchy was created for use with children. She defines intrusiveness as “the degree of counter-control, choice and consent, for the learner.” Another way to say it is that control is making the world behave; having power. In layman’s terms, can you teach something with a minimum of choice and control given up by the learner?
Now, here is where this gets real. Think of food on one end of a teeter-totter and control on the other end. If I take away control – but add food – to get a behavior, is that enough? What happens?
Taking away control is probably experienced as aversive by the dog. Is offering him a food reinforcement more valuable to him than what you took away? With LGDs, as my specialty, the answer is no, very often. So, think about this. I have had people tell me that their LGD is not food motivated. It is true that being motivated by food in training may be something that needs to be taught, for a variety of reasons. The presence of food may be novel, which LGDs may be suspicious of. Eating has a social context, so that may be a consideration for the dogs. But. What if, the owner put a leash and collar on that LGD, took him to an area to train, and got out that clicker and treat pouch. The dog gave up a lot just to be in this physical position, never mind being asked to learn something in this situation. And dogs that are worried are unlikely to eat anyway. But what if, the owner walked out into the field the dog works in, clicked the dog for some offered behavior, and tossed that same piece of chicken on the ground in front of him. I think the dog’s answer might be very different, and it may have nothing at all to do with the quality of the food reinforcement.
In my life as a trainer, there are many, many times that I make training decisions because they feel right to me. I try what I think will work, and I ask the dog. If the dog gave me the behavior I was hoping for, I figure his answer was yes, but why?
I know that my dogs prefer to be handled with a harness rather than a collar. I know they prefer that I use a long line instead of a six-foot leash. I know that they prefer training done off-leash. I know that they prefer working in a large area. I know they prefer to work with me on the ground they are responsible for. I know that if I walk away from something that frightens the dog, rather than toward it, the dog will feel safer. All of this makes a lot more sense to me now that I have this other part of the puzzle, that an animal’s physical autonomy is a primary reinforcer.
I have wanted to write about Sasha for such a long time! I will share with you here just a piece of her story; I will write it up in detail in one of my upcoming books.
Sasha is an eighteen-month-old Sarplaninac dog who was in training with me for three months. Weaned at six-weeks-of age and raised as a single dog on a property without perimeter fencing, by necessity, she was confined more than she wanted to be. She came up with a lot of creative ways to deal with the stress this created for her, and this intensified as she matured. Sasha’s owners adore her. They knew she wasn’t happy in the environment they had for her, so with much regret, they brought her to me so that I could evaluate her with an eye for re-homing her.
Very quickly, it became clear to me that Sasha was not a good candidate for re-homing. I let her owners know that – and you will not believe what they did for this beautiful dog. It makes me teary to think about it.
They fenced three acres of their property for her and built a beautiful barn. They bought her cows, after I taught Sasha how to be trustworthy around cows. Bonnie completed the Foundations course, and she came to the ranch weekly to train with me with Sasha.
After three months of living here with me I was able to send Sasha home with the people who would love her the most for the rest of her life, to safety, and I included two Maremmas of similar age, gifted to them. What is really true, though, is that I also ensured that I get to keep Sasha and her family in my life because my dogs are there! But this isn’t the story I meant to tell!
When Sasha arrived here she was unrecognizable from who she would show me she really was about thirty days later. Take a look at the graphic below.
Transitions and Shaping – Let’s Not Test the Edges https://www.bensonmaremmas.com/post/transitions-and-shaping-let-s-not-test-the-edges
Sasha hovered, almost all the time, just below that reactivity line. Her eyes darted everywhere, her body was so fast. She jumped on me with her feet on my shoulder blades, down, mouth on my arm, up, and around we would go again. She was minimally safe to be near. I used to walk near fences just so that I had better odds of staying on my feet. For Sasha, just calmly breathing was more than she could manage. She had NO canine social skills; even from a distance, she barked at my white dogs with such frenzy that they all returned that energy and were generally spooked by her.
So, here I am, positive reinforcement trainer – and I am almost invisible to this dog. Her autonomy, space around her, and the freedom to charge everywhere are what was reinforcing to her. If I got in her way, in an attempt to slow her down and redirect her, that mouth went around my arm. She would look up at me with a plea in her eyes; no malice, just stress. So I used what I had; I used space, what I now think of as choice and control, because that is what mattered to Sasha.
Many times a day, I would walk into the field with her and just walk, just be with her. Sometimes I didn’t even speak to her because I didn’t want to influence her behavior with things that had previous history; words had been connected to commands. Then I would walk to a gate, and much to her surprise I would open it wide and invite her through. It blew her mind! She’d glance over her shoulder at me, and dash away, and I walked with her.
For two weeks, this is what we did. She sometimes liked to be touched, if she could be still that long. Sometimes she would play ball with me, something I have never seen an LGD breed do. For Sasha, having physical freedom, particularly near a human who wasn’t trying to curtail her activity, was what was the most reinforcing thing that I could give her. Once she started to trust me about that a little, I put a harness on her. That set her back! She was used to being handled from her collar and it took two people, being dragged by her, to move from one place to the next. I put a harness on her, with a long line – and I opened that gate; she shot through it like a rocket. I just tried to keep up with her as she led me all over my barnyard.
Eventually, she paused long enough for me to put a hand on her back, and then walk away from her. She followed, clearly confused, and off we went again. We did this for about a week. She finally figured out that a harness meant we would go on an adventure together. Finally, finally, she paused long enough for me to pop a piece of baked chicken into her mouth. She didn’t spit it out. It wasn’t a big deal to her, but she considered it, and after a few more days, she accepted that chicken showed up on our walks, with no strings attached. Within a few days, I could reinforce her offered behaviors, and we began the language of positive reinforcement training, added to her autonomy.
Over the next few weeks, her body started to relax. I learned how to keep her under threshold so that she didn’t feel she needed to put her mouth on me, and boy, did I have to look back to my science to do this! As she became calmer, and could think, I was able to introduce white dogs – tolerant white dogs – and then cows.
My point here is that even though I did not yet understand that control was a primary reinforcer for all animals, I absolutely knew it was for Sasha, and I knew that if I took control away from her there wasn’t a steak large enough to matter to her.
In my training education, I was taught that in order for a reinforcer to be used to reinforce a behavior, it must be linked to the behavior through repetitions. It is easy to click, and then poke food in a dog’s mouth, but it is more difficult to link freedom and control to a dog’s behavior in a way that they understand is connected to their action. I worked and experimented with how to create that with Sasha. It was fascinating. She is, by far, the most difficult dog to work with that I have ever known, or she was. Now?
Sasha lives with the two white dogs; we call them “Sasha and her bookends” because the dogs flank her and travel everywhere with her. She travels calmly, with a relaxed frame. She rarely puts her feet on anyone and I haven’t seen her use her mouth for a long, long time. She is a sweet, happy dog who is a pleasure to know.
Before Sasha came to me, she had been exposed to three other trainers. Her owners were advised to use a shock collar, a prong collar, and the list goes on. They knew better than to follow this advice, so they kept looking for a solution that honored Sasha. I don’t even remember how they found me, but I am so glad they did. I learned so much from Sasha, and I learned about love and dedication as I watched the choices Sasha’s owners made for her.
The Kim puppy
This is a short story, and then I will wrap up this super-long blog post!
Two of Kim’s pups came to me at eight weeks of age. My plan was to partner them with the Shelby pups, blog post below. When all the pups were about ten weeks of age, visitors began to come to take a look at the Shelby pups.
The stranger walks in, one pup shoots into the field next door. I didn’t know he felt this way about strangers; if I had, I would have orchestrated this encounter differently. But he bolted. I sent the stranger away from the pup to visit with the group, and I went to the scared pup. I walked near him and then past him, headed away from the stranger, and I simply stood near him. I didn’t speak to him because I didn’t want to influence his behavior. I wanted him to be able to think his way through this puzzle, knowing that he was physically safe and that I would make sure of it. And then I asked the stranger to head out the gate.
The next time a stranger visited, I was ready. The stranger came in slowly, the pup moved away slowly. I went to him and then past him – the pup glanced up at me and back at the stranger and made the decision to go investigate the stranger rather than stay with me. How cool is that?! An animal that feels safe can afford to be curious. I honored this pup by supporting his choice and control, and he felt safe enough to learn confidence around new people because of it.
Clicker Training Up Close – The Shelby & Kim Pups https://www.bensonmaremmas.com/post/clicker-training-up-close-the-shelby-kim-pups
Positive reinforcement training can be coercive. It isn’t always appropriate to mark and reinforce. It isn’t the only training option, and the more I learn about control as a primary reinforcer the less often I get out my clicker. If I can change the antecedents to change the behavior, that is my first choice, and I can almost always find a way to do this. There are times, lots of them, when clicker training is appropriate and fun for everyone, but with LGDs I look for ways to be a working partner first, if I can, especially with mature dogs.