Exploring A Concept

[Rom]

Okaaayy......thinking cap on.....

So could it be said in the instance of instinctive chase behaviours that it is too uncomfortable for the dog not to engage in the chase? That the intensity of the negative punisher delivered by not chasing (or the quarry withdrawing itself or acting to prevent the dog from catching it) that the dog has to change behaviour in order to relieve that discomfort....and so it engages in the chase in order to catch the reward. I guess what I'm asking here is by what function does instinct provide motivation? Is it through the motivation matrix? Through both intrinsic and extrinsic punishers and reinforcers? After all, instinct is not robotic behaviour and learning and behavioural change can still occur within the realms of instinct....if it couldn't, then a dog couldn't adapt to changing environments and hunting circumstances....what is training in drive if it is not harnessing hunting instincts and teaching a dog that an obedience routine, for example, gives you the best possibilty of having a successful 'catch'? So have we not super imposed an obedience routine over hunting instincts and influenced how the hunting behaviours are expressed?

Also we know that negative reinforcement cannot exist without positive punishment......negative reinforcement is the removal of something unpleasant, right? But how did the unpleasant thing get there in the first place for us to be able to remove it? An aversive has to have been applied first before you could take it away to deliver the negative reinforcement. So, if the aversive is applied at the time of an unwanted behaviour and then removed when the dog engages in a wanted behaviour then you have applied both positive punishment and negative reinforcement in order to both weaken the unwanted behaviour and strengthen the wanted behaviour.

So what I'm getting at is the opposite to the above and I feel that in line with the above positive reinforcement cannot exist without negative punishment....the dog has to want the positive reinforcer and if he's not getting it he will change his behaviours in order to get it because there is some discomfort for the dog in not having that reinforcer. So negative punishment weakens those behaviours that do not successfully deliver the positve reinforcer and the positive reinforcer strengthens those behaviours that do deliver the positive reinforcer.....

[SkySoaringMagpie]

Barking, digging, fence jumping, fighting, stock chasing. My theory is that these kinds of self rewarding behaviours are extremely difficult to extinguish without the use of aversives.

I have yet to hear of anyone who has managed to cure stock chasing with positive reinforcement.

Some of those behaviours I would not bother trying to deal with through training. While changes to infrastructure can admittedly be expensive, a good secure kennel run where you have the ability to properly separate dogs deals with all but one of those behaviours without the risks associated with failures in training.

Also, with fighting, I think that aversives can make the problem worse.

[Rom]

I've had an idea.......I think I understand why there are conflicting opinions in this thread. There is a clue in the below

Denis Carty:

......... the flaws and weaknesses of operant learning theory interpretation of learning/teaching which excludes the internal state of the animal at any given time, the animals psychology etc........

This is not the first time that I've seen a trainer refer to operant theory as being flawed and I don't know if its right or not, but here's where I figure my thoughts are originating...I'm talking in terms of operant theory in my examples and operant theory, as it was originally written, has at its basis the four quadrants of the motivational matrix. But I'm thinking in terms of there actually being 8 octants (for want of a better word) in the motivational matrix when you take into account the terms intrinsic and extrinsic. Some of the negative punisher that I'm referring to are intrinsic negative punishers......does that make sense? Am I trying to apply operant theory and take into account the internal state of the animal?

[SkySoaringMagpie]

This is not the first time that I've seen a trainer refer to operant theory as being flawed and I don't know if its right or not, but here's where I figure my thoughts are originating...I'm talking in terms of operant theory in my examples and operant theory, as it was originally written, has at its basis the four quadrants of the motivational matrix. But I'm thinking in terms of there actually being 8 octants (for want of a better word) in the motivational matrix when you take into account the terms intrinsic and extrinsic. Some of the negative punisher that I'm referring to are intrinsic negative punishers......does that make sense? Am I trying to apply operant theory and take into account the internal state of the animal?

My theory on the differing opinions is that the internal state of the human warrants much more investigation than the internal state of the dog.

Many people, including me, tend to rationalise what they have already decided, and what they have already decided is, as often as not, an emotional or ethical matter, not a scientific one. A person who has a heavy investment in thinking of themselves as perfectly rational is usually one of the most irrational of the lot.

As someone else mentioned, people are often very unaware of what they are doing with dogs. Heaps of positive only people use correction but don't recognise it as correction because it is verbal for example. Then there are the pro-correction people who tell others they need to be "tougher with that dog" while letting their own dogs stomp all over non-enforced boundaries.

I'd actually be interested in seeing a study along the lines of a myers-briggs survey - what personal human attributes translate into preferences for different training methods? Because that informs why a person insists on "rewarding" a dog by doing something the dog really dislikes, or "punishing" the dog by doing something the dog thinks is pretty good.

[Diva]

I'd actually be interested in seeing a study along the lines of a myers-briggs survey - what personal human attributes translate into preferences for different training methods? Because that informs why a person insists on "rewarding" a dog by doing something the dog really dislikes, or "punishing" the dog by doing something the dog thinks is pretty good.

That would be an interesting study.

[m-j]

So could it be said in the instance of instinctive chase behaviours that it is too uncomfortable for the dog not to engage in the chase?

It certainly would be if they were restricted, but I really don't think they give it a thought. They have an urge they need to fufill and I don't think that urge is negative they are doing it because at that moment there is nothing else they would rather do. ????

cheers

M-J

[Rom]

M-J, picture yourself in this situation....another human example:

You are a passenger in a car and your wallet/purse is sitting on dash. The driver takes a corner and your wallet suddenly starts to slide along the dash at speed heading for the open window. Think about your internal feelings at that moment.....are you feeling "Yahooo, I'm about to be rewarded", or "Shyte, I'm about to lose my money"(read opportunity to catch). There is discomfort for you in the fact that your wallet is sliding out the window (read getting away from you)and under those circumstances few would just sit and watch that happen without changing behaviour suddenly and explosively to try and catch the wallet and alleviate the discomfort.....and you felt that discomfort, even if momentarily, even though you were not being prevented from trying to catch your wallet.

Did you act instinctively and just respond to the stimuli as they presented themselves, or did you have to rationalise and make a decision to catch your wallet?

The discomfort that you felt was an intrinsic negative punisher (ie something that you valued being removed/withheld) that caused you to change your behaviour and move to try to catch the wallet.

I believe that it is a similar process that is happening with your greys. And like your greys who showed no interest in the skin when it wasn't moving, your wallet sitting on the dash without moving was a neutral stimulus.....until it started heading for the window, so it didn't cause a change in behaviour until that point.

[Erny]

Rom ..... I love the way you 'deep think' things through. And the above is an excellent analogy. :rolleyes:

[Lablover]

:rolleyes:

Am I trying to apply operant theory and take into account the internal state of the animal?

Now, that takes experience and knowing the individual.

[Rom]

Rom ..... I love the way you 'deep think' things through. And the above is an excellent analogy. :rolleyes:

Thanks :p

I'm grateful to those who are contributing so that I can nut it all out

Am I trying to apply operant theory and take into account the internal state of the animal?

Now, that takes experience and knowing the individual.

Or testing the balance of drives and understanding the instinctive goals behind those drives, maybe?

[Rom]

Intuitively, we associate something bad as punishment but in operant conditioning, a punisher is something that decreases a behavior.

A positive punishment is something bad happens. A negative punishment is something good gets taken away.

Driving in heavy traffic is a bad thing. One morning, I left a bit earlier than usual and didn't run into traffic jam. The next morning, I left earlier again to avoid heavy traffic. My behavior of leaving earlier is strengthen by the consequence of avoiding heavy traffic.

In this example, there is no punishment in the sense of operant conditioning. My behavior was negatively reinforced.

Edited for spelling.

Still thinking

See how the behaviour was instigated by a punisher though....to you driving in heavy traffic is a bad thing....something that you want to avoid, so you changed your behaviour so that you could avoid it. In this instance the move is from +P > -R.

[Ah Ngau]

Okay, I see what you mean.

...

So could it be said in the instance of instinctive chase behaviours that it is too uncomfortable for the dog not to engage in the chase?....

In this case, you are defining a positive reinforcement as a punishment.

AND

...

Driving in heavy traffic is a bad thing. One morning, I left a bit earlier than usual and didn't run into traffic jam. The next morning, I left earlier again to avoid heavy traffic. My behavior of leaving earlier is strengthen by the consequence of avoiding heavy traffic...

...

See how the behaviour was instigated by a punisher though....to you driving in heavy traffic is a bad thing....something that you want to avoid, so you changed your behaviour so that you could avoid it. In this instance the move is from +P > -R.

In this case, you are defining a negative reinforcement as a punishment too.

You are right to say that all conditioning are punishments because in essence, you are defining ALL four quadrants of operant conditioning as punishments (i.e. drive of uncomfortable desires, actual punishments, threats to punish and denial of good things).

Logic dictates that you are right by definition and I kinda suspect that you are a "glass half empty" person.

[Rom]

Okay, I see what you mean.

...

So could it be said in the instance of instinctive chase behaviours that it is too uncomfortable for the dog not to engage in the chase?....

In this case, you are defining a positive reinforcement as a punishment.

AND

...

Driving in heavy traffic is a bad thing. One morning, I left a bit earlier than usual and didn't run into traffic jam. The next morning, I left earlier again to avoid heavy traffic. My behavior of leaving earlier is strengthen by the consequence of avoiding heavy traffic...

...

See how the behaviour was instigated by a punisher though....to you driving in heavy traffic is a bad thing....something that you want to avoid, so you changed your behaviour so that you could avoid it. In this instance the move is from +P > -R.

In this case, you are defining a negative reinforcement as a punishment too.

You are right to say that all conditioning are punishments because in essence, you are defining ALL four quadrants of operant conditioning as punishments (i.e. drive of uncomfortable desires, actual punishments, threats to punish and denial of good things).

Logic dictates that you are right by definition and I kinda suspect that you are a "glass half empty" person.

Not quite. I'm saying that punishment is the instigating factor in you deciding to change your behaviour and you changed your behaviour in order to deliver the negative reinforcer of missing the heavy traffic. I'm saying that we are actually working on two behaviours. The first is you leaving at a time that meant you getting caught in the heavy traffic. Leaving at this time was weakened by a +P. The second is you leaving earlier to miss the heavy traffic...this behaviour was strengthened by -R.....and as dog trainers we would also add +R if the dog got it right.

It depends on whats in the glass on rare occasions the glass hasn't a hope of staying half empty

Edited for clarity

Edited February 28, 2008 by Rom

[m-j]

until it started heading for the window, so it didn't cause a change in behaviour until that point.

That's true but my motivator wasn't instinctive, the action was, may even be a reflex ie messages get short circuited, my motivator was the possibility of loosing my money, lots of repercussions there, none for the greys if they didn't chase as they know they don't chase for their food.

cheers

M-J

[Rom]

What do you consider to be a punisher? Not having food and making the dog hungry is not a punisher, its a stimulus that will then cause the dog to do XYZ behaviour and get XYZ result- some kind of consequence that will then determine if that behaviour will be repeated next time the same stimulus is presented. I don't consider dogs to be in a constant state of 'punishment' but maybe my definition of punishment is different to yours.

Stimulus or motivation is different to punishment IMO

Cosmolo, thanks for these statements....I get where you're coming from now.....its not so much that the concept that was wrong, but the language that I was using?? Eg, if an event occurs before a behaviour then it is labelled a trigger or a stimulus if it occurs after a behaviour then it is labelled in the terms of the motivation matrix even though the event could be exactly the same before or after the behaviour and produce the same biological stress in the animal?

Perhaps Akitaowners wording was better by the use of the words 'pressure' or 'aversive'?

ETA: the word aversive in last sentence.

Edited March 5, 2008 by Rom

[Lablover]

Or testing the balance of drives and understanding the instinctive goals behind those drives, maybe?

Sure.

I do think it is important to learn of operant and classical conditioning, but I certainly do not dissect every behaviour I personally exhibit through the day.

I am just like my dogs eh, I do what works for me - trial and error.

I was actually speaking to a friend yesterday who mentioned, IHO, Balabanov as the conditioning king.

I should stay out of these topics, as I feel the most important criteria is an active dog, with good temperament. Not many of those!!!!!!!!!

[Rom]

until it started heading for the window, so it didn't cause a change in behaviour until that point.

That's true but my motivator wasn't instinctive, the action was, may even be a reflex ie messages get short circuited, my motivator was the possibility of loosing my money, lots of repercussions there, none for the greys if they didn't chase as they know they don't chase for their food.

cheers

M-J

This statement took me on an interesting line of research.....

Simple reflex is basically automatic responses to protect the body from direct harm....eye blinking, yawning, knee jerk reactions, withdrawing from something hot etc. There are more complex reflexes, but they are conditioned....kind of along the lines of Pavlov's work with the salivating dogs.

I can't remember the exact path I took to get from there to the next point (although some of it was prompted by conversations with Erny), but I ended at the endocrine system and homeostasis. The root of the word homeostasis means resistance to change and the function of homeostasis is to maintain the internal environment at optimum balances for survival.

If there are imbalances diseases develop or death can be caused depending on the imbalance. And imbalances of some hormones can affect/effect behaviour. While the production of hormones is necessary for some internal functions, the body also acts to destroy those hormomes once they have acted on the target organ or receptor.....generally they are destroyed by the target organ/receptor itself so that a state of homeostasis can be returned to.

There are two ways that the internal system recieves feed back that it must act to return the body to a state of homeostasis in order to maintain a healthy internal environment.....they are negative feedback loops and positive feedback loops. The majority of feedback loops in animals are negative and there is only one positive feedback loop identified and that is the effect that a suckling whelp has on the production of the hormone that triggers milk let down.....if this was a negative feedback loop then the suckling action would cease or reduce milk let down rather than increase it.

Returning to topic, the interesting thing about stress hormones and adrenalin...the production of which are triggered before fight, flight or persuit behaviours is that the only efficient way for the body to destroy/metabolise these hormones so that it can return to a state of homeostasis is through physical activity. These hormones also have an effect on emotions.

So...in the case of instinctive behaviours, not only does the animal act in order to gain a percieved reinforcer/reward, but it also acts to relieve an uncomfortable internal state which occurs before the expression of the behaviour.

LL:

I do think it is important to learn of operant and classical conditioning, but I certainly do not dissect every behaviour I personally exhibit through the day.

Touche!

Under different circumstances I probably would have been one of those annoying little kids that kept asking "But why?"

BTW curious that you mention Ivan Balabanov....does his work cover some of this stuff?

[Lablover]

LL:

I do think it is important to learn of operant and classical conditioning, but I certainly do not dissect every behaviour I personally exhibit through the day.

Touche!

Under different circumstances I probably would have been one of those annoying little kids that kept asking "But why?"

BTW curious that you mention Ivan Balabanov....does his work cover some of this stuff?

Hey, I am an adult and still ask BUT WHY............but very little nowdays.

ROM, I was with a group of dog trainers and breeders over the weekend, one of whom, whose passion is with Malonios. This particular fellow has travelled the world.

We worked our dogs and finally as USUAL as dog people tend to do LOL, discussed training (for hours!!).

I have the Ivan tapes, even though I am a retriever trial based person.

My ears pricked when the guy mentioned Balabanov, as the operant king. So many topics were covered into the night and next morning.

As I am house bound today, I might slip Ivan's DVD on, the watch him more closely.

[m-j]

There are more complex reflexes, but they are conditioned....kind of along the lines of Pavlov's work with the salivating dogs.

Hi

Long time replying as I have been busy, sick dogs and litter of pups. I can't find my book that has the info on this stuff but this conditioning is the type that would make me reach for my purse as i had to learn that money has value.

I still have a hard time coming t terms with this as actions and drives are governed by emotions. Can't see where the negative emotion for chasing an item would come in unless the dog is restricted from completing the prey drive sequence must find my book and have a read.

Although long winded , it has been an interesting thread.

I do think it is important to learn of operant and classical conditioning, but I certainly do not dissect every behaviour I personally exhibit through the day.

Hi LL

Gosh we would never get anything done if we did

cheers

M-J

[Rom]

Hi M-J,

When I was talking this over with a friend she gave me an example in regards to the wallet on the dash of the car that I'll admit gave me pause for thought....Her example was what if instead of a wallet, it was a piece of paper....many would react the same way in order to try and prevent the piece of paper from going out the window......

I pondered on this for a bit and then thought 'Why would we be less likely to react in that manner to a random piece of paper blowing past us down the street?'

I'm still digesting that one......

In any case, a member of a UK forum requested that I post this topic over there....its taken some interesting twists and turns.....I'm not sure that I can post a link to it here given the forum rules, but will be happy to PM a link if anyone is interested.