Training dogs is not an easy job. While Southeastern Guide Dogs’ dogs are bred to be uber-intelligent, it still takes approximately six months of hard work before they are ready to be guide dogs. And that’s on top of the approximately 12-16 months the puppy raiser spent preparing that puppy to come back in for formal harness training. So, how do you teach a dog advanced skills like guiding? Well, you do it through repetition, praise, repetition, praise and repeat infinitely.
Recently, we have instituted a new (to us) training method – clicker training. In this method, the dog is asked to perform a command and when they execute the command properly a clicker sound is made, shortly thereafter followed by a treat and much praise for the dog. As most of our dogs are Labradors or goldadors (Labrador x golden retriever), much of their drive comes from their stomachs, so clicker training works really well for them.
We use the clicker training only in the early stages of their training for a couple of reasons. First, probably the most important thing guide dogs do is to alert their handler of changes in elevation that could trip them up, literally. Since this is such an important part of their training, the sooner they can learn it, the better. We use clicker training when teaching the dogs to stop at curbs and they seem to pick up the concept quicker when rewarded with a treat and praise as opposed to praise alone.
Here’s the drawback though…we have to wean them off of the clicker/treat method pretty quickly because we wouldn’t want them to be guiding their handler and expect a treat every time they stopped them at the curb while crossing the street – that could be detrimental to the team.
So in the subsequent stages of their training, the trainers return to praise or correction. Using this method, the dog is given a command or put in a situation where a certain behavior is expected and if they do it correctly, the dog is praised, if not, the dog is corrected, shown what was expected and the command is repeated until he gets it right.
For instance, if the team (guide dog and trainer) are coming to a street crossing, the dog is expected to stop before going into the street -- whether there is a drop in elevation into the street or a blended curb with no drop off. If the dog stops before going into the street, it is given lots of praise and then given the next command. However, if the dog misses the curb and continues on, the trainer will give a leash correction, bring the dog back to where they should have stopped, the trainer then shows the dog what it missed, they turn around with the dog and attempt it again. Usually the dog will get it the second time around, but if not the process is repeated until it is done correctly at which point the dog is given lots of praise.
One of the most amazing things the dogs learn is something called “intelligent disobedience.” Intelligent disobedience is when the dog disobeys a given command because it would endanger the team. For example, say a team is at a street crossing and the visually impaired handler has listened to the traffic pattern and believes it is clear to cross the street so they give the “forward” command, but the handler doesn’t know that a silent hybrid vehicle is coming. The dog will refuse the “forward” command, staying in place until the danger has passed.
Guide dog training works on a series of increased responsibility for the dog. The dogs are expected to learn certain commands first and the training builds on itself. Once they learn one behavior, they are expected to remember that one and the next one is added, and then the next one, and the next one, until they are completely responsible for their handler’s safety. Once they have mastered all the commands, the final test before being deemed “class ready” is to lead a blindfolded trainer through a route with another trainer shadowing to make sure everything the dog does is spot on. If they make it through then it’s just a matter of finding the perfect match for them in class.