|
Location: West Sussex UK
Joined: Jan 2010
Posts: 2,044
|
|
Nikie ~ I think I get where you're coming from I also agree about the "inadvertant myth throughout dog training that operant learning theory is just to do with dog training applied to dogs by humans". That somehow detracts from the fact that the dog is learning from its own responses to certain stimuli. If a behaviour weakens, then the action/behaviour that resulted in that stimulus is a punisher to that animal. If the behaviour doesn't weaken, then that stimulus is not perceived as a punisher. Similarly with reinforcers ~ the actions/behaviours that result in the reinforcers are strengthened ~ if they aren’t strengthened then they are not reinforcers.
Whilst it easy for dog trainers & dog owners to say we only use +ve reinforcement to train our dogs, we need to be careful about the theory behind what we are doing. There is a tendency to assume that just because we’ve got a pocket full of tasty treats to dole out when the dog does the wanted behaviour, that we are ONLY using +ve reinforcement. This is not necessarily the case because unless the wanted behaviour increases, then the dog is not being positively reinforced. And then there is the other side of the coin in that if treats are on offer & the dog knows that it has to do something to obtain the treat, then by withholding the treat, we are using –ve punishment.
As I’ve read it, (& please correct me if I'm wrong) I don’t think Nikie is commenting on the pros or cons of using +ve/-ve punishment or +ve/-ve reinforcement, just the theory behind operant conditioning as researched by B F Skinner. The moral or ethical reasons for using these different ways to cause behavioural responses to strengthen or weaken, is a totally different argument to the theory of how operant conditioning works ~ and it is that moral & ethical stance that is really being debated on this thread i.e. the use of ecollars.
And whilst operant conditioning is relevant to much of what animals learn ~ there are other forms of learning too, that tend to get overlooked, especially in the world of dog training. Classical conditioning, superstitious learning, observational learning & other forms of learning from non-rewarding events.
I don't subscribe to the "3 repetitions & the dog has learned" view. Perhaps Skinner found this to be true under controlled laboratory conditions where no other factors could possibly influence the outcomes of the experiments, using pigeons with their somewhat limited behavioural repertoire when compared to dogs ~ then this might hold true. But in the dog world, there are too many other influencing factors to be sure that the expected outcomes from operant conditioning alone are going to hold true.
BTW Nikie ~ you earlier quoted one of my posts & referred to my quote from the AVSAB site ~ that site was talking about the moral & ethical ways in which operant conditioning is applied in the world of dog training. It is not a site on the principles of Skinner’s laws of operant conditioning.