Victoria Farrington, Bob Bailey and Gary Wilkes all describe using a NRM (no reward marker) during shaping sessions.

Victoria’s explanation from the agility.uk forum really made it clear for me. Why waste time and energy during a shaping session letting the dog try every trick in the book. A simple NRM will help get the dog on track without a huge amount of handler interference.

I worked one NRM session with each dog today using Bob Bailey’s description. They both caught on right away (Katy was a whiz kid, of course). The only confusion that I have is that Bob seems to indicate that a cue is already in use for the initial hot spot. I’m wondering if the same cue (Sd) is then used for the *new* hotspot. Guess I’ll have to ask the experts…

For the session, I used pyrex refigerator bowls - a green round one and a red square one. I used un-unh as my NRM, but I think I’m going to use “not that”. It’s more concise.

I just read an article by Gary Wilkes where he suggests using a “working cue” while shaping. Once the behavior starts to look like something specific, start using the working cue. Use this while the behavior is refined and then when the behavior gets to 99% reliability, add the “performance cue”.

Clicker Retrieve
From Deb Jones
with modifications for using a tennis ball in preparation for flyball.

TRAINING STEPS:

I. With tennis ball in your hand, close to you

1. Look at ball
2. Move towards ball
3. Sniff ball
4. Touch w/nose
5. Touch with mouth/teeth
6. Put mouth over ball
7. Close mouth over ball
8. Hold in closed mouth
9. Increase hold duration

II. With ball placed on floor within 3 ft. of you

10. Look at ball
11. Move towards ball
12. Sniff the ball
13. Touch w/ nose
14. Touch w/mouth/teeth
15. Put mouth over ball
16. Close mouth over ball
17. Pick up off of floor
18. Increase vertical distance off floor
19. Turn towards trainer w/ball in mouth
20. Move one step towards trainer
21. Increase number of steps towards trainer
22. Bring ball to trainer

III. Adding distance

23. Place ball four feet away. CT for pick-up
24. CT later & later on return
25. CT when dog reaches handler
26. Keep increasing the distance that the ball is placed away from the handler
27. Yo-yo distances
28. Place ball on a raised surface - like a chair seat, ottomon or overturned laundry basket.
29. Repeat steps 23 through 27 with the ball on a raised surface.

IV. Adding requirements

Retrieve over jump
Proof for send anticipation (stimulus control)
Proof for distractions

Solutions for common problems:

Playing with the object, including keep away: 2 treats/toys, trade, on leash work. Shorter, calmer retrieves. Place object rather than throw it.

Mouthing: size & type of retrieve object. Position of head.

Anticipation: NEVER CORRECT, just don’t reinforce. Reinforce the stay instead.

7. Dog sitting in a solid upright sit, paws moving off of the ground. Click for either paw anywhere off of the floor or collection of body towards the upright position (straightening of back, backwards lean).

OK, maybe not. As anticipated, the transition from extending the time of the sit to expecting some sort of back collection didn’t go very well. I think I need to assist in getting the next step. I don’t want to lure with food, but I could use either a hand target or a target stick. Which then I’ll have to fade.

Without a target, after a wonderful 8 second, straight sit and stare, Ellie broke from position, started to lick the carpet, then her butt. Lay down. Then started doing sloppy sits interspersed with more carpet licking.

Time to go get some advice from the experts! Then back with more planning.

Update: Answer from Sue. Use the hand target as a lure, but fade it ASAP. As soon as she gets into the “sit-pretty” position, click for position before her nose actually hits my hand. Then fade the hand target fast.

Here are the steps that I came up with for “Have you washed your paws”. This list is the second iteration of the steps. As I typed them up from my handwritten notes, I added at least two additional steps.

For each step, the trainer should identify the criteria and then write out exactly what they are going to click for. All with the goal of eliminating the “thinking” that we are tempted to do during the training session.

(1 is the finished behavior, then each prior step proceeds down the list. I would start my training with step 9)

Finished Behavior: 1. Dog sitting back on his hind end, paws up at his eye level, pads facing handler. Holds for three seconds.

2. Dog sitting back on his hind end, paws up at his eye level, pads facing handler. Hits the position, time in position doesn’t matter. Click for feet in position for any amount of time.

3. Dog sitting back on his hind end, paws moving off of the ground and up toward the dog’s eye level. Click for paws above chest level.

4. Dog sitting back on his hind end, paws moving off of the ground and up toward the dog’s eye level. Click for both paws anywhere off of the floor and dog holding upright position for up to three seconds.

4a. Dog sitting back on his hind end, paws moving off of the ground and up toward the dog’s eye level. Click for both paws anywhere off of the floor and dog holding upright position for up to two seconds.

4b. Dog sitting back on his hind end, paws moving off of the ground and up toward the dog’s eye level. Click for both paws anywhere off of the floor and dog holding upright position for up to one seconds.

5. Dog sitting in a solid upright sit, paws moving off of the ground. Click for both paws anywhere off of the floor.

6. Dog sitting in a solid upright sit, paws moving off of the ground. Click for either paw anywhere off of the floor.

7. Dog sitting in a solid upright sit, paws moving off of the ground. Click for either paw anywhere off of the floor or collection of body towards the upright position (straightening of back, backwards lean).

8. Dog sitting in a solid upright sit, eyes on handler, Click for upright sit - three seconds.

8a. Dog sitting in a solid upright sit, eyes on handler, Click for upright sit - two seconds.

9. Dog moves from any position into a solid upright sit. Click for upright sit - any amount of time.