If you (all of you) really want to question learning theory, can you READ the preceeding posts properly.
- positive reinforcement (adding something the dog likes)
- negative reinforcment (taking away something the dog doesn't like)
- positive punishment (adding something the dog doesn't like)
- negative punishment (taking away something the dog likes)
Punishment is something you do, by either adding or taking away, that reduces the frequency of the behaviour.
So yes, if your dog pulls, you stop each time he does so, and the behaviour of pulling reduces in frequency then that IS punishment. In this example it is negative punishment, you have taken away something the dog liked (forward movement).
If your dog pulls and each time he does so, you give him a jerk to the neck that he dislikes sufficiently to try to avoid (from which we can conclude, it hurts), and as a result the frequency of pulling reduces then that IS punishment. In this case, its positive punishment, you have added something the dog wants to avoid.
Using aversive punishment, positive punishment such as jerks to the neck, harnesses that tighten in sensitive areas, shock collars, prong collars, slip leads... suppresses behaviour.
We have not addressed the cause of the behaviour, we have just punished it out of existance.
Negative punishment on its own, tends not to work well - if you just stop each time your dog pulls, the chances are you will actually teach your dog to dash back and forth to the end of the lead and back to you, rather than teach him to walk beside you.
So the better way to teach this is that forward motion occurs when the dog is beside you (positive reinforcement), AND that forward motion is ceased if the dog goes ahead (before he pulls, to avoid that back and forth issue).
Positive reinforcement 'balances' with negative punishment.
Positive punishment, 'balances' with negative reinforcement.
However, if you try to combine, as 'balanced' trainers do, positive reinforcement, with positive punishment, the positive punishment outweighs the reinforcer, and any rewards or praise words are not rewarding in themselves but are actually markers for 'you avoided punishment' which in itself IS a positive reinforcer because relief is hugely rewarding... and thats not good, anyone working for the relief of NOT being punished is really not a happy or mentally stable individual.
Just beacuse the dog in this example now appears to behave well, does not mean that this has been achieved in a safe, appropriate way.
The end, does not always justify the means, particularly when there is a better, safer alternative way of getting to that end.