Training Theory Never Correcting is Never a Good Idea

From Service Dog Training

Service Dog Training Theory: Why Never Correcting is Never a Good Idea

Recently I’ve been seeing an increase in the number of service dog schools and trainers who advertise themselves with such phrases as “positive training only” “no corrections” or “our dogs never hear the word ‘no.’” This trend concerns me. Service dogs are working animals, not pets, and the approach to training must have a different focus.

The basic training philosophy behind positive training, marker training, and other similar training methodologies is sound. It is based on the concept that a dog learns better, responds better, and is generally better adjusted when training is guided by positive reinforcement in the form of a treat, praise or a toy. This is in contrast to forms of training used earlier in the history of dog training which involved physically forcing a dog into positions and harshly correcting it for failing to achieve and maintain the goal position.

The problem with positive training for service dogs arises when correction is never used, even after the dog has thoroughly learned the command. Now, in pet dog training, if the dog becomes distracted, is being stubborn or otherwise does not obey a command after the dog has learned it well, it is possible for the owner to take a few steps back, and re-teach the command using purely positive methods. Usually, there is no risk to the handler if the dog disobeys, nor is there any pressing reason that the handler cannot take the time to re-work the command.

On the other hand, if a service dog disobeys a well known command, the handler may not be in a position where it is feasible to stop, get out a treat, a clicker or a toy and re-teach the command. An unreliable dog can cause the handler anything from an inconvenience to a serious danger. It is important for a service dog to realize that not only are there rewards for appropriate behavior, there are also consequences for inappropriate behavior. When a dog is asked to perform a command, the dog must weigh the value of the reward gained by complying against any potential value in refusing the command. If there is a stimulus competing with the command, a cat to chase, food or something else desirable distracting the dog, the benefit of pursuing this distraction may outweigh the reward of obeying the command. This is where corrections come into play.

Ideally, each command should be taught in three phases. The first phase is the learning phase. In this phase, the dog is taught the command through positive training methods with lots of rewards and no corrections. This should be done in a quiet place with no distractions. Once the dog has learned the command with constant reinforcement, the reinforcements should be phased out until the dog only receives an occasional treat or toy.

The second stage is the proofing phase. In this stage, the reinforcements are reintroduced in the presence of increasing distractions. At this point, corrections are still not used as the dog learns that the command means the same thing and will be rewarded even in the presence of distractions. Again, once the dog is doing well, reinforcements are slowly phased out.

Once the dog is confident performing the command in the presence of high levels of distraction, the dog is ready for the third phase, the correction phase. At this stage, the dog is still asked to perform the command in the presence of distractions. However, instead of just reinforcing the correct response, the handler corrects inappropriate responses with whatever type and level of correction is suitable for that particular dog.

Ideally, if the stages have been accomplished successfully, by the end of the third phase the dog should be responding immediately to commands with neither positive reinforcements or corrections necessary except during practice sessions or on the rare occasion the dog becomes so distracted a correction is necessary. In the long run, this method of carefully using both positive reinforcements and corrections will produce a happy, reliable dog. If the service dog has been trained with both positive reinforcement and appropriate, fair corrections, the dog has two layers of incentives to obey the command. The dog will learn that responding correctly to the command will earn a reward, and that failing to respond to the command will result in a correction. This adds one more layer of reinforcement to the behavior to ensure the dog performs the command reliably.

--Tiffany Huggard-Lee 09:35, 5 September 2009 (CDT)