SKoth wrote: I consider a well rounded trainer to be one that is has a complete understanding of all facets of behavior and training. And, the whole point of the dialogue to improve our understanding - to learn - which is an essential part of being the best we can be.
But why does "all facets" of training have to include aversive methods? I'll never understand trainers who insist that positive reinforcement can't work with certain difficult cases. Can you give me an example of a case that you've worked with where properly applied positive reinforcement techniques have failed and aversive methods have been needed instead? I really am curious because in my own layman's experience the opposite has been the case.
Wouldn't you want to base your training techniques on the latest science of exactly what motivates dogs to learn and how they learn in the first place?
Absolutely. You're missing the point. This dialogue isn't about what I want or do. Instead of making this type of statement why not post the scientific studies so that the forum can learn and understand? No pointing fingers - contribute, teach, share....
There have been pages and pages of links to studies showing that aversive methods are unnecessary at best, harmful at worst. The trouble is, those who advocate "well rounded" training dismiss them and then keep asking for more studies. How about those people start posting scientific studies that support aversive training?
I'm not trying to point fingers. I'm trying, as someone who has dealt with a difficult dog who, by the standards of many members here, should've been put down years ago, to understand why the techniques that messed my dog up continue to be practiced when there is a much better alternative available. I'm not an expert dog trainer but I am a trained scientist and I can recognize a sound scientific argument when I see it.
Do you disagree that dogs learn through a combination of classical and operant conditioning?
Operant conditioning (R - SRF)
â€¢ A voluntary response (R) is followed by a reinforcing stimulus (SRF)
â€¢ The voluntary response is more likely to be emitted by the organism.
â€¢ A reinforcer is any stimulus that increases the frequency of a behavior
â€¢ To be a reinforcer stimuli must immediately follow the response and must be perceived as contingent upon the response
Classical conditioning (S - R)
â€¢ An involuntary response (UCR) is preceded by a stimuli (UCS), or
â€¢ A stimulus (UCS) automatically triggers an involuntary response (UCR)
â€¢ A neutral stimulus (NS) associated with UCS automatically triggers a conditioned response.
â€¢ The NS becomes a conditioned stimulus (CS).
What have I said that would lead you to believe that I don't believe in operant or classical conditioning? Negative reinforcement and positive punishment are both part of the operant conditioning model.
I asked because I wanted to get a baseline as to where you were coming from. If you do indeed believe that this is how dogs learn, why do you think they cannot be taught alternative behaviors that are incompatible with aggression while being desensitized to the stimulus that sets off the behavior? Why would you introduce positive punishment when the dog is already over its threshold? At what time is this justified when there is clearly an alternative?
So, you begin to see my point? Being the best we can be, understanding the most we can understand includes understanding the entire classical conditioning model. Negative reinforcement, positive punishment and negative punishment are important to understand to apply operant conditioning.
I'm not sure why you posted the definition of classical and operant conditioning when it seems you are trying to justify the use of positive punishment and negative reinforcement.
I have no problem with negative reinforcement in certain instances. For example, a dog who jumps on you when you come home wants your attention, you'd turn your back and ignore him taking away something he wants. IMO, that's a completely appropriate use of negative reinforcement, as I understand it. I'm sure an expert will chime in if I'm wrong.
My issue with positive punishment, esp. when dealing with aggression, is that it may suppress the behavior, the dog won't chew your shoes in front of you because he'll get hit or yelled at, but leave him alone for a few hours and see what happens. My own dog wouldn't lunge or go for other other dogs with a prong collar on but take it off and it was all you could do to keep her from getting to any dog she'd see.
If you understand and accept the concept of classical and operant conditioning, why would you think that positive reinforcement is ineffective?
Or that a dogs behavior is essentially motivated by the promise of a reward, in whatever shape or form it might come?
Absolutely. But lets consider that there are many types of reward. Primary, secondary, positive, negative and self. If we don't understand this paradigm how can we manipulate behavior with the presence of a reward.
I agree there are many types of reward which is why I phrased my comment like I did.
You'll notice that many members with dog aggressive dogs cannot call their dog off another dog. For many of these dogs biting another dog is the ultimate reward. It is a self rewarding behavior. So, if that's the ultimate reward and we cannot replicate or reproduce that reward then we cannot override that behavior with a positive reinforcer. And this is the point where we consider a reinforcement model and consider the application of a positive punisher or negative reinforcer to balance the dogs motivation.
OK, so in this case, (a dog who is truly motivated by simply wanting to bite the other dog, assuming fear, anxiety, previously reinforced aggression, etc. are not the underlying causes ) why would desensitization and counter conditioning not work? You still run into the issue of addressing the behavior without modifying the reason for the behavior. You've done nothing to change the dogs desire to bite, only to control his behavior. How is that safe or fool proof? It seems to be to be a band aid treatment that is bound to fall off sooner or later.
Positive reinforcement is the method of choice by most trainers and behaviorists who base their methods on science. Using positive reinforcement and working within the limits of the dogs threshold to whatever stimulus, you reduce the stress and thus allow the dog to actually LEARN an alternative behavior, associate the negative stimulus (another dog, for example) with something positive, and then you can begin to change the REASON for the dogs behavior in the first place.
Is that your opinion or a fact? We're being objective here right? My personal experience with trainers and behaviorists has represented more of a Antecedents - Behavior - Consequents (positive or negative). If you are familiar with Karen Pryor's Don't Shoot the Dog
you can see a similar reinforcement theory model.
Antecedent-Behavior-Consequences just seems like traditional aversive methods to me. Nothing ground breaking there. The dog is afraid of people, sees a person (antecedent), barks and lunges (behavior), and then gets pinched with a prong collar (consequences). Am I understanding you correctly? Trust me, I've had personal experience with that too, nearly a decades worth and it was not only ineffective but my dog ended up with generalized anxiety issues on top of aggression issues.
I've not read Karen Pryor's book yet but I have it here and plan to get to it sometime.
Using aversive methods, you may suppress a behavior but it does nothing to address the underlying reason the behavior is occurring in the first place and there is the chance that you may be doing harm to the dog....physically or psychologically!
I guess our understanding of classical and operant conditioning differs. The correct application of positive punishment and negative reinforcement are effective in shaping behavior.
No, I think we both understand the definitions correctly. My point is that it should not only be about shaping behaviors. Unless you deal with the underlying reasons the behavior is occurring in the first place, you will never be treating and "training" the dog, only suppressing the undesired behavior.
Like I said, I'm no expert and I know that reading a book and intellectually understanding a concept is much different that actually putting said concept into action. Boy do I know, because I'm doing both and the latter is much much harder! But my experience and background does enable me to understand and dissect an argument and I still have not heard an argument from those who advocate aversive methods that has convinced me that it is necessary or more effective than what can be accomplished by positive reinforcement training techniques.