Monday, July 25, 2011

Seminar, corrections, attitude and lots of other stuff

This is long. Sorry. See what happens when I don't blog for a couple of days? My head gets so full of stuff I have to do a brain dump.

The Laura Romanik seminar was great! Possibly the best part was her use of detailed handouts that reflected the topics she covered so I could actually pay attention to what she was saying and doing instead of trying to scribble 100 pages of notes that I probably wouldn’t be able to read after the fact anyway.

Got some great new ideas to try. Of course, I’m not going to run out and change the way I train everything but there are a few things I can implement and I'm looking forward to working them into my training with Phoenix.

Ironically - or coincidentally - she did a great segment on corrections: when and how to use them and why they are important. I can’t repeat her entire seminar here but if you get a chance to see her, it’s worth the time and money.

Interesting comments from readers on the “Are corrections really necessary?” post! Honestly, every single point you guys have made has been bouncing around in my head during the last month, since I’ve overhauled the way Phoenix and I are training.

I’ll address some of those comments and try to make my thoughts clearer. I stand by my belief that corrections are an important part of obedience training if you want to achieve anything beyond a CD or if you want to achieve competitive scores at any level. Like many of you have said — and it bears repeating — they're not harsh or abusive. They're just an information route.

A correction is information about what you’re doing wrong and how to do it right. Imagine trying to learn a new job. You’ve been working at it for awhile and think you understood just how it should go. Your boss looks at your work, throws it out and says, “Start over and maybe you’ll get it right this time.”

Not too helpful, huh? If your boss had pointed out the place where you had made the mistake, you could remember not to make that mistake again. In the future, if you made different mistakes, your boss would be there to point them out, allowing you to eventually master the task at hand and be able to do it confidently and correctly, knowing exactly how it should be performed. That’s what I want from my dog: I want him to know exactly what I expect and for him to be able to do it truly independently, with only one command - no repeated commands, do-overs or cookie waving. And yes, I totally admit I’ve fallen short on this with Phoenix. Lessons learned. My previous dogs were incredibly tolerant of my bad training habits!

Different dogs need different types and levels of correction. I believe there ARE dogs out there who need little or no correction while others will question everything you ask them to do for their entire career. My sheltie Connor was one of the former. I am starting to think Phoenix is one of the latter. Jamie fell somewhere in the middle.

One reader asked how will corrections solve the problem of a dog losing his “support system” when we get in the ring where neither corrections nor treat/toy rewards are allowed? By correcting the dog for errors in training, the dog learns what is the right response and what is the wrong response. It erases the gray area of confusion. A dog who knows how to do the job will be confident when he goes into the ring and won’t NEED a support system to help him perform beyond what the handler can give with verbal praise, petting and body language. I think this is a state of nirvana that a lot of dog and handler teams never reach. They go into the ring in a constant agony of worry, lacking the trust and confidence that YES, their dog totally understand what to do.

Another reader mentioned the genetic factor when it comes to a dog’s trainability. She was right on! It’s no accident that many of the top OTCh. dogs in the country are from the same kennels or that when ultra competitive trainers look for a puppy, they go to kennels known for producing high-achieving dogs. That’s not saying these are going to be “easy OTChs” but it’s going to be easier to put an OTCh. on a golden retriever with three generations of OTChs on both sides of the pedigree than on one from, say, conformation lines only and no history of performance titles. Not saying that can't be done but the dogs from OTCh. lines have traits that make them highly suitable to the demands of training for that level of competition.

Most of us trainers, OTCh. and non-OTCh. alike, simply train the dog we have. We buy a dog because we like the breed, we have established a relationship with the breeder, someone else recommended the breeder, they have healthy dogs, etc. Sure, we put some time into researching pedigrees but the bottom line is we get our dogs not because we want an “easy OTCh.” but because we want to share our lives with this particular breed, for whatever reason.

None of my dogs have been chosen because they had a long string of performance titles behind both the dam and sire. Knowing that, I realize it’s largely up to ME to produce the motivation, compulsion, desire, etc. that will turn us into a winning team.

One more question from the comments: what if Phoenix makes a mistake during the non-food stage of our training? If it’s not an NQ-ing mistake, right now I’m letting it slide. Our bigger picture is stringing together a successful, passing performance in all exercises. I’ll get back to the precision element later. Yeah, I care about heeling bumps and crooked fronts but I don’t care about them right now!

If he makes an NQ mistake (doesn’t drop on signals, for example) I have to ask myself: was he not trying (result - correction) or is he really confused and doesn’t know what to do (result - pull this part of the training OUT of the non-food work and work on it individually. A bare minimum of food can return at this time but only if the dog shows me genuine effort. Flooding him with treats is not going to suddenly make everything clear in his mind.)

BTW, I have brought toys and play back into our training picture on a limited basis. Phoenix has several exercises that are rock solid (did I really type that out loud? what am I thinking!?) and after a successful sequence of 2-3 exercises (done formally, front, finish, the whole works), I will release to a toy and play. But this is not a mid-exercise release and I’m not doing it with the intent to reward or build enthusiasm. It’s just play because I enjoy playing with my dog. More on that in future posts, too.

Oh, and several of you mentioned attitude: I absolutely agree - I do not want to show a dog in the obedience ring who does not want to be there. I want my dog to be happy when we train and show. No, correcting a dog is probably not going to make him “happy.” But think about this - if you were doing a job that you were very uncertain and hesitant about because you really didn’t understand how to do it properly, how happy would you be? What if someone stepped in and corrected you when you made mistakes? Believe me, I’ve been there! Being corrected didn’t make me happy either but it helped me learn the job. When I got good at the job, I was much happier and could truly enjoy the work and perform it well.

Please understand that my corrections are followed by genuine, sincere, heartfelt, honest, appreciative (help, I’m running out of adjectives) praise when he gets it right. I am not browbeating my dog. I don’t expect him to work like a golden retriever who wags his tail nonstop and is delirious about obedience. Phoenix is an insane nut around the house (flower pot on his head, stealing socks, etc.). He's serious when we go to train. He CAN be nutty in training, depending on the exercise. I think he'll get nuttier when our understanding of each other improves. I can live with that. If he were truly miserable, we'd stop this obedience game and do something else. But he's not. The journey continues.

Something I’d like to point out from Laura’s seminar and this is only MY interpretation, but if you have to correct for something more than twice in a row, you need to do something different with that particular skill - don’t just keep correcting the dog. Either make the exercise easier or go back and strengthen whatever part of it the dog is having an issue with. Repeatedly correcting without getting better results is going to create learned helplessness where the dog thinks, “I do not know what to do. I cannot win. I give up.” This is very sad.

Okay. There’s more to come. I just have to get my mind wrapped around what I want to say next.

7 comments:

  1. This is very timely for me as well. My Scorch tends to be a very over-the-top, make a million-efforts-in-2-seconds kind of dog. Corrections can sometimes seem to center and calm him. He craves clarification and I will use whatever option seems most efficient at getting the message across.

    Your posts about always throwing cookies at your dog hit home though... I'm working on integrating play with ME into our training. Scorch generally does not like petting when he's amped up (i.e.- in the ring) so coming up with more interactions is a challenge.

    ReplyDelete
  2. Again, thank you for sharing your journey with us.

    As I read the blog and the comments, I wonder if we have a problem with semantics. When I am in the world watching obedience training, I see the word "correction" used to mean everything from withholding a cookie to "pain compliance". In traditional obedience, I'd say "pain compliance" is the more common usage, even though it's rare to hear someone be that blunt. Words or phrases like "pain compliance" or "compulsion" don't sound as nice as "correction". I'm not saying corrections have to be pain compliance, but I'm more interested in what people are actually doing than in what they are calling it.

    You have mentioned that a correction should not involve physical pain or fear and intimidation, so I am very much looking forward to seeing how you will apply this to Phoenix's training to achieve an effective result.

    Right now, there is way too much physical pain and intimidation in obedience, and that needs to change for the sport to survive.

    ReplyDelete
  3. "I stand by my belief that corrections are an important part of obedience training if you want to achieve anything beyond a CD or if you want to achieve competitive scores at any level."

    Completely, utterly, categorically, and adamantly disagree with this. Yes, there are few clicker trainers in obedience who are really lighting up the scene. I believe this is due mostly to reasons other than "you must correct." Foremost, there is little support system in place for the handlers. By and large, they are going it alone with a seminar here and there and some e-mail lists. Obedience is not a clicker friendly environment, in part because of the must-correct attitude, and many prospective competitors defect to agility which has a stronger history of R+ methods.

    "A correction is information about what you’re doing wrong and how to do it right. Imagine trying to learn a new job. You’ve been working at it for awhile and think you understood just how it should go. Your boss looks at your work, throws it out and says, “Start over and maybe you’ll get it right this time.”

    Not too helpful, huh? If your boss had pointed out the place where you had made the mistake, you could remember not to make that mistake again."

    Nothing in the clicker training zeitgeist would lead to this. You would end at the point of error, probably with a NRM which can be viewed as a correction. I think of it as mending a hole in a pair of pants. Start outward, reinforce the edges and work your way in so the problem area has support around it.

    "In the future, if you made different mistakes, your boss would be there to point them out, allowing you to eventually master the task at hand and be able to do it confidently and correctly, knowing exactly how it should be performed."

    And in an alternate interpretation of this metaphor, you have a boss that is constantly hanging over your shoulder and you're on pins and needles waiting to find out what he's going to jump on you about this time. *I've* been *there.* Yes, this is largely a matter of balance and reading your dogs and I don't mean to insinuate that you cannot find the correct balance with your dog, but this is just as good a way to confuse your dog as any.

    "If he makes an NQ mistake (doesn’t drop on signals, for example) I have to ask myself: was he not trying (result - correction) or is he really confused and doesn’t know what to do (result - pull this part of the training OUT of the non-food work and work on it individually."

    And you aren't even using your own metaphor. You aren't correcting wrongness, you're correcting lack of effort (this is a topic I would like to hear more on, because I hear traditional trainers go on about it all the time, but they never explain it). You are specifically *not* correcting confusion, which would correlate to getting red marks on your returned report. *This* is what's wrong, *this* is how to fix it.

    What I think you are doing has more to do with building value for the work itself and the things in the ring than any magical power of correction and showing the dog what things are incorrect.

    " if you have to correct for something more than twice in a row, you need to do something different with that particular skill - don’t just keep correcting the dog. Either make the exercise easier or go back and strengthen whatever part of it the dog is having an issue with. Repeatedly correcting without getting better results is going to create learned helplessness where the dog thinks, “I do not know what to do. I cannot win. I give up.” This is very sad."

    Hear hear, on all counts. I think I will put LR on my seminar wishlist. I imagine I'll be all kinds of outraged and indignant, but come away with a lot of knowledge.

    ReplyDelete
  4. Hi Raegan,

    I think the best we can do is to respectfully agree to disagree. I don't want this to become a battle of clicker trainers vs the world of obedience, blah-blah-blah. Everyone chooses their own training methods and there's enough room for all of us. But that's not to say we can't learn from each other and I do appreciate your thoughts. I helps me organize my own!

    Neither can I explain every detail of my thoughts with crystal clarity in this blog so there is never any confusion about anything. This is a tough subject to communicate about face to face with total understanding - let alone in words over the Internet. I am sure some things are getting lost in the translation.

    Just a note: you can't correct confusion with a correction. (Confused? See what I mean!) In cases where I feel my dog is truly confused about what to do, there IS no correction. Just strengthening of that skill and/or maybe some re-training.

    I will try to write about wrongness/lack of effort, which ARE all the same thing, at least in my book. But you have to be really honest with yourself when asking is my dog making mistakes because he isn't trying on an exercise he KNOWS how to do . . . or maybe he doesn't really know it as well as I think and he needs help, not correction because I assumed a level of proficiency that really isn't there.

    ReplyDelete
  5. I can respect that. :) A lot of the verbiage and concepts are totally new to me, my background includes practically no traditional methods. Well, I did read a Barbara Woodhouse book when I was 9. So I guess I'm just going through an extinction burst ;)

    ReplyDelete
  6. Like Denise, I would like to hear some specifics of how you translate "a correction is intended to reduce confusion and provide information" into action. This would really help me understand what "correction" means to you. I had begun to think you meant physically positioning the dog or something like that until you mentioned correcting for lack of effort but not for confusion.

    In the clicker training of my older Malinois, who is very sensitive to my mood, I found that she reached a point where she clearly perceived the mere fact that I did not click a particular trial of a behavior to be punishing. The only thing I could do about this was break the behavior down into smaller pieces so that her success rate would be higher, since you can't click just to make the dog feel good about itself. This problem went away as she gained a more sophisticated understanding of the process.

    My only point is that something as mild as witholding a click could potentially be perceived as a correction by the dog, and I am quite interested to hear what you mean when you say correction.

    ReplyDelete
  7. "Repeatedly correcting without getting better results is going to create learned helplessness where the dog thinks, “I do not know what to do. I cannot win. I give up.” This is very sad."

    I see this with a few friends dogs and I haven't been able to put it as eloquently as you have. May I borrow your words? It is sad to see a dog fail because the trainer isn't' getting the point across. Heck, I know I did it early on in my training with my NovA dog! Now if my dog acts confused after a "correction" I step back and do something easy to build him up to the understanding of what he did wrong and how WE can fix it.

    Very interesting stuff and I'm using it to motivate me in my training. THANKS!

    --jill

    ReplyDelete