My dog was about 11 at the time and she’s generally good with other dogs in “real life,” but will bark and lunge at times on leash. What struck me about this approach was the idea that by having the decoy walk to a point and stop, the gift I’m giving my dog is the opportunity to “pause and consider” while she’s still able to think and make a choice about what to do.

It’s not an escalating situation where the other dog keeps coming and my dog is busy gearing up for what she’s going to do when the dog gets to her.

Even though we didn’t do that great of a job trying it all out, I still found a great improvement that has lasted. I don’t really have to manage with click/treat very much at all. Often, she’ll see another dog and choose to turn away and sniff a bush and not want/need anything from me to make it OK for her. She’s mostly very happy and relaxed.

I agree with your point that a dog that is allowed to remain over-threshold is harder to work with and often more sensitized to the decoy. Plus, the dog is kind of wrung out and it’s harder to keep going in the session. I like the idea of the owner/trainer being partner in “conversation” with the dog: “That’s too hard for you? OK, let’s try it again a different way and see if it’s better.”

Good point about the Old Bat thing…

Can you tell me how it would be best said, in guy-speak? How about “BAT is the use of environmental rewards in a low-stress environment.”

Regarding it being a “micronized version of Premack” – it’s not really just the Premack Principle. That principle states that a high probability behavior will reinforce a low probability behavior, and that’s only part of the picture. When you Permission to do X as the reward for doing Y, and X is more probable than Y, you’re using Premack. Permission to chase squirrels (high probability) can reinforce heeling (low probability). That’s Premack.

When you use “walk away from the scary monster” as the reward for signs of friendliness, that’s probably Premack, too, although in order to be Premack, “walking away” would have to be more probable than “turning head away from scary monster” I guess if you look at it as “walking away when mom says let’s go,” then that *is* usually more probable than “turning head away from scary monster” unprompted. I still think it’s less likely to be a Premack thing than a negative reinforcement thing – relief of social pressure.

But if the dog’s own behavior is not part of the reward, it’s not Premack. “Scary monster leaves” is not a behavior the dog itself does, so the fact that it’s rewarding is definitely not because of the Premack Principle. It’s just regular old negative reinforcement.

The fact that there haven’t been any comments from the other gender team might have some bearing on the way methodology descriptions are being presented and ideas exchanged. No negative criticism intended; this may simply be akin to what others refer to as the Venus/Mars differences in communications. Where are all the other men who are still not yanking and cranking anyway?

My thought and perception on the BAT commentary – and CAT as well – is that it is being expressed in an over-complicated or micronized version of what was simply stated by Premac a long time ago and practiced by many behaviorists and trainers in the field for some time. I think this can be substantiated through some of the posts by others. The basic principals are there but I suppose one could make, “the devil is in the details argument” to some degree.

All the above aside, I agree that, no matter whether you want to call it CAT, BAT or RAT it is a valuable set of behavior modification/conditioning tools to share and promote. And I do applaude you for your efforts to break it down and sort out some key points to consider for a most successful implementation. … Keep up the good work!!

You might want to hold back on that “Bat Girl” thing unless you’re willing to be known as that “Old Bat” in a few more years.. haha Jim

We worked at his friend’s home, outdoors. It was a one hour drive. When
Dusty got out of the car and saw the decoy, he was very fearful and tucked
his tail and wanted to jump back in the car, which surprised me b/c the
decoy wasn’t even near us and Dusty is usually not bad with people at that
particular distance. So I walked him around the street a bit until his tail
untucked and he seemed interested in his surroundings.

When Dusty seemed better, we started to work. We did several things and
mixed them all up. The decoy walked towards us; we walked towards the
decoy; we did parallel walking; we did pass-bys. Sometimes, my friend (who
Dusty doesn’t bite) walked with the decoy on the parallels and pass-bys.
Most of the time, the decoy walked alone.

We did each option 5-7 times before moving to another option. Each trial
showed improved demeanor on Dusty’s part.

We rewarded with several things. Dusty and I moved away either walking or
running. The decoy moved away while Dusty and I were stationary. Sometimes
Dusty got a treat or several treats while he watched the decoy moved away
(retreating people are a trigger for him – if I can change that with CC or
whatever, I will; and I don’t care if I’m mixing protocols if it works for
my dog). I often praised Dusty for making good decisions (to me,
communicating with my dog is real world stuff, and I don’t care if I’m
mixing things as long as I get results). Distance was always part of the
reinforcement and it was usually the sole reinforcer.

Behaviors that were rewarded included every visible degree of relaxation
(unfurrowing his forehead, relaxing the jaw, soft eyes), relaxed blinking,
eye contact with me, watching the decoy move without stiffening, air
scenting in the decoy’s direction, passing the decoy without any sign of
lunging or trying to pull me toward the decoy or past the decoy (in other
words, relaxed passing by).

We took a break and then did a few more trials with Dusty and me standing
next to my friend and his dog near the front porch as the decoy came out of
his house, walked past us, and around a corner.

During the trials, Dusty started his lunge routine a few times and we
repeated the same style of trials at slightly more distance. For each set
of trials, there was never more than one slight lunge attempt and I always
said, “Nope. Try again.” and then he seemed to know that we needed a
different behavior. The decoy never stopped, so the lunge attempt was
actually rewarded with distance, but that didn’t stop Dusty from abandoning
that behavior for a better one on the next trial.

At one point we stopped fairly near the decoy and we three humans were
discussing the trials, and Dusty sat calmly next to me and suddenly started
to play LAT with the decoy. It was pretty funny, and I let him play and
treated him for each round of LAT.

What I loved about this entire session was that Dusty was totally engaged in
the work and was happy to try to figure out the right responses to get to
run away, to get the decoy moving, to get a treat, etc. This whole
experience was so obviously a very good one for him. It’s what I had hoped
for when I first tried CAT.

Another amazing thing I noticed is this. After every CAT session we did,
even though there were very few times he had outbursts, Dusty was super
reactive the rest of the day and the following day. I attribute this to
being tethered; and even though he had walking around breaks during the
sessions, the tension mounted from staying still with me practically
ignoring him. Today, we stopped at a park on the way home. There were a
bunch of kids and their parents on the playground. We walked the dogs
around the perimeter and Dusty was absolutely relaxed around the people
including those walking past us (I kept at least a 15 foot distance for
safety).

Suddenly two teenagers ran straight for us from only about 30 feet away, I
braced myself for Dusty to lunge and bark and was about to do a fast about
turn and run away. But Dusty looked at them, stood still, and calmly looked
at me as they were VERY quickly approaching. His look said, “Can we run
away now, Mom?” and we did just that. I was shocked that he figured this
out and remained calm with just the little work we had done an hour earlier.
Yesterday, people running towards us at that distance would have sent him
over the edge instantly.

Needless to say, I am absolutely thrilled with everything that transpired
this morning. I have a happy dog who apparently had some fun figuring out
ways to work the environment to his advantage and comfort. What more could
I ask for? Can’t wait to do this again. Thanks for the great ideas,
Grisha.

Jude
NJ

I like that you:
1. Give frequent breaks, both when you decide to have them and when the dog needs one.
2. Are creative in your approach to keep the dog sub-threshold. Have to think about the vibrating collar idea. It’s definitely a quick way to signal an immediate retreat. I also am a big fan of rewarding the dog for the first signal, rather than continuing to a line that I’ve drawn in the sand. To teach persistence, I do deliberately ignore the first response and keep going until we get the second (or hit the Point B line), but I like giving the dog the clear feedback that this works before adding that in.
3. Do ’synchronised swimming’ in the midst of it. They do need to move!
4. Do pre-training to get the dog in learning mode and build up a set of core behaviors. I need to do more of that.

Walkie talkies or cell phones with headsets are a must for this, at least for the dogs than need big distances to get started. It definitely takes more distance than CC/DS, for example, because there isn’t a constant stream of food to distract the dog. I like something with a headset – the walkie talkie scared my dog when we first got started. It adds in unusual stimulus conditions.

I started thinking about employing CAT earlier this year, but like you had some serious reservations with the levels of stress for the subject dog, (and the actors when dealing with dog/dog.)

I do clicker training so typically for aggression cases I will use a combination of classical/operant conditioning, D&CC. The client dog always learns sub-threshold, so I didn’t see why CAT should be any different.

I think the modifications I have made are very similar to yours, but just to summarize:

I give the client a pre-treatment plan; this is what I call laying the foundations, so for 3 months prior to treatment the subject dog will learn:

sitting and giving handler eye contact makes good things happen
impulse control through games
a small repertoire of cued behaviours, hand targeting, sit, bow, away (retreat), and head turn left/right

During this 3 month period, I will also accompany the client on a regular dog walk so that I can be aware of the dogs pre-stress signals, or rather low level stress signals.

To be honest I do this foundation work for many cases not just aggression, but I could see how applying these during a treatment would be beneficial in keeping the client dog in thinking mode, and when a dog is in thinking mode, it’s not reacting.

I’m fairly new to C.A.T. having only done 6 so far, but all have been successful. I’ll briefly outline modifications:

I use walkie talkies with headsets to communicate to the owner handling the subject dog, and the actor handler. This allows me to give them space.

I will have a close up of the owner and subject dog simply by using wide focus binoculars

A colleague will do same with actor and handler

The actor handler will also wear a vibrating collar, yes the human not the dog, (and I’m told they quite like it!) This is a measure I take to ensure an immediate stop and retreat as we progress in proximity. The slightest change in how the subject dog is holding his/her muzzle, a change in the pupils, a shift of the front legs, averting eyes, a blink, slight turn of head etc. I need the actor handler to stop immediately, I then signal to retreat as soon as an alternative acceptable behaviour is offered.

So I use this to communicate a stop, therefore prevent any outburst, and to retreat. I want the reinforcement to be immediate. I don’t rely on the actor handler being able to notice the subtle changes in body language of the subject. So I give all the cues, the buck stops with me.

So these are the safety measures I take to make sure the subject dog is learning sub-threshold, and is therefore able to offer an alternative behaviour.

Throughout the process if I think the subject dog is asking for a break, I will ask the owner via the walkie talkie to first take his/her dog through a few of the pre-trained cued behaviours. I do this because the dog finds these fun to do, it gives some positive interaction with the owner, and it engages the dogs mind. Both dogs will then retreat and have a break.

I don’t always wait for the subject dog to ask for a break, if I think a break is due, we take a break, but again I ask the owner to go through a few cued behaviours.

When I start to see a pronounced change toward crossover (but not quite there yet), I will actually introduce a few short breaks. Again the subject dog will go through some cued behaviours with owner, but the actor will be doing same. So it’s a bit like synchronised swimming, but on dry land, at a distance

What I find interesting at this point, is that by adding a few short breaks, the subject dog starts showing an eagerness to get back to it.

So really, I’m not doing CAT at all, the only similarities are, operant conditioning, long sessions, numerous trials, reinforcer is the same, behaviour changes, emotion changes.

So if I may, I will call it BAT

I do teach my client’s dogs retreat as a first step. And I do use food. One of the “games” I play is to toss treats for the dog to find – initially without triggers. Once the dog wants to play, I toss a treat away from the trigger and the dog retreats controlled. With many dogs, treats can tossed toward the trigger soon, and approaching becomes part of the game.

Like you all, the comfort of the dog is priority and building on successes makes most sense to me – and often leads to real quick progress

I realized that this wasn’t a “real” CAT training session, but a demo that was being filmed. As such, I didn’t want to behave like a “real” client, but rather just intended to let Jesus do whatever he wanted to do, as long as I felt Sagan was safe. I was surprised and disappointed when he failed to have any contact with us before we started – I got no introduction, no instructions – and therefore he got no information about my dog.

During the session, he was far across the room from us. I could barely make out what he was saying. I kept receiving mixed messages – interact with my dog, no don’t interact with her. During the two hour session, I was not asked once, not a single time, for my interpretation of my dog’s behavior or my opinion as to how the session was progressing. I felt ignored at best, and blamed for any failures (“you’re sitting too close to the dog”) at worst. It was an incredibly frustrating experience.

Not that there wasn’t ample time for adjustments. The session was very, very, very long. Jesus could have come and talked to me at any time, or during one of the several, lengthy breaks. He never did. As a person who had lived with my dog for nearly 9 years, my thoughts and feelings and feedback was apparently irrelevant.

After the session, Jesus came over and said he thought it went well and that I should notice an immediate improvement in my dog’s behavior. Again, he didn’t ask me for my feedback. I left the session, feeling exhausted and disappointed and frankly, a little used.

And during the long ride home back to Seattle, it was apparent that the demo session actually had some damage. My dog was barking at everything that moved outside our car. She was hyper-aroused, unusually so. I now regret putting her through what was a very stressful and unpleasant experience…for both of us.

Would I do it again? No. I think Jesus was much more interested in talking to the audience and making a video than actually doing anything beneficial for my dog. I applaud him for continuing to work on alternative training methods – but if you’re not going to do it right – don’t do live demos. First, do no harm.

I love that you’re praising the dog’s choices – and still leaving them time to notice they’ve gotten the distance reward.

Does Bat Girl have some great phrase to say, like ‘may the force be with you,’ only different?

Thanks for your comment! Can you put a link to your blog post here? i haven’t read it and would love to do so.

Our experience has been using the “go” signal to the actor dog/person as a “click” that predicted for the dog the leaving of the actor. We also encouraged the handler to congratulate the dog for their awesome choice following the dogs’ awareness that their behavior did result in the space they were asking for. It felt right – and your putting it in writing so clearly makes it obvioius why. GO BAT GIRL!!