New BAT theory

Valuable training articles posted by Victoria and other Positively members.

Moderators: emmabeth, BoardHost

Post Reply
runlikethewind
Posts: 1166
Joined: Fri Apr 02, 2010 12:48 pm

Re: New BAT theory

Post by runlikethewind »

So what you are saying is when you are treating dog-dog reactivity, the behaviour modification games can be dangerous because you have spent ages rewarding look there's a doggie (LAT, look at that), only to have it ruined by a dog barging up or being aggressive itself to your dog, thus losing all trust in you and the game.

Dog - person reactivity - you think they're good?

Editing this: this is in response to Nettle's post: and to add to this, I still think look at the doggie (LAT) is a good programme - because you still have to vigilant to block unruly dogs like with just protecting them for ever more.. but by desensitising them over time to dogs being close as well, you give your dog a better ability to handle the unruly dog, if it does arise.
Last edited by runlikethewind on Tue Feb 15, 2011 1:14 pm, edited 3 times in total.
User avatar
GundogGuy
Posts: 230
Joined: Wed Jan 19, 2011 5:01 am
Location: Scotland
Contact:

Re: New BAT theory

Post by GundogGuy »

have to go to puppy class.. i'll come back to this :D
"Oh what gold there is to find when one is blessed with an open mind" - me, not five minutes ago :-)
User avatar
Mattie
Posts: 5872
Joined: Tue Jan 09, 2007 5:21 am

Re: New BAT theory

Post by Mattie »

No I am not saying that, please don't take this as an attack on you but I think you are trying to overthink what dogs are doing because you are thinking as a human and not a dog. Dogs don't try to think why something is happening, why has something worked etc. We have all done this, tried too hard to see the reason for why things happen when to dogs, they don't think this. A dog will do what works but won't try to think why it has worked.

When a dog came towards Gracie she didn't think that stupid owner is bringing the dog to me, all she sees is the dog walking towards here, yes it did set her back a bit but not right back. By keeping things as low key as I could for the next few days she was able reduce her stress levels and by taking her back a few steps, she was able to make progress again.

When dealing with dogs we have to try and think the same way as they do or we won't be able to make the improvements that we do make.
[url=http://img.photobucket.com/albums/v312/Nethertumbleweed/PIXIE.jpg][img]http://img.photobucket.com/albums/v312/Nethertumbleweed/th_PIXIE.jpg[/img][/url]
runlikethewind
Posts: 1166
Joined: Fri Apr 02, 2010 12:48 pm

Re: New BAT theory

Post by runlikethewind »

Thanks Gundog. I do follow the theory and I understand that the dog wants distance as a reward under normal reactive circumstances... but I had trouble understanding why the reward should be distance in this context when all the conventional counter conditioning is about reward for looking at the trigger or reward for sticking around if the dog stays within threshold. I do like the idea of running away though - it also keeps the dog excited and focused on you. :D

Just a thought - some have said BAT is difficult if distance is NOT what a dog wants as a reward ie wants a fight (very unusual) or to play. But I think in general, they accept that dogs in the main do want distance.....

So do you think BAT is more powerful than LAT - look at that?
runlikethewind
Posts: 1166
Joined: Fri Apr 02, 2010 12:48 pm

Re: New BAT theory

Post by runlikethewind »

Hi Mattie = my 2nd to last comment was in response to Nettles.
User avatar
Horace's Mum
Posts: 1129
Joined: Wed Feb 06, 2008 5:10 pm

Re: New BAT theory

Post by Horace's Mum »

I have lost track of who said what, but I have two thoughts on this conversation.

I use "watch" a lot, and have done from the very start. It has helped me tremendously to work through many of our issues, as well as being something to fal back on when we get stuck in a situation and have to wait it out - like a strange dog coming towards us that is going to have to pass close by, there are not other options. I spent a very long time making sure that the watch command was very strong, to the point where I could maintain eye contact for 3 minutes or more without breaking, no matter what was going on around us. I know that for us it may be slightly different, because once I have eye contact then Horus is relatively oblivious of what is going on, but surely that should make it harder for him to trust me, when he knows something is behind me but I will keep him safe if he does a watch for me?

It works in two ways I think. The first is along the lines of Mattie, but not quite the same. I can use it if he is getting fractious waiting for something, like talking to a passing neighbour. It is an alternative behaviour to reacting/barking/staring. But it also allowed us to enter into situations that would have been impossible without it - the Watch is such a strong command of trust for us that by asking him to watch he visibly relaxes and stops worrying - it creates an instant "timeout". So I could take him places that I knew would stress him to a degree, allow him to look around, see what is there, try to work through it by rewarding a good look at the object and back at me to make sure i have seen it. But when it gets too much, I asked for a watch for a minute or so, which allows him to regain control of his emotions (can't think how else to put it). When we have done that, I release him at a moment when there is nothing too scary directly behind him, and we can start again, rewarding the looks. This may well be something that we have developed that is fairly unique to us, given that I have a dog who relies so heavily on eye contact and facial communication.

I am not sure that walking away is necessarily the best option, I think it depends a lot on the dog. I know as well that if Horus sees something then looks up at me, he is asking whether I have seen it, and does he need to worry about it or have I got it under control? A quick smile, thumbs up, and "on you go" is enough to reassure him that I am aware of it but it is nothing to worry about. If it is something a little scarier then I have the option of smiling, asking him to present in front of me in a sit, and then either "watch" the scary thing past, or just leave him in a stay - the first would be if it is a threatening thing like another dog, the second would be something like a horse coming past, he is not scared but the horse might be. If the thing is immobile and we have to pass it, then I have to read whether he is worried enough to react despite my reassurance, in which case I "work" him past with heelwork, or whether he is just checking I have seen something, in which case once acknowledging it we are able to walk past on a loose lead as usual. It is a constant conversation.
So, if the handler can keep the dog outside of its escalation zone then the reward is given for anything other than escalating. The reward given is a 'life reward' (like allowing a Spaniel to hunt) which is of course, putting distance between the dog and the object it will be reactive towards.
The dog is rewarded for doing something outside of the zone which will normally be something that can only be accomplished if the dog is relaxed, i.e. look away, sniff the ground, a wee shake or whatever... So, if the dog is relaxed, its not tense, and if it's not tense, it's not being reactive and it's not giving off reactive signs (or early signs of reactiveness). So, as the dog gets rewarded for doing something relaxing, the distance can slowly be lessened so that the dog can get closer to the object without showing the signals that make it or the other dog reactive (because very often reactiveness is a two way street. This is why some dogs react to each other and some don't)...

I've heard of a gundog trainer using a similar 'tool' with a dog that went crazy every time it saw water. This dog loved to splash and swim so much it became quite unmanageable when it saw water. The trainer took in excess of 30 minutes to get this dog to the waters edge in calm frame of mind by continually approaching the water and as the dog started to escalate, turn around and reward the dog for being calm... Similar technique for a different stimulus...

Basically, (I think) the dog 'learns' that it's more rewarding to be relaxed around the stimulus than to be reactive, coupled with a heightened sense of trust in the handlers ability to cope... Not rushing in there and immediately asking the dog to comply to 'watch me' or even to expect the dog to progress with the clicker is basically getting the dog into a rewardable/trainable frame of mind and so setting the scene for success.
I am not sure that sniffing, turning heads away or shaking IS always a sign of being relaxed. Sniffing and looking away are classic calming signals, which would indicate stress. I know several dogs (including mine) who shake when they are anxious about doing something they are being asked. So this would need to be considered when using BAT.

I think your example of the water is a different scenario altogether. Rather than the water being a bad thing, so walking away is a reward, the walking away is being used in the same way many teach loose lead walking. The dog can only reach its goal (the water) when it manages to maintain calmness on the journey towards it. It is teaching self control by removing the reward, not by giving a reward.

As for the last comment, again I think that Watch can indeed be used as an aid to learn relaxation - there are many many dogs who would simply be unable to instigate a look/sniff/shake of their own accord, and the watch can help indicate to them that there is an alternative behaviour. This may well apply more to the herding breeds such as collies and there crosses, due to their propensity for staring things down, but I know if I had tried Bat with Horus we would have been there forever waiting for him to quite staring and decide on another behaviour for himself, that I could then reward by walking away. And the flip side is that the more he stares, the harder it then is to persuade him to do anything else at all, other than physically drag him out of sight (and even then he will keep checking behind). This may be exaggerated by deafness, but I believe it to be more his collieness than his deafness.

I hope I have said what I mean to say, and not something totally different!!!
User avatar
Nettle
Posts: 10753
Joined: Sun Apr 13, 2008 1:40 pm

Re: New BAT theory

Post by Nettle »

Just a thought - some have said BAT is difficult if distance is NOT what a dog wants as a reward ie wants a fight (very unusual) or to play. But I think in general, they accept that dogs in the main do want distance.....




The reward is in feeling safe. The distance is the means to feel safe initially. Then the reward is being able to rely upon the owner to keep the dog safe from the feared object. Which leads to being able to be closer to the scary thing without reacting any more than "Hey lookit that scary thing. We're going to be okay here aren't we? You'll take care of it won't you?" Eventually the dog will see scary thing and simply trust the owner to deal with it.


If a dog wants a fight - or to "play" then it isn't scared. So we are on a different subject. Which I can develop if you'd like me to.
A dog is never bad or naughty - it is simply being a dog

SET YOURSELF UP FOR SUCCESS
jacksdad
Posts: 4887
Joined: Thu Dec 10, 2009 10:48 pm

Re: New BAT theory

Post by jacksdad »

nettles wrote: You cannot train away fear. You can only manage it.
The trainer/behaviorist that helped me a bit last spring explained that fearful/reactive dogs are like Alcoholics, once a fearful/reactive dog always a fearful/reactive dog. I have never brought it up because I could never think of a good way to explain that without leaving someone with the feeling of no hope. But if you understand that that doesn't mean there isn't hope, rather you must always give some thought about what your going to do with your dog, what situations you will be asking it to deal with, then it actually helps you NOT put your dog into a situation that it can't deal with thus leading to a fearful/reactive response.

The same trainer/behaviorist runs what she calls K9 cinema for her former students. A small number of us get together an watch a DVD on training or behavior or some other continuing educational topic for working with our dogs, then we can chat with her, get advice etc. 3 weeks ago she showed the BAT DVD.

What I took away from it, and there was a lot to process so I really need to watch it again, is that at a high level BAT asks the dog to make the choice. With Watch or Look we teach a skill, then ask for it, then reward when it's accomplished. Watch (in addition to how mattie describes how she uses it) is go for teaching your dog to look away and to prevent them from just staring and getting worked up. this is for dogs like Jack who would just see a dog and flip out. Look. is good for when a dog avoids the scary thing. teaches it is ok to look. it's also a nice progression from Watch once you have built that trust nettles talks about.

If I am fallowing the BAT explanation the difference is....

With WATCH and LOOK we train/teach/ask for these. With BAT, we let the dog figure it out. The significant difference is we are not teaching the dog to look away by use of a lure (ie lure to sit), rather we are reinforcing the choice to NOT react fearfully to scary, and making that choice highly rewarding thus increasing the odds of it being repeated.

again, a very high level and simplistic explanation of my understanding of BAT using a dog/dog fear problem. You allow your fearful dog to observe the source of their fear, in this case a dog from a safe difference. you then wait for the dog to offer the calm response and some sign they are "done" watching. initially you might need to click and treat, but the goal is to get quickly to a point that your dog is self rewarded by turning away or walking away after having checkout the object of their fear and realizing there is no need to go "reactive". Grisha calls it "gathering information".

in the beginning, you may need to do some click/treat work and build up from just the dogs eyes turning away to look off to the side. then progress to asking for a little bit of a head turn, then a full head turn, then a full body u turn and walk away.

But for all this to work, you HAVE to know your dog, what it looks like calm, working up to a reactive episode, just checking things out but on alert etc. You have to KNOW their threshold for not reacting and work right at that, you go over threshold BAT isn't going to work any better than any other method. At my current understanding of BAT I am concerned that the subtleties of reading your dog, knowing it's threshold etc. might be too much for someone without that knowledge and experience with their dog or professional help. And thus might not make BAT a good first step for people who can't work with a professional or who not yet learned to read their dog's emotional states.

I am probably not doing it all justice, like I said I have only seen Grisha's DVD once.

Since watching that DVD I have been trying out the very basic idea...let Jack look at his trigger from a safe distance and reward him for choosing to turn away. after only a few sessions of this I have gotten more "auto watches" and "turn and come to me" than before trying BAT from him when we run into people. Two examples. two weeks ago we were walking on a fire road and came around a corner and ran into a lady and her dog, Jack was ok with the dog which was good because there was no where to "escape" to and the dog insisted on exchanging sniffs. ok, no problem. then the lady (just tying to be nice) went to bend down and tried to touch Jack. BIG No No with Jack. Soon as I saw her try this I asked her not to. Jack noticing it rather than going into a barking fit, just keep looking from her back to me, back to her, back to me...in seeing this, I immediately lead him for distance and gave him a treat reward. The second example was last night. Jack is always a bit extra leery of strange men, but is uncomfortable with men (other than me and my son) in general. last night we ran into one my neighbors and his dog. again, no issue with the dog, but Jack is always unsure about him. last night he went up to my neighbor, looked right up at him, turned and came back to me not a peep, not a single change in his body language completely relaxed...JACK POT TIME. I have also been using it with dogs hoping to help Jack "get comfortable" with the last few dogs in the neighborhood that he still struggles with. Just to be clear for those new to this topic, the goal in "getting comfortable" is not Jack being best buds or even exchanging a calm greeting but to just be ok with the sight of them, to past them calmly that kind of thing.

I don't really see BAT as a revolutionary discovery or an all new way of doing things. rather it's just another approach to applying the basics. but one that depends very heavily on timing, knowing your dogs trigger, threshold and ability to read their emotional state. which you could say applies to all methods...but it seems to me that BAT requires just a little extra than the "older methods".

Again, this is just my thoughts based on watching the DVD once and trying the absolute basics of the concept out over the last 3 weeks.
runlikethewind
Posts: 1166
Joined: Fri Apr 02, 2010 12:48 pm

Re: New BAT theory

Post by runlikethewind »

Horace's Mum wrote:I am not sure that sniffing, turning heads away or shaking IS always a sign of being relaxed. Sniffing and looking away are classic calming signals, which would indicate stress. I know several dogs (including mine) who shake when they are anxious about doing something they are being asked. So this would need to be considered when using BAT.

I agree with you on this one and it would be interesting to challenge BAT on that basis.
runlikethewind
Posts: 1166
Joined: Fri Apr 02, 2010 12:48 pm

Re: New BAT theory

Post by runlikethewind »

I like this 'gathering of information' idea. I have so far been following LAT (look at that) with improvements. My new rescue has taken to sniffing the legs of strangers as they walk past, using LOOK AT THE PERSON, click/treat and then we step back (not a proper walk away) - only until the unpredicable happens and they turn around, I try to make a quick retreat.

I think a distinction could be made between a LOOK AT THAT (as in the LAT programme) and Jackdad's version of LOOK which he describes to be used when a dog does not want to look. I'd never have considered there'd be an occasion for a dog NOT to want to look at something. I'd have thought they always want to look at their triggers.
runlikethewind
Posts: 1166
Joined: Fri Apr 02, 2010 12:48 pm

Re: New BAT theory

Post by runlikethewind »

To Jack's Dad so you think BAT is working well for Jack - better than the previous steps you've been trying?
jacksdad
Posts: 4887
Joined: Thu Dec 10, 2009 10:48 pm

Re: New BAT theory

Post by jacksdad »

runlikethewind wrote:So do you think BAT is more powerful than LAT - look at that?
Watch, Look, BAT all equally powerful tools when used at the right times. There is NO ONE "cure". I think that is also what your stumbling on. There are many tools and paths you can take to help your dog. There isn't a one size fit all, all powerful cure. There are tried and true starting places (example avoid trigger, protect dog, teach watch, look etc), but you do have to adapt the ideas/concepts to your dog and as your dog make progress, grows more confident, less reactive you have to adapt to their needs at that phase and not get stuck at "step 1".

In Patricia McConnell's latest DVD on treating reactive dogs she starts to demo Watch with a dog, but with this particular dog she switched to look very quickly. is look more powerful? no, it was just more appropriate for THAT dog. Other dogs she demoed with just the sight of the "stoodge" dog was too much for them, and watch was very helpful for them. it just all depends on the individual dog what tool is going to work best.
jacksdad
Posts: 4887
Joined: Thu Dec 10, 2009 10:48 pm

Re: New BAT theory

Post by jacksdad »

runlikethewind wrote:To Jack's Dad so you think BAT is working well for Jack - better than the previous steps you've been trying?
working well at my level of understanding of BAT???....ya, I think it is showing promise.

Better than previous steps... that isn't so cut and dry. I don't think Jack nor I would have been good candidates for BAT a year ago. He would flip out over dogs 100+ yards away, I had only had him for a couple months so there was minimal to no trust yet that I would deal with the other dogs for him. I had no idea what a calm Jack looked like and was still struggling to read his emotional state based on his body language.

I think for us, WATCH and avoidance were spot on for that point in the journey. about mid year, we progressed to Look as the primary tool. BAT might be just what the "doctor" order to address the last hurtles. however, watch continues to be a powerful tool for us in some situations, look in others. it just all depends on the situation.
User avatar
Nettle
Posts: 10753
Joined: Sun Apr 13, 2008 1:40 pm

Re: New BAT theory

Post by Nettle »

I think it is a tremendously important point you make Jacksdad when you remind us that there is more than one way to help the reactive fearful dog. :) None of the positive methods is more or less "right" than the others of that ilk: we have always to suit the situation and the response to the individual dog.
A dog is never bad or naughty - it is simply being a dog

SET YOURSELF UP FOR SUCCESS
runlikethewind
Posts: 1166
Joined: Fri Apr 02, 2010 12:48 pm

Re: New BAT theory

Post by runlikethewind »

Yes I agree - lots of methods work ' in the toolbox' - but if two conflict, then they shouldn't be perhaps used together. That's all I'm thinking.
Post Reply