Friday, June 4, 2010
Skimming Drills with Time-outs
For the last three days, Laddie and I have continued to work on the Skimming Drill, using our Bumper Boy for setups with both high cover and water as obstacles.
Each day, I've experimented with increasing distance to the entry point of the obstacle, and narrowing the width of the section cut off, the two factors that seem to have the most effect on difficulty. Laddie has been making steady progress for both kinds of increase in criteria.
I've done little handling. If Laddie attempts to skirt the obstacle, in most cases I call him back. Today, for example, I used no handling at all.
In today's Series A, Laddie did a great job on two retrieves running thru a corner of high cover. The distance to the entry was relatively short compared to some of our earlier set-ups, but the amount of cover to be cut off was smaller. In other words, I relaxed one criterion while raising another.
For Series B, we moved to the pond. Again, our distance was a little shorter than yesterday's at the pond (50 yards to the entry versus 70 yards yesterday), but today's slice of water was narrower than yesterday's.
As Series B progressed, I ended up making one change to our procedure that I had not planned in advance.
For the first three throws of Series B, Laddie tried to run the bank the first time I sent him, then took a good entry when I called him back and sent him again. When he tried that two-try pattern again and attempted to run the bank on the first send-out of the fourth throw, I called him back and made the snap decision not to give him a second opportunity to perform that particular retrieve. Instead, I walked him quietly to the van, put him in his crate, and closed the back gate. All the windows and a side door were still open for ventilation, but I wanted him to have a clear sense of a time-out, a procedure with which he has almost no experience, but which seemed appropriate in these circumstances.
After reloading the BB, I brought Laddie back out and we ran the same challenging retrieve again. This time he took the correct entry the first time I sent him after the throw, breaking the chain at last. Yay!
However, when I fired another bumper and sent him again, he again tried to run the bank. So for the second time today, I called him back, we walked to the van together, I had him get in his crate, and I closed the back gate of the van.
After reloading the BB one more time, I brought Laddie out and we ran the retrieve once more. Again he took the correct entry the first time I sent him after the throw. And then I fired the other bumper and sent him again, the only difference that I ran him from my other side, the more difficult side. It did no harm. This time, too, he took the correct entry.
I was pleased with the results of using the time-out for this situation. I think I could have taken a different approach, which would have been to move closer once I saw that he wasn't ready to run the first send out correctly four times in a row.
However, the fact that Laddie ran correctly the second time in each case seemed significant to me, as though he were learning a behavior chain of running the obstacle incorrectly on his first try, and then running it correctly after being called back. Not efficient, but not so easy for him to fix himself because of his strong impulse to run the bank and get to the fall that much faster. I felt that showing him a specific undesirable outcome for the two-try pattern, the lost opportunity to complete that retrieve at all, might be more appropriate in this case than simply lowering criteria by backing off on distance or increasing the width of the slice.
Why? Because I sensed a risk that if I didn't show him an undesirable outcome, the two-try pattern would remain in his system only to reappear in less convenient circumstances, such as group training or an event. In group training, the chain would probably be reinforced because I'd be reluctant to give up the entire series, and would be more likely to call him back and run him again. In an event, I wouldn't take any action at all and would hope that he could correct his line and complete the mark despite running the bank. If I was going to try to put an end to the two-try pattern, this seemed to be the right time to do it.
I don't know whether making it easier would have been better in this case, or if the time-out was better. I would not want to use time-outs very often. I can imagine undesirable side-effects of a time-out, such as Laddie someday intentionally running a bank if he was tired and wanted a break. But today, as a novel procedure, it seemed to work well.
It occurred to me later that perhaps another way of addressing the two-try chain would have been to use a Walk Out rather than calling him back. WOs have been quite effective for other difficulties in the past, and they appeal to me much more than using a time-out. I think that's what I'll try if it comes up again in the future.
Perhaps I shied away from it today because I'm currently suffering with a gout attack, which makes walking difficult. It's just something I'll have to deal with.