A game is a challenge, created by the rules that govern it, bound by the cooperation between the players of the game who all have the same purposes, intentions and focus; all of which results in a quantifiable goal.

The Goal of Shaping

Shaping is about going through the steps of an action and reinforcing each step until the final action is achieved. Shaping with markersworks best with those fine motor skills and intricate movements that conprise a simple behavior. Shaping with games can increase the amount of behavior that can be learned in short period of time.

Timing, criteria and reinforcement are critical in marker shaping

Timing: Are you marking at the exact moment of the action or non-action? One of the nice things about marker training is that you don't "ruin" a dog if you make a mistake. But to actually teach a behavior, the timing of the mark must be nearly perfect. This is one of the reasons clickers are thought to be best. It's a sharp, short, unique sound that can be (according to clicker trainers) done faster then any other form of mark. Personally, I'm all thumbs and never did get the timing right. My thumb would press too fast and hurt me, too slow and not make the sound, off center and the clicker went flying. But there are many great clicker trainers out there who either have better manual dexterity then I do or they took the time to get it right. Timing a "yes" may (and that's debatable) be slower, but for me it works. I've trained things as intricate as an ear flick using only "yes" as the marker.

Criteria: This is where many trainers seem to not understand. You MUST know what the final behavior should look like and the actual actions that build that final behavior. Planning is one of the most important steps in shaping. It doesn't matter whether you are shaping with markers, lures, targets or games. Break the behavior down, understand each step. Then when it comes time to increase criteria as you are working with the dog, you know what comes next. Without the plan, your dog with be stuck in a guessing game as you try and determine whether the action he made with lead to the final behavior or not, which often causes you to miss marking a specific movement until you figure it out.

Reinforcement: This is another area, especially with beginning marker trainers, that needs work. Reinforcement should be fast and abundant; especially in the beginning stages of teaching a new behavior. Reinforcement should be appropriate. If you are teaching an ear flick, you don't want hard to chew large pieces as you will be reinforcing tiny movement which would get lost in the effort of chewing. But if you are teaching "go around", treats could be larger and more rewarding for the larger movements of the dog.

Timing, criteria and reinforcement while shaping with games.

Timing: When shaping with Games, timing is not that critical. Reinforcement generally happens at the end of the game, occassionally inside the game if there are a lot of steps to the game, and it's always what would be considered a jackpot in marker training. The entire sequence of actions is reinforced, not a series of progressive approximations. As you play the game, the movements should flow one from the other for both the human and the dog. Flow is what is critical in shaping with games, not timing.

Criteria: Criteria are planned out. Just like marker training you must know what the steps are to creating the final behavior. The difference with shaping with games is that the criteria become separate games. For instance in training a "go around". The first game would be runing to the object and running back. The second game could be putting the reinforcement on the ground next to you and by using game one, the dog would run up to the object, run back and be rewarded with what's on the ground. By manipulating the environment, walking the dog through the steps of the game and rewarding each step, the dog gains understanding from the beginning. There is no guessing game.

Reinforcement: Reinforcement is large when shaping with games. You want the dog to learn that the action just performed is very valuable and could serve on it's own and not just a bridge to a final behavior. The placement of reinforcement when shaping with games can be vital. You are asking your dog to play a game and showing him the "prize" that can be won if he plays the game to a win. Rate of reinforcement isn't as crucial here. If you have broken down a complete behavior into simple parts and created a game for each part, reinforcement automatically happens very fast. One of the biggest differences between shaping with markers and shaping with games is that with games there is no need for variable or differential reinforcement schedules to create motivation. The game itself takes on the value of the reinforcement and because it's own secondary reinforcer. This can be seen when you teach a dog to get "up" on objects. After less then a week of "up" on various objects, your dog will offere this behavior everywhere. It becomes something that is so reinforcing, it can even be the reinforcement for other games.

How Shaping with Games Differs

End Goal: First identify the goal of your end behavior. What are you trying to train, what does the end behavior look like in your mind's eye? What is the criteria; how fast/how high/how many repetitions/how much or how little movement?

Stationary or Motion: Distinguish if you are shaping a stationary behavior or behavior of motion. The answer to the question will determine your reinforcement strategy. For example if you are shaping a stationary behavior your goal will be to build duration for that response. Rewarding the same response but working towards extending the interval of the dog maintaining that response (a sit or down stay or holding a sit pretty). By way of contrast, if you were to consistently reward the same moment of time for a behavior of motion (always rewarding the exact same number of steps away or number of rotations around), you run the risk of turning that behavior of motion into a stationary response (the dog will predict always getting rewarded at the same place and will want to stop when he reaches that place).

Know What You Don't Want: Almost as important as having a vision of what you would like your dog to do is clearly knowing what you DO NOT want your dog to do. There are some responses that can be toxic when trying to shape your dog. That is, if these responses were inadvertently rewarded they would make it difficult for your dog to do what you are really after. For example, imagine you are trying to shape your dog to back up away from you. You see his paws move back so you mark the response, but at the same time your dog sits. If you feed the sit, you make it difficult to get the back up; how can a dog back up when he is sitting? By knowing ahead of time what responses you do not want your dog to offer, you will be prepared to not mark or reward the dog should your dog offer one of those responses.

Map it Out: All good training sessions start with a written plan. It is plan of action of how to get your end or "goal" behavior. Bob Bailey suggests to us all that we "Be a splitter, not a lumper." By that, he is encouraging us to chunk the vision of our larger behavior down into smaller responses that we can build separately and eventually put together to give us our targeted response. This sort of planning gives me the flexibility to move ideas around and be flexible with the elements of my plan as they move from my mind's eye to the computer planning page to the dog training session.

Manipulate Your Environment: Your first step towards you actually training the dog should be to manipulate the environment in order to set the dog up for success. Environmental manipulation reduces the number of failures your dog has, increases the success of your session as it decreases the options your dog has while working with you. In my opinion there is nothing else that is unrelated to the dog himself, that has more impact on your dog training than environmental manipulation. You can manipulate your dog training environment by:

Space: Limiting the space your dog has to work within. The larger the environment you train in, the more potential distractions to your training and the more options your dog has available to him. Keeping the environment small and sterile (without exciting distractions) helps to keep you as the focus and your training session as the focus of the dog's attention!

Your last exercise for this week. Don't forget to post videos for each of the exercises you've done

Potty Training

  • Put a towel on the closed seat of your toilet and a leash on your dog
  • Walk into the bathroom with your dog on leash, tap the towel and say "up"
  • If you are using a chair instead, still put the towel on the chair
  • Once your dog is up, move your body around until your dog faces you
  • Play a tiny tug game on the seat
  • Ask for "off" and walk out of the bathroom with your dog
  • Repeat without the leash until your dog is just leaping up there
  • Put the lid up. Start at the beginning and repeat all steps

You have now potty trained your dog

In Training




Standard Parkour

Tucker

Brinkley


Jumpers Parkour


Creative Parkour