4 The Love of DOG

As we prepare for Halloween I ‘ve got to wonder how many people out there put their pets in costumes and do they like it? Our Scrappy seems not to mind to much and likes his sweat shirt when it is cold and rainy out. Send me some pics of your pets and what you think

mark@doggies411.com

When training a dog to sit the best method to do this is, tell the dog to sit and if they do not sit then give them a harsh correction like a leash pop. Does this sound like a logical way to train a dog? Does the dog even know what sit means? This paper will compare correction based training, better known as traditional training, to positive reinforcement based training or clicker training. It will answer many questions on how dogs learn and show that clicker training is much more reliable and fun for everyone involved.
Correction based training is also referred to as, “Traditional training”. When using this style of training commands are given. If the dog does not respond correctly a correction is given, usually through a leash pop or leash correction. A leash pop quickly tightens a choke collar which also pinches the dogs neck where the loop hangs. Traditional training is based on compulsive methods, also known as positive punishment and negative reinforcement. By delivering an unpleasant or even painful consequence when the dog performed an undesired behavior and pairing praise with the end of discomfort when the dog performed the desired response, the dog learned that praise indicated the end of or avoidance of pain or discomfort. (Mullinax, 2010) Does this sound like an enjoyable way to learn?
Clicker training is based on positive reinforcement which is based on Classical and Operant Conditioning. This is also how humans learn, “One of the many ways in which we learn is by trial and error. In Unit4 Psychology, this type of learning is called operant conditioning because we learn to operate or act on the environment to effect some outcome” (Jory, 2004). This is basically saying if the outcome of an behavior returns a desirable result the behavior is more likely to occur again.
Now compare the two ways of teaching an dog a new behavior. With correction based training the dog is told, “sit”, while pushing down on the hind quarter and pulling up on the leash. When the dog sits the tension on the leash is released and praise is given. Once the command is learned if the dog does sit when the command is given then a leash pop is administered to the dog.
In clicker training to teach the same behavior you would use a clicker. The clicker is a conditioned reinforce. A conditioned reinforce is a meaningless signal, a sound, motion, or light that is deliberately presented before the delivery of the reinforcer. (Pryor, 1999) So the first task is creating a conditioned reinforcer or charging the clicker. This is very easy to do and can be fun for both the trainer and the dog. What is done is the click or whatever conditioned reinforcer is desired is given and immediately following the reinforcement is given, usually this is food but can be a toy or anything the dog desires. Do this for about 20 repetitions in three or four sessions, once the dog expects the reward when the conditioned reinforcer is given then this phase is done and now you have a way to really communicate with the dog; when the behavior occurs that is wanted the dog can be rewarded letting them know that is the desired behavior.
Now the real work of training a behavior can begin. With clicker training capturing or shaping is the method used for training. As soon as the dog performs the desired behavior the clicker is clicked followed by the reward. More complex behaviors are broken down into steps and are shaped one step at a time. To start training a new behavior you should always start in a place where the dog is least likely to be distracted, like a quite room in your house. If the desired behavior is to sit then as soon as the dog sits click and treat. A good way to reset the dog so they may sit again is to throw the treat so the dog must get up to get the treat. Then wait for the dog to sit again, click treat. Pretty soon the dog realizes this is a fun game and can earn food by sitting.
Now a cue is added. In contrast to traditional training where commands are used, and taught first, the command is given and then the dog is forced into the desired position. After many repetitions the dog learns the behavior to avoid being pushed around. (Pryor, 1999) Clicker training uses cue’s, and cue’s are added after the behavior is learned. This makes much more since then trying to tell an dog to do something they do not understand yet. This would be like telling a person to do something in a language they do not understand. But now there is something the dog does understand and that is the conditioned reinforcer, the click. Adding a cue is simple, as soon as the dog starts the desired behavior the cue is given and then the click and treat. What happens is the cue becomes a green light for the dog to earn a reward. Once the dog understands the cue, the click can be removed so now instead of cue, click treat it is just cue and treat. If the dog is not responding to your cue then back up a step and continue capturing. The dog may not understand the cue or may be to distracted to perform the cue. What is distractions?
Distraction are things that will take the dogs attention off you. For instance a car driving down the street while you are in the yard with your dog. A good way to deal with distractions is after you have captured the behavior but before the cue is added, capture the behavior some more in the presence of these distractions. Start with the least distraction and move up to higher distractions as the dog learns to perform the cue in the presence of each. A good tip is to create a list of distraction from least to highest. It may look something like this
Example of Distraction Hierarchy
1. Back yard
2. Front yard with cars passing
3. Front yard with neighbor out in yard
4. Toy present
5. Favorite toy present
6. 100 feet from other dogs
7. 50 feet from other dogs
Each individual dog will have different things that are more distracting than others. It is good to assess the dog and then make a list to keep so as you train each behavior the list will be utilized in proofing the behaviors before the cue is added. Once the dog performs well in all distractions then start adding the cue in a quiet place and then move through the list of distractions again.
In correction based training there is no way to proof distraction, the command is given and if the behavior is not performed then a correction is given. Is this a logical way to train? Imagine a person seeing a something they really desire while another person is asking some mundane question. Chances are the person asking the question will be ignored but if that question was, “how would you like to have ten million dollars”, the person would most likely hear that. This is how clicker training works. The cue is associated with something the dog desires so when it is given it has value.
Does the dog always have to be rewarded in clicker training? This is a question frequently asked and a false belief by correction based trainers about clicker training. After the cue is learned you should start a random reinforcement schedule. Think of it like a slot machine, humans love them and so do dogs. Sometimes you get nothing, sometimes a little something and then you may hit the jack pot! Bancroft explains in her paper how random reinforcement creates better responses than fixed reinforcement schedules because the subject never knows what they are about to receive. (Bancroft, 2008) In dog training this works great because it keeps the dog excited and always trying to earn the jackpot even when they have done ten behaviors for one small reward, the next one may be the jackpot! It is a good idea to find a number of different things to use as rewards, toys, praise, food, play with other dogs. Some dogs like to sniff so much even permission to go sniff can be used as a reward.
In conclusion correction based training is based on fear and avoidance. Correction trained dogs are always fearful and commands predict a correction. Clicker training is based on positive reinforcement. Clicker training builds excitement into the dog, always looking to earn a reward and cues are happy sounds that are the green light to earn a reward. Corrections must always be administered when the dog does do the desired behavior but in clicker training once the behavior is learned treats and rewards come randomly. Clicker training also gives you a wide variety of rewards the dog may earn where with correction based training all the dog can do is avoid punishment.

Just wanted to do a quick post and say thank you to the people who visit my site. Please help spread the word and get others to come log on to doggies411.

Recently my brother asked how to get his dog to stop pulling on leash. The first piece of advice is, don’t let them advance when they pull. Get them to focus on you, reward for lose leash and then advance. also teach them to follow leash pressure rather then go against. I would love to here any other advice from my readers and please send questions.

Thanks
Mark Dog Video

Well I just wanted to tell everyone who visits my site how sorry I am for not being more up to date and posting helpful dog training tips. If you know me you know that I am currently in school taking IT classes at Kaplan University. I am also looking for work which has been tough here in the Lynchburg area. Anyway please keep sending me your pictures and posts and I will try to work harder.

Mark

Deegan

Two of the dogs at AllAmericaMuttRescue that think they are lap dogs!