Establishing operations

How do you figure that's classical and not operant?

Still though, that sounds like an SD. In the presence of "where's the ball," + command, the ball is available as a reinforcer. In the absence of "where's the ball," the ball isn't available. He can't see the ball, but in the past, in the presence of this, the ball has been available. You could have multiple SD for the same behavior, for example 2+2, and 3+1, different antecedents, but ask these of most people, you get the same response. I'd be curious to know what happens if you just say where's the ball.

griffinej5

How do you figure that's classical and not operant?

I never intentionally trained it or worked at it, I just always said the exact same thing in the exact same position with the exact same result.  I say it, dog's drive increases.  No actual behavior is being learned; nothing is actually being presented or taken away. The heeling behavior would definitely be operant, but that's separate because I can still heel my dog without this cue, and I can give this cue and notice the dog's response without also heeling.  It's more about the dog's state of mind than a behavior.

What is an SD?

Kim_MacMillan

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.

Hi 

Well we have something in common. I can't get around wheter what we do is an EO or classcial conditioning with teaching our dogs that something is good that they are unsure about. When i got both of my poodles  they would go zoomy over a thrown toy which has a couple of problems. It is away from you, the reward is quite short in duration and not the greatest motivator i have ever seen. It is dam hard to manage.

I think the first part is a bit like diet change. People love Macdonalds often but hate what is good for them unitl they get used to it. I guess that it we shut down the alternative (McDonalds) it beocmes an EO In my case i reduce the zoomie play.

So i use a fluffy toy on a bout 4ft of very small dia climbing rope (2mm in dia very strong not so visible as to destroy the illusion) so what happens now is my dogs go for a thrown toy which they chase around trying to catch in a small circle close to me.At this point i sue "Yes" whiich is my start cue.

I can start tug by using the rope when they catch the toy.

I use the toy directly in tug games

I change to a tug toy but yo yo back and forwards. I still use the fluffy toy on a rope for an end it all jackpot for absolutely brilliant work.

My dogs do not have access to toys during normal times.

When we are in the ring we are asked by the judge "are you ready?". Some clever people use this as the charge up cue.I say yes and off she goes.

You can see it visibly in http://s587.photobucket.com/albums/ss311/deniscody100/LuciLuci%20UD%20training/?action=view¤t=seekback.flv about 3 seconds in.

If i don't get the little anticipation, i can say no and reset my dog. 

Luci will steal tug toys out of my pocket and goes for them any day anyhow. Sam is getting there but not enough that i will trial him yet.

If you have a look at Luci heeling during the signals, her legs move quite differently than food only rewarded heeling. Her body posture is quite different. I had to resort to this mode for quite a while when i was dealing with a noise phobia that she had. .She is nowhere as obviously keen. I could post the clip but am concerned that she be judged by it but will do so if it is useful.  Again it is a sample of one but many top poodle trainers told me the same thing. Aint got toy drive aint got a working poodle... Now Luci is a poodle from a particlular line, my Lab had great heeling and he worked to Lab rules. No matter what the problem just throw me more food. He would kill his mum for it.No need for EOs for him...

Classical conditioning involves stimulus-stimulus pairings, and has nothing to do with the reinforcer.

In classical conditioning you have this(the arrow means results in):

Unconditioned stimulus ---->  unconditioned response

through repeated presentations of a neutral stimulus with the unconditioned stimulus you get

conditioned stimulus ---> conditioned response

You have to put these two together in order to get this.

So for example

Loud noise ----> startle

presented with "lookout"

lookout ---> startle

 An SD (discriminative stimulus) is by the simplest definition,  a cue. In the presence of an SD, the behavior it signals is reinforced, and in the absence, it's not reinforced. In the presence of weave poles, a certain type of body motion is reinforced. In the absence of weave poles, this type of motion isn't reinforced. You don't see dogs walking like they're in weave poles all the time? Nor do I constantly define and give examples of an SD. I know that doing so only gets reinforced in certain situations, so I only do it in those cases.

 Found some notes I was looking for in regards to setting events.

Conditioned Stimulus (setting event)

[Discriminative Stimulus (SD)--> Response---> reinforcer]

for a dog we might have one like this

"weave"

weave poles --> weave through the poles --> reinforcer

In a very simple sense, the setting event is the context. Without the context of being told to weave, in the presence of the weave poles, it's not the correct response. On that note, in the context of this forum, my definiton and example given for any one term is different from the ones I gave during my oral exam, or the one I would write on a test.

 Hi

You are filling in the theory for things that i do and get away with :)  I say that it is pointless using a specific cue for many of our obedience exercises as it is blindingly obvious by context as to what the dog need to do next. We just need to start them. I have found that with Luci if i use a specific cue she would become confused as if i was asking her to do something different.She would offer other behaviours. Some of that might be that i am a chronic late cuuer. I have been on the way to a trial and realised that i don't have a formal cue for an excercise.

An example is the retrieve. When i throw the dumb bell it becomes the setting event. She really resists heavy efforts such as hands alongside her to show her where it is , she already knows. So something it different is it?  Shouldn't i do it?

All that she needs to do is go. Sometimes too much information is just as confusing for dogs and humans as too little :)

griffinej5

Classical conditioning involves stimulus-stimulus pairings, and has nothing to do with the reinforcer.

In classical conditioning you have this(the arrow means results in):

Unconditioned stimulus ---->  unconditioned response

through repeated presentations of a neutral stimulus with the unconditioned stimulus you get

conditioned stimulus ---> conditioned response

You have to put these two together in order to get this.

So for example

Loud noise ----> startle

presented with "lookout"

lookout ---> startle

Thanks for your response. it is helpful and illuminating.

I was reflecting on how simple it was to get my Lab to heel . None of this swapping toys about, multiple EOs , it was just clicker food and fading. Now how can you get that wrong?  Well you can.... I see it once a week watching a yank and crank class. Something so simple got so wrong.

 I guess in macro terms that my poodles need an EO to change zoomies with a ball to catching a toy on a rope and tugging.

Then catching a ball on a rope and tugging to playing tug with a tug toy.

Meanwhile we had been establishing  that heeling results in for poodles a lowish value reward (food)

we now change the reinforcement to a higher value manageable toy based reward with occasional food.reinforcement.

All we do with Labs is create this equation heeling ===food sometimes. And they go for it!!!

griffinej5

 Yeah, Anne, I know that. Often times, there is a slightly different topography in behaviors maintained by two different reinforcers, but not always. I experience that more with humans that with dogs. I was trying to keep my example to dogs and realistic. 

The question we're asking here in dealing with EOs is not how can I reinforce, but how can I make my reinforcement more effective.

Yup, that too:-))

Always, of course, based on what the dog considers reinforcing.

spiritdogs

Always, of course, based on what the dog considers reinforcing.

Yup, too some extent. Deprived enough of something, like food, we all get a little more willing to accept something we wouldn't. I don't suggest though that anyone starve his or her dog in order to get him to accept a lower value food as a reinforcer.

Poodleowned, I guess if their are no choices in terms of what to do, having a verbal makes no difference. If it is always to do the same thing in that situation, I don't think you have to say anything. I always want my dog to sit upon approaching the edge of a curb. I say nothing, because I don't want him to rely on my behavior to do this. I want to be always, in the presence of street corners, sit. I want it this way because If I use him one day with any kids I work with who don't stop at the end of the street, I might want to use his sitting at the end of the street as a prompt to stop and look. The weave pole scenario though is based on a scenario in which a choice is possible, and sometimes, in the same situation, you would want a different behavior to occur. 

I was in the mood to mess around with drawing a little picture

So, you have weave poles, a jump, and an A frame here. The black line is his response, going to the weave poles, and I couldn't draw him going through, but he does. He gets to the end and is reinforced.

I don't suggest though that anyone starve his or her dog in order to get him to accept a lower value food as a reinforcer.

Yeah, I don't think that would have occurred to me to do.  I'm hearing enough whining today from my hound because of that cone on his head. 

griffinej5

Poodleowned, I guess if their are no choices in terms of what to do, having a verbal makes no difference. If it is always to do the same thing in that situation, I don't think you have to say anything.

There are choices sometimes . In formal obedience anticipation is a no no. For example the dumb bell gets thrown, they need to wait until you tell them to go out on instruction from the judge. That is why i use a start cue "go".

In heeling i just move.