Returning to the Tug after Treats

Warning: This post could also be called "Gitchi the Guinea Pig" since I'm kinda making it up as I go... but I just had to do something to make an attitude adjustment once I realized the following correlation:

Gitchi's tug drive was inversely proportional to the value of the treats she got. 

It became pretty obvious when I started giving her Crate Game yummies to snarf down before she ever even came out her kennel. When I was putting way more energy into our games of "motivational" tug than she was, I concluded it would take less energy to actually <gasp> train her to tug.

First we went back to "playing" tug but with the mindset that it's not a game; it's a command.

The goal was to get tug on cue, so I began to always use that cue to instigate tugging when I knew that she was about to tug. No food involved yet - the game was the reward. I stylized the cue both verbal and physical: I hold the tug between two hands where she can grip in the middle and simultaneously give the verbal "GET IT!"  The caveat is that once I give that cue, she IS gonna tug, dammit! I started with her favorite toy at the time (it's Pinky Dread in the video below, too) and if she didn't tug I'd "beat" her with it until she got irritated enough to engage with it/me. (I'll wait while you look up the # for the SPCA.)

Then I started to sneak in food. After she was engaged and tugging good, I would click (or say "yes") and offer a low-level treat, like a kibble. Once the release was reliable on the click, I started pairing it with a release word, "Break!". I had to learn to wait for that one treat to be chewed and swallowed and all residual molecules in the mouth to be fully processed - to prevent distraction as well as asphyxiation from inhaled cookie particles. Then I would present the toy in the previously-conditioned-without-food stylized manner with an excited cue to "Get it!", maybe also running, dragging, twitching, beating the dog, etc -whatever it took to get her to engage. When she engaged in tugging again, she'd get a click and JACKPOT! (Cesar, dinner, whatever. I just had to make sure I didn't show it to her or let her smell it until *after* she engaged in tug that 2nd time.) And that would be the END of that session. In other words:

["Get It!" - tug, tug - "Break"/click - low level treat] - "Get It!" - tug, tug, - "Break"/click - JACKPOT (end of session)

When we'd done lots of reps of this, we started repeating from the [ to the ] above before ending with the Jackpot. I try to ping-pong how long we tug before the Break/treat, sometimes just a few seconds, sometimes a tug-a-thon. But you can see in the video below, I tend to be too consistent. (That's a not oft uttered  phrase!!)

The idea is to build a record of success returning to the toy after food and highly rewarding that return.

So now I have just begun to throw in a couple of high level treats DURING a session, demanding that she return to tugging afterwards. You'll see in the video below that I misguidedly emphasized the "demand" after a high value treat, forgetting entirely the "whatever it takes to get her to engage" that had worked so well in previous sessions. Predictably she disengaged, sniffed, ignored me, etc. But she's a tractable pup who actually does love to play tug, so she quickly forgave me and rewarded my next more appropriate effort with an enthusiastic response. (Wait, who's training who here?)

So here we are, practicing tugging after treating:

 http://www.youtube.com/watch?v=iYafuQjry90

Suse 'n Gitchi