Loopy Training

Why do we not want you to repetitively toss treats when your dog isn’t tracking? Here’s the long answer. PLEASE as questions below. This is, hands down, my favorite training topic.

Training loops.

It’s my favorite training topic.  Honestly, it is.  We, as dog (or human, or animal) trainers, accidentally spend a LOT of time reinforcing exactly what we don’t want.  We do this because we do not understand the loop.

A=antecedent.   B=behavior.  C=consequence. If you’ve trained anything, you’ve seen this.  And it’s generally taught as A—>B—>C.  What happens after C?  Does the animal stop learning?  Does learning EVER stop?  Is A independent from B? Is B independent from C?  Is C  independent from A????

NO!!!!  Let’s look at it another way. 

A (the antecedent) results in B.  And B (the behavior) results in C (the consequence).  Usually, A is the cue, B the ‘thing’ and C the reinforcer.

Your dog is tracking (B). A dog barks (C).  What’s your response?  Your response is the antecedent (cue) for further tracking.  You toss a treat….awesome.  Dog resumes tracking.  Dog stops and looks at you.  You toss a treat.  What behavior are you reinforcing???

While it’s perfectly fine to toss one treat to get the dog back to work, it’s not fine to toss treats every time the dog disconnects from the track…cuz here’s the thing.  The C is the antecedent for the A—that is the C predicts the A.

Confused?  I get it, this is a hard concept, and a VERY important one.  No one step in the process stands alone—the B acts as an antecedent for the C (the behavior is an antecedent for the consequence ((eating is a behavior))—, and the Consequence (eating) is an antecedent for the A—eating allows the dog to respond to a cue (in tracking, the presence of scent), which allows him to ‘behave’ (track) and earn a consequence (food)

.

Before you throw your hands up and consider shooting me…..let’s simplify it.

You ask your dog to ‘sit.’ (Antecedent).  He sits. (Behavior).   You give him a treat. (Consequence).  The opportunity to sit again—the space between C & A, is reinforced by the cue ‘sit.’  So the C (eating) is reinforced by the next cue…because that gives him the opportunity to earn more reinforcement.

OK, harder.  Everything that happens between A & B is reinforced by C.  Everything that happens between B & C is reinforced by A.  Everything that happens between C & A is reinforced by B. Each step promises reinforcement, or opportunity for reinforcement. Sitting on cue ‘promises’ reinforcement (sort of, but not going there now). Eating offers the opportunity to be asked to do another behavior…with it’s associated reinforcement. And so on.

Tracking. Your dog is tracking, and then he stops and stares.  You wait, and then you toss a cookie. You just reinforced stopping and staring. I’m NOT saying it’s not the right thing to do…I AM saying it has fallout.  So use it carefully.  Your dog turns and barks at you, and you toss a cookie.  Reality is, you just reinforced the barking. Again, might be the right thing right then and there…but it’s not a long term strategy.  Your dog gets to the article.  You ‘wait him out,’ he eventually downs, you throw food.  Did you reinforce the down? Sure.  Did you reinforce the time between his nose touching the article and the down? You sure did!  Does that mean we shouldn’t wait them out? No…it means wait them out with a plan.  COUNT to yourself.  1…2…3 “down.”  Dogs learn through anticipation—if they anticipate reinforcement, they will offer the behavior sooner and sooner.  What about that duration between when you cue ‘food is coming’ and actual food delivery? If he doesn’t anticipate food is coming, the target behavior will weaken. He ‘thinks’ the food was for everything he did between finding the article and your food delivery—ie, standing and waiting. Does your dog anticipate feeding time?  Does he start reminding you 10” early?  That’s learning through anticipation.  Does your obedience dog come before you call him?  Yup…anticipation.  He knows the loop and is short cutting the route to the consequence.  If you correct him….you’ve made it more likely he will remember the error—and if remembered, it’s more likely to recur.

Brain on fire. If you’ve trained carefully, your dog LOVES tracking. If your dog has a positive CER with tracking….tracking becomes a reinforcer all by itself (since tracking is an instinctive behavior, this happens more easily than say, a stay). This is why we are such nags about articles & foundation behaviors—if you train articles well, the reinforcement for indicating one is the opportunity to continue tracking. Does your brain hurt yet?😂

Hopefully Anne Morris will get her last track with Mila up for you all to see.  You will see Anne tossing food every time Mila is distracted by the dog across the street.  Then you will hear me say, “remember Loopy Training.”  (Anne has taken my Loopy Training class.)  And after that, Anne tosses the food before Mila disconnects; she keeps Mila moving forward. And after 20 yards or so, Mila realizes she CAN track with a barking dog, and she takes more tracking steps. And Anne anticipates her discomfort (reading your dog!) and keeps the cookies coming before Mila stops to look for the dog.

Understanding training loops is hard—I get it.  Understanding training loops is “the keys to the kingdom.”  If you understand each step in the loop is reinforced by the next one—regardless—….you can train your dog to do anything he can physically do.  The magic is in the loop!

Please ask me anytime about loops—don’t care if they are tracking or other dog stuff.  And, if you really want to blow your mind…start thinking about the loops within the loop!🤯. Can anyone tell me what is the smallest possible loop?