Sunday, January 23, 2011

Puppy Training or What I Know Now that I Wish I Had Known Then!

Having a puppy around is always a time of reflection for me on my training techniques - what I want to do differently with this puppy, what this puppy needs done differently than others and what I can learn from this puppy.

Right now at ACTS we have a lot of puppies in various stages of training to be agility dogs. It is an exciting time. We are working on improving and growing our "sports foundations" program to see how we can best prepare students for the intricacies and demands of agility training as well as for the rigors of daily life training. So right now I'm spending a lot of time listening to my students, getting their feedback, tweaking the program and watching students and puppies. We are trying to do our training as much as possible without training aids such as prong collars, gentle leaders or choke collars. However there may be times when these will be needed in certain circumstances or for short periods of time. Our goal is to help students wean off of these training aids as much as possible. They can't be used in competitive sports so they need to be weaned off at some point and sooner is better. They can become crutches for the handlers if used too long and can be a poor substitute for good training if used for too long.

We try hard to help students master the art and science of shaping or what is more popularly known as clicker training. Too often students use the clicker as a positive marker for existing behaviors that have been lured. To really develop the skill of capturing a behavior takes a lot of practice. It is a really fun thing to do once the skill is developed. Thinking about an end behavior and then breaking it down into the component pieces is very helpful for training but a difficult skill to learn. It really makes you think about how many stages there are in learning a particular behavior. Getting stuck in a spot is a common problem and there are many ways to work out of it. Using tools like strategic placement of reward can be very helpful.

It is really important for my puppies and my student's puppies to learn to eagerly go and interact with objects. I find that dogs who are willing to do this and are not afraid to make mistakes will have a much easier time learning agility and learning to do it independently. If a dog has to be lured across a board, or lured through weave poles the dog is going to be thinking more about the food in the hand than about what they are doing. The food then has to be faded. A dog that is eager to go offer behaviors is more likely to be willing to walk across a board in order to get a treat or try to go through weaves to get the treat or toy. These dogs will be easier to teach to do obstacles at a distance when they are trained this way from the beginning. Will their performance be a little sloppy at first - yes. But with any of my training I would rather have a lot of enthusiasm and sloppiness than a lot of worry and carefulness. It is much easier to refine a dog's behavior who has a lot of enthusiasm and sloppiness than it is to excite a worried and careful dog. The worried dog may gain confidence over time and with that gain speed with the right training but the training has to be done with careful planning to ensure that happens and is more work for the handler.

I've been shaping behavior with dogs since the early 90s. I graduated from college with a degrees in Psychology and Biology with an emphasis in Animal Learning and Behavior and then went on to get a Masters in Educational Psychology focusing on Adult Development and Learning. Before getting into dog training I had a lot of experience with the "science" of learning. It is different from real life learning and training but it has been helpful to have worked with animals like rats, pigeons, mice and chameleons and to apply what worked with them to dogs. The big difference is that dogs live with us every day and all day. They don't live in cages 24 hours a day. So it is really important for dogs to have structure and boundaries in their lives. While some of this can be taught with operant/shaping techniques, much of it can not be. So there needs to be a balance and a thoughtful use of different types of training techniques for different situations. It would be nice to think we could do everything with shaping - click and treat but the reality in my 20 years of experience working with dogs and their humans is that does not provide enough structure for most dogs. Dogs do have free will. Dogs do have minds of their own. There are times when a dog will choose for very good reasons (in their mind) to not follow a cue/command. It is our job as the human part of the team to discern whether that reason was due to a high level of stress, due to high level of distraction, due to a lack of understanding of the cue, due to the dog's desire/preference to do something else or due to the dog's physical inability to do it (due to soreness or illness). It is this analysis that a human has to learn to do very quickly in order to determine how to respond. Sometimes we get it wrong and sometimes we are totally confused. If a dog is choosing to not follow our cues/commands because of free will then there may need to be consequences. However I never impose consequences on something UNLESS and UNTIL I am sure the dog understands the cue and can do the cue with fluency and with distractions and/or distance. Too often students want to introduce corrections before a dog has even learned the behavior by saying "no" or "eh eh" to the dog while they are teaching them something. Not only is this not fair to the dog but it can also lessen the impact of your negative marker when it is over used. I use clearing my throat as my negative marker - it is hard for that to slip out accidentally but I don't use it until I'm very certain the dog can do what I ask in the situation I am asking for it. If the distraction level is too high then I need to work more on that behavior with distractions before I can correct it.

I spend a lot of time analyzing my dog's behavior in many situations and trying to determine the underlying cause of a behavioral problem so I can determine how to best "fix it." This is when it is important to spend time getting to know a puppy. Finding out who they are, how they think, what they like, what they don't like, how they learn and how they feel physically is so important. I don't expect my dogs to respond the same even if they are the same breed or even closely related. They are all individuals. It is important to distinguish whether a dog/puppy is stressed or not. Sometimes people are too quick to say a dog is "blowing me off" when the dog is really stressed and doing what dogs do when they are stressed - sniff! Sometimes a dog is too excited in the situation to be able to focus on the desired behavior. A student needs to look at how much have they trained and proofed the dog to be able to do that behavior with a lot of distractions. Often we don't look honestly at our training programs to determine how well have we taught a given behavior. Dogs don't generalize well so they need to have lots of different experiences in order to generalize. Dogs get excited and need to know how to have self-control in stimulating environments. This all starts when they are puppies. Learn the signs of stress in your dog. Licking lips is another common sign of stress. I see it a lot in dogs on agility courses but the table, teeter and weaves. If you slow down a video of a dog in the weaves very often you can see them licking their lips. This is a huge sign of stress and should be a concern. A very low level of stress is common and should not interfere with learning but it also should not produce a lot of outward signs. The more visible the signs of stress the higher the level is.

Having this litter of 9 puppies was one of the most educational dog experiences of my life. I could see first hand what kinds of things come "hard wired" in a dog by how they responded to things within the first two weeks. No formal learning from a human was taking place - only their own learning and interaction with the environment and processing that information was happening. I learned so much about how puppies develop and how they think and how they interact with novel stimuli and familiar stimuli from day to day. I was able to see one of the puppies was very sound sensitive and easily stressed in new places with lots of people. I immediately did a lot of intervention with this puppy to help him overcome these things. I made sure his owner new of my concerns so she could work with him from the very beginning to build his confidence. I knew which two puppies were almost over the top in energy level as they would be the only two still running and playing after two hours at a puppy party. I knew which ones were going to be easy to handle and which ones didn't like to be touched physically and were going to need lots of work with this. I knew which ones were more independent and which ones were likely going to be "one person" dogs and which ones would be good with kids and families. It was really an incredible experience. Over the course of 9 - 10 weeks I could see as they got older how they learned. I could see which ones were leaders in exploring new things, which ones were followers and which ones were watchers. As I get to see a couple of the puppies on a regular basis and work with their owners I can see how different they are and I can also see similarities to their mother - but different similarities between the puppies.

While it may be hard to see now, many people have forgotten what Sinco was like for the first two years. She was a boisterous puppy. She had a hard time with mistakes and would run around and act silly when she thought she had made a mistake or she would freeze up. However every experienced person I took her to for lessons said she was "high drive." I often shook my head because I couldn't see it since she would freeze up in learning because of her fear of being wrong. Even though I had never used even as much as a negative marker with her in training. She had a hard time with shaping. She would be reluctant to offer behaviors. I would get frustrated and lure her a couple of times and then she would get going. When she was older, about three, I started shaping games with her again and she was much more confident and much less worried and now I can shape things and she will eagerly go interact with an object if I tell her "show me something else."

Now when Carmine and I started shaping I saw her do some of the same things as Sinco. Fortunately I was at Michelle Pouliout seminar when I saw it happen. Michelle had some great ideas (I wish I had known when Sinco was a puppy). She suggested tying a string to the box or object I was trying to get Carmine to interact with and move it a bit and when she showed interest click and treat for it. She also suggested changing the object frequently and have 2-3 objects around at a time to keep it interesting. Within a week of using these different things I had a puppy who was wildly interacting with any object and not at all afraid of making a mistake. She is now an incredibly fast learner with a clicker in hand.

The other lesson I learned from my Border Collie Bryce is to put shaped behaviors on cue as soon as possible. He would offer all kinds of behaviors to me in rapid succession. He was very fun to train but I had to get things on cue quickly to prevent him from always offering his repertoire. I am having to do that with Carmine now and I've reduced how many things I teach her at a time and put things on cue before adding new things to the "list."

Tay reminded me that dogs can learn differently and have different kinds and levels of intelligence. Tay loves shaping exercises. However she is very dependent on kinesthetic learning and having an object present is a much more salient cue to her than any thing verbal or my body language. She has a hard time focusing on my body language or my verbal cues to do a behavior. She can learn very well a set behavior with a certain object. It is very hard to get her to do something different with the same object. For example I taught her to jump on a chair as an exercise to do after a run so she can get her treats. It is almost impossible to get her to do anything else with a chair present and she has generalized this to ANY chair she sees ANY time. She has a very hard time with behaviors that stand alone without any objects in the vicinity. Stay is a very hard concept for her except on a platform or a mat. She has the most reliable and awesome door behavior of any of my dogs - EVERY door she goes through she turns toward me. I rarely reinforce it with a treat but she does it almost every time.

Spring has taught me about "show me the money" games. Michelle Pouliout in her seminar talked about not having dogs see you put the food in your pockets when you are getting ready to train. Spring is definitely one of these dogs. If he sees me go for the treat jar or my pocket he will be right there with me. But he plays games at home and at trials where he won't come when called. It pushes my buttons to have a dog not come when called. It is especially strange because he and I have a great relationship otherwise. He is very fun to train and to run at trials except for the table in AKC. He has been taught "go leash" which he will do and 90 percent of the time it is rewarded with treats pretty soon after at trials or in training. So as long as I say "let's go leash" at the end of the run he will come running over and put his head in his slip lead. I make him go to the treats on leash. At home, only when he knows I'm loading up the van to go somewhere, will he play "I am going to run outside through the pet door when you call me" game. It is ironic because he loves to be a lap dog, loves to be cuddled, loves to be held, loves to have his neck scratched and he loves to go places with me. If he knows I have treats he won't play the game. But if he knows I don't have treats like in a trial, he plays the game. So I have taken to being very careful around him about what I am doing with regard to my treats. It is clearly a game. It is very different from when other dogs won't come. He is watching me the entire time, if I go out of sight he will slowly follow me and watch me and if I turn toward him he darts away. It is in his mind a very interactive game of cat and mouse. I have never seen anything quite like it and now it has carried over to trials where the table is involved. So I am trying to connect these two things and I think if I can resolve the problem game at home then I will be able to solve it at a trial. I've had a lot of advice from "walking him down" and grabbing his ruff to using air cookies. I am not fond of any of these because I don't think it will solve this kind of problem. He will go into his crate any time and any where, he will come when called in many other situations very readily. He is at my feet most of the time wanting my attention.

So I am extremely careful with puppies now to make sure they don't know when I have treats and when I don't for recalls. Often I call a puppy in and they have to run into their crate and then I will go and get a treat for them. I don't want to ever show them I have a treat when I'm calling them.

Agility training is difficult to prepare for because there are a lot of complex skills that are also somewhat abstract. An ideal agility dog will be able to have focus and speed at the same time and to have self-control and a willingness to work independently (meaning not overly dependent on a handler for cues or to have to be in close proximity to the handler). These can at first appear to be conflicting skills. They are in fact difficult skills for young puppies to master and definitely difficult for them to have all of them operating at the same time. All of this must be kept in mind. I play a variety of games with my puppies to turn them "on" and turn them "off." I want them to learn to self-modulate depending on various cues from me. I rev them up and then turn them off (very briefly at first) and then rev them up and turn them off. Some dogs have more natural self-control and need more help revving up and others get revved up easily and need more help with self-control. That is why it is important to work on both of these skills early and often. The games we teach in sports foundations help to develop these skills. I also have some games that teach focused speed too. I had a dog with a lot of speed and very little focus and it was a lot of work out there on an agility course to get through a course. He would see his own set of obstacles and he couldn't focus on his handler for cues at times or focus on his job in performing obstacles correctly. So focus with speed is important to develop early.

Most puppies in our classes are on average 9 months old when they move from "sports foundations" to "agility foundations." In agility foundations we do a lot of flatwork for the first 8 weeks. We do shadow handling, targeting/distance exercises, more focus and speed exercises, backing-up and learning to work with distractions. Actual agility equipment is not introduced until they are close to a year old. They have seen wobble boards, played on narrow boards, gone through hoops and around gates at this point. Developing a good working relationship and laying the groundwork for handling skills is so very important. Teaching the individual obstacles is the easy part. Building teamwork takes much longer so it is important to start that early and work on it often.

Usually by 7-9 months I can tell if a given puppy is going to have issues reacting to the motion around them of an agility class. I manage the puppies in agility foundations to minimize the stimulation and keep arousal levels down. However I am watching them for signs of problems with that and I want to intervene as early as possible if I see a puppy having trouble focusing with dogs running around. Then I feel it is best to stop agility training and work on some of the Control Unleashed/Click to Calm exercises and maybe even take our Focus in Motion class to really work on that. If a dog is getting aroused around motion there is no point in training them. They will have too much adrenaline in their system to be able to learn anything properly.

The other thing is that puppies do not fully physically mature until at least a year of age - no matter what breed. Some tiny breeds may physically be done growing earlier than that. The mental maturity in most dogs takes even longer to develop. There is no point in doing actual full size agility equipment until a dog is both physically and mentally mature. Too often I see dogs who started weave training at six months of age. They will come to me when the dogs are 12 months old or 18 months old wondering why their dog can't weave. Their bodies change, even though they may be done growing in terms of height, they still will need to fill out, their chests "drop" and muscles get stronger. These changes seriously affect things like weave poles and running contacts which rely on muscle memory. So I see no point in starting the training of these obstacles early - they will only have to be retrained in a few months or a year because their bodies have changed along the way. You may also cause undue physical and emotional stress on a dog training them this young to do complex behaviors that require a lot of balance and coordination.

Right now my puppy knows a number of tricks, she is great at interacting and offering behaviors with any object, she loves to retrieve any object (not on cue yet), she knows "ready 1-2-3" and "ready-steady" games, she has a stay on a mat, she can do a stand stay on a platform, she does recalls through my legs, she knows "sit" and "down" from a stand, she can send to a target 20 feet away, she does hand targeting, she is learning "choose to heel" and loose leash walking, she has done some cone turns (to stay one step ahead of her contemporaries so she can demo!) she is learning to be calm when greeting people and their dogs, she loves crate games, she has awesome toy drive, she can be quiet in a crate while I'm teaching class and she is learning to ride a skate board and play on the wobble board. She is over 7 months old. That is all she knows. She has yet to get on a formal piece of agility equipment.

What she will be taught in the next few months is the foundation for a two on/two off as well as the foundation for running contacts. I plan to try to train her to do running contacts but will teach her two on/two off on a board as a back-up plan. She has a very long body and long stride so I'm not sure how well the running contacts will go but I need a new challenge. I've taught two small dogs so now to try a big dog and she is a big girl! She will learn some cone turns and shadow handling. She will do some shadow handling with cones and then with hoops. She will do some puppy jump chutes. She will learn "out" with gates and targets.

I know I will learn new things from Carmine and make new mistakes... That is what makes dog training fun for me. Sometimes the learning is harder than I want and sometimes it is not as fun as I want it to be but when we come out the other end of the tunnel it is almost always very rewarding. So I can hardly wait to see what lessons are in store for me now!