The "Art" of reinforcement

Art? Is proper reinforcement really an art? I believe it is, for some it comes incredibly naturally, for others it is something that requires much thought and effort, and, for a small number of individuals it is almost impossible to master. I am one of the lucky ones, reinforcement comes very naturally to me, whether human or animal it is very clear to me that learning occurs much faster in a rewarding environment. It goes much deeper than that though :-) The reasons for reinforcement are many: we reinforce to motivate, we reinforce to maintain motivation, we reinforce as a catalyst for learning. I am fairly certain that everyone understands these points, where the problem lies is truly understanding the "placement" of reward to truly reap the benefits of reinforcement as a training tool :-) Before I go any deeper into this blog however, let's deal with the elephant in the room… The portion of the dog training population that feels that rewards are unnecessary in dog training. I know that over the years many dogs have been, and, are still trained without the use of tangible reinforcers, that does not make it right, without tangible reinforcements it is incredibly difficult for a dog to learn to make correct choices based on clear understanding rather than fear of failure. I think there is a lot of learned helplessness that occurs when we expect dogs to perform without a true understanding of the behaviors we are asking for. Learned helplessness is not having the ability to choose the right or wrong decision because the difference between the 2 has never been taught, the dog is in a perpetual state of confusion. Different temperaments will deal with this condition in different ways, from total shut down, to taking control and running amuck. So, we not only need to be willing to use tangible reinforcers, we must have them readily available to mark proper responses, and, most importantly, know where and when to reinforce. In training situations it is not unlikely to see a handler withhold all reward until the end of whatever sequence they are working on, the problems with this are many. 1. You are missing so many fantastic opportunities to reward great things that occurred prior to the end of the sequence, i.e.. a turn that was tighter than before, a perfect contact performance, etc.. 2. By ignoring the great things that happened on course and only rewarding the end you are reinforcing exactly that: the end! There are many dogs that are a bit droopy all the way around the course and perk up amazingly at the end of the run, why do you suppose this is? 3. You are missing opportunities to build value for the obstacles which translates into speed, drive, and, enthusiasm. Once you have truly built value for the obstacles your dog will never notice that the toys and treats are not showing up at trials. I have even heard people say that they will reward the dog when "they" (the handler) gets it right! I think the whole wrongness of this statement is pretty self explanatory and needs no further discussion. It is also argued that you cannot use toys and treats at a trial so the dog needs to learn to work without them. I have a huge problem with this statement, as stated above, one of the benefits of frequent properly placed reinforcements is that they create drive, and, build value for the whole process. In reality if you think bout it you are training at least 75% of the time and trailing maybe 25%, with those differentials how sad to be missing all of those opportunities to create the fast, confident, amazing agility partner we are all looking for! On a happier note, let's talk about when and where to reinforce :-)Proper placement and timing of rewards will in and of itself help build drive and maintain motivation simply because it breeds clarity! You should be rewarding your dog for all behaviors that you would like to continue, avoid reinforcing behaviors that you would like to extinguish. That might seem overly simplified, however, if we could keep our training simple it would be far easier! When you make a mistake, you should reward your dog! The behaviors that your dogs are offering are based on the information you are providing, if that information is contradictory to what your intentions were, the dog is still correct as they were reading what you were saying/showing not what you were thinking! When your dog has a particularly difficult time with a behavior and they finally get it, REWARD your dog! There are definitely places that reinforcement is not indicated, one of the most commonly mistaken places to reward is when the dog deflates and we try to use food to get them motivated, if you are feeding the dog for slowing down, or, for doing nothing you are perpetuating the problem. Reward the dog for working, just be sure that the paycheck not only reflects the level of work you are expecting, but, also what the dog considers fair wage. When you lose a dog to distractions, especially sniffing, hunting for food, etc.. they are telling you that they are not getting paid enough, listen to your dog they have very valuable information to share with you that will make your job as trainer and handler much easier! I realize this blog barely scratches the surface where reinforcement is concerned, it is a topic that could use a class of its own! Maybe that is what I will do a class on proper placement and timing of reward :-) Until then I hope it helps at least a little bit to clarify some of the questions about the value and necessity of tangible reinforcement in dog training!