Visual Signals before Acoustic?

“Puppy nerd” asked a great question in his or her comment: Given how visual dogs are, should one start an exercise with visual signals to help the dog get it right, and then switch to acoustic ones, or avoid visual signals altogether if you want your dog to pay attention to your voice?  Well, this could keep us all busy for the next few months. I know this is a loaded issue, with people strongly advocating one or the other (mostly the latter in my experience.)

There’s no ‘right’ answer, at least not in my opinion.  But then, I’m not a big advocate for there being one way to train. There are many roads, as they say, to the top of the mountain. I think what’s most important is to be aware of the advantages and disadvantages of starting with visual signals.  The first obvious advantage is that dogs learn them readily, and thus you can create ‘wins’ fast and start reinforcing dogs right away. That’s good for dogs, and it’s good for novice owners too, because people get reinforced when their dog does it right, and are more likely to keep training when it works relatively fast.

I also think it’s a good thing to help dogs understand what we want right off the bat; I don’t think it’s always “positive” to not give a dog any clues at all about what he’s supposed to do next.  The other reason I like visual signals early in training is that I love using them in daily life. I love being able to “call” my dog to come, lie down and stay while I’m on the phone, or motion one dog to do one thing and another to do something else.

The disadvantage of starting with visual signals is that if you want your dog to lie down to a verbal command, without relying on a verbal prompt,  you need to carefully and thoughtfully eliminate the motion during training.  This can be tricky, because non-professional dog trainers tend to be relatively unaware of the movement of their body, and end up often using a movement as a prompt. They think their dog is lying down to “lie down,” but he’s really watching to see if their head dips forward. Their dog never really learns the verbal signal, and is always waiting for the owner to give the salient signal. The salient signal to the owner is the phrase “lie down,” but the salient one to the dog is the  head nod. That leads to obvious confusion and frustration on both sides.

For whatever good it is, here’s what I do.  I’d be curious how others handle it; I look forward to your comments.

To train sit, for example. I use the tried and true ‘lure/reward’ method advocated by Ian Dunbar, but I combine it with a lot of operant principles.  I’ll lure the dog into a sit with a treat in my hand, but not give the verbal cue at all. Over the next few sessions I”ll modify the movement of my hand into less of a ‘lure over the  head toward the tail’ to a upward sweep of the hand.  Once I can predict the dog will respond to the motion with a sit 80% or so of the time (you know we all really make those numbers up, don’t you? I’m just estimating.) I’ll add in the verbal cue, being careful to say it BEFORE I move my hand.

After a few sessions of lots of saying “Sit” right before I make the visual signal, I’ll say “Sit” and not move. Now the dog is being asked to sit just to a sound, not a movement. I’ll wait 2 seconds or so. If the dogs sits within that time frame I’ll jackpot big time, with lots and lots of treats, and then try again.  If the dog doesn’t sit and just stares at me like I”m an idiot, I’ll turn away, wait a few more seconds and then try again. If I get no response just to the word for 3 tries, I’ll add back the visual and end on a ‘win.’

So, what do you do?