Titles Achieved to date...

Monumental A to Z High On Liberty
NW1, RATI, RATN, RATO, NW2, L1I, RATS, L1E, L1C, L1V, L2C, L2I, L2E, RATM, R-FE/N, PKD-TL, PKD-N, ADPL1, ADPL2, TD, UWP, ADPL3, NTD, TKN, L2V, ADPL4, SDS-N, ADPL5, ADPCH, ADP1(2), ADPL1(GC), ADPL2(2), ADPL2(GC), VPN, AP, UWPCH, ADPL3(2), ADPL3(GC), NC, NI, NE, SCN, SIN, SEN, CZ8B, NV, NN, ADPL4(2), ADPL4(GC), ADPGCH, ADPL5(2), RATCH, CZ8S, AI, TKI, AV, AE, AC, AN, R-FE/X NW3-V, NW3-E, SI, RN, R-FE/NS, CZ8G, SC, SV, SE, SN, SEA, SBN, SWN, SIA, SCA, ADP-1(Th), ADP-2(Th), ADP-3(Th), ADP-4(Th), ADP-5(Th), and ADP-CH(Th)... 81 and counting...

Sunday, April 30, 2017

Wait For The Cue

Recently it's become really apparent what an issue it is for Gimme to be in charge of over-offering behaviors which stems from not listening, all rolled up with problems about stimulus control. It has been a weak area all along, I just haven't put the focus on it. Its not an issue in some sports, where it's perfectly alright for Gimme to be in charge, such as: nosework, barn hunt and tracking (all sports which rely on her special nosey talent for success). Naturally, I don't blame Gimme - she is just doing what has worked for her all along. It's my job to wrestle to controls from her and be the grown-up here.

This last month we've been looking at cleaning up the alignment in her "take-a" bow behavior, since she's developed a tendency to swing her butt out 45º. I think this might be because she also offers "take-a" as her tracking indicator, where she swings out so she can look back and see if I've noticed and am dutifully trotting up to reward her find. In any case, her tendency to go right to offering when she thinks she knows what we are doing made it hard to get more than one or two reps in before she started offering. I needed to start her from a balanced stand to get a clean "take-a" and when she's offering, she's not balanced.

Anyway, efforts to teach her the value of being still as suggested on the list weren't working, so I decided to revisit a wait-for-the-cue game from yesteryear. To get ready I listed seven behaviors she knows well and can do in a small space. Then I used a random number generator to create a long sequence of numbers from 1 to 7, replaced the numbers with the behavior names and made up behavior lists... starting with 2 in a row (8 sequences), then 3 in a row (8)... and so on.

We couldn't get correctly through the first one! With all the work we've been doing with "take-a", Gimme decided any "down" from a "stand" (the first sequence of 2) really means, "stand"-"take-a". So I had to teach her she really can do "down" from "stand". The time it took me to realize what was going on gave her time to get frustrated, so she started throwing every behavior in her book at me - especially "bacon" (her fave). This game involves a lot of treats for waiting for the next cue, but she wasn't waiting long enough for me to get treats in her so I could interrupt the offering. Once I achieved teaching her she can "down" from "stand" I broke off the training and thought about it.

I decided to do short mini-sessions with the plan of focusing on one two part sequence at a time. I planned to start with a constant stream of treats and then gradually start to space them out from 1 nano-second to 1 milli-second to 1 micro-second to 1 centi-second to 1 deci-second to 1 second and so on...

So after 45 minutes, we did another session, with thirty treats and just the two behaviors "stand" and "down". I cued "stand", clicked it when she assumed it and then rapid fire fed her 8 treats... then cued "down" (lured and waited for the rump to drop) clicked and then rapid fire fed 10 treats... then cued "stand", clicked it when she assumed it and then rapid fire fed her the remaining treats...

This went better, she didn't have a chance to offer other behaviors and the behaviors I got were clean, even though I had to lure the down and wait on her rump. Then I cued "all-done" and went back to my evening project. She stood there dumb-founded for several minutes, "what the................." I finally had to call her to get on the couch to convince her it really was "all-done".

An hour later we did another quickie with the next sequence of two behaviors, "sit" and "turn" (CW spin). This went well and I was able to insert tiny pauses between treats. Gimme was a little more accepting of my "all done", though I still needed to call her to the couch.

Much later we did another sequence, "touch" (nose touch to hand) and "sit". This went really well and I was able to begin to take my treat hand away from her lips briefly, while she remained pretty still. When I said "all done", Gimme sighed, but then was ready to go to bed. I was nearly 1:00 a.m. after all.

I think this is promising and will likely improve as I work out some more bugs. I notice my click timing is not as good as it should be. Fortunately Gimme is pretty tolerant of my ineptitude.
 

No comments: