Investigating Learning Theory Further: Beyond P,R,+, and -

Cassidys Mom

spiritdogs

I like that idea, too.  I'm especially interested in the application of classical conditioning with aggressive behavior, a la Jean Donaldson's suggestion about feeding the dog even while it is growling at a scary looking human to change the dog's basic perceptions about scary humans.

 

When we started using that technique in Cassidy's Difficult Dog class, it seemed really odd to me at first. Everything I'd learned or read up to that point (obviously not all that much!) stressed that behavior rewarded is behavior that will continue. So to "reward" a leash reactive dog with treats seemed completely counterintuitive. We don't WANT them to keep doing this, right? So why are we rewarding it?

But then when it was explained in class, and I read further, it started to make sense. When a dog is reactive because it's stressed or fearful, simply issuing corrections may eventually stop the behavior. But do you want your dog to not go ballistic at the sight of another dog because he fears the consequences, or do you want your dog to not go ballistic at the sight of another dog because he longer feels stress and fear in the presence of other dogs?

I do not consider Kenya to be truly reactive and/or aggressive, she simply has a very fast mind with high-strung, nervous tendencies, but I've also found that doing positive things even in the presence of a trigger that puts her on edge has helped.  For example, today we had our first basic obedience lesson with Coke.  Kenya came along b/c we had permission to crate her during the class (practicing crating for shows).  After the class, this very hyper puppy wanted to meet Coke and Kenya.  Kenya is not dog aggressive or particularly defensive BUT she doesn't like bigger  dogs coming head-on and jumping on her without a few minutes of sniffing first.  I say that she just has a smaller range of what she considers acceptable dog behavior, and when the other dog is outside of her range, she will show fearful signs and try to leave, which is good, but the other dogs keep coming at her.  Instead of just sitting there and waiting to either correct her, remove her, or reward her, I put her in a sit and massaged her cheeks.  She loves this.  I do it whenever we are in a situation that I think *might* be stressful for her.  I've found that it helps ME to relax as well, to be doing something positive with the dog rather than both of us thinking "omg, omg, is she going to curl her lip?  will she get snappy?...".  I don't look at her, I don't speak to her, I simply grab a jowl in each hand and massage her cheeks in circles and she sits there with her tongue lolling, rather than standing up with her tail tucked and her mouth tense.  It seems to communicate, "I'm OK, you're OK, everything is OK."

I use the same attitude regarding warning signs.  I never correct a short growl, lip curl, baring of teeth, or air snap.  I agree that to correct these warnings will not address the cause of them and only teach the dog to stop using them and escalate directly to a bite.  I'd rather observe them and look for a pattern so I can figure out why they are happening.  When we first brought Coke in, Kenya would curl her lips if he came within 2 feet of her.  I watched VERY closely for a few days and noticed that Coke paid attention to this and gave Kenya her space.  The more respectful he was of her, the more she allowed him to be in her space, and now they will basically lay on top of each other.  If Coke had not paid attention to her signals, I would have stepped in and moved HIM back, not corrected Kenya.  I feel like there is a big distinction between normal dog warning signs and outright aggression. 

Observing and participating in more multi-dog interactions has taught me to be more observant of what the dogs are doing and why and less adhered to corrections/rewards based on what I want.  It's different than simply training a behavior like a sit, a retrieve, weaves, etc because for the dogs there IS some underlying emotional baggage attached and it involves dog-dog communication, not just human-dog communication.
 

I see classical conditioning as establishing habits. I think it's a powerful tool when you live with an animal, and it can help you shape all sorts of behaviours over time. I've found my rabbit and hare to both be creatures of strong habit, and if a habit develops that I don't like, I don't have many options to address it but to break that habit and create a new one. IME, those two things have gone hand in hand. If I break a habit, it is unlikely it will stay broken if I don't establish an alternative that addresses the same thing. I think it's much easier to break a habit when you provide an alternative outlet to the behaviour. The lovely thing about my dog is that she's so food driven that it's wildly uncommon for her to be more interested in her behaviour than food. For a while there I was using food to distract her when the dog next door started throwing herself at the fence when we walked past. Penny has a VERY strong association with anything in that yard considering she was attacked a couple of times by the dog that used to live there before the fence went up. She has been conditioned to get anxious about walking that part of the driveway. I had to leave before my counter-conditioning program was done, but even so, sometimes she stops and looks at me expectantly, remembering that once I gave her some tasty treats on the driveway.

I think dogs and other animals remember by association. That gives us a window of opportunity where we can help things become set in their minds by influencing their emotional state when something happens. When I first got Kit, my hare, he didn't find my soft tones and quiet, slow words very soothing. But after living with me for a few months, he learnt that those sounds mean I'm going to back off and any pressure on him will be removed. When our relationship was at its height (before I had to leave him in the care of my mother for 18 months), I could use those sounds to calm him if he was still thinking. I could convince him there was no need to panic and flip out because he was so used to those sounds being accompanied by his feelings of relief as pressure was removed. So I think classical conditioning could be equally useful with dogs that are frightened. Although, I'm yet to find anything that takes Penny's mind off a thunderstorm or fireworks. I can minimise that one, but I don't think I can change it.

science may be able to determine how the painting that i posted i my signature was done but can science predict what anyone thinks of when they view it?

lostcoyote

but can science predict what anyone thinks of when they view it?

Behavioral science would have some general guidelines as far as color and patterns go. How should people interpret the Navajo quote that accompanies it? And might the perception of it's inclusion and meaning change with each person that views it? And I'm having a blonde moment because I'm not sure how discussion of the painting relates to the discussion of behavioral sciences. I get it that yes, art is subjective, that we each view the world through our own eyes. But it's not yet creating a common language through which we can exchange views on behavioral sciences.

For the actual topic, I think what seems to be different approaches in behavioral science are facets of the same thing. But other things outside of that discipline may also have importance. For example, a dog that seems to not obey may not be disregarding the human. They might be partially deaf. Or not register the pitch of your voice. Trying a command at a different pitch and/or volume may achieve what is necessary. So, then, the dog interacts with the world or human through the physical limitations. We already know that they don't have the same visual acuity that we do.

 How much of what we can accomplish with a dog should take into account the abilities of the dog to perceive?

I'm not sure how discussion of the painting relates to the discussion of behavioral sciences.

what do you think is the basis in which OP relies upon? what is one of the primary forces driving a dogs behavior inside of the dogs mind? it has everything to do with how people respond to art, or someone in a relationship..... just like the book you referenced "beyond physics"... just what is "beyond physics" anyways? the answers are not found in hyperbolic trig functions.

I use the same attitude regarding warning signs. I never correct a short growl, lip curl, baring of teeth, or air snap. I agree that to correct these warnings will not address the cause of them and only teach the dog to stop using them and escalate directly to a bite. I'd rather observe them and look for a pattern so I can figure out why they are happening. When we first brought Coke in, Kenya would curl her lips if he came within 2 feet of her. I watched VERY closely for a few days and noticed that Coke paid attention to this and gave Kenya her space. The more respectful he was of her, the more she allowed him to be in her space, and now they will basically lay on top of each other. If Coke had not paid attention to her signals, I would have stepped in and moved HIM back, not corrected Kenya. I feel like there is a big distinction between normal dog warning signs and outright aggression.

lets say that someone reading this decided to use your methods of observation - but it was a strange human or another unfamiliar dog that got into their dogs personal space. i've witnessed how a short growl, lip curl, or baring of teeth can turn into a lunge and/or bite within quarter of a second or less - which is not even enough time for a human to react. within that short time frame, a dog can cause serious injury to another dog or human being... and if it is as simple as "snapping at the air", if that snap lands on flesh, a simple puncture wound can occur (see pic below)..... and if it is another human that get's injured, that puncture wound can land their dog into being put to sleep or worse, a lawsuit.

i believe it's best to prevent the escalation of agression in the first place - unless one's dogs live in isolation.... because it can happen just that fast. i think it's really good that dogs can be trained to allow others into their physical space... but yeah, when it comes to food, a short growl is probably OK as that's the way dogs communicate..... but if the dog's growl turns into that snap... that's where i draw the line personally..... probably a touchy subject here tho.... but yeah, as the saying goes - "sheet happens."

Observing and participating in more multi-dog interactions has taught me to be more observant of what the dogs are doing and why and less adhered to corrections/rewards based on what I want.

In my experience, having spent roughly 365 almost consecutive days at a dog park, dog culture varies fairly widely.  Misunderstandings happen. What one dog considers a crystal clear communication of "Get out of my face NOW" another dog sees as a yellow light, or even nothing to worry about at all. 

So there is a big, big, big difference between dogs that know each other, and dogs that don't. We can't assume two strange dogs meeting share the same vocabulary.  I watch dog vocab, because Sasha is a great communicator - when the other dog has a shared language base. Her language is a pretty standard dog language, but it isn't universal. She also takes insults really badly, which is her baggage, and another reason I pay mucho attention to the way dogs communicate with each other.

I don't correct growls, 99% of the time. (Exceptions exist) I do correct staring contests, gang signs, and other punk behaviors that escalate tension. My rule of thumb is: you can tell someone to back off, but you can't tell him to F off.  

lostcoyote

but rather, to spur whether or not science can pin down the QUALITY of what a feeling is or the QUALITY of what a thought is.

So, your question is how can science determine that a dog finds a treat "rewarding" or of "high value" if it can't predict the subjective value of a piece of art to future viewers? Or is the dog responding to +R out a need for survival by using a primary resource, such as food? IMHO, it would have been clearer to ask that question, initially.

So, I would ask the question in return. How is a dog's need fullfilled by punishment ("corrections";) when there's not that much analog in the dog's world? Or, another version of a previous question, how do you use social pressure on a dog such as an Akita, that does not seek the same social bond as other breeds? And most dogs received their punishment from momma as far as playing too hard and bite inhibition, not where they walk in the group. In a related group of canids, there is distinct evidence that the alpha male does not walk in front. In fact, while being released back into the wild, the alpha hung back until he could do a "roll call", ensuring all members of the group were accounted for. So, in what analog is it necessary to promote leadership by walking in front and assume that the dogs understand that as leadership?

Or maybe, how is it disproven that OC terms do not describe the behavioral sequence just because it can't go beyond the ability to note that a reward has a high value and that living organisms work to achieve that reward?

Are you not receiving a reward from engaging in this debate and are willing to work at replying in order to continue said reward? It doesn't matter what I think of what the reward means to you, it is only important that you find it rewarding. My dog finds steak rewarding, carrots not rewarding. I could spend a lot of time wondering why that is, but training is advanced by simply accepting that the steak is the higher reward.

As for most dogs not actually trying to achieve dominance but seeking a balance, as suggested in the non-linear theory, there is some proof of that, as far as edited video can show. That favorite tv guy couldn't have a 40-dog pack unless the dogs do, in fact, value having balance over superiority, per the non-linear theory and some other theories, such as those supported by Ian Dunbar. He feels that much dog aggression is mislabelled.

lostcoyote

just like the book you referenced "beyond physics"... just what is "beyond physics" anyways? the answers are not found in hyperbolic trig functions.

The title of the book is "The End of Physics" and my mention of it is to show that a number of theories abound, all to describe one aspect or another but each one not quite complete. The end of physics is a euphemism for the Unified Field Theory. If one were to find a unified theory that explained everything in physics, then the active quest of theoretical physics would have reached its conclusion. And there are some fantastical theories that explain one phenomenon at one scale really well but beyond that one scope, fall apart rapidly.

lostcoyote

the answers are not found in hyperbolic trig functions

I'm not implying that they are and most specifically, the Lorentz Group treatment was simply another way to describe Einstein's Space-Time continuum thingy and was popular in the 90's. Einstein actually disproved his own theory in the EPR Event in Quantum Mechanics. But we'll go too far afield if we keep on that track. Also, most theoretical physics are unproven. As opposed to the traditional scientific method of observation, theory, test the theory. In some theoretical physics, it's theory, observe, alter data to fit theory, even if it requires applying Lorentz's transforms in hyperbolic trig functions and then mash those into a topological space.

The value I find in Learning Theory and OC is that it does not assume a lot. Through evidence, it seems apparent that an organism finds one thing rewarding and another thing either not as rewarding, not rewarding at all, or punishing. And that rewarding things get repeated. Non-rewarding things tend to not get repeated. By using this limited scope and observable principles, a dog can fit in well in human society.

Dog_ma

I do correct staring contests, gang signs, and other punk behaviors that escalate tension. My rule of thumb is: you can tell someone to back off, but you can't tell him to F off.  

Absolutely. We go to off leash parks too, but only weekly, not every day. I watch my dogs and their interactions with other dogs carefully. If mine are acting in a way that is, or MIGHT, make another dog uncomfortable I step in and redirect. If another dog is acting in a way that is or might make MY dogs uncomfortable, I step in and redirect. But around the house, with each other, there's pretty much nothing that's too much - they each give as good as they get, and seem to enjoy it mightily. 

So, your question is how can science determine that a dog finds a treat "rewarding" or of "high value" if it can't predict the subjective value of a piece of art to future viewers?

i was asking if science can pin down the QUALITY of what a (dog) is feeling or thinking.... and i was speaking here in generalities. i only used the picture as an example of what science might be able to do versus what it probably can not do.

i was not asking about treats, rewards, and punishments... but since you brought that up, what do you think these things hinge upon within the context of a dogs mind?

is there a unification in which S,W,+,- operate from? if so, is it related to the artwork in the sense of the way the utilization of offering treats or punishments utilize a dogs emotions to the trainers advantage? if there is a unification, then that, i would think, would add value to the topic at hand - moving beyond OC theory.

and i was being metaphorical with my comment "the answers are not found in hyperbolic trig functions" as i'm not sure how that entered into the topic in the first place.

Great post, Dogma. My dog is phenomenal with other dogs, at least to me, but I think she's just an old hand and knows what she's doing. She knows what she's doing because I never took the opportunity to learn away from her by intervening too early. Yes, she's been to vet a couple of times because an unsocialised dog attacked her. Yes, I would have stopped that if I'd seen it coming. And that's the crux of it. When things like that happen, there's very little you can do and I've seen it happen with no provocation. I wouldn't stop everything on the offchance it was going to turn nasty. Pyry is 5 and Jill is 3 and neither have ever been seriously hurt in a dog fight. Our previous dogs have never been hurt, either. Penny just happened to live with a psychopath for a couple of years. We all learnt from that, though. I think it's uncommon for dogs to be hurt by others. And my dog is a snapper, but she's an air-snapper. She snaps at dogs I would rather she stayed away from, but she's never provoked an attack with warning signals. Never. Not even on the dog's own territory. If there's a dog that might attack at warning signals, she bloody well stays away from it. She didn't need to be bitten to learn that. She just needed me to trust her. She almost never makes contact when she snaps. I let her snap when she growled and curled her lip a fair bit first. I scold if she goes straight to a snap, and send her to bed. I don't expect her to put up with everything, but I like it when she gives plenty of warning that she doesn't intend to put up with something.

I don't think it matters whether science can tell the quality of an emotion. When it comes down to it, a computer model with enough information plugged into it would most likely be able to predict behaviour pretty well, and we can trace that back to some idea of the emotional state that produces that behaviour, but we can't get inside the head of another and experience their feelings for ourselves. We can't even quantify our own human feelings, because each person experiences it in a different emotional landscape and got there through different associations. That's what makes life and interacting with others so fascinating. It wouldn't mean anything to us if science could describe or explain the quality of an emotion, and I don't think science can do a thing with anything that isn't quantifiable. Emotions aren't very easily quantified, even for us with our complex language.

lostcoyote

was asking if science can pin down the QUALITY of what a (dog) is feeling or thinking.... and i was speaking here in generalities. i only used the picture as an example of what science might be able to do versus what it probably can not do

Believe it or not, I'm not trying to be a twerp but what are your conditions for quality? Every time you respond, I get a different idea of what you are trying to say, which means we aren't communicating. Is it science's job to determine quality or does each individual do that? I like strawberry ice cream more than chocolate ice cream. Which one has the lower quality or are they equal? As far as science is concerned, at one level, they are equal and "quality" would only relate to parameters such as a recipe and even then, the "quality" is arbitrary.

Or are you saying that going much beyond or away from learning theory and the precepts of OC maybe futile as the language and ability of the sciences can't go to the place you imagine or suppose is beyond them? I may be wrong but I think the original reason for this thread was to also examine other models, some of which do have some scientific structure and see how they relate to or are part of learning theory.

Perhaps the title of the thread supposes too much. More likely, it can be easily misinterpreted. I don't see the topic as a rejection of science or as purely a catalog of perceived limitations.

lostcoyote

i was not asking about treats, rewards, and punishments... but since you brought that up, what do you think these things hinge upon within the context of a dogs mind?

is there a unification in which S,W,+,- operate from? if so, is it related to the artwork in the sense of the way the utilization of offering treats or punishments utilize a dogs emotions to the trainers advantage? if there is a unification, then that, i would think, would add value to the topic at hand - moving beyond OC theory.

I believe that rewards and punishments are subtextual to a state of mind. But an individual dog can define quality for itself by means of its response. And there may be a cycle of cognition that happens in training. A dog simply likes drippy chicken. Then it learns that it can get that by doing this or being this and gets its reward. Done enough times, it no longer consciously makes the decision as the complex behavior chain now becomes classical. Maybe nearly as classical as the first response to drippy chicken. The initial response might be classed as part of a behavior system, to always value food. As a scavenger, one never turns down food, if it smells like food. But that would be dog-specific.

I am not convinced that other viewpoints such as behavioral systems, non-linear dog theory, etc., are unrelated or "beyond" the mechanics of OC. And who's to say that emotions are not a contextual thing related to one's ability to survive. Have you ever been homeless and without food? I've been homeless a few times and have gone as long as 5 days without food. I'm "happier" (more apt to survive) with a home and some food. I will work doing what I can and sometimes putting up with things you can only imagine for that piece of paper that says "Pay to the order of". From a biological standpoint, money, or whatever it is that secures home and food, conditions or trains.

Now, if we are to suppose that there is a spiritual aspect to dogs that is inadequately described by science, that's fine, too. IMO, we are now dealing with another soul, and soul is, also imo, inherently sovereign. Or soul may be an evolutionary off-shoot of an organism that has reached a certain level of sophistication in it's central nervous system. In any case, it is another entity. In which case, I don't see what's wrong with engaging another entity to do something in a way that is profitable to that entity.

I fear I may not have answered your questions fully but then, again, I am a blonde.

are you able to define what a soul is now? up on page 1, you said you could not define it.... but it's being used so i think a definition may be useful if we're going to use it within the context inferences made. when we say something like "that person really puts his soul into his music" what does that mean really? that's sorta the way i define it... not esoteric at all... but for me, it's how a being expresses his true self and in this sense of defining it this way, i'd say that dogs do have it.... maybe you just call it personality - but it's intrinsically linked to the dogs spirit (quality and quantity of mental energy)

i also believe that the whole trick in training whether it be +S, +W, or whatever, is how that spirit is manipulated by man to bring the dog into a sense of well being and balance. clickers, treats, touch, to me, are just one of many tools one may choose from to bring an unbalanced dog into the state of balance... but these tools are not the driving force behind the movement towards balance. feeling is.

just my opinion. 

Okay, this thread has evolved into many different conversations, all of them equally interesting but I think it's a bit much for one thread to handle. By reading through, it seems that classical conditioning is something that a lot of people are interested in learning/reading more about. I say let's go for it, it's a very fascinating world of science and one that I have intimate relations with, living with two special needs dogs.

However, due to my schedule at the moment with finals at university, my activity level is quite sporadic until exams are over. Would anybody be interested in taking on responsibility for leading a discussion and starting a new topic for it? It might be easiest to begin explaining what it is, and defining some of the terms (US/UR/CS/CR, as an example), before moving on into more complex issues such as what things become classically conditioned, and how we use it in changing behaviour and emotions.