When dogs perform behavior after we cue it verbally and/or through gestures, we are providing an antecedent to that behavior. When dogs bark because of an external trigger (like a dog or person walking by, or a sound), then the antecedent is the trigger itself.
In neither case can the dog associate an antecedent given after the behavior, with the behavior already performed — the classic example is when the dog sits and their handler says, “good sit.” The dog may perceive the “good” as praise, thus it is a positive-reinforcement consequence of the behavior; but the “sit” part is just an antecedent again, for something they already did — which at best does nothing, and at worst, confuses the subject. This is why using an antecedent after a behavior, while attaching praise (positive reinforcement), or correction (positive punishment) is not effective. Another example is “no bark,” after Dog barks. He may perceive the “No!” as punishment, reducing the barking (I don’t recommend this, due to punishment fallout), but if he does know what “bark!” means, then he is literally being cued to bark immediately after being punished for doing so. Assuming the subject recognizes knows what “bark” means, then saying “no bark” creates this sequence:
Antecedent: sound or visual event triggers autonomic responses, which trigger barking
Consequence: verbal punishment, “No!”
Antecedent: verbal cue, “Bark!”
To illustrate the issue further, imagine using a hand signal for sit, and no verbal cue. Will your dog sit? Likely — most dogs recognize visual antecedents better than verbal ones. So, cue the dog with only a gesture. After the dog sits, say, Good! and then repeat the hand gesture. Would repeating the hand gesture after the dog sits help the dog’s understanding? The gesture as an antecedent fills the same position as the word would — repeating it, or the word, won’t (can’t) help.
I hope that kinda makes sense. I know people recommend saying “good sit” and “no bark” but these approaches are not effective. When they do seem to work, it is in spite of the repeated antecedent, not because of them.
If Dog performs a behavior we want to reinforce, well-timed general praise (good job!! Yay!! or whatever) and/or other reinforcers (food! toy, play, etc), are quite effective and do not risk confusing the subject.