Tuesday, December 29, 2009
"What a hoot!" Lure Coursing
We took the dogs to a friend's house to lure course and swim today. They all had a blast and are sleeping like puppies. See for yourself: Here's a video and here's another.
Sunday, December 27, 2009
Squirrel Woes
I've been really careful with Pilot to try to keep her focused on object play (ball) and not on the rodentia family. For the first time in 33 years of living in Capitola we are having regular squirrel visitors. The terriers have been racing out to the yard to see if they can see one as it is climbing the trees. It is very much like a scene out of the movie "Up!" where everything else stops when the most exciting word "Squirrel!" is heard.
The second piece to this is that we went out walking on Christmas and Pilot took off under a fence into a pasture with a long horn steer in it. (Why would someone have a long horn steer in Soquel?) Amazingly, Pilot did not bark at the steer and she came back within a minute or so, but it could have been faster. I had visions of coyotes dancing in my head, as a friend's Jack Russell got grabbed by the neck by a coyote a couple of weeks ago. Yes sir! I do want a faster recall.
I sallied forth today under a cloudy cool sky. I put Pilot on a retractable leash and we spent about 30 minutes doing "yo-yo" recalls, out and back over and over again. If she did not respond, I quietly said "Oops!" and I reeled in the leash and limited her scope for 30 seconds, then I called her again, marked her head turn, asked for a sit, rewarded, praised and scratched her neck (collar grab desensitizing), played some tug and then released her and the leash with a "Go Play!".
I use two releases. "Break!" means you are released from this behavior but stay with me because we are still training. "Go Play!" means stay close but go do whatever you want to until I call you again. Pilot immediately went off to check out the "grundoons" (gophers) and I'd wait until she lifted her head up and call her. We got about 85 to 90% sucessful responses, which is reasonably good. Even so, I will be practicing this daily. Practice makes perfect.
Another way to look at this is that we were using the Premack Principle or life rewards. Premack is also known as Grandma's Rule "Eat your vegetables and you can have desert." in Pilot's case this is "Come when you are called and you can go check out the gopher holes". Salient consequences are important in getting the behaviors you want. I've had people tell me that you are playing with a fire when you are trying to use Premack to trump genetics with food or toy rewards, but I've seen it used effectively with prey drive. Rachel Sanders allows her Jack Russell, Trump, to go after gophers as a game, but she can also end that and get her dog back on a toy any time she asks. It is effective if you are deft at controlling resources as rewards and your relationship and training is stronger than the arousal of prey drive.
Genetics and breed play a big role in how your dog reacts to the world, but training and management can trump genetics as long as you are willing to do the practice - and in our case it is all about play!
The second piece to this is that we went out walking on Christmas and Pilot took off under a fence into a pasture with a long horn steer in it. (Why would someone have a long horn steer in Soquel?) Amazingly, Pilot did not bark at the steer and she came back within a minute or so, but it could have been faster. I had visions of coyotes dancing in my head, as a friend's Jack Russell got grabbed by the neck by a coyote a couple of weeks ago. Yes sir! I do want a faster recall.
I sallied forth today under a cloudy cool sky. I put Pilot on a retractable leash and we spent about 30 minutes doing "yo-yo" recalls, out and back over and over again. If she did not respond, I quietly said "Oops!" and I reeled in the leash and limited her scope for 30 seconds, then I called her again, marked her head turn, asked for a sit, rewarded, praised and scratched her neck (collar grab desensitizing), played some tug and then released her and the leash with a "Go Play!".
I use two releases. "Break!" means you are released from this behavior but stay with me because we are still training. "Go Play!" means stay close but go do whatever you want to until I call you again. Pilot immediately went off to check out the "grundoons" (gophers) and I'd wait until she lifted her head up and call her. We got about 85 to 90% sucessful responses, which is reasonably good. Even so, I will be practicing this daily. Practice makes perfect.
Another way to look at this is that we were using the Premack Principle or life rewards. Premack is also known as Grandma's Rule "Eat your vegetables and you can have desert." in Pilot's case this is "Come when you are called and you can go check out the gopher holes". Salient consequences are important in getting the behaviors you want. I've had people tell me that you are playing with a fire when you are trying to use Premack to trump genetics with food or toy rewards, but I've seen it used effectively with prey drive. Rachel Sanders allows her Jack Russell, Trump, to go after gophers as a game, but she can also end that and get her dog back on a toy any time she asks. It is effective if you are deft at controlling resources as rewards and your relationship and training is stronger than the arousal of prey drive.
Genetics and breed play a big role in how your dog reacts to the world, but training and management can trump genetics as long as you are willing to do the practice - and in our case it is all about play!
Thursday, December 24, 2009
Swing to Heel
Swing to Heel is such a pretty behavior! We've been working on shaping this. Shaping means getting closer and closer approximations to the behavior we want. I had a friend who said she got a swing by using a stick to push the dog into place and I thought that sounded a little too coercive for my taste. I'd rather get it without forcing it.
I'm having to break it into a lot more steps than I thought would be needed when I initially started training. Then again, I know the goal behavior and the dog does not. We train in 2 - 3 minute increments a couple of times a day - keeping sessions short - which allows me to work with the other dogs in the small amounts of time that is available. When shaping, I cut the dog's food back by 30 to 50%. I use Natural Balance (available locally at Pet Pals and Aptos Feed) mixed with a variety of 4 or 5 other treats to reward the dog.
First the dogs are to put their 2 front feet on a low box or an air cushion. I click and treat when they sniff the box at first then put one paw on, then finally 2 paws. I'll throw a small crumb to get the dog to "reset" by going away from the box, then I will click them when they come back and put 2 paws on again. I repeat this for a couple of sessions. I want the dog to be sure of what is being clicked before I move on. All of this is done with very little talking. If the dog is unsure, I will praise them as I deliver the reward. If the dog doesn't get it, then I have to go backwards to lower my criteria or I have to break it down and make the task simpler. If I were to constantly correct the dog then the dog would get discouraged. I keep the reinforcement rate quite high to build the dog's confidence and to make sure they have a reinforcement history that will keep the learning curve going up.
I worked with Pilot's sister, Remi, on this step today. She has never done shaping and it was a new concept for her. She caught on relatively quickly but I ran the session a little too long and her ability to concentrate started waning. Note to self: Use kitchen timer when training.
In the second phase of training a swing we click for every small motion they make with their back feet. I do not want to have to use my body motion to get them to move, but it does jump start the process, especially if the dog has not done a lot of shaping before. If I step across with my outside foot I've found I need very little motion to influence the dog to move it's back feet. If I use motion, I know that I am pressuring the dog, the dog is not the one who is doing the thinking and the dog will become reliant on that movement to accomplish the task. My goal is to mimimize the movement and get rid of that prompt as quickly as I can. I found this stage went a little faster when I did some line-ups for regular heeling just prior to and after each training session.
In the third phase, once I have the motion down in one direction, after one or two training sessions, I start working the other direction. Pilot was much better in both directions than KK. KK goes well in the counter-clockwise direction (to heel position) than clockwise (to side position). One of the best reasons to work on this behavior is to improve coordination and proprioception in both directions.
I keep a training journal where I make a couple of lines of notes on each training session with drawings if needed. It helps me in planning the next session.
Each day I am getting more motion and Pilot's heel (left) and side (right) line-up positions without the training props has become much better as well.
I love watching the dog's figure it out. It takes patience to let them work it out for themselves, but it is sweet when they really learn the behavior!
Take a look at this video on training a swing. I think this pup has been working on this behvior for a couple of weeks. This pup is a student of Sylvia Trkman, a world class agility competitor who uses tricks to train her dogs body awareness, muscle development and pure fun! A friend who has a giant breed (over 100 lbs) and did not want her dog leaning against her leg as the dog does in the video. You'll have to make a decision about this part of the behavior for yourself. I asked Sylvia about it recently when I attended her workshop and she said she wanted her dogs to have the tactile sensation of touching her leg - then again her dogs are under 20 lbs.
All in all, it is very fun to see the dog's learning so quickly!
I'm having to break it into a lot more steps than I thought would be needed when I initially started training. Then again, I know the goal behavior and the dog does not. We train in 2 - 3 minute increments a couple of times a day - keeping sessions short - which allows me to work with the other dogs in the small amounts of time that is available. When shaping, I cut the dog's food back by 30 to 50%. I use Natural Balance (available locally at Pet Pals and Aptos Feed) mixed with a variety of 4 or 5 other treats to reward the dog.
First the dogs are to put their 2 front feet on a low box or an air cushion. I click and treat when they sniff the box at first then put one paw on, then finally 2 paws. I'll throw a small crumb to get the dog to "reset" by going away from the box, then I will click them when they come back and put 2 paws on again. I repeat this for a couple of sessions. I want the dog to be sure of what is being clicked before I move on. All of this is done with very little talking. If the dog is unsure, I will praise them as I deliver the reward. If the dog doesn't get it, then I have to go backwards to lower my criteria or I have to break it down and make the task simpler. If I were to constantly correct the dog then the dog would get discouraged. I keep the reinforcement rate quite high to build the dog's confidence and to make sure they have a reinforcement history that will keep the learning curve going up.
I worked with Pilot's sister, Remi, on this step today. She has never done shaping and it was a new concept for her. She caught on relatively quickly but I ran the session a little too long and her ability to concentrate started waning. Note to self: Use kitchen timer when training.
In the second phase of training a swing we click for every small motion they make with their back feet. I do not want to have to use my body motion to get them to move, but it does jump start the process, especially if the dog has not done a lot of shaping before. If I step across with my outside foot I've found I need very little motion to influence the dog to move it's back feet. If I use motion, I know that I am pressuring the dog, the dog is not the one who is doing the thinking and the dog will become reliant on that movement to accomplish the task. My goal is to mimimize the movement and get rid of that prompt as quickly as I can. I found this stage went a little faster when I did some line-ups for regular heeling just prior to and after each training session.
In the third phase, once I have the motion down in one direction, after one or two training sessions, I start working the other direction. Pilot was much better in both directions than KK. KK goes well in the counter-clockwise direction (to heel position) than clockwise (to side position). One of the best reasons to work on this behavior is to improve coordination and proprioception in both directions.
I keep a training journal where I make a couple of lines of notes on each training session with drawings if needed. It helps me in planning the next session.
Each day I am getting more motion and Pilot's heel (left) and side (right) line-up positions without the training props has become much better as well.
I love watching the dog's figure it out. It takes patience to let them work it out for themselves, but it is sweet when they really learn the behavior!
Take a look at this video on training a swing. I think this pup has been working on this behvior for a couple of weeks. This pup is a student of Sylvia Trkman, a world class agility competitor who uses tricks to train her dogs body awareness, muscle development and pure fun! A friend who has a giant breed (over 100 lbs) and did not want her dog leaning against her leg as the dog does in the video. You'll have to make a decision about this part of the behavior for yourself. I asked Sylvia about it recently when I attended her workshop and she said she wanted her dogs to have the tactile sensation of touching her leg - then again her dogs are under 20 lbs.
All in all, it is very fun to see the dog's learning so quickly!
Subscribe to:
Posts (Atom)