Shaping
Tricks, Obedience, Agility, Balance Discs
From the time we get a dog we are thinking about behaviors that we want them to do such as sit, down, wait at the door, wait for dinner, come when called, hand target body awareness work and perhaps tricks. The skills we want to teach our dogs are behaviors that do not occur naturally. Shaping allows us to start at ground zero and create these behaviors that we want.
Here are a few tips for successful shaping…
Use high value reinforcers that are only used for training. For example I use dehydrated steak when training obedience work. This is the only time my dogs get this treat and they value it greatly as well as look forward to our time training together. Both my border collies work much harder and with excellent focus when they know they will be getting this favorite reward. I want my dogs to put in maximum physical and mental effort by having to work for that high value treat.
For less challenging training, such as balance disc work, I use a treat of lesser value.
Treats are the best reinforcers to use for training involving the clicker or verbal marker. When a behavior has been trained such as heeling, the use of toys can also be a reinforcer. In agility toys are often used when training and are excellent reinforcers for building drive. However most agility trainers will say that dogs need to be reinforced with both toys and treats since both have their place in training.
From Karen Pryor's "Getting Started with Clicker Training for Dogs"
Another example would be the Utility Go Outs. It is very important to mark the behavior of the touch to the stantion, wall or pole. This is the behavior you are training and want to be understood by the dog. Therefore it is very important to click when the dog lifts his paw to touch or touches the stantion. Clicking after the touch to the stantion is late. Then you ask for the SIT, walk in and reward ON THE STANTION. By rewarding the exact place you want the dog’s behavior to be, you are helping them to understand exactly what they have done and the position they must be in to be correct. I have noticed that in teaching the touch for go outs in obedience that placing the reward on the stantion grew the understanding of this behavior much faster and with fewer wrong attempts.
You will need to identify the end behavior and then think of all the ways to break this behavior down into tiny component parts. The video that follows is one example of how a final behavior of 4 feet in a bowl was broken down into small parts and taught over time.
I have noticed that some of the tricks that I teach take many months to perfect. An example of this is SIT PRETTY and HOLD A BOTTLE or TOY. Myst had to learn to sit pretty which took a few weeks; hold her balance in the sit pretty which took longer and then learn to hold an object which took the longest for her. We just kept working on this a little at a time and rewarded tiny steps for holding the object in sit pretty.
What toys or tools do you need? What reinforcer is appropriate for the training? How can you manipulate the training environment to help the dog be successful? AND, when
In my sessions I work both dogs; one sits and waits while the other trains. This way I can alternate dogs and keep training sessions short. It also provide an excellent opportunity to reward the good behavior of a sit and wait.
Sometimes I will lure with a treat for the first few times but the lure is faded very quickly. You will see this in the video below of how I taught Myst the 4 feet in the bowl for the first 2 times she was encouraged to jump into the boxes.
An example of this is when teaching fronts and finishes in obedience. Once the dog understands what is a correct front or finish, they should not be rewarded for anything less than perfect otherwise they will never know what is that perfect performance. Many novice trainers are so excited to see their dogs just performing a behavior CLOSE to what they are looking for, that they reward. The dog never learns to do this skill with perfection.
In obedience this negative reinforcer might be for a slow drop or no drop on recall; a crooked front or finish when the dog knows what is correct. A Utility Go Out where the dog turns and sits and does not touch the stantion or does not perform the touch to the stantion. If these incorrect behaviors are not marked the dog has no way of knowing what is the correct or perfect performance.
Most of all, HAVE FUN AND KEEP SMILING to let the dog know you are having fun. End each training session with a fun interactive game.
HAPPY TRAINING
VIDEO 1: Shaping - How I Taught the 4 Feet in the Bowl Trick
Tricks, Obedience, Agility, Balance Discs
From the time we get a dog we are thinking about behaviors that we want them to do such as sit, down, wait at the door, wait for dinner, come when called, hand target body awareness work and perhaps tricks. The skills we want to teach our dogs are behaviors that do not occur naturally. Shaping allows us to start at ground zero and create these behaviors that we want.
Here are a few tips for successful shaping…
- You will need to use a clicker or a verbal marker such as Yes or Good to let the dog know that they have performed the behavior you are asking for. The use of the clicker or verbal marker is known as a conditioned reinforcer. The clicker's value is in shaping a new behavior or refining details.
- You will reward the behavior immediately after the conditioned reinforce; this is called the primary reinforcer and is usually a high value food treat.
- In order to determine what reinforcer reward you will use for your dog, think about what favorite food motivates them. Or perhaps a food you have never used before. Dogs are motivated to try to do a behavior if the value of that performance is associated with a high value reinforce such as a piece of steak or cheese.
Use high value reinforcers that are only used for training. For example I use dehydrated steak when training obedience work. This is the only time my dogs get this treat and they value it greatly as well as look forward to our time training together. Both my border collies work much harder and with excellent focus when they know they will be getting this favorite reward. I want my dogs to put in maximum physical and mental effort by having to work for that high value treat.
For less challenging training, such as balance disc work, I use a treat of lesser value.
Treats are the best reinforcers to use for training involving the clicker or verbal marker. When a behavior has been trained such as heeling, the use of toys can also be a reinforcer. In agility toys are often used when training and are excellent reinforcers for building drive. However most agility trainers will say that dogs need to be reinforced with both toys and treats since both have their place in training.
- The timing of the click or verbal marker is very important. The click of the clicker is a acoustic marker that goes into the dog's nervous system with the message, "what your body is doing at this instant has just paid off." Use the clicker at the moment that counts. What your dog is doing when the treat arrives is NOT important. The dog will remember what it was doing WHEN IT HEARD the click. After you click you can let the behavior stop, give the treat, and then go back to training.
From Karen Pryor's "Getting Started with Clicker Training for Dogs"
Another example would be the Utility Go Outs. It is very important to mark the behavior of the touch to the stantion, wall or pole. This is the behavior you are training and want to be understood by the dog. Therefore it is very important to click when the dog lifts his paw to touch or touches the stantion. Clicking after the touch to the stantion is late. Then you ask for the SIT, walk in and reward ON THE STANTION. By rewarding the exact place you want the dog’s behavior to be, you are helping them to understand exactly what they have done and the position they must be in to be correct. I have noticed that in teaching the touch for go outs in obedience that placing the reward on the stantion grew the understanding of this behavior much faster and with fewer wrong attempts.
- Although Karen Pryor writes, "What the dog is doing when the treat arrives is not important" it is important to reward when you can in the correct position you are asking for. The above example of rewarding on the stantion is but one example. Another example of the importance of where you reward is if you are teaching a recall to heel position, your HIGH VALUE food reinforce should be at the correct heel position. Or if you are teaching long sits and downs do not come back to your dog, release them and then give them a treat. Reward in the sit or down position since that is what you want them to know they are being rewarded for otherwise they think that the break from that position is what they are getting rewarded for.
- Only click once. Resist the temptation to click over and over for a behavior you like or for a duration behavior. Karen Pryor stresses that when you click you are telling the dog that at that moment you like what they are doing. If you use multiple clicks the dog has no way of knowing which click is the meaningful one.
- Behaviors that we are teaching need to be broken down into their smallest component parts. The dog needs to start with the first and easiest step and be able to be successful quickly.
You will need to identify the end behavior and then think of all the ways to break this behavior down into tiny component parts. The video that follows is one example of how a final behavior of 4 feet in a bowl was broken down into small parts and taught over time.
I have noticed that some of the tricks that I teach take many months to perfect. An example of this is SIT PRETTY and HOLD A BOTTLE or TOY. Myst had to learn to sit pretty which took a few weeks; hold her balance in the sit pretty which took longer and then learn to hold an object which took the longest for her. We just kept working on this a little at a time and rewarded tiny steps for holding the object in sit pretty.
- Find a training location free of distractions. Ideally this would be in the house, in a room where no training has taken place before. By controlling the environment you have limited other distractions that may seem like fun for the dog. Their focus can now be on you entirely. A bathroom is ideal or a room in your home not previously used for training. I have a training room that was a guest bedroom.. you can see this in my GoPro video, “Two Cool Border Collies Go To the Gym”.
- Have a plan of what you are going to teach and what responses you would like to see. Know exactly what the behavior is you are looking for so you can mark that exact behavior with a click or verbal marker. REMEMBER.. THE DOG REMEMBERS WHAT IT WAS DOING WHEN IT HEARS THE CLICK.
What toys or tools do you need? What reinforcer is appropriate for the training? How can you manipulate the training environment to help the dog be successful? AND, when
- Begin each training session with a game such as tugging with your dog. This helps your dog associate the new location with the fun they are going to have with you. You will see in my GoPro video “Two Cool Border Collies Go To the Gym” at how much fun both dogs have as we go to their favorite place later in the day to shape tricks, work on behaviors and do balance disc work.
- Keep your training sessions short. Depending on the age of the dog and what you are training, this might be 1-2 minutes or as long as 5 minutes. Always end your training sessions on a happy note with the dog wanting more… and end with a fun game such as tugging.
In my sessions I work both dogs; one sits and waits while the other trains. This way I can alternate dogs and keep training sessions short. It also provide an excellent opportunity to reward the good behavior of a sit and wait.
- Don’t try to help your dog. Help is interference with the dog learning to become a problem solver on his own and figuring out the new behaviors you want. If you lure your dog or jump in with prompts or cues, you are teaching your dog to wait for your direction and not problem solve on his own.
Sometimes I will lure with a treat for the first few times but the lure is faded very quickly. You will see this in the video below of how I taught Myst the 4 feet in the bowl for the first 2 times she was encouraged to jump into the boxes.
- At the end of each session evaluate what went well, what was a challenge and what your work needs to be in the next session to build on what you have just done.
- This principle is very very important to understand – Reinforce for average or better. After you have worked on a behavior and have gotten at least 5 correct responses, consider this to be “average.” Do not reward for anything less than average. As dogs begin to understand and perform the correct behavior with more accuracy then stop rewarding for average and only reward for 100% correct.
An example of this is when teaching fronts and finishes in obedience. Once the dog understands what is a correct front or finish, they should not be rewarded for anything less than perfect otherwise they will never know what is that perfect performance. Many novice trainers are so excited to see their dogs just performing a behavior CLOSE to what they are looking for, that they reward. The dog never learns to do this skill with perfection.
- Establish a conditioned negative reinforcer to let the dog know that what they just did is not correct. This negative reinforcer should be a word such as "Opps" that cannot be said with a lot of negativity in your voice. It is just a marker to say, "I don't think so, I am not going to reinforce you. Let's try again." An example of this would be with weave pole training; if the dog misses the first pole or pops out early, then it is important to let them know they did not do this behavior correctly. Do not let them finish after making the mistake. Mark the incorrect behavior with the "opps" and then ask them to do it again.
In obedience this negative reinforcer might be for a slow drop or no drop on recall; a crooked front or finish when the dog knows what is correct. A Utility Go Out where the dog turns and sits and does not touch the stantion or does not perform the touch to the stantion. If these incorrect behaviors are not marked the dog has no way of knowing what is the correct or perfect performance.
Most of all, HAVE FUN AND KEEP SMILING to let the dog know you are having fun. End each training session with a fun interactive game.
HAPPY TRAINING
VIDEO 1: Shaping - How I Taught the 4 Feet in the Bowl Trick