Establishing operations

I have to say that i really find it difficult to accpept a behavourist only explanation of behaviour and the particular example here is almost perverse. Self harm in humans is often attached to other issues and is commonly associated with "meltdowns" in Apsergers for example. It might be that a similar phenomina occurs outside of Aspergers but is outside my knwoledge zone.

The primary goal of self harm during a melt down is to obtain quiet and space and to sort of clear the mind. It acts in my opinion as a disruptor to unbelivable stress and angst usuully from difficulites in communicating. 

I often read articles like this and hope like hell that some recent graduate doesn't read it, and think that they can suddenly fix' self harm issues in other human beings from some station on high. The consequences are often an example of the abuse of power. I have observed some of the *******  via my wife's wrok in this and allied fields and the reality is very chilling.

So if we aside the very poor example, and get back to preparing reinforcers , i will say that for a while there is been an almost underground of R+ trainers who have observed different topographys in their dogs behaviours but  have almost been afraid to put up their hand and say "look when i use this reinforcer my dog's beahviour is different". It is like we have got stuck in the systems view of behaviour that was put to us by Skinner so many years ago. It is as if we have got lost in the very straight jacket that Skinner sought to free beahvourial scicnece from.

Leaving aside whether Panksepp is too Anthromorpological , we can use information from his work that he used to present his arguements. It is fairly certain that the part of the brain htat responds to operant and classical conditioning is different to the part of the brain that responds to play. The Neurochemical picutres seem to be different. Sure we can go on saying  that play can be presented as a reward and i certainly presnet it as that, but the activity of a dog playing is different than a dog consuming a treat.

We also know that the prefered pleasure for many animals is the chance to seek or look for rewards. This topography is different to the consumptive phae of most food rewards.

Many great trainers have instinctively used that knowledge. Most good trainers that i know of these days spend an enormous amount of time just playing and being with theri dogs.  When i climb in to the ring, i don't even think about what kind of reinforcement that i am going to give my dog when i get out of the ring, i am thinking about what reinforcement that i can give her right then in the ring. How i as a human can be reinforcing for my dog.

EOs for this is "Most good trainers that i know of these days spend an enormous amount of time just playing and being with theri dogs".

Now in one of my first responses i showed a video of my two dogs playing. It is a bit stilted becuase of the video. That is EO ing.

I'm going to leave the discussion after this, because honestly, if you're going to talk about EOs, you're dealing with behavior analysis. The name comes from behavior analysis, so does the meaning. It is difficult to have a discussion in which few of the participants actually have any knowledge of the subject.

Additionally, meltdown is a poorly defined behavior, *removed by moderator*. One of the reasons we analyze behavior is so that we can provide effective treatment. We don't just decide that most of the time, behavior X is reinforced by Y. Each person, or animal is an individual. To decide that this particular person's self injury is maintained by whatever because that's what it is for most, we do a major disservice to the ones who aren't in that "most" group. We might subject that person to ineffective procedures, and we might make their problems worse. This is a reason why we don't just decide on a function based only on the behavior. I do unfortunately work sometimes in cases where someone above me attempts to provide a course of treatment without determining the function of the behavior. It's never effective, and sometimes makes things worse. These people though, are not behavior analysts. They have degrees in Psychology, they maybe took one course, or a part of a course about behavior analysis. We're lucky if they have a decently trained dog. They tend to assert the same type of thinking that you do though. If he is doing X, it must be because of Y (without necessarily even observing the person in question), and we must do Z.

griffinej5

Additionally, meltdown is a poorly defined behavior, *removed by moderator*.

I thinik that you haven't read my post properly  and discovered that i placed restrictions on what i why saying to my direct experience

I qoute

"Self harm in humans is often attached to other issues and is commonly associated with "meltdowns" in Apsergers for example. It might be that a similar phenomina occurs outside of Aspergers but is outside my knwoledge zone.

The primary goal of self harm during a melt down is to obtain quiet and space and to sort of clear the mind. It acts in my opinion as a disruptor to unbelivable stress and angst usuually from difficulites in communicating. "

You are actually managing to refelct the arrogance (i hope accidnetally) that my nightmares are  made of. There is a fair chance that  i was feflecting my experience as an Aspie and the community of Aspies that i am in contact with.

*removed by moderator*

griffinej5

They tend to assert the same type of thinking that you do though. If he is doing X, it must be because of Y (without necessarily even observing the person in question), and we must do Z.

*removed by moderator*

Aww, please don't go. I'm learning stuff!

So, what is it when you use cues to tell your dog that you're gonna do something awesomely fun with them in a moment? Not like a marker and then a reward, but like when I say to my dogs "Ready?" and they know that even if I don't even have a toy on me I'm going to play a fun game with them eventually. They might have to do several things first, and it might be a full minute before they finally get to play, but they stay excited that whole time because they are anticipating the play reward.

I'm thinking the same thing happened with Erik and his meals by accident. He figured out that when we pick up the bowls, we will then go to the fridge, get out some meat, find a spoon, open the containers of food, dish out the food, mix it up, ask for quiet and a sit, then feed him. So he just started getting crazily excited when we picked up the bowls. We spent some time derailing his predictors of imminent meals and the behaviour has gone away. What do you think? Can cues of imminent rewards that may arrive several minutes after the cue be an EO? Assuming that the cue creates such anticipation for the reward that when the reward does come it is better than if there is no anticipation?
 

 EOs for dummies

Hi 

I have just looked up a couple of references that help understand what an Eo is for non professionals like myself. The original paper that Anne quoted also explains CEOs which are conditoned EOs.

An EO is

“The term establishing operation (EO) refers to an event that alters the reinforcement effects of a

particular stimulus.”

www.auburn.edu/~jec0018/8550/WC1998.pdf

In Michaels paper, much of which is opinion, he says quote 

"A num-
ber of my statements about EOs are a form
of conceptual analysis that I consider quite
defensible, but they often go beyond well-
established empirical support. Such state-
ments, when given without qualification or
extensive justification, may seem to be dog-
ma, but I would like the reader to assume
a parenthetical ‘‘from my perspective’’ or
‘‘in my opinion’’ after many such state-
ments. "

and is tough going for us non pros.

He is trying to work out the nature and reason and relationship of Eos and CEOS just as we have fialed ot do so miserably in this series of postings.

If we go back to the first more accesible statement, my arguement is that play may form an EO if built and developed so that it is more reinforcing for the dog. This has always been my contention. Play is tricky for many dogs and takes patient effort to be built as a reinforcer.

During play i use my start and heeling cue shamlessly and often so that it becomes a CEO and used by the dog to pair play activity with a less rewarding activity.

Does it work? well with a sample N=1 i woudn't have the foggiest. But my goodness she is one fiesty happy dog in the ring.

In MIchael's article, some note is made of unpairing CEOs. I did this with my dogs so that my car coming up the drive , door opeining etc ceased to become the predictor of me coming to greet them simply by expanding the time  between those actiites and me greeting them.

 I have one overarching concern. I realise that the name Panksepp is like yelling bomb on  a plane, but one of his statements resonated heavily with me. I have a strong background in the physical sciences and it holds true there... It basically said that with overspecialisation there was little communication between fields. I have found that the same is true . There are scientists running around yellig at each other saying the same things in hard physcial sciences.

If we only look at behaviour from a system point  of view then our view is very restricted.

Moderator Speaking

Just a reminder that there's nothing on this board that warrants personal attacks on other forum members simply for debating a topic.  There just isn't.  If you feel you have a personal problem here, contact a Moderator if you can't work it out privately, but in-thread is inappropriate.  The only folks who have the right to tell another member to leave are us Mods/Admins, for violations of rules, ok?  If you feel a rule has been broken, report the post.

Carry on.... in a respectful manner, please. 

 Corvus, I think, but I am not quite sure and I will have to open my book and do some more reading, that you are describing a setting event. The basic description of that which I have here is, "a stimulus response interaction that will affect stimulus response interactions that occur later,"  It's used sometimes interchangeably with Establishing Operation, but they're not really the same. (Newman Et. Al., 2003) I'll have to do some searching to see if I can get a really good example that really differentiates it from an EO. It might be a transitive CMO, but this is another area of confusion, as this is often interpreted as an SD (discriminative stimulus, a cue if you will, signals that a certain behavior will be reinforced). I think though, given that a CMO-T is a ," stimulus that alters the value of another stimulus" this might be it. (Michael, 2007)

One of the examples Michael gives in Applied Behavior Analysis, is regarding flashlights, which are typically available in a home. However, you don't usually look for one until the power goes out. The darkness increases your flashlight searching behavior, and also makes the light from the flashlight more valuable to you.

I'll have to go see if the Michael 1982 paper has the full description of the screwdriver example mentioned in the article Anne linked to.  There's a pretty nice description of it here in my textbook, but I'm not sure if that's available somewhere that everyone else could see it without hunting one down.

I use toy prohibition as an example of EO's. There are certain toys that my dogs really enjoy that are put away and only used in certain situations. In the beginning I don't use them for any training, I just build interest and drive for them. The more you "play" with them and don't use it as a reinforcer for anything, the more the reinfocing value it develops. And then eventually, once you have increased value for it, you start to use it in training to aid as a reinforcer.

I also have a word for Gaci that immediately is an "on" switch for her, and it's an EO. I just created it by using it before reinforcing situations. I use it now to build excitement and increase drive when I want it. It's "Ready", although it generally comes out as "Reeeeeeaddddddyyyyyyyyy.........readyyyyyyyyyyyy........GO!!!!" Now if I even use that word like "Going to get ready to go", I have to be careful not to over-use it.

I'm so glad we get to actually have "real" behaviour discussions...we didn't used to get in depth into these types of behavioural terms. Makes me happy!!!!

Kim_MacMillan

I also have a word for Gaci that immediately is an "on" switch for her, and it's an EO. I just created it by using it before reinforcing situations. I use it now to build excitement and increase drive when I want it. It's "Ready", although it generally comes out as "Reeeeeeaddddddyyyyyyyyy.........readyyyyyyyyyyyy........GO!!!!" Now if I even use that word like "Going to get ready to go", I have to be careful not to over-use it.

Nikon has one like this too.  When I started fading the ball reward for heeling, I would always say "where's the ball?  where's the ball?" and he gets all excited and even barks at me.  Now I say this to him if I intend to go for a longer period of heeling before I reward, to remind him what is coming.

I think his track command ("search";) works the same way.  It's supposed to be a command, but he is trained to track with food on the track, so basically the word cues him into looking for the food.  It may very well be that a trial is the only time the dog ever tracks without any food reward.  For me, tracking is rather boring and time consuming, but the dog loves it.  I think it's because the communication is very clear.  The food is already in the footstep, so if the dog is tracking correctly, he's automatically getting rewarded.  If he's not tracking correctly, no reward.  Slightly off topic.

 Kim, (and everyone really), one of the really difficult parts of this really is that not everyone agrees about this. Some people don't even believe in MOs, which is the more broad category under which EOs fall. I'm not sure that what you're describing is an EO, and not a setting event, because I don't really know that it alters reinforcer effectiveness.

Liesje

Kim_MacMillan

I also have a word for Gaci that immediately is an "on" switch for her, and it's an EO. I just created it by using it before reinforcing situations. I use it now to build excitement and increase drive when I want it. It's "Ready", although it generally comes out as "Reeeeeeaddddddyyyyyyyyy.........readyyyyyyyyyyyy........GO!!!!" Now if I even use that word like "Going to get ready to go", I have to be careful not to over-use it.

 

Nikon has one like this too.  When I started fading the ball reward for heeling, I would always say "where's the ball?  where's the ball?" and he gets all excited and even barks at me.  Now I say this to him if I intend to go for a longer period of heeling before I reward, to remind him what is coming.

I think his track command ("search";) works the same way.  It's supposed to be a command, but he is trained to track with food on the track, so basically the word cues him into looking for the food.  It may very well be that a trial is the only time the dog ever tracks without any food reward.  For me, tracking is rather boring and time consuming, but the dog loves it.  I think it's because the communication is very clear.  The food is already in the footstep, so if the dog is tracking correctly, he's automatically getting rewarded.  If he's not tracking correctly, no reward.  Slightly off topic.

Not meaning to nitpick, but just because I like to argue the other side. Does saying where's the ball change the effectiveness of the delivery of the ball, probably not I think. I'm guessing it's just acting as verbal R+. It's pretty common to condition verbal praise as a reinforcer so that you don't have to constantly use food or something else. I don't think it's the verbal praise that changes the value of it, it's the deprivation.

Having had the dog in questions for six years, and having only implemented the "ready" cue this year, I can say that it does indeed alter reinforcer effectiveness. My prime example is my experiences with trying to get Gaci to treat a tug as a reinforcer. People who know me know the troubles we have had trying to use it as a reinforcer. It used to be touch and go (and sometimes still is, it's a WIP), but once I taught a "ready" word her value for the tug increased dramatically. I also use it in agility and it really does increase the reinforcing value of doing agility (jumping specifically). You can see clear differences between when you use it, and when you don't. That to me is the direct definition of an establishing operation.

Then again, some texts treat setting events and establishing operations as the same thing, or similar things....almost subheadings. Wherein establishing operations and abolishing operations can both be types of setting events.

griffinej5

Not meaning to nitpick, but just because I like to argue the other side. Does saying where's the ball change the effectiveness of the delivery of the ball, probably not I think. I'm guessing it's just acting as verbal R+. It's pretty common to condition verbal praise as a reinforcer so that you don't have to constantly use food or something else. I don't think it's the verbal praise that changes the value of it, it's the deprivation.

It either doesn't matter, or makes it even better (in that I'm more likely to reward because the dog got that cue, he figured out right away what we were doing and popped into drive without me having to spend a lot of time using the ball to amp him up and get the "game" going).  However the verbal cue itself doesn't really work without the ball as follow up, if that makes sense.  I could drop the verbal cue and still progress very quickly with the ball, but I cannot quit using the ball and only use the verbal cue, so I don't think the verbal cue is pure R+, do you think?  I use the verbal cue at the beginning, and possibly if I'm "losing" the dog (he looks away from me or the heeling gets flat), so also in that sense it's not really a reward by itself.  If the dog is heeling nicely, he gets verbal praise or the ball itself.  I think the "where's the ball" is more of a motivator than an actual reward.  For me it sometimes takes the place of doing a bit of prey play with the dog before popping into heel, since this takes a lot of time and energy on my part and also requires that the dog very quickly out the ball in order to build the drive and then momentarily cap it as we start to heel.  It's no secret my dog usually will not quickly or cleanly out the ball so building in the "where's the ball?" verbal cue sort of has the same effect without all the extra work. 

 Nope, given that description, I don't think where's the ball is R+ at all. If I get this correct, at this point, it sounds like he's not doing what you want, and when you say it, he reliably will do whatever the behavior is that this is associated with. Given that as it is, it sounds like it's an SD. Just because you are more likely to reward doesn't make it necessarily that the value of the reward is any better, and leads me more to think that it's an SD. In the presence of the SD, a certain behavior is rewarded, and in the absence, it's not. He could choose to heel any time he wants, and you could choose to say four all day long. However, without an SD, the dog isn't going to heel, and you aren't going to just walk around saying four.

I thought of getting really funny here, and I was going to say, that you could walk around any time you choose in a halloween costume, and go trick or treating, but you don't do that either. 

I think it is a cue that has been classically conditioned to help get the drive I want.  It's not that he's not doing what I want as far as the command that was given (if that is the case I will stop, correct, break into play, etc), but I want a more aroused state of mind as he's doing it.  I mostly say it as we start.  The dog is sitting in the basic position (at my left) with his eyes on me face.  I ask him "where's the ball?", give the command, and go forward.  The dog cannot see the ball but when I give that verbal cue, he then knows that the ball *is* there and he's going to (eventually) get it.  In Schutzhund we like the dog to "push" the handler, even in obedience and this gets him "pushing" me, asking me (literally, by salivating, rocking on his font feet and barking) to do the work.