Non-Reward Markers are forms of correction, such things as telling your dog NO, AHH, OFF, LEAVE IT, applying leash pressure -- things you do that let your dog know to cease what they are doing. They shape behavior without food or toy rewards. Some consider them negative, even harmful. Susan Garrett posted a discussion on the use of NRM's and asked for commentary. She says she never uses them. But if that is so, I think that is only because she has so much confidence and certainty in her training methods, they don't seem born of frustration.
I read a bunch of the guilt-ridden comments from uncertain trainers, and when I couldn't stand it any more, posted the following:
"This whole discussion raises the hair on my neck because it reminds me of the total guilt trip society has done on parents ever correcting their children for fear of warping their character, demotivating them, depriving them of their God-given rights, etc. So now we have a world full of smart-mouthed, spoiled, un-ruly children who can barely function on the job yet don't respect the adults who support them, and they think they rule the universe.
A little "punishment" in the form of leash pressure or an AH or a NO is no way shape or form dog cruelty, and if you agree with Michael Ellis's work, puppies should learn right from the get-go not to fear a few well placed, well deserved NRM's. Puppies come into this world pliant and adaptable, they don't scar that easily, they recover from being corrected. With that assurance, your well-placed, well-meaning, properly delivered (without hostility), confidently delivered NRM's should be an integral part of their character-building training.
So I urge you one and all, don't abdicate your authority. Don't be afraid to correct your dog. You are the one who knows what they ought to do, in some cases absolutely need to do. You absolutely HAVE to be in charge. And I agree with Lori and Loretta, don't leave your dog confused about how to please you. Ask for what you want, reward lavishly, of course play the shaping game (where the dog guesses what you want and when they accidently deliver it, you treat, until they are doing it deliberately over and over) as far as it will take you (different for different dogs), but as needed give clues, including NRM's. And always, as any self-respecting teacher or scientist does, observe the results of your experiments, and if you're not getting what you want, try something else. Don't be afraid of your dog.
My observation has been if you deliver NRM's with ANY shade of guilt, your dog will quickly pick up on that and start playing the worst head games with you--wilting at your slightest criticism, looking sad or guilty, hanging their head down, running away, tail between their legs, whining, looking afraid to participate, etc. And that is them dishing out their NRM's on you, shaping your behavior!
In conclusion: if you want to diminish the need for NRM's, be confident in your training. You'll get a much more confident, happier dog."
This comment seemed to flush a different set of commentators out of the woodwork, trainers who aren't afraid to give commands. I found this most satisfying, and chalked one up for the gipper!
P.S. The next day, Susan came back to clarify she meant not using NRM's during a shaping game where D is guessing at the behavior you're looking for, trying all sorts of things to see what will earn them a reward. At that point, you don't punish wrong choices because you aren't giving them any clues what to do. You want them to feel confident exploring and continuously build up a positive training relationship with you. That's quite a bit different from never, never, ever using NRM's!