In psychology, aversives are unpleasant stimuli that induce changes in behavior through punishment; by applying an aversive immediately following a behavior, the likelihood of the behavior occurring in the future is reduced. That is mostly true in dog training, but a significant exception is forcing, during which a behavior is made
more likely to reoccur through the application of an aversive stimulus. Aversives can vary from being slightly unpleasant or irritating (such as a disliked color) to physically damaging (like a 2x4!). It is not the level of unpleasantness, but rather the effectiveness the unpleasant event has on changing behavior that defines the aversive. Aversive tools apply ‘unpleasant stimuli’.
The above description includes the word ‘punishment’. But that isn't specifically accurate in all dog training applications. I’m not going to launch into an Operant Conditioning discussion (yet). But what we’re really talking about here are aversive tools. Ear pinch, heeling sticks, e-collars, et al. They may punish, correct, or merely compel. But they do so by being used as implements that apply unpleasant stimuli (pressure/force).
I thought this might be a worthy discussion, what with hunting season coming up and all! I don't know about you, but I'm sharpening my dog up for hunting!
Evan
It's a very worthy discussion Evan, especially when it isn't accurately described in the very first post.
"Forcing" is far from the exception to the rule in terms of negative re-enforcement strategies used in retriever training. We use both punishment and negative re-enforcement every single time we train. Every single time. Even if we're using steak!
Here's what I mean, starting with aversives. When a dog is standing and we command sit, followed by pinch collar pressure, we are punishing the previous behavior (standing) and re-enforcing the new behavior (sitting). If we command here, followed with the e-collar, we are punishing whatever else he was doing (sniffing dog poo) and re-enforcing here. On heel with a pinch collar or a stick we are punishing him being out of position and re-enforcing him being in position.
This is actually true in positive re-enforcement training. If a dog is standing and we command sit, we withhold the reward, creating frustration (a form of punishment), until he sits, relieving the frustration (negative re-enforcement) and earning the reward (positive re-enforcement).
I could go through all of the other examples as well but it isn't necessary. Hopefully people get the idea, supporting my assertion that we are ALWAYS discouraging a behavior and encouraging a behavior. Regardless of the re-enforcement strategy, for behavior to change we have to discourage the previous behavior and encourage the new (desired) one. Logic says so.
The only question then becomes, how strong the aversive treatment happens to be. Even standing there withholding a treat the dog desperately wants is very frustrating and stressful for the dog. It is, in fact punishment. We're teasing him and no one likes to be teased. I have asked a +r trainer, "Did you like it when your brother ran around with your favorite toy saying "nanny nanny poo poo" when you wanted to play with it?" No, you didn't and you probably chased him, then ran to Mommy to try and get the toy. That may or may not have worked but eventually, you found a way to get that toy back, didn't you? Maybe you ignored him and he got bored with your lack of reaction. In the future then, you immediately ignored him and got your toy back more quickly. He's trained you to ignore him when he picks up a toy of yours now, hasn't he? He give a signal (picking up the toy) and you respond appropriately (ignoring him). This example is no different than a dog running through various behaviors trying to get a reward. You're teasing them, they're frustrated and trying to figure our how to relieve that frustration. This is an aversive experience for the dog, whether these trainers want to admit it or not.
I have seen "positive only" trainers stand on a leash many times to keep a dog from moving away. They will tell you this isn't punishment, to which I have replied more than once, "OK then put the collar on and I'll tie you to a post". Same concept as the above, only a bit more aversive to the dog.
We go from there all the way to using a heeling stick! It's all a matter of semantics but there is ALWAYS an aversive of some form used in training, period, end of story. The only question is what level of aversive treatment you're willing to inflict on the dog to get what you want.
Bottom line, training is training. In order to "change" behavior one has to be discouraged and the other encouraged. You ALWAYS use an aversive to effect that change and some form of reward to re-enforce desirable behavior. It's just a matter of degrees.
Defining aversives in terms of what we believe to be socially acceptable causes a lot of misconceptions among the general public and I believe is a terrible idea.
If we want to talk about learning theory then let's talk about it in full.