poodleOwned
Posted : 9/9/2009 5:24:10 AM
corvus
I really think bias does no one any good. It gives the people who disagree fuel for the fire and the people that agree either jump on the bandwagon with guns blazing or just feel frustrated that the methods discussed would have spoken for themselves much more convincingly without the biased spin. '
Hi
I am a bit sorry to all. I find it hard not to be a little firey commneting on another forum that suggested that it would be "cool" to use an e collar at 15 weeks on my now 3 year old poddle and no one challenged the proposer other than myself. The suggestion of course was based on an "Alpha" dog understanding of pack structure. The need to dominate and correct was fairly rampant. BTW my dog was then and is now somthing of an angel. :(((( You will see the link at the end...
In real life as in dog life i am much concerned with the adequacy of models. I guess that in a way that we have models that don't quite get there in terms of describing the full gamut of dog behaviour. As trainers, we fill in the gaps with more and more complicated statements and ideas, or throw up our hands in horror and decide that it all is too much for us and use the dreaded "common sense".
It is obvious to most that have worked with more modern models that the dominance model is deficient. I am not going to repeat the original statement form the AVA.As Temple Grandin reports in her latest book and this statement does, it just fails to describe dog behaviour as it really is. One of the tests of a model is to see whether it can predict the behaviour of the data that was collected to make it. It fails.I think though that the dominance model as Temple reports may describe the behaviour in some stressed forced packs. Behaviour that is traditionally associaited with this model may be described as the behaviour of a group of stressed animals, and to me that pretty much describes how it works as a training model. I think this is a key insight as to why it continues to exist. It does self perpetuate the conditons required to validate itself.
I like sometimes to look at the different models and see where they may lead me. we often say emotions beat classical beats operant. we now have an exception on top of two models that can work together that are based on data collected from many trials.
My dog Luci is trialling in UD. (Yep, we call it Futility dog here too) I have so focused on having her in a happy focused state but there is one part of one excercise that i have never been happy with. At the end of our heeling in Signals, we leave our dogs in the standing position, walk a way a few paces turn around and signal for them to Drop, then sit then come to us.She has always been slow to drop in this excercise.
if we use the dominance model, i am absolutely certain that she would drop. I am not going to describe what would happen, but IMHO it is inexcusable and not requred. I also believe that in a very short time that she would not be a top trialling prospect.
If i use R+ (which i have done ) i really couldn't shape the speed down. There was something else happening. Using clasical in the drop didin't work either, and besides that she has a fast rapid drop everywhere else. I was reading Temple Grandin's latest and a model caught my eye. It is based on the idea that a whole lot of behaviour is emotional (it is a pretty interetsing model) and i used the model to test what is happening. I noted that when she stood besides me, her tail went down which is an early sign of fear. While she has a steady stand for exam, it isn't that exciting for her , weighing 7.5Kg wringing wet. I decided to work on the emotion of the stand by counter conditioning her so that her tail was up and she was smiling in the stand... and when i did ask her to drop again after she went like a rocket. So a new model provided me with better data to look after my dog.
I worry that this AVA statement has been issued many years after data existed that disputed the dominance model, and that some of the new information which links brain function to behavior such as the papers alluded to above from Temple Grandin it will be an absolute struggle to communcicate to the wider dog audience.
I have been having great fun working with the corrolaries of this theory with biomedical people. We have been playing with diagrams and concepts and to say that we are excited is an understatement.But i despair that it will get out to the wider dog world at all.
This brings me to a neccessary concept for trainers to understand too when we try to communicate concepts of positive training. It will often not work with trianers who may not live lives that are that greatly rewarding themselves, and have needs for control to feel different about themselves. It may also be that the need for peer acceptance is greater than the need to train their dog well. If the group norm is to be a bit rough and gruff then that is what most others will do. Much of this is predicted within the literature on group behaviour. So i have linked the start of my post to the finish... and this time not strayed to much off topic!!