Love. It Works.

One of the things I enjoy about training dogs is that -- like people -- one size doesn't fit all. Training is a constant and thoughtful process of assessment, planning, implementation, and evaluation. But it is not just the dog that is constantly in this change process -- to be a good trainer, the human must also continually engage in a parallel process. Life with Dogs is all about becoming a better human being -- or finding out that you are missing the mark so you can try again.

Yesterday I wrote about a Missoula-based trainer behaving badly when a dog she was going to show in Utility would not do the down signal during a practice; I observed that her behavior was both mean and dumb, and I stand by that assessment. I also shared that Zoey went through a spell when the Down Signal -- to her -- seemed to be interpreted as a signal to become a statue. Today I want to discuss the strategies I utilized to work through the Statue Phase with Zoey.

1. Assessment

We cannot always know why a dog or a person does what she does, and so assessment should be flexible. That said, I observed that Zoey seemed a bit stressed or worried or maybe distracted in the ring upon the first down signal, but would go down immediately with a repeat signal or verbal cue.

Hmmm...

I assessed whether my dog really understood the down signal even when something was competing for her brain cells. I found she struggled to go down if I was holding her ball or food. Good info. 

I discovered the signal was setting-dependent  -- Zoey was 100% reliable with the down signal in some places but not others. The presence of people had less impact -- it seemed to be setting that was the bigger issue. 

And so my assessment was that Zoey had an inadequate understanding of the down signal when pressure (ball, food, certain settings) was applied.

2. Plan

I did not wish to turn the signal exercise into: Signal/Stand/engage brain/Signal/Down -- that is a danger of repeat commands. And missing the first signal is a flunk anyway so what is the point of a second command, besides establishing an undesired pattern? And so I decided to stop double-commanding in the ring or in practice.

I also decided to leave the ring when Zoey failed an exercise -- after doing something well so as to end on a positive note -- to avoid making unnecessary withdrawals from the training bank, so to speak. You do not train by showing a dog.

I decided to only show Zoey where ring rentals/matches were offered before the show - this was to reinforce the ring as her Happy Place before showing in it.

In terms of training, I considered the goal (UD title) and decided I did not care if she could down all over the county -- she only needed to down in the ring. My plan was to apply the pressure of a ball or food to the signal but in familiar places, and to not worry much about training the down signal in hard places. (Reason -- because I did not want to potentially flood the down signal with stress as we worked the "pressure" piece by also adding a hard setting).

My plan for "failed" down signals in training was to turn away/ignore it, and to look for data to inform the ongoing assessment. NO Positive Punishment (i.e., adding an aversive for undesired behavior) -- hopefully that goes without saying.

I also decided Zoey and I should take a break.

3. Implementation

I took the winter off from any real training.

After the Sparklers left, I started up again -- I trained the down signal in "easy" places and used her ball (and/or food) as both pressure and reward (Premack Principle).

Except for the Specialty, I selected shows based on ring rental/match opportunities.

I did not "correct" her for failed down signals -- I simply viewed it as her missed opportunity to play with her ball, did something else that was easy, and tried again.

4. Evaluation

We showed in Bozeman in March -- she did the down signal just fine, and earned her second UD leg.

We showed at the Specialty in April -- she did the signals and then  had a poop episode in the ring (that was a major signal to me that something was wrong with her).

Zoey, almost nine, was then diagnosed with B-cell Lymphoma, and chemotherapy commenced. Talk about putting the down signal in perspective! 

The chemotherapy protocol is four weekly treatments and then a week off and another four weeks and so on; with her veterinarian's approval we extended the time off to two weeks so that Zoey could show in Utility again.

We showed in Great Falls twice -- down signals were perfect (go-outs were not).

We showed in Missoula twice -- down signals were perfect again, and she earned her third leg to complete the title.

GCH Kaibab'z Forever Bright UD TDX AJP AXP OA OAJ NF RN DD BDD VCD3

BMDCA Versatility Dog Excellent

BMDCA Working Dog Excellent

BMDCA Top Producer of Working Title Dogs (in a single litter)

Heart Dog

P.S. After I wrote this, I decided to take some photos to go along with the content. I opened the article box  for the first time since the show and found this note:

Wow. What an incredibly wonderful thing that someone did to leave me that note to find. THANK YOU, ANONYMOUS DO-GOODER. 

And I want to add: My dog brings out the best in me heart

Have a Happy Sunday -- and spread the love, just like the Anonymous Do-Gooder did. It is pretty powerful stuff, that love thing.

P.P.S. And speaking of that love thing and in case the message isn't as obvious as I hope -- my dog got her UD and all her other titles without the use of aversives (positive punishment).

Love works. I promise.

3 comments

by Jill on Sun, 07/02/2017 - 11:35

Love, fun with dogs, & Do-Gooders. Good stuff!
Congratulations on that UD and best wishes to Zoey as she goes through treatment.

by Alison J on Sun, 07/02/2017 - 15:29

I love the Premack Principle and agree with much of what you write, Mary-Ann. However, I offer another perspective: that ignoring a dog's response to your cue and/or withholding a proffered reward are actually both aversives. Certainly they are substantially different aversives compared to the extremes you described from your shows, yet they are still aversives.

Example: I really like strawberries (and they are in season right now - yum). You are trying to teach me to do a deep knee bend (I need more exercise) on a signal from 40 feet away while you hold a pint of gleaming, red, juicy strawberries - my distraction and reward. For some reason I do not respond to your signal. Perhaps I was distracted by a Great Horned Owl flying past behind you, or your signal wasn't clear to me today or I just dozed off as you walked away from me : ) If you respond by soundlessly walking away I WILL be disappointed as well as baffled. Why did I not earn my strawberries? I do not know what I did or did not do to fail to earn my strawberries, nor do I have any new information to help me succeed the next time. Oh no, next time it may be a Sandhill Crane! I want my strawberries!

As you very appropriately state, strangling and choking are plain mean, however I prefer to end an attempt by providing my dog with information on how to succeed and then praising and rewarding him for that success.

Another example: Since a key aspect of the signal exercise is that the dog not move forward for the down I have taught a 'walk-back' game. Dropping into a sphinx position is much easier for a dog while moving backwards. So, if my dog does not drop I would move closer, ask him to 'walk-back' and then signal him to drop. He drops, I praise and reward and we have ended on a success and the dog has information to go forward with on how to succeed.

Another related game, taught up close, is to teach the dog that, once down, a repeated down signal means to scoot back in the down position. Likewise with the sit, once sitting a repeat signal means a scoot back in the sit. Great indoor game to teach on rainy or, heaven forbid, snowy days. This enables me to play games with my dog during signals. He drops well, 'good dog", another drop signal, he scoots back, extra good dog and he has been doubly reinforced for thinking backwards not forwards during this exercise. Same game with the sit. This also helps avoid pattern training stand-down-sit-come when additional down-backs or sit-backs can be interspersed. Of course there are also release games for doing well as well.

So, a different perspective for consideration.

by Mary-Ann on Mon, 07/03/2017 - 06:49

Your post emphasizes the importance of the assessment part of the training process -- thank you! 

I use the term "aversive" for positive punishment, a term that means ADDING (hence the positive part) something in the face of an undesired response; the goal of positive punishment is to decrease the undesired "thing" (i.e., standing like a statue). An aversive, in my mind, can also be a negative reinforcer, which is all about stopping something unpleasant (think shock collars) by offering desired behavior. 

Negative punishment is taking something away -- again with the goal of decreasing the undesired behavior. A "positive" trainer may well take away the ball or attention to get something to stop. It is all about the goal.

On the other hand, positive and negative reinforcement are utilized to increase the likelihood of a desired behavior. Here is my quick and easy summary of the differences:

Punishment = Make it stop!

Reinforcement = Make it happen!

My focus in training is typically on "Make It Happen" and so positive reinforcement is the way I prefer to achieve desired behaviors. Telling a human or dog to STOP may create anxiety because s/he doesn't know what to DO, and anxiety interferes with learning. I find it most helpful to train the DO and save STOP for such things as jumping off cliffs ;)

I see a break in training (walk away and try again) as neither punishment (because my goal is not really to stop anything -- except maybe to stop me from having criteria that is above the current capacity of said dog) nor reinforcement (because I am not really trying to make something happen by walking away) but as a a reset of sorts. It allows me to try again with lowered criteria to create a circumstance in which my dog can gain a positive reinforcer, and work up in baby steps to the completed, desired behavior. 

That said, if my goal is to stop the stand like a statue behavior, then walking away is Negative Punishment. It can be very hard to: 1) make the distinction between various punishments and reinforcers; and, 2) truly understand our goal, which is needed to describe what we are really doing in terms of punishers and reinforcers.

I want to amplify one of your points -- making a dog "guess" what we want may be problematic. I do believe it can sometimes be very helpful to let a dog puzzle things out, but I also think showing desired response can be extremely helpful in the learning phase.

And again -- you point to the importance of assessment. Zoey seemed to know the down signal -- the issue was she was not responding to it with pressure, which maybe is a certain way of not knowing...

At any rate, my assessment was that pressure was the issue -- not really the down signal. Therefore, I needed to use baby step with the pressure and not really worry about the down signal, if that makes sense? In other words, I had to titrate the pressure in training until I had increased her ability to respond to the down signal under ring condition pressure.

Had I insisted on response to the signal with full pressure applied from the get go, Zoey's concerns would have increased and/or I would continue to have to double command, and I do not believe I would have the reliable down signal response in the ring that I now have (knock on wood) -- hence the importance of being able to accurately assess the dog in front of us, and pull apart the strands of training.

Thank you for sharing your thoughts - and giving me a chance to clarify mine!

 

 

Post new comment

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
By submitting this form, you accept the Mollom privacy policy.
Contact © 2011 Website built and maintained by G.Sontag.