Monday, December 2, 2013

My Training Philosophy

Something that has always interested me greatly is how different dog trainers and dog people train their dogs. What methodology do they use? What methodologies even exist? There are the so-called "Purely Positive" trainers, there are correction based trainers, there are trainers who believe you need to be "dominant" over your dog. As a first time dog owner and an individual who is new to the wold of dog competition I struggle with what exactly my dog training philosophy is. I knew right off the bat that I was not going to be one of those "dominate your dog" type of trainers. I just do not believe that a training program full of fear and intimidation will bring out the best in a dog. When I first started out I thought I was going to be one of those "Purely Positive" dog trainers. It sounded wonderful - nothing, but rewards - I would have the happiest dogs in the world. Then I started learning more about training and what exactly went into the four quadrants of dog training. I feel like the Positive Gun Dog Yahoo group does a great job of describing the four quadrants so I am using their definition of each quadrant:

R+ :  In positive reinforcement, a behavior is strengthened by the appearance of, or an increase in the intensity of, a stimulus.
    Example: Fido is cued to sit, sits, and immediately receives a treat (thus increasing the liklihood that he will sit on cue in the future)

P- :  In negative punishment, a behavior is decreased by the removal of, or a decrease in the intensity of, a stimulus.
    Example:  Fido barks, is immediately put in his crate (thus decreasing the liklihood of future barking)

P+ :  in positive punishment, a behavior is decreased by the appearance of, or an increase in the intensity of, a stimulus.
    Example:  Fido barks, is immediately spanked (thus decreasing the liklihood of future barking)

R- :  In negative reinforcement, a behavior is strengthened by the removal of, or a decrease in the intensity of, a stimulus. Negative reinforcement usually involves avoidance behavior.
    Example: Fido is cued to sit, doesn't sit, receives a stim from the e-collar, he sits and the stim is removed (this is actually a combination of P+ and R-, as the onset of the stim (P+) will decrease the behavior Fido was doing before the stim and the removal of the stim (R-) will increase the liklihood that he will sit on cue in the future)

I started to get confused about what exactly a "Purely Positive" trainer was. Do they use R+ only? Do they use R+ and P-? It did not seem possible to only train with R+, or at least it would be very difficult. If you are only ever giving rewards, how do you eliminate negative behaviors? Simply ignoring a dog when it barks and rewarding with treats and praise when it is quiet isn't necessarily going to be enough to get it to stop the behavior. Barking in itself is a rewarding behavior to a dog, kind of like digging. If a dog can practice self-rewarding behaviors it will never stop the behavior. There has to be some sort of P- (removal of reward) to get the dog to stop, right?

I did not want to jump straight to punishing a dog for unwanted behaviors though. In my mind, it is not fair to punish or correct a dog if they do not understand what you are asking of them. Dogs are not born knowing how to live in our human world. We usually have, high and unrealistic expectations for how dogs should behave. It is not to say that these desires are not attainable, but a lot of the time we expect dogs to act a certain way just because - without ever having shown them to behave that way. This brings me to the main reason why I do not want to use any sort of correction based training with my dogs - more often than not, any unwanted behaviors my dogs exhibit I (or others) have unintentionally created and rewarded this behavior. Jumping, for instance, is a good example. Every time anyone allows my boys to jump on them and gives them any sort of attention (even negative attention - pushing them off and saying "Off") is rewarding them for jumping. In my eyes it is not exactly fair to "reward" my boys for jumping and then immediately punish them. That's awfully confusing to the dog. This happens a lot more than I'm probably even aware of and it is the main reason why I don't want to use corrections with my dogs. Maybe one day, when I am more experienced, and know how to use correction tools the right way I can add them effectively to my training program, but I am definitely not at that level yet. I have seen many trainers use these tools with great success, they have happy dogs who absolutely love training. Bridget Carlson comes to mind. Her dogs and her working relationship with them is admirable. She uses prongs and other training equipment. The difference, though, is she knows how to use them to enhance her training. They are added tools in her toolbox, not the foundation of her training. 

This brings me back to the main purpose of this post, what is my training philosophy? I'm not "Purely Positive," if that even exists, I'm not using fear-tactics or dominance, and I'm not quite at a level where I can use corrections appropriately. So where do I fall? I'm sure it is bound to change over time, but as of right now I really like the idea of following the LIMA principle - Least Intrusive, Minimially Aversive. As stated by the Positive Gun Dogs group "This means that all reasonable R+ methods have been tried or carefully and thoroughly considered before moving to P- methods, and all reasonable P- methods have been tried or carefully and thoroughly considered before moving to P+ or R- methods." I think this methodology will allow me to encorporate the largest amount of positivity into my training. I see no reason to use corrections or aversives to teach behaviors if they can be taught in a positive way. I think it will definitely lead to challenges down the road, since the majority of training strategies out there use corrections, punishment, and aversives (none of which should be considered as one in the same - they all have different meanings). But as Susan Garett says, lets live in the "Land of Do". Lets get creative and teach our dogs what we want them to DO instead of always telling them DON'T. I absolutely love this idea. Instead of telling a dog "No" because they jumped up on a guest, teach them what you want them to do (like "Go to Place" or "Sit"). So I am off to live in the Land of Do and train using the LIMA principle. Wish me luck!

Positively Taught Retrieve

I have decided it is time to get up off the couch and really dive into our hunting training. We have a lot of obedience work to do before we're ready for actual field work, but that is no reason to put off getting our formal retrieve down. We started our formal retrieve in hunt class, but we have picked up some bad habits. Bernie mouths the bumper, carries it on its end or by the string and Oliver doesn't hold or bring the bumper back. So to sum up, we have almost no elements of the formal retrieve. All we have is a very solid "Out". 

I'm not interested in using a force-fetch method as my dog training philosophy ties in with LIMA principle – Least Intrusive, Minimally Aversive. I do not see why a solid retrieve cannot be established using positive methods, so here I am - taking on the challenge of a positively taught hunting retrieve! I am using the following as a training program: I had already started doing something similar as I had watched some videos Denise Fenzi had posted on her site about clicker training a formal retrieve (I believe obedience was the end goal in mind though). As of right now we are probably around Step 9 - "Hold the dumbbell out and click for the only thing in ring seven: mouthing/gripping the dumbbell." I want to repeat this step multiple times in training though because both dogs tend to gravitate towards the end of the bumper when its placed in front of them. I won't be moving on until we consistently have mouth grabs in the center. Right now I have been using a bumper for the training, but I really need to get some dumbbells. It is clear that the dumbbell is a key tool in teaching the dog to grab for the middle of the object. I'm also in the process of really building value for the bumper and taking objects. Bernie has no issue with getting excited for the bumper, the minute it is pulled out he is bouncing off the walls - excited. Oliver, on the other hand, had no interest at first. If I wiggled the bumper around he would get excited and chase it for a short time, but he got bored with it very quickly. I am hoping that doing multiple Click/Treat (C/T) training sessions will build value for the bumper. I believe this is something referred to as the pre-mack principle. Oliver is more of a food-driven dog than a toy/play-driven dog, so every time we train with the bumper and he is getting treated for correctly grabbing it, it is my hope that some of that value he has for food will be transferred to the bumper. It seems to be working. When we first started training with the bumper Oliver was so disinterested I could barely get him to look at it. He knew what he had to do, but he had this look of disgust on his face. Now, today, we are at the point where he enthusiastically grabs for the bumper, no hesitation, no look of disgust. It is going to be a slow process, as we are all very new to this, but I am excited!

To-Do List:
1. Continue with Step 9 of C/T formal retrieve
2. Build Tug-Drive with both boys, especially Oliver
3. Play Susan Garett Recall games on a daily basis - nothing's more important than a solid recall in field work!