Hi Kathleen, I've gotten a number of private posts questioning this thread, so please share this with the list if you think it will add to understanding of the process of eliminating undesirable behavior. Hi Kathleen & List, Thanks for keeping me up-to-date on your progress with Marcie's growling. Since I've gotten a number of private posts asking a variety of questions about this thread, I thought I'd offer some clarification that might help sort all this out. The behavioral approach (clicker training) focuses on *behavior*, not what the animal may or may not be thinking. It's not that I don't care what a dog is thinking (anyone who loves dogs certainly wants to think there's good stuff going on in there), however, the problem is that we don't truly *know* what the dog is thinking -- we can only guess. Sometimes we probably guess correctly -- someone who is good at reading and interpreting body language most likely has a reasonable handle on it -- but not everyone can do that, and how does one assess whether we're right or not? After all, we can't ask the dog <g>! Anyway, the bottom line is not to get bogged down with what the dog is thinking -- it only matters what the dog is *doing*. For example, when you call your dog and he's barreling toward you on a recall, do you really care if the dog is thinking: "Darn it, I was having such a good time playing with Spot. I wish Mom hadn't called me away just yet."? Of course not. While we all want our dogs to "be happy", we are honestly more concerned that they behave in a desirable fashion. By focusing on behavior, you are then in a position to determine what affects it -- triggers, increases or decreases. As you have learned by experience both with Marcie and with your dogs in the past, using aversives reinforces growling behavior. How do you know this? Simply by looking at the frequency or strength of the behavior. Whether a behavioral effector is a reinforcer (strengthener) or a punisher (reducer) is determined by *its effect on the behavior*, NOT on how we feel about it. In other words, while we humans intend that aversives such as a collar pop, scruff shake, bellowing "NO!", alpha roll, etc. are "corrections" (which is just a politically correct reference to "punishers"), the truth is that when they are used in an attempt to reduce growling behavior, the vast majority of the time they not only don't reduce growling, in fact growling (and aggressive behavior) escalates. Therefore, by definition, if the behavior increases, the use of an aversive in this instance served as a reinforcer. The effector is judged by its effect on the behavior. Now, because you know (from experience, reading, study, etc.) that growling often increases by using aversives, is there another approach that will work to reduce growling? Well, experience (trials, reading, study, etc.) has proven to me that ignoring growling and concentrating on reinforcing *another* behavior works to reduce or eliminate growling behavior. I can make educated guesses as to why this works this way, but from a behavioral perspective, my personal theories and thoughts don't matter (although they make fun conversations over a few beers <vbg>). The only thing that matters is that the dog's behavior changes, growling is reduced, and another, acceptable behavior replaces it. Has the dog changed his thoughts? Who knows? Who cares? All we can know is what we observe -- that the dog's *behavior* has changed. Some of the private posts I've gotten question the correctness of clicking a behavior that you want to reduce. How come, some have asked, Marcie's growling doesn't increase when Kathleen clicks it? Since I'm not actually seeing what Kathleen is doing, I can only surmise that Marcie is not continuing to growl beyond the first or second click, and that the behavior that is in fact being reinforced is *not* growling but something else -- perhaps even eating the treat. Marcie is doing lots of other stuff, and she's quickly dropped growling as one of the behaviors she's throwing out to see what is getting clicked. Kathleen is astute and experienced enough to know that when and if Marcie growls to test whether or not *this* is the behavior Mom is clicking, Kathleen would ignore that one and click something else. On the other hand, if you want to reinforce growling, set up the session to get the behavior, and reinforce it. Here's another example that might help clarify. Last night at an instructor's meeting, we helped one of our assistants sort out a problem with her adolescent GSP who recently started showing a fear response to being examined in handling class (clicker). (Before speculation begins -- Jenny is well-socialized, comes from good genetic temperament, etc. etc.). Her fearful behavior started a couple of months ago (around 8 months of age -- for which there's a likely explanation, but that's another topic unrelated to clicker training), and Jenny has gotten progressively worse as the weeks went on, to the point that as soon as a person approaches her, she starts shivering, tail down, ears back, and tries to back away. Pauline (owner) wanted explanations as to why this happened -- what is Jenny thinking? Is she really afraid? Does she think the people will hurt her? Was it Pauline's fault? Did something bad happen that Pauline didn't know about? Did she unintentionally reinforce Jenny's fear? The only question I could answer with any assuredness (although I answered quite a few speculatively <g>) is that something was reinforcing Jenny's *behavior*. How do we know this? Because nothing Pauline has done to try to eliminate the behavior has worked, and the behavior has increased. So we started reinforcing different behavior to modify her fearful behavior. Jenny looked at someone - click. Didn't matter if her tail was up or down -- we just needed to get started. And then we reinforced any positive behavior toward strangers -- looking alert, standing confidently, having tail up, putting ears forward, taking a step toward someone, standing still if they approached, etc. Very quickly Jenny was into the game and began boldly approaching people. Interestingly, Jenny's behavior deteriorated if the person became solicitous -- crouched, talked baby talk, or did any of the typical behaviors people engage in when trying to reduce fear in a dog. Clearly Jenny had learned an association between people's solicitous behavior and her own fearful behavior. I think this is a clear example of what happens when you try to deal with the thought process (reducing her fear) and not the behavior (acting confident). It often backfires. I hope this makes sense. Again, thanks for keeping me posted on your progress. Give that calendar pup Marcie a hug from me. Regards, Gail -- Gail Fisher The All Dogs Gym (tm) Manchester, NH 03103

Home