Esteban Fernandezlopez says:

Replays for all Q and A sessions will be put into the Academy in a section Sarah will add today. Yes, you are very close, and your issue has been caused by your creation of a “gray area” in the dog’s criteria. We need to turn this “gray” into “black and white” so the dog understands exactly what he needs to do in order to get the food, every single time. You CAN NOT change criteria from time to time (where you sometimes ask for a retrieve/tug before feeding and other times you give the food without retrieve/tug).

In your situation, tugging is a behavior, so think of it as an obstacle. Your dog finishes an 8 obstacle sequence, but you can’t reward until the 9th obstacle is done, which is tugging. If your dog refuses to bring the toy and tug with you, it’s like he refused the 9th obstacle or popped out of the middle of the weave pole, or missed his contact. Would you reward that sequence? Hopefully not.

Handlers often feel pressure to reward “effort” or in your case, a difficult obstacle like the weaves, but this is not the right way to approach this issue. First, the tugging chain (tugging as an additional behavior) must be taught on the side with a single easily performed obstacle, like the tunnel (and then a jump). You should NOT start with a full sequence or an obstacle like the weaves. On the side, you should tug before (as you already are), cue the tunnel, throw the toy and then get the retrieve and tugging, and when there’s a strong moment of tugging or weight shift, mark it with a “yes!” and whip out the food. Note that we are saving our “yes!” for this moment, so at the end of the tunnel do not say “yes!” but say “get it!” or whatever your cue is to take the tug and bring it to you.

Even before we add the tunnel, your dog should be able to retrieve and tug with you for food, as shown in the video (the retrieve is not shown, just the tugging is shown). The Dinner Bowl Protocol can help you here as well. The key is you may NEVER reward with food unless your dog tugs with you, and you can “selectively reward” the best efforts.

Now you can go back and add the tunnel. If the dog refuses to pick up the toy or drops it at your feet (remember, he shouldn’t because you have already taught him to bring it to you), you refuse to treat him because he has NOT MET CRITERIA. This is your dinner bowl moment, and when the light comes on, he will pick up the toy and tug with you and then you can treat. Then in the next few reps, you will RAISE the criteria so he has to pick up the toy and tug with you immediately, or you will send him back through the whole process, which at this time is the tunnel plus tugging. All the other times you encouraged him to go find his toy before you will reward him has taught him this is acceptable –> do a sequence, run to mom for food and if she doesn’t give it, then go bring the toy to her. We need to change that to –> do a sequence, grab toy, and tug with mom because that’s the ONLY way I ever get food.

So you have to work very hard on your retrieve in isolation, and once you have a good one (even if you require two toys), you can add in a single obstacle and polish the routine before slowly adding more obstacles. Never reward for non-tugging, no matter how good the agility sequence is–with the tugging behavior, the sequence is incomplete.