The Certification Council for Professional Dog Trainers uses the least intrusive and minimally aversive procedure to succeed in training or changing behavior. It is called LIMA, or "Least Intrusive, Minimally Aversive."
Why not, as a body of leaders in behavior, instead of striving for the least intrusive, minimally aversive behavior intervention, why not strive for the most reinforcing, optimally appetitive behavior intervention?
I realize the ask is big. We must rely on available education, intelligence, and hopefully our desire for the best possible relationship with our dogs. Our most respected organizations, including the American Veterinary Society of Animal Behavior (AVSAB), the American Veterinary Medical Association (AVMA), and the American College of Veterinary Behaviorists (ACVA) do not endorse using pain or fear as a standard for animal training, and have clear position statements that aversive methods are not necessary in dog training, including treating behavior problems.
Rewards Are Preferred
I want to be honest about the fact that I use punishment and know what I am doing, rather than just hoping I always use reinforcement. I also want to recognize when I am using a punishment, not just with dogs but with human students and colleagues, as I am about to do here by voicing my disapproval (a positive punishment [P+] reprimand) of the target behavior defined as "using fear or pain."
Ironic, because I am engaging in the P+ of trying to reduce the frequency of using P+ by voicing my disapproval of P+. I am intentionally trying to be self-aware, civil, thoughtful, and scholarly in tone as I do so.
In general I agree with Dr. Ian Dunbar's teachings, especially that rewards are always preferred and appropriate behavior must always be taught to the dog and available to produce to avoid punishment. Dr. Ian Dunbar says repeating punishment of any kind beyond three repetitions is ABUSE. This seems a very good principle.
We learn by scholarly study and in real-life practice, that for punishment to even be effective, it must be immediate and certain. Unless one is 100% able to deliver the punishment 100% of the time the target behavior occurs, then don't bother. Otherwise it becomes an intermittent reinforcer, thus strengthening the very behavior we wish to diminish!
Punishment Has Fallout To Become Abuse
Considered in light of scholarly study, this makes me re-examine punishment. If indeed one cannot follow through with delivering the punishment every single time and not repeating beyond three repetitions, then one must strongly consider stopping the use of that particular punishment, ever. No matter how un-painful or fear-free one considers it to be.
See Dr. Ian Dunbar's eighth rule of punishment and his idea that punishment beyond three repetitions is abuse. If an educated, respected, well-known, and scientifically sound expert scholar says it's abuse, it probably is.
I strive to only use punishment that is not painful and not scary but is effective. However unless I can honestly apply it 100% of the time, and repeat it less than three times, I will consider not using punishment for that behavior.
Examples of my use of punishment follow:
Positive punishment: physically preventing my dog from performing a behavior that will hurt him by physically grabbing him (providing a consequence [a physical grab] that decreases the frequency of the target behavior [hurting himself, say getting run over by a tractor if he cannot hear me]). P+ is to be used only at the utmost end of need.
Negative punishment: crossing my arms and looking up if a dog tries to jump up on me when the dog's guardian does not want the dog to jump up on people (removing the reinforcer [my attention] to reduce the frequency of the behavior [jumping up]).
Negative reinforcement: walking away from my dog if he gets excited while in the crate while I approach to release him from the crate (providing aversive consequences [I walk away, something my dog will work to avoid] to maintain or increase the behavior [Gunner stays quietly lying down in his crate until released]).
Positive reinforcement: where I choose to spend most of my efforts.
For my part, I do not want to incur the heavy responsibility, as Sue Alexander rightly puts it in her Applied Behavior Analysis course, of using pain to teach a dog.
Confidence and Pleasure (No Fear or Pain)
I prefer to rely on the four quadrants of reinforcement and punishment (without fear or pain), as well as my intelligence, experience, and if necessary consulting a colleague whose opinion matters, to develop a plan to attain the behavior I desire that's in the best interest of the dog and our relationship.
If we find ourselves in a situation where we are considering using fear or pain, we must re-examine what we know, including what is known by respected and scientifically sound experts. When necessary, reach out to a colleague or expert peer for the appropriate help.
Yes, even for fence jumping, life-saving off-leash recall, harming chickens, and the like. Modern science and our most respected scientific communities support positive reinforcement and optimally appetitive behavior intervention.
Given enough thought, effort, and the right brain power, there is almost always a better alternative than using fear or pain.
No comments:
Post a Comment