ClickerSolutions Training Articles

"Reinforcing" Fear

I'm basing this observation on experience. Many times I've seen dramatic changes in the dogs immediately after the owners stop "poor baby"-ing and start "cheerleading." I've been able to get dogs who are fearful of strangers and don't know me calmed down by using a gentle but cheerful tone of voice. However I will agree that every dog is different, and there are some dogs who may respond better to the soothing tone.

I too see these differences, but I explain it that the two approaches are fundamentally the same and the dog just responds to one better than the other, rather than saying that one is "telling the dog fear is ok" and the other isn't.

See, both are "rewards." Poor Babying is, in some people's minds, a reward for the fear, and to be avoided at risk of increasing fear. But if you look at it carefully, by that rationale, so is jollying. That is a reward, too, but no one tells you not to do it or you might increase fear. We know it is a good intervention. But they work by the same classical principle, and do not work operantly.

So, that means that we *do* understand that you need to work classically here and that operant doesn't really work there. If you jolly, you're using classical. If you soothe with "poor baby" you're using classical, too. So why choose one over the other? Why do we see one work more than the other commonly?

Two things:

  • Some people "poor baby" with tons of fear and nervousness in their own affect. It isn't the "poor baby" or the soothing that is at fault, but the addition of the owner's arousal and fear that the dog is picking up on.
  • Also, soothing just isn't that meaningful for lots of dogs. They often don't give a hoot about the calming attempts at all. Often this is because they're too close to what they're afraid of, but frequently it is because of plain old individual differences in what works for the dog. Changing the picture for the dog and getting their mind off the problem by jollying just may work better for an individual than providing a calm, supportive presence.

So, when it doesn't work, it doesn't work because the soothing means *nothing* to the dog, or because the owner is doing it really nervously, not because it just plain "increases fear," or "tells the dog that fear is ok."

The point about soothing a scared dog is this: find what soothes your scared dog. If it is jollying, go for it. If it is a quiet petting session, do it. If it is cooing and babytalk, fine. My own dog can handle many more things if in physical contact with me and doesn't particularly like the jollying as it makes her more hyper. That is why we don't train by recipe, but we seek to understand the underlying principles.

Classical works, and you have to find the channel that speaks to your dog, be it jollying or cooing or something else. But don't think that one is operant and the other classical, or that they're both operant and "telling" your dog what behavior is "right." It will just mix you up in the end.

Part II

And if the emotion elicited by your response is congruent to what the dog is emoting, then won't that increase what the dog is emoting? For example, if the dog is happy and you give it food, the happy feelings the dog emotes will increase.

Right. Since we were specifically talking about fear and I didn't want to make my email a tome, I didn't make the distinction.

Classical works by imparting the associations already made with the UCS (say, food) onto the situation at hand. If you have a happy situation and you add some already-made happy, you have, well, happy.

If you have a neutral situation and you add happy, you have happy, eventually.

If you have a bad situation and you add happy, you'll have happy, but only with a lot of work, and attention paid to threshold and such, since a bad situation makes for a challenging learning environment and a slow transfer of happy.

If you have a happy situation and you add bad (say, a knee to the chest), you'll get all sorts of things depending on the dog (arousal, appeasement, avoidance, aggression), but none of them should be an increase of happy, technically.

For example, if the dog is fearful and you give it food, the fear behavior decreases the because food elicits 'happy' emotions and these counter act the fearful emotions.

There is also another mechanism at play here. If you add a happy UCS generally, you get happy based on what I said above. But if you specifically add food as your happy UCS, you get the assistance of the parasympathetic nervous system counteracting the sympathetic, and helping you change the physical state. Which in turn, likely, helps the dog calm down. It is not a benefit you get with using tug or chase as your happy UCS.

Amy Cook
copyright 2007 Amy Cook


| Training Articles Contents || Site Home |

Copyright of all posts is the property of the original author. Please obtain permission from the original author before copying, quoting, or forwarding.

List and Site Owner: Melissa Alexander, mca @