Level 2 - Week 1
Hi guys, I am going to start you all off with a bunch of theory this class. You are now advanced trainers for your pups, so I want you to understand what you are actually doing. :)
VERY IMPORTANT INFO TO READ IN ORDER TO UNDERSTAND HOW TRAINING IS DONE
POPPING A BEHAVIOUR ON CUE
There is different ways of teaching a behaviour. Some, like the 'Default Settle' are taught by using 'free shaping'. Most behaviours are easiest taught with a lure. This type of training simply involves using a food reward to guide the dog into the desired position or behaviour.
We have done this in class for the Sit, Down and lots of other behaviours, like last week the 'Chin Rest'. Now, obviously we don't want to have a dog that will only perform those behaviours if we have a treat at their nose. So, we quickly need to get rid of the lure and introduce the cue. As soon as your pup performs the behaviour reliably, you start using the same hand movements as before, but there is no treat in the hand. You then mark the right behaviour with your YES and bring a treat out to reward. Your pup will learn there doesn't need to be a lure at its nose because the reward comes anyways. Once this behaviour is reliable you would add the cue, so the word 'chin' in this case. You say the cue before the routine of making the hand movement and then YES and reward. You repeat this until you see that your puppy starts going into the desired position without you doing the hand movement. Keep the routine up of saying the cue and showing the hand movement. We want the pup to know both as a cue, the verbal and the body language.
Now, I had to use the 'Chin Rest' as the example, but actually I wouldn't pop a verbal cue on this particular behaviour. The reason for that is that I want VOLUNTARY vet behaviour to be as voluntary as possible, so I don't like having a command. The way you hold your hand is cue enough in this case. But think about all your other commands whether they are all on cue or whether you need to fix them a bit.
DO I HAVE TO KEEP GIVING MY PUPPY TREATS FOR EVERY TIME THEY DO AS I ASK THEM TO?
Treats are great to use when teaching a new behaviour! You want to start with a Continuous Reinforcement schedule which means EVERY TIME your pup responds correctly they get a reward. Once the behaviour is offered in a timely and consistent manner in response to the cue, you need to change the Reinforcement Schedule. Continuous reinforcement allows a high rate of reinforcers to be presented which keeps the pup highly interested and the frustration level low. Continuous reinforcement makes the newly learned behaviour resistant to change, however, if you continue to use the continuous reinforcement for too long, the behaviour will become less precise and timely. Therefore, we need to introduce a different reinforcement schedule once the pup is fluent in the behaviour (8 out of 10 correct responses). You can start rewarding just the best responses (criteria could be speed or preciseness for example), BUT you need to make sure your standards are right. Be aware that if you do not maintain a high enough rate of reinforcement for the majority of the time, the behaviour will cease to be offered by the dog. So, make sure to raise your standards slowly. Find out what your pup finds rewarding. At the end of the day, that doesn't just have to be treats. Toys are my favourite and I hope you work on getting your pup super duper keen. But pats and praise also cut it for some pups in certain situations. The higher the distractions, the higher value the reward needs to be. It is recommended to keep the reinforcement for the Recall, your 'COME' cue, incredibly high!
POISONING THE CUE
If we have poisoned a cue we have basically destroyed the strong reinforcement history that we originally created with our continuous reinforcement schedule and therefore our pups are less likely to perform the behaviour to the given cue. This can happen in different ways, one is certainly by punishing the dog for not performing the behaviour. The cue now no longer seems safe and therefore the pup doesn't have the automatic response to follow it anymore. The way of 'correcting' the fact that the pup didn't perform the behaviour should therefore be to take a step back and work with less expectations, less distractions or higher valued rewards in order to build up to this specific situation. Remember what you learned about dogs and their ability to generalise! Another way of poisoning the cue is by stopping the continuous reinforcement schedule too early or by not maintaining a high enough rate of reinforcement for the majority of the time. One way of doing this is by overusing cues. "Fluffy, Sit, Sit, Sit, Sit", "Fluffy, Come, Come, Here, Come".
Now, get ready to have your mind blown. ;) Once a cue has a strong reinforcement history you can use the cue to reinforce a previous behaviour in a chain. What does that mean? If I ask my dog to 'Sit' and once she has done well, I now ask her to 'Shake' I am using the 'Shake' cue to reinforce the 'Sit' cue. The more you work with chains the less you have to reward every single behaviour because the next cue is reinforcing enough. So long the cue is well-established positive.
See you all next week. Remember to bring yummy, long lasting chewies to make everyone's lives easier. :)