Thoughts and observations from someone who has been repeatedly introduced as "Nicole Silvers, that dog whisperer lady I was telling you about" I don't whisper to dogs; I eavesdrop on their conversations with each other.

Search This Blog

Friday, July 24, 2009

Elicitation-- Step 2 -- Getting the Dog to Do It

This step of teaching is, by far, the most controversial and misunderstood.

One effective approach, especially for beginners, is to actually start here. The human must learn how to draw out the desired behaviors. Once you are certain you can reliably get your dog to do a behavior, THEN go back and put it on cue by presenting your cue (hand signal, verbal command, context clue, or other), get the dog to do the behavior, release, and reward.

For non-beginners, this approach is unnecessary, because they can already reliably get the dog to do certain behaviors. Generally, people have ideas about what works for them personally. As long as what works for you personally is working for your dog, I'm generally ok with it.

There are a few basic approaches to eliciting the behavior. THEY ALL WORK.

Waiting for the behavior
PRO:Absolutely no added stress to the dog.
CON: Time-consuming, can add frustration to the owner.

Luring (leading with a bit of food)
PRO: Does not require physical contact with dogs uncomfortable with contact
CON: Lure can become part of cue if not faded quickly

Physical guiding
PRO: Dog is getting right answer more quickly. Can teach dog to accept handling as communicative.
CON: Bite risk for dogs uncomfortable with contact. Touch can become part of cue if not faded quickly.

Gimmicks/Band-Aid Solutions/Equipment
Baby gates, headcollars, visual barriers, prong collars, anti-bark, crates, muzzles, squirt bottles, etc.
PRO: Immediate results.
CON: Dog learns nothing unless gimmick is part of long-term training program. No long-term results if gimmick is removed.

Feedback (Tertiary Reinforcers)
Using a continuous stream of markers: "good. good. good. no. good. good. no. good. good. good." followed by release "OK" & motivational reward.
"Good" = one treat. "No" = no treat and/or acceptable punisher. OK = 5+ treats fed one at a time.
PRO: Once communication system learned, all future learning is significantly more rapid. Exquisitely clear communication.
CON: Demands intense concentration, ultra-precise timing, rapid decision-making from owner. Human brains tend to opt for option that involves least thinking. That would not be this option.

Punishment (Traditional training)
I don't like to do things that make me feel icky. Punishment that consists of more than a brief discomfort makes me feel icky. Punishment that is unfair (the dog doesn't know what caused it) makes me feel icky. "Punishment" that consists of owner venting emotions onto the dog makes me feel icky. What makes me feel most icky is application of ineffective punishment. I can't honestly say that all punishment is ineffective. I also can't say that it is free of side effects. But, for me, immediate effectiveness defines humane punishment. ANY punishment, no matter how mild, that requires dozens of repetitions is a red flag. More on punishment in another post.
Traditional training uses pain as both cue and feedback. Pain/lack of pain is used as the primary communicator. While this works, it is a) unnecessary, and b) risks very serious side effects.
PRO: Works.
CON: Slower response times. Relationship built on avoidance of pain. High potential for mis-communication. Extremely high risk of "breaking" the dog -- pushing frustration level beyond dog's ability to tolerate = very serious intentional "you will stop that now" bite.

No comments:

Post a Comment