Thread: Don Sullivan
View Single Post
  #30  
Old January 23rd, 2013, 02:21 PM
doggirl doggirl is offline
Senior Member
 
Join Date: Feb 2005
Location: Ontario
Posts: 153
@ripped -

"I used the methods taught by the behavourist without any success."

"In less than a week of using corrections my dog no longer lunged and barked at people"

"I don't believe that bribing a dog to get results is the answer."

Learning theory is not simple or easy. It actually requires a decent amount of theoretical background and skills/ability. Positive reinforcement, desensitization & counterconditioning, learning by association - they work, that's not even up for debate. The variable is the handler. If they don't understand what they're doing (eg thinking that food rewards are "bribing") they are very likely to be unable to carry out even a simple task (like teaching a dog to sit) using food rewards. Timing is a huge culprit. Many dog trainers don't understand timing - it's not difficult to find dog owners who say their dogs won't learn but the dog is learning things no problem as soon as they're put with a handler who has the knowledge and skills to train dogs. They also take time - because you are changing the dog's emotional response to a trigger, so that they choose on their own not to produce the undesired behaviour. It's fixing a problem, not suppressing a behaviour that's really a symptom of the problem. The methods work - that's tried and true (and if you say that's not true, you are saying that virtually 100% of all the experts in the field of animal behaviour are wrong). That doesn't mean that someone can do it wrong, and get results. If you used these methods for a year, then I might take another look from scratch and find out where you're going wrong.

Aversive-based methods on the other hand are very unsophisticated. They all can be summarized pretty quickly - "correct" the dog every time it does something you don't like. Pretty basic. Hit, jerk a collar, yell, threaten, kick, choke, drag, etc. Dog barks, yell or "correct" (punish). Dog pulls on leash, yell or "correct". Dog jumps up, yell or "correct". Everything is treated the same. Very easy. It requires zero ability to figure out why the dog is engaging in an undesired behaviour (fear, territory, prey drive etc). It requires zero understanding of even the basics of behaviour modification and extinction (e.g. the fact that how you cure - not suppress - fear aggression, and how you cure - not suppress - dominance-aggression, are different). Just yell or "correct". All undesired behaviours are treated the same. The dog does something you don't like, punish it (corrections are punishments). You are not changing the dog's negative association with the trigger, you are not changing the dog's emotional reaction. You are simply wanting them to stop xyz right now. In most cases, aversion-based methods make the negative association and the emotional reaction stronger. However, fear is a powerful thing. Being hit, kicked, even yelled at is something distressing for most dogs, and they will do a lot to avoid it. So their fear of punishment is greater than the negative emotion that has caused the behaviour - they learn that humans are volatile and yes, they will often cease the behaviour you don't like. The behaviour is not extinguished, it's suppressed. The dog's fear of the consequence is greater than the emotional response that's causing the behaviour - which should tell you something.

There are many different techniques in training. Aversion-based trainers (eg Cesar Millan, Brad Pattison, Don Sullivan), who are not respected by the behaviour community, basically use one. Flooding and punishment. Cue them to do something you don't want the dog to do so that you can punish it for doing what you don't want it to do. Very easy. There are still people raising kids that way too. A human equivalent would be luring a 3YO onto the road by putting it's favourite toy in the middle of the road, then giving them a spanking for going on the road. The child will probably stop going on the road pretty quick. The question is, is the child educated? Or are they scared of another spanking? Have they learned that roads are dangerous? It takes a lot longer and isn't as easy as just spanking, to explain cars/traffic to them, get them to understand that roads are dangerous, and show them the behaviour you want (look both ways, stay in a defined area). They are avoiding punishment. When the parents aren't around, is that child going to stay off the road the same way as the child who was taught about road safety? This is what behaviourists mean when they say that these methods cause aggression. You yell at a dog for growling, he may stop growling at you. Then when the 6YO neighbour is over and grabs for the dog's rawhide, they are MORE likely to get bitten - because the dog has not learned that they don't need to guard their resources from humans, they've only learned not to growl at that guy who hits them for it. If this is "these methods work" to you, then you have different goals for your dog than I do. To me, "working" means extinguishing, not suppressing a behaviour. Doing it in a way the dog enjoys (like the child being read a road safety book as opposed to the child lured onto the road and spanked) is also important to a lot of dog owners.

"Dogs are animals not humans and human phsycology has no place in the animal world. It's almost as if you are seeing human emotions in the animals and trying to find the underlying cause of those emotions to correct their behaviour."

Well, I don't know any animal behaviourists who use human psychology, although there are a ton of similarities. They do recognize however that humans are animals, and learning theory works the same way for humans and animals. With the obvious difference that humans are capable of more sophisticated cognitive functions, we do still learn the same way. Google "exposure therapy" or "systematic desensitization and counterconditioning". The experts in human behaviour modification and dog behaviour modification are using the same methods. It doesn't appear you have any background in psych or behaviour modification so I'll leave it here and not get too complicated but animal behaviour is animal behaviour, learning is learning, with many more similarities than differences between species - the main differences being simply levels of complexity. This is one reason that one of the TV "trainers" recently had behaviourists shaking their heads when he said chicken camp had nothing to do with training dogs and was "stupid". Just wow. That should tell you tons.

Good behaviourists know that dogs do not have the same emotional range that humans do, but a pretty significant cornerstone of dog behaviour is understanding the root of the behaviour, and while their emotional range may be more rudimentary than ours, it's naive to think they don't exist, or that they aren't the motive for various behaviours. To not understand/accept that is like treating a medical condition by medicating symptoms and not doing anything about the actual diagnosis. For example, treating a dog with advanced heartworm for his shortness of breath, etc. You may get a suppression of that symptom temporarily, but if the heartworm is not treated, it's still there, and treating symptoms will only get you so far. The shortness of breath will come back. It's not "fixed". This is not a perfect analogy however, because heartworm is a fatal disease, whereas for example fear of strangers isn't, and you can suppress behaviours more easily and longer than serious medical symptoms.

When people defend fear-based training methods with "but they work!" and I answer as above, I usually find what comes next is "my dog is not scared/damaged at all, he loves me". It is the nature of many species, a few in particular (humans, dogs) to be subject to even horrific abuse and to continue to try to earn approval from their abusers. Abused dogs continuing to return to owners for abuse, children clinging to and even defending the parents or others that have abused them. Because the dog doesn't have a sign on it saying "obviously, I'm hit" does not mean they're not abused or even happy. For those who understand behaviour (behaviourists - not actors with dog training shows on TV or people who say they don't need to go to school, they "watched dogs" and do what dogs do) it is very obvious when dogs are stressed, anxious, fearful, avoidant - and it's very obvious that those are the primary body languages you see in dogs trained using aversion-based training methods.

Last edited by doggirl; January 23rd, 2013 at 04:13 PM.
Reply With Quote