Sunday, March 25, 2012

Is barking a behavior?

After many months of trying to find a way to cast Laddie during a water blind without him vocalizing, I have finally come to the conclusion that Laddie does not comprehend barking as an operant (voluntary) behavior.

This is consistent with other vocalizing I have worked on with both Laddie and Lumi before him, in Lumi's case, barking while backing up as a freestyle move.

And perhaps more importantly, I've concluded that trying to train the dog to perform a desired behavior, while suppressing the vocalizing that has come to accompany the behavior, runs the risk of damaging performance of the desired behavior without even suppressing the barking, other than if the behavior itself becomes suppressed.

I'm not talking about vocalizing at the line. I suspect -- I'm not sure, but I suspect -- that that can be suppressed, especially if dealt with early in the dog's career, by taking the dog off the line instantly any time she barks, with 100% consistency.  At least I don't see the dog having some other, desired behavior suppressed by that approach.  I feel it would be safe to try, worst case being that the dog could never figure out why she was being taken off the line sometimes.

But I have repeatedly tried various versions of separating the correct response to a cast, with the vocalizing that accompanied it, only to see Laddie become increasing uncertain whether to move at all when cast.  If vocalizing is less than desirable, "ignoring" a cast is disastrous.  I do not believe a trainer can afford for the dog to have any doubt in his mind whatever whether to take a cast.

So is there any solution to the vocalizing? Perhaps.  Here's an approach I had some success with today.

The general idea is to train the behavior itself, adding some element that removes the dog's inclination to bark. After that version is well trained, slowly, gradually, fade the extra element.

Here's today's example of the concept. I placed a duck (highly desirable retrieval article) at the edge of the shore, letting Laddie see me do so from the start line we would be using up the shore from there. Then I walked halfway back to him and threw an orange bumper well out into the water. Returning to Laddie, I lined him up to pick up the duck.

Then I sent him, and he ran along the shoreline toward the duck. At the halfway point, I stopped him with the whistle and cast him into the water toward the bumper.  That's a cast Laddie will take without vocalizing, because her can see the bumper.  After he was a few feet into the water, I stopped him again and cast him back to the duck.  This again was a cast Laddie could take without vocalizing, since he could see the duck or at least had a clear memory that it was there.

Hooray, Laddie just performed a difficult water-handling maneuver involving two casts that would normally be accompanied by vocalizing, but without a sound.

We repeated this several times, always sending Laddie initially toward the bird, both targets always known and more or less visible. Sometimes I let Laddie  finish the original outrun and bring back the bird. Sometimes I let him complete the cast to the bumper.  And sometimes I cast him twice, first over, then back, much as you might need to do in an advanced on-and-off-the-point blind retrieve.  With the known and visible targets, all of that could be done without vocalizing.

But the bird wasn't that visible from the distance. And neither was the bumper if I threw it out far enough and the current carried it even further.  Increasingly, Laddie was taking his casts on faith that the article must be where he had seen it propelled toward. And also that it must be there because it's been in that same location throughout the season.

I ended the session with one send toward the bird's location even though no bird was planted. At the halfway point, I stopped Laddie and sent him "over" to the bumper, far out in the water and possibly not visible to Laddie from shore.  I had laid the groundwork for the double-cast maneuver with the final target not visible, and no vocalizing. That was enough for today.

But I can imagine that on another day, in the same or another location, I will eventually be able to call for the double-cast maneuver without a bumper in the water, as long (initially) as Laddie thinks a bumper is still out there, he just can't see it because it has drifted too far out to see from shore.

And then, perhaps someday, all that practicing of that maneuver, increasingly with invisible targets, will enable Laddie to execute that maneuver in a trial, where there really isn't a bumper out in the water, and he can't see the blind till he gets to it, all without vocalizing. Not because he learned to suppress the barking per se, but because he learned to perform the maneuver without barking.


No comments:

[Note that entries are displayed from newest to oldest.]