Today Chelse, from Dog's Best Friend, came and did a CAT session with Gustav at the park. I brought Dottie over to my mom and dad's house so she wouldn't feel left out.
Chelse started from really far away and just walked up to a certain point and stopped. Then when Gustav offered something we liked, like sniffing or head turning, she walked away. She worked waaay under threshold, even more than I expected, but I can see how that would be a good idea. He definitely noticed her but never reacted more than closing his mouth. No hackles, no growling, and no super high tail. We talked about what to reinforce and shape for. Chelse said the best thing to shape for is signs of actual friendliness: soft eyes, a relaxed approach, friendly air-sniffing, maybe even a little tail wag. None of those were displayed too much, but it's good to know what to look for.
My verdict: Gustav definitely never got reinforced for being anything other than totally relaxed. The big question is: did he begin to associate his relaxed state and/or specific behaviors with Chelse walking away? That's the goal. I couldn't say for sure whether he was making that connection or not, it's hard to tell.
Chelse said the biggest draw-back to CAT is the impracticality. It has to be a pretty set-up situation. I also pointed out that, to the untrained eye, it basically looks like nothing. Expensive nothing.
I was really happy to see a professional do it, because she's got a great sense of body language and was very careful and precise. I video-taped Gustav during the procedure. It basically looks like he's standing with Justin in a field, since I couldn't get both him and Chelse in the shot.
Next up I'd like to try it out on my own. I'm hoping I can get a friend who Gustav doesn't know to volunteer. For me, I will believe in it when I see Gustav make a new friend and change his emotions towards a particular person. Now I'm trying to think of someone who wouldn't mind hanging out in a field with me for half an hour or an hour a few times, who Gustav doesn't already like (that cuts out Dad, my brother, and everyone in the band. Basically all the people I would normally ask to do something like this.)
I'd also love to try it out with Dottie and a dog, but I need a good decoy dog. Chelse said her dog is a great decoy dog and would be willing to do it. I'll think about whether I can afford it or not. If it's hard for me to think of someone to do it with Gustav just as a person, it's double hard to think of a person AND a bomb-proof dog who would be willing to do it.
So, it wasn't a dramatic success (yet?), but it wasn't a letdown either.
Subscribe to:
Post Comments (Atom)
break out the Tequilla! or Keg or Margaritas!
ReplyDeleteTHAT'S EXACTLY where you start...those non events are when you celebrate!!!!
Even using eachother(YOU/JUSTIN) as decoy's or Dottie for that matter. You don't even need a stranger or strange dog...as the whole point of the exercise is for Gustav to be in a relaxed state and for you to click, recognize and praise him for those subtle cues so he is able to realize that yup, there's a person/dog/cat/squirrel, and wow I'm impressed that you Gustav are one cool calm guy. The progress comes quickly after he has that lightbulb moment.
For fosters that are extremely reactive, I will literally take them from home to carport and in car then drive somewhere remote to practice these exercises (with a known subject) so that they catch on faster. It's all for them to be in a relaxed state long enough to understand that they've been able to hold things together and you noticed their rock star performance.
I've read that it takes 30-60 days for something learned to go from short term to long term memory. For Wags after watching him lunge at people in the distance, for 3 wks I did not walk him in neighborhood, I took him to remote spots (if you notice our videos) so his body could remain in relaxed state and I used my dogs as decoys and Darrell and a friend. That I think is one of the big reasons that he made so much progress in the 4wks of classes. I made things easy on a daily basis for his body and mind to learn a new behaviour.
I'd love to see the video! It's always so interesting!
way to go!!!!!
So Sonny and I are moving back to Chicago this summer and I'll be closer to Madison. I know this is ridiculous, but I'd be happy to help once I'm back :) We can find times for me to come up and I'll help! The more I can learn the better. :)
ReplyDeleteKate: Thanks for your words of encouragement! I was really pleased to see Gustav's relaxed state. I remember just a few months ago I'd never even seen him wag his tail at anyone outside. When I came back from talking to Chelse, he wagged it furiously.
ReplyDeleteI have a question about using friends as decoys: I understand that it works well in some circumstances, for example when I go in and out of the door while the dogs sit-stay in the kitchen. But with CAT, we're doing the straight, original version with no treats or interaction from the owner, all just negative reinforcement. In this protocol, it seems like using someone Gustav likes wouldn't work, since seeing them walk away wouldn't be a reinforcement. It might even be possible to shape undesirable behavior that way, like barking to bring people closer.
I definitely still do tons of counterconditioning and look-at-that while out on walks, since a straight CAT procedure would be impossible all the time. So Gustav gets a click and treat for looking at people (I say "who's that?") and not being aggressive. In that instance, though, he can even be a bit stiff, as long as he's not totally lunging and barking, and still get reinforced with a treat.
I'll post a video and the results from our first DAP-sprayed bandanna walk a little later, once the videos are done uploading.
Erin: Hooray! Welcome back to the midwest! I've been reading Finley's blog, sound like you'll have a lot to teach me! Congratulations on working so hard with him. And when you come up to Madison, we should definitely have some dog-training parties. Maybe someday Gustav and Dottie and Finley could all be friends.
My food for thought...in a lab setting if you were not there, I agree whole heartedly that negative reinforcement would be in place.
ReplyDeleteBut there are two variables in this scenario where positive reinforcement does come into play, in my opinion anyways.
1 Chelese is not actually a stranger to Gustav.
2 you/Justin are on the other end of the leash, or in the same area.
I think with any given scientific theory is used in real life...there will always be an overlap.
In life there is balance to every interaction and situation. Positive and negative. It's never just bad, there's always some good in there.
Positive reinforcement I believe takes place even if it's with you being quiet and NOT reacting to Gustav. So Gustave receives two very important tidbits of information from the exercise
1- being relaxed makes things that are uncomfortable go away
2- being relaxed provides biofeedback from you by not reacting to his behaviour and he gets to be with you.
Also at the core of all the inappropriate reactions is the fact that these dogs/ or people succumb to their internal chemistry over riding what would be a non-event for someone else.
Performing the same exercise with a known person, you are allowing the dog to work on their emotional self control.
Using the same technique allows their motor skills to pick up the pattern faster.
So for that reason CAT is always done with strangers and known people / dogs.
Sorry I meant to say that I use CAT with both strangers and known.
ReplyDeleteEven a stranger if the leave and then come back, they're no longer a stranger. They're not a novelty any longer. Come to think of it if you want to split hairs... we all walk, look, act, smell like humans....
that's just my two cents worth, what do I know.