Bat - Behavioural Adjustment Training

[Lollipup]

I just bought a book called BAT - Behavioural Adjustment Training, which is written by a trainer who I think is in the US, and it is her term for her particular way of using counter conditioning for reactive dogs.

I find the method sounds interesting and she claims to have had a lot of success with it, but some of it does contradict what I have been taught and so I am curious for others opinions.

Using BAT, the 'environmental rewards' are used instead of treats. The dog is marked with a verbal marker or clicker when it shows calming signals or looks confidently at the stimulus and then is moved further away from the stimulus as a reward(as long as this is what the dog wants and is therefore rewarding). Calming signals could including looking away and I was always taught that if you reward the dog for looking away, even for focus on the handler, you are just teaching the dog avoidance. The writer of BAT claims that rewarding the dog for calming signals teaches the dog that it is control of the situation, therefore instilling confidence and alleviating the anxiety. She has videos on youtube and her website about it, I don't have them on me but a simple google search should find them. She runs a training business called Ahimsa Dog Training.

So, what do you think? Have you used this method for reactive dogs, would you use it and why/why not? Do you think it would teach the dog avoidance or confidence as claimed? She seems to have achieved good results with her own dog. She claims it can be used no matter the reason the dog is reactive.

Edited August 14, 2012 by Lollipup

[megan_]

I think BAT is far more advanced in the US and you don't find many trainers who know a lot about it in Australia (even fewer than those that know about LAT).

I find the concept interesting and could see how it could work. One issue with traditional classical conditioning is that the dog can learn "if I stay calm the threat stays". Often the threat outweighs the reward, even if they do take it. I think I accidentally taught my girl this lesson.

I incorporate some of the principals into my training now, but they aren't formally BAT. For example, I do LAT, but I make sure not to "push my luck" and I try to keep stressful sessions very short and sweet (you behave, the threat goes away).

I use LAT because I had the resources to learn it, but if I had my time again and had access to a trainer who was experienced I'd most probably try BAT as a first port-of-call, with LAT as a back-up. I imagine you'd need to do it in a fairly controlled environment though, which can be very hard. For example, if your dog was scared of men, you'd need to ensure that you could remove your dog away from men. This is very hard to do in a park/in public with people wandering around.

[BCNut]

There's a trainer that blogs on DogStarDaily that does something fairly similar I think, Drayton Michaels:

http://www.dogstardaily.com/blogger/992

If you go back through his stuff (theres a few pages) theres a few studies on reactivity in some pitbulls which yielded good results, these were filmed and put on youtube.

I tried a bit of counterconditioning, not for reactivity but for keeping focus on me in distracting environments and found it very effective :)

Not sure if this is exactly the same methodology but similar and interesting nonetheless :)

Whats the name of the author of your book?

[Lollipup]

I think BAT is far more advanced in the US and you don't find many trainers who know a lot about it in Australia (even fewer than those that know about LAT).

I find the concept interesting and could see how it could work. One issue with traditional classical conditioning is that the dog can learn "if I stay calm the threat stays". Often the threat outweighs the reward, even if they do take it. I think I accidentally taught my girl this lesson.

I incorporate some of the principals into my training now, but they aren't formally BAT. For example, I do LAT, but I make sure not to "push my luck" and I try to keep stressful sessions very short and sweet (you behave, the threat goes away).

I use LAT because I had the resources to learn it, but if I had my time again and had access to a trainer who was experienced I'd most probably try BAT as a first port-of-call, with LAT as a back-up. I imagine you'd need to do it in a fairly controlled environment though, which can be very hard. For example, if your dog was scared of men, you'd need to ensure that you could remove your dog away from men. This is very hard to do in a park/in public with people wandering around.

Can you give me an overview of how LAT works? I have heard of it but not very familiar with. The way I was taught was to makr and reward for looking calmly at the trigger without reacting, keep dog underthreshold and correct for any signs of reacting/prevent response. Maybe LAT is similar to the mark for looking at stimulus part?

I like that BAT doesn't really use treats because I have seen dogs that get protective of their food in the presence of the other dogs (when thats the trigger) so I think taking food out of the equasion would be useful in that situation.

There's a trainer that blogs on DogStarDaily that does something fairly similar I think, Drayton Michaels:

http://www.dogstardaily.com/blogger/992

If you go back through his stuff (theres a few pages) theres a few studies on reactivity in some pitbulls which yielded good results, these were filmed and put on youtube.

I tried a bit of counterconditioning, not for reactivity but for keeping focus on me in distracting environments and found it very effective :)

Not sure if this is exactly the same methodology but similar and interesting nonetheless :)

Whats the name of the author of your book?

Her name is Grisha Stewart :)

[fuzzy82]

I think BAT is far more advanced in the US and you don't find many trainers who know a lot about it in Australia (even fewer than those that know about LAT).

I find the concept interesting and could see how it could work. One issue with traditional classical conditioning is that the dog can learn "if I stay calm the threat stays". Often the threat outweighs the reward, even if they do take it. I think I accidentally taught my girl this lesson.

I incorporate some of the principals into my training now, but they aren't formally BAT. For example, I do LAT, but I make sure not to "push my luck" and I try to keep stressful sessions very short and sweet (you behave, the threat goes away).

I use LAT because I had the resources to learn it, but if I had my time again and had access to a trainer who was experienced I'd most probably try BAT as a first port-of-call, with LAT as a back-up. I imagine you'd need to do it in a fairly controlled environment though, which can be very hard. For example, if your dog was scared of men, you'd need to ensure that you could remove your dog away from men. This is very hard to do in a park/in public with people wandering around.

Can you give me an overview of how LAT works? I have heard of it but not very familiar with. The way I was taught was to makr and reward for looking calmly at the trigger without reacting, keep dog underthreshold and correct for any signs of reacting/prevent response. Maybe LAT is similar to the mark for looking at stimulus part?

I like that BAT doesn't really use treats because I have seen dogs that get protective of their food in the presence of the other dogs (when thats the trigger) so I think taking food out of the equasion would be useful in that situation.

I noticed during the block training for the NDTF course the referred to giving the dog a correctioon for reacting as "response prevention". But it's not. That's response interruption. Response prevention is when you keep the dog at a distance where he's not reacting.

And what you're doing sounds like LAT, you basically just keep the dog below threshold and click and reward for looking at the stimulus. The person who came up with LAT (can't remember her name now) actually said somewhere that she uses BAT as well, she clicks and reward for looking, and then moves away to increase distance between the dog and the stimulus.

Grisha Stewart is coming to Australia in October/November, and doing seminars on BAT in Sydney (I think) and Cairns. I'm going to the Cairns one, can send you the info if you like. There will be demo dogs.

[Lollipup]

I think BAT is far more advanced in the US and you don't find many trainers who know a lot about it in Australia (even fewer than those that know about LAT).

I find the concept interesting and could see how it could work. One issue with traditional classical conditioning is that the dog can learn "if I stay calm the threat stays". Often the threat outweighs the reward, even if they do take it. I think I accidentally taught my girl this lesson.

I incorporate some of the principals into my training now, but they aren't formally BAT. For example, I do LAT, but I make sure not to "push my luck" and I try to keep stressful sessions very short and sweet (you behave, the threat goes away).

I use LAT because I had the resources to learn it, but if I had my time again and had access to a trainer who was experienced I'd most probably try BAT as a first port-of-call, with LAT as a back-up. I imagine you'd need to do it in a fairly controlled environment though, which can be very hard. For example, if your dog was scared of men, you'd need to ensure that you could remove your dog away from men. This is very hard to do in a park/in public with people wandering around.

Can you give me an overview of how LAT works? I have heard of it but not very familiar with. The way I was taught was to makr and reward for looking calmly at the trigger without reacting, keep dog underthreshold and correct for any signs of reacting/prevent response. Maybe LAT is similar to the mark for looking at stimulus part?

I like that BAT doesn't really use treats because I have seen dogs that get protective of their food in the presence of the other dogs (when thats the trigger) so I think taking food out of the equasion would be useful in that situation.

I noticed during the block training for the NDTF course the referred to giving the dog a correctioon for reacting as "response prevention". But it's not. That's response interruption. Response prevention is when you keep the dog at a distance where he's not reacting.

And what you're doing sounds like LAT, you basically just keep the dog below threshold and click and reward for looking at the stimulus. The person who came up with LAT (can't remember her name now) actually said somewhere that she uses BAT as well, she clicks and reward for looking, and then moves away to increase distance between the dog and the stimulus.

Grisha Stewart is coming to Australia in October/November, and doing seminars on BAT in Sydney (I think) and Cairns. I'm going to the Cairns one, can send you the info if you like. There will be demo dogs.

Yes I would love that info, thank you.

I think in the NDTF course the wording they used was along the lines of prevent and/or correct desired responses but I get what you mean. If BAT is effective I would much rather do it that way. Especially for dogs who become worried about the food in the presence of other dogs.

[Aidan3]

I just wrote a long reply then lost it, so here is the pared down version.

Before BAT we used to have discussions on one of the old listserv forums about the use of functional reinforcers (and extinction procedures) in behavioural modification for aggressive and reactive dogs. One of the moderators of one of these forums was a lady by the name of Kellie Snider, who with her post-grad supervisor, came up with a protocol they called "CAT" which is an acronym for Constructional Aggression Treatment.

Their argument (supported by evidence from the study they undertook) was that because aggressive behaviour has a "distance-increasing" function, the use of other rewards (e.g food) motivates the new behaviour and doesn't necessarily change the old behaviour (aggression).

Grisha diverged from this idea and was happy to use other rewards as well. She called this BAT.

The key difference is that BAT will use other rewards (e.g food) and it fudges the extinction side of the equation a bit (BAT doesn't necessarily wait a barking dog out without changing the environment). BAT also lacks support from empirical data, but this doesn't necessarily mean that it doesn't work.

A key idea in the early discussions (and in CAT) is functional analysis. This is a tool that applied behaviour analysts use to determine what is reinforcing the unwanted behaviour. You look at Antecedent, Behaviour, and Consequence. Then play around with the Consequence, and maybe the Antecedent, measuring changes in Behaviour; and hopefully you figure out what triggers and motivates the behaviour and form a plan of action for modifying it.

I mention all this because I think it makes a big difference to how well you can apply these ideas when you understand the background. I think it's especially helpful to do the functional analysis bit, because then you understand why the dog is doing what they do and how it benefits them.

[Lollipup]

I've got the BAT book but haven't read it yet but as I understand it, it works on principles similar to those used for LAT, Click to Calm etc.. Justice and I have started with a new trainer who works using these principles to counter-condition dogs and teach new emotional responses to triggers. She was really happy when I told her I was about to start reading BAT and also recommended another book for me to read when she found out we have some of the same books, like Brenda Aloff's Aggression in Dogs. We've only had the initial assessment so far so I can't report on how effective it is for us yet (although I think it will be fantastic and is just what we've been looking for) but another DOLer who recommended this trainer has had wonderful success with these methods. Her DA dog has been taught to turn away from dogs he is unsure of, go behind her legs when he is afraid and has been building positive associations with dogs to the point where he has been able to play off leash with at least one of the trainer's dogs and if I remember correctly has been able to go on walks with both of them. If it's not raining tomorrow the plan is to go and sit outside a dog park with Justice (so he's safe and none of the dogs can get to him) and work on the LAT game and we're also working on conditioning "look", "let's go" (for getting away from something quickly but calmly) and "leave it" to the point that they're so ingrained he can still respond to them when he's extremely anxious or aroused.

ETA: Grisha Stewart has just released a new book that I've ordered but haven't received yet, although I'll go through BAT first.

Yes I am intested in her new book as well. Was waiting for the kindle version as my bookshelf is overflowing now, although I do prefer to have the real thing too! It looks good.

[corvus]

I have used it with my Erik. I use LAT more and have a lot of success with it, especially after teaching Erik to take a breath on cue (TAB). TAB is incredible. Nothing calms Erik faster short of a Thundershirt. Erik is a very active dog and he's a heeler by nature, so tends to rush in where other dogs freeze or avoid instead. BAT is HUGELY useful for figuring out what he really wants, distance or proximity. Also very useful in figuring out critical distances. Often he wants information, but needs to be given a chance to calm down and seek it in more appropriate ways. BAT or LAT can give him the rule structure for that, but where LAT is most useful for reassuring a dog that they won't be asked to interact with something, BAT is a way to allow a dog to interact with something in small increments so they don't become overwhelmed. So I would use LAT if I don't think Erik needs or wants to interact with the thing he's bothered by, and switch to BAT if I think he does need or want to interact with it. I'm more inclined to use LAT for triggers he's familiar with and BAT for novel things he's not sure of. There's never an expectation even with BAT that he will interact with the thing he's bothered by, but the nice thing about BAT is it gives you a lot of information to make good decisions about how close you will take them and what they are comfortable with. LAT gives you similar information, but it comes quicker and in smaller doses! Much easier to misinterpret or miss all together.

I learnt BAT from Grisha's first dvd. I am a huge fan. It has changed the way I train in a big way. I don't use BAT formally much, but I use functional rewards a lot. I used the concept to teach Erik coping skills with troublesome dogs at the dog park, and even to teach him to swim. :p Grisha is speaking at the APDT conference this year, and I believe is doing a seminar in QLD afterwards.

I think that BAT and Control Unleashed exercises are very compatible and I switch between them pretty seamlessly whenever I feel the situation is better suited to one over the other. If you're using BAT in a simple manner, I think it's pretty easy as long as you remember to err on the side of more distance and slower approaches while you watch your dog very carefully. Just tedious. Using it to teach coping skills at the dog park was very effective, but took some skill. But it just goes to show the broad uses the idea of functional rewards can be put to. Well worth learning IMO.

[Lollipup]

Thanks Corvus. It's interesting what you say about the thundershirt. I have tried it on 3 dogs - my 2 and a foster. I have noticed no difference whatsoever so I just keep saying they are in the other 20% since the thundershirt site says it works on 80% of dogs! Or maybe there are just subtle differences I haven't picked up on, but seemed to make no difference to me.

Do you think BAT is particularly useful for herders since they find movement so rewarding? Would you choose BAT or LAT or other method in particular when focusing on reactivity to other dogs? You say you would use BAT if you thought the dog needed to be able to interact with the stimulus, do other dogs fall into that category? :)

[Kavik]

How does the thundershirt work?

[Guest Panzer Attack!]

I'm using a lot of the CU exercises with Scooter at the moment, as he's been squashed by a few different off-lead dogs (twice now at class!!) and we went from having the most horrible class in his 18 months (obedience two weeks ago - he completely shut down and I spent the entire time holding back tears as it was our assessment - we didn't pass obviously!) to having the B E S T session he has EVER had (last week at agility - there are 4 dogs in our class, 3 are DA, and he held his own just as well as the non-DA dog did!). I have the DVDs and both the CU adult and puppy books but at the moment I'm getting the most use out of the Puppy Program. It's a heck of a lot more easier to sit down and absorb than the first one (much more smoothly written and in short, easy to understand chapters so you can just flick through and read what you like).

As corvus said, Grisha Stewart is touring Aus later this year - [link]http://www.dogtrainingseminarcairns.com/[/link] (why does QLD get all the good ones?!)

If this kind of thing interests you have a look at Dr Karen Overall's Protocol for Relaxation (vet behaviourist and Leslie's mentor) and the CU homepage. Leslie works with a lot of stressy and aggressive performance dogs, and uses simple (awesome!) methods which can be translated anywhere if the owner is willing to do the work. I can't rate her highly enough :)

[Lollipup]

Thank you!

I do find it very interesting but it's hard when different things I research conflict with each other too. I will look at those links :)

[Lollipup]

I won't be able to make the seminar unfortunately, I will only just be getting back from overseas those days and won't have any extra time off. Oh well maybe can get it on DVD if Tawzer does it.

[Lollipup]

http://www.thundershirt.com/

ETA the video add on the main page of their site is a little creepy!

Edited August 14, 2012 by Lollipup

[corvus]

Thanks Corvus. It's interesting what you say about the thundershirt. I have tried it on 3 dogs - my 2 and a foster. I have noticed no difference whatsoever so I just keep saying they are in the other 20% since the thundershirt site says it works on 80% of dogs! Or maybe there are just subtle differences I haven't picked up on, but seemed to make no difference to me.

It might have to do with how aroused the dogs you've tried it on are. We usually find with Erik it works wonders in fairly specific circumstances and is not real effective in others. It works best when his arousal is higher than usual, but there's a point where it gets too high and it won't do much. We don't use it so much anymore, but when we do it is, I think, more effective now than it was to begin with. Not sure why. I asked the company if there was any research into why or how it works and I gather from their response that there isn't.

Do you think BAT is particularly useful for herders since they find movement so rewarding? Would you choose BAT or LAT or other method in particular when focusing on reactivity to other dogs? You say you would use BAT if you thought the dog needed to be able to interact with the stimulus, do other dogs fall into that category? :)

I think it really depends on the dog and the handler and what they have at their disposal and what they want. I use LAT with Erik with dogs, but Erik is most of the time quite good with other dogs. He gets intense when he sees them playing fetch. I use LAT because I don't want him running over to them and doing his heeler thing around them, which most dogs HATE. And it's such a high arousal situation with a fast-moving, unpredictable trigger that I don't think it really suits the careful approach of BAT. But if you can get a dog and handler on board to help you, BAT for dog-reactive dogs can work, and lots of people have had success with it (and CAT). You have to consider how many dogs you can get, though. It's unlikely just one will be enough. You might also be able to use it in controlled situations like fenced dog parks if you stay outside the fence. If your dog is very sensitive something like a fenced dog park will probably be too busy, even from a distance.

I think that for reactivity towards something in particular, you have to consider why the dog is reactive. If they only have one or two triggers, it might be they are just plum scared and some classical conditioning will solve the problem. If they are emotionally reactive by nature, you're looking at ongoing management. Erik is a pro at CU stuff like LAT. He's turned around and run 10m back to me to start offering LAT before without a word from me because something is bothering him. It's very powerful if used consistently, but part of its power comes from the handler viewing it as a conversation. Erik uses LAT to tell me when something is approaching too much for him to handle, and I use it to tell him he can handle something that is bothering him using this rule structure, or to warn him that something bothersome is coming so he isn't surprised. If I don't listen to him when he's doing LAT and respond to what he's telling me, it will lose a lot of its meaning. It's not just a system for treating reactivity. It's a system for management and opening a channel by which you can communicate to each other as well. There's certainly an exchange in BAT as well, but it's more about treating a specific problem than establishing rule structures that can help you for a variety of problems. That's my view, anyway. As for whether the dog is going to interact with other dogs or not... That's your call. Some dogs don't want to interact, and that's why they are reactive. I think if they don't want to they should have that choice honoured and be allowed to stay well clear of other dogs. There may be exceptions, though.