The following article is a discussion from the ClickerSolutions mailing list. It begins with a description of how Doug redirected the prey drive in his Bouvier puppy. In the conversation that follows, Melissa and Doug explore the differences between redirection and desensitization and discuss when to use each.

My new Bouvier, Tasha has some pretty high prey drive. As most of you know from other posts, my daughter has a pet rat. Well, from the second we got her home, Tasha fixated (not in a good way) on Joxer (the rat). Obviously, working to gain total control over a dog's natural drives is the hardest thing to do without +P or -R, but I thought I'd use kind of the same approach I did with the bitework program I had set up, give the dog what it wanted (the rat) as soon as it stopped messing with what it wanted (the rat). Kind of a zen thing, don't you think?

Because they are drives, you can't extinct them, so the way I see it is you only truly have two choices regarding driven behavior, suppress it, or redirect it. (Let me know other opinions, I'm using you guys as a sounding board) I'm really against suppressing behaviors because I want more control than suppression can give me. So, I figured I would have to find a way to redirect the inappropriate behavior to an appropriate object, thereby making it an appropriate behavior. I would use the environmental cue of the rat moving as the "cue" or "command" to redirect to the appropriate item.

So anyway, I set up an exact duplicate of joxer's cage, with a fake cat mouse toy with a couple of strings sown onto it. Then, when Tasha started messing with Joxer (real rat), I would tell him to "stay" (that just means stop moving and look at me) and would just hook her onto a leash and walk her into the bathroom and just stand there for about a minute. After about three times of this, she went over to Joxer's cage and then immediately looked at me, sort of just to see if the -P was coming. As soon as she did, I called her over to the fake rats cage and started moving the stuffed mouse erratically in the cage. Then, when she saw that, she fixated on the fake rat cage. Every time she moved physically on the real cage, I walked her away for about 20 seconds. Every time she focused on me after looking at the real cage, I redirected her to the fake cage and moved the stuffed toy around and then tons of praise.

Well I did this for about two hours in small training sessions and it worked great, she stopped really focusing on Joxer, even when he was zipping around the couch, she would look at him, then just look at me and then walk over to the fake cage. So then I took the fake mouse out of the cage (too small she could choke on it and I wanted her to mouth it for this next phase) and replaced it with a tennis ball with a little string in it.

Next session, when she saw Joxer and went over to the fake cage, I zipped the tennis ball out of the cage and around the house while she destroyed my living room like a black tornado going after it (new rule from the Missus, no more prey work in the living room). We did this all day Sunday, so now when she sees the real rat move around, she goes and gets her tennis balls with strings and drops it either in a)person's lap or b)on or near rat (now that's CC at it's best huh). This morning she noticed Joxer darting on the couch, ran over to him and they touched noses and then she dropped into a play bow and BUZZED (that's what I call that "butt down running at full speed" thing dogs do) over to a tennis ball with a string.

It seems to have conditioned really fast, so for her at least, I think this worked really well to redirect inappropriate behavior to an appropriate item. Since I already have a release under mediocre stimulus control, when I want her to stop, I just give her the release command and redirect her to a less "active" activity, such as staying or "attention"!!

What other ways could I have used? Any other ideas or other avenues?

Doug Johnson
Good Dog Behavioral Training, L.L.C.
johnsond@usa.net
copyright 1999 Doug Johnson

Melissa's response:

Hi all, I've been doing some thinking about Doug's post about Tasha and trying to compare his method to desensitization. I'm kind of forming a theory (which may well be something everyone else says "No duh" about, but which I never really thought about before), and so I'm going to use you guys as a sounding board.

Please let me know if I'm on the right track, or if this is fundamentally flawed somewhere.

Situation: A stimulus triggers an undesireable response. In Doug's case, a moving rate triggered a chase response. Other examples are a dog who chases squirrels (or cats), a dog who is frightened by an approaching stranger, and a dog who reacts aggressively to approaching dogs.

In all of these situations, we are dealing with emotional or reflexive behaviors. Is that pertinent?

The current response is undesireable. Possible solutions: The response can be:

• Suppressed through punishment (which doesn't address the motive -- only the actual behavior, correct?).
• Redirected into another activity (which may or may not satisfy the original motivation).
• Altered(?) by desensitizing the response to the stimulus.

The "best" solution is different in different situations. For example, Doug *wants* Tasha to have a high prey drive. But he also wants her to know that it is unacceptable to exercise the prey drive on the rat. So he uses negative punishment to teach her "don't chase the rat," then redirection to give her an opportunity to USE that prey drive appropriately.

If a person has a dog who aggresses toward approaching dogs, a desensitization program would be better. In desensitization, the dog is never pushed to the point that he exhibits the inappropriate behavior. In Doug's method the dog learns that exhibiting prey drive in one instance is unacceptable but exhibiting it in another IS acceptable.

Here are the generic steps that Doug used. Doug, please correct me if I extrapolated the step incorrectly.

1. Present stimulus in controlled circumstance. At the moment dog exhibits inappropriate response, apply negative punishment. (One minute time out in boring location.)
2. Repeat step one until upon presentation of the stimulus, dog hesitates or looks away from stimulus.
3. When dog looks away, reinforce that behavior by redirecting it to an activity which stimulates the same drive. (Reinforce giving up what you want by giving you what you want.)
4. Repeat, repeat, repeat until the stimulus cues the dog to perform the alternative behavior.

Here are the steps of desensitization:

1. Establish "attention" in the environment minus the stimulus. (For example, work on attention in a park without squirrels before attempting to desensitize to the presence of squirrels.)
2. Present the stimulus in a controlled situation at a distance far enough that the inappropriate response is NOT triggered. Move closer to the stimulus until the dog begins exhibiting interest -- NOT close enough to trigger the response.
3. Establish attention at that distance and begin rapidly reinforcing with high- value treats until the dog relaxes and "ignores" the stimulus.
4. Walk away, then repeat steps one and two, trying to move a little closer to the stimulus with each repetition. Each time, move only to the point that the dog begins to exhibit interest but not to the point that the stimulus triggers the inappropriate response.

In each repetition, the dog is learning to relax and focus on the owner in the presence of the stimulus and that the stimulus equals "good things for doggy."

Am I perceiving these situations correctly? All feedback gratefully accepted.

Melissa Alexander
mca @ clickersolutions.com
copyright 1999 Melissa Alexander

Doug's response:

Melissa,

This is getting pretty deep, but here we go:

In all of these situations, we are dealing with emotional or reflexive behaviors. Is that pertinent?

I think it's definitely pertinent. It is with these types of behaviors that we have the most problems manipulating through pure +R. That's why you have to come up with all these kooky ways of redirecting using the dogs actual drives and instincts as reinforcements.

The current response is undesireable. Possible solutions: The response can be:

• Suppressed through punishment (which doesn't address the motive -- only the actual behavior, correct?).
• Redirected into another activity (which may or may not satisfy the original motivation).
• Altered(?) by desensitizing the response to the stimulus.

Yep, but desensitizing doesn't usually work if it's a truly instinctive or driven behavior. For example, put a border collie with good drives but no control training with sheep and it will literally herd them until they die of exhaustion. Keep allowing it and it will only stop to rest and then start back, It won't stop, or be "desensitized" to the sheep. You can use OC to train it to not follow it's instincts, but you have to have some form of control.

Now, I know there are levels of this, obviously the strength of the drive or instinct is extremely important.

The "best" solution is different in different situations. For example, Doug *wants* Tasha to have a high prey drive. But he also wants her to know that it is unacceptable to exercise the prey drive on the rat. So he uses negative punishment to teach her "don't chase the rat," then redirection to give her an opportunity to USE that prey drive appropriately.

Actually Melissa, I don't think the -P would have ever STOPPED her fixating on the rat cage, I only used it to get attention so I could redirect. The drive was so intense that a mere noise or something else to get her attention wasn't working. This was the least aversive method I could think of to get her to look at me right when the drives kicked in. I didn't use it to teach her not to go after the rat though, that comes when she realizes the "other rat" is more fun. I'm not actually training her to leave the real rat alone, just tricking her into doing what I want by using her instincts.

If a person has a dog who aggresses toward approaching dogs, a desensitization program would be better. In desensitization, the dog is never pushed to the point that he exhibits the inappropriate behavior. In Doug's method the dog learns that exhibiting prey drive in one instance is unacceptable but exhibiting it in another IS acceptable.

I think your statement about when to use desensitization is spot on. But again, what made Tasha chase the ball was not the fact that she learned that chasing the rat was unacceptable, it was that the "other rat/ball" was ALWAYS more fun. It engaged her drives/instincts MORE than the real rat. The -P was only used to get the ball rolling. Were I to get rid of the other rat/ball, but continue using just -P, she wouldn't stop. I think the value of the second prey item is very important in realizing that we are manipulating internal desires.

Here are the generic steps that Doug used. Doug, please correct me if I extrapolated the step incorrectly.

1. Present stimulus in controlled circumstance. At the moment dog exhibits inappropriate response, apply negative punishment. (One minute time out in boring location.)
2. Repeat step one until upon presentation of the stimulus, dog hesitates or looks away from stimulus.
3. When dog looks away, reinforce that behavior by redirecting it to an activity which stimulates the same drive. (Reinforce giving up what you want by giving you what you want.)
4. Repeat, repeat, repeat until the stimulus cues the dog to perform the alternative behavior.

Perfect except the key to it working is in #3, you have to make the activity you're redirecting too BETTER than the one you are redirecting from. That's why it works. The dog redirects because of the value of the second item stimulates the drives MORE than the first item.

Now, you CAN use redirection, but have the animal redirect mainly because of the -P aspect of the formula. Redirection through compulsion though is less and less effective as you move to dogs with more and more drives. That is EXACTLY why those methods didn't work with the Malinois. We had to find a way/reason to make them DESIRE to let go/disengage.

Here are the steps of desensitization:

1. Establish "attention" in the environment minus the stimulus. (For example, work on attention in a park without squirrels before attempting to desensitize to the presence of squirrels.)
2. Present the stimulus in a controlled situation at a distance far enough that the inappropriate response is NOT triggered. Move closer to the stimulus until the dog begins exhibiting interest -- NOT close enough to trigger the response.
3. Establish attention at that distance and begin rapidly reinforcing with high-value treats until the dog relaxes and "ignores" the stimulus.
4. Walk away, then repeat steps one and two, trying to move a little closer to the stimulus with each repetition. Each time, move only to the point that the dog begins to exhibit interest but not to the point that the stimulus triggers the inappropriate response.

In each repetition, the dog is learning to relax and focus on the owner in the presence of the stimulus and that the stimulus equals "good things for doggy."

I think you got it perfectly, as usual Melissa.

Doug Johnson
Good Dog Behavioral Training, L.L.C.
johnsond@usa.net
copyright 1999 Doug Johnson

Melissa's response:

Okay, okay... brain whirling... coming to an epiphany...

Yep, but desensitizing doesn't usually work if it's a truly instinctive or driven behavior.

But it DOES work if it's an emotional response.

For example, some dogs are dog aggressive because they're afraid of or feel threatened by other dogs. That's emotional. Other dogs are simply hard-wired to be aggressive. If that "drive" is triggered, that dog wants to attack other dogs.

A dog that is fear aggressive can be desensitized. We can change his emotional reaction to the other dog. Taken far enough, a fear aggressive dog may get over his fear aggression entirely and be completely safe around other dogs.

A dog that is hard-wired to be aggressive can't be "cured." He can be taught self-control in controlled circumstances -- a dog show, an obedience ring, walking on leash -- but we will NEVER change his genetic drive to attack other dogs. He will never be reliable off leash at a dog park.

Melissa Alexander
mca @ clickersolutions.com
copyright 1999 Melissa Alexander

| Training Articles Contents || Site Home |

List and Site Owner: Melissa Alexander, mca @ clickersolutions.com