Aidan3

The argument that I'm putting forward is based on the fact that we don't have all the answers and neither does Cesar Millan. You can either blindly accept what you see on a television show, or you can think about it and discuss it like an adult. It's called rational debate, but some people can't seem to grasp that, preferring to let their emotions and beliefs rule their thoughts. Hence, if you do want a rational debate on Cesar's methods, you are labelled as a "Cesar basher". Political correctness gone mad. All sorts of things "work". Chloroform works. Starvation works. Why wouldn't a clicker trainer starve a dog? When does deprivation become abuse? I'd like to thank Nekbhet and sandgrubber (and a few others) for being able to have an adult discussion on this topic.

If they did think like that (and I know you know they don't), they might think "It doesn't matter where I turn, it's dangerous everywhere!" Is it really awesome staying with a handler who is liable to kick you in the stomach then hoist you up on a slip collar when you're already frightened out of your mind? It's a bit hard to believe. Remember here that I am debating the methods seen on Cesar's show, not what others might do. If he's not exhibiting the dangerous behaviour, it's a moot point. Will he exhibit it in the future? We don't know. How do we find out? Two ways: 1. elicit the response and punish it. Then we can test again, and punish again if it crops up. Our level of confidence decreases every time we successfully elicit the response, and increases every time we fail to elicit the response. 2. push just shy of the response and reinforce the alternative response. Then we can push a little more the next time, and little more. Our level of confidence decreases if we elicit the unwanted response, and increases every time we reinforce the alternative response. So is there a difference in our confidence? I would argue that there isn't. Is there any experimental data that suggests there is a difference? Some, but it goes both ways and leads to new arguments. Well I think you know I'm not arguing against corrections, sometimes that is the only practical way and I'm certainly not debating your methods. I'm debating Cesar's methods. I also have reactive dogs who have never been punished "self-correct", in fact it's what we're always looking for, for them to make the choice and learn that it's a good choice to make.

It's the lower abdominal area. Try putting some pressure on the fold between your dog's hind leg and abdomen, then put the same pressure on the fold between your dog's front leg and abdomen and see if you get the same response. Normally (unless your dog is habituated to this sort of pressure), you won't. The abdomen is protected by virtue of being underneath the dog, it is otherwise poorly protected having no bones and little muscle protecting it from underneath. Damage comes from pressure minus whatever resistance is offered. I really doubt Cesar is physically damaging these dogs and that is not the point. Pain is poorly understood, despite being one of the largest industries (pain management) in medicine and psychology. We do know that areas that require greater sensitivity (e.g areas that house important organs but are not protected by bone) produce more sensation of pain. Pain perception increases or decreases depending on arousal, the reason why I feel nothing until after the game of rugby has finished (unless I get an elbow to the kidneys or a palm to the nose). Good psychological research into pain attempts to "triangulate" subjective experiences with objective data. In the dog's case we really can't ask them how they feel on any sort of scale or comparison, but we can observe their responses. Attempting to escape and responding with aggression (both clearly and repeatedly shown in the video) are good datum points to begin to form a picture. Seeing this over and over again is a form of test-retest reliability of this measure. If kicking in the abdomen didn't elicit a stronger response than anywhere else, why doesn't Cesar kick anywhere else (the tail, for e.g)? Is it purely superstitious behaviour on Cesar's part? I doubt it. Slapping a dog about on the rump or shoulders can be reinforcing. We can prove this by increasing the rate of responding of an operant that precedes this consequence. I've never seen anyone reinforce a dog by slapping him on the stomach or throat, both areas which could accommodate the full area of the hand thus distributing the force in the same way. You could try it, but not on my dog Perhaps more important is the psychological dimension, the effect it has on current and future behaviour. Unfortunately understanding this requires some background and it's pretty technical. One of the researchers in the field of "affective neuroscience" (the science of emotion) attempts to make this field more accessible by referring to circuits in the brain such as "PANIC" and "FEAR" and "RAGE" that correspond with physical circuits in the brain that we can measure in a variety of ways. The aforementioned are not good states for learning, and until I see some objective data that shows a benefit in deliberately eliciting high levels of activation of these areas of the brain, I will remain skeptical. Even just some fair tests on the show would sow the seed of doubt in the data for me, or some follow-up demonstrating the changes. We believe all sorts of things, remember how much everyone loved John Edwards the psychic? There is something I would like to add, not pertinent to my response above or directed at anyone in particular (certainly not sandgrubber). I'm not "Cesar bashing". I'm debating his more controversial methods on rational grounds. I think until people start to get an understanding of those "rational grounds" it looks like I'm attacking the person and not his argument but that is not the case. A handful of people here get that already, but I think it bears mentioning. I think Cesar does a lot of good things and has a lot of good advice and I'm of the understanding that he has saved a lot of dogs from being put down. So does your local obedience club. His special talent is getting it on TV, it's just a shame that it's also a platform for some very dangerous advice.

Well if you can get a photo of it, you can click it

Normally no - Elbie does it when he's excited and waiting for the ball. When I've praised him madly he looks at me like I am bonkers - has no idea why I am praising him. Hoover, too Be ready with the clicker, see if you click as soon as it happens. Then toss the ball.

Would they do it often enough to capture it? Then push a little for it before throwing the ball the next time?

One of those Danes was pretty scary. I would have to know a lot more about the back stories to judge any of the other dogs, but they didn't look any different to the reactive and aggressive dogs I imagine we all see. Are there any stand-outs you think I should take a closer look at? I'm willing to reconsider.

Indeed! As you know, I don't accept this premise, but even if I did - why is it necessary to use his foot? It's not hard to put a dog over threshold, especially one of these "red zone" dogs that only Cesar works with apparently. What benefit does an entirely new stimulus offer? If we're able to put the dog into more and more challenging situations then we can demonstrate some level of reliability by the dog not responding with aggression. Sure, we can never say the dog will never respond with aggression ever again, but we establish a probability. If we elicit the response, then correct it, what extra measure of reliability does that give us? What principle are we applying that gives us confidence that the response is now gone? How is that probability any different to the probability I describe above?

The special, unique dog seems to be ubiquitous

He attempted to fulfill the criteria with no less than 33 Standard Poodles without success. He got lucky with the first cross-bred litter, and quickly realised that the probability of obtaining suitable stock this way was also very low. I wonder what would have happened if he'd used slightly older dogs. They mature later than Labs but they live longer too. The main problem was not their trainability, interestingly enough. The client's husband was allergic to nearly all of them. Interestingly, he was not allergic to three of the cross-bred litter.

He attempted to fulfill the criteria with no less than 33 Standard Poodles without success. He got lucky with the first cross-bred litter, and quickly realised that the probability of obtaining suitable stock this way was also very low.

I like them to be able to explain the difference between operant and classical conditioning, using the Rescorla-Wagner model and Herrnstein's Matching Law to illustrate the difference. If they can do that, I let them have their car keys back. Failing that, if they can show me they know when to click and treat, and how to satisfy themselves that the response is actually increasing and how to back up if it isn't, I'm happy. Good question, btw, worth thinking about. Good idea I would like a new car, cept ma math is worst than ma ingleesh. It's not a new car, they get the keys to their old car back! With just enough petrol to get home.

That's a really, really good practical example of positive reinforcement and extinction procedures (or "reward" and "ignore it" procedures ) I'll be honest with you, I've never really "got" why a lot of writers and instructors teach classical conditioning in relation to the clicker. The only thing I'm worried about when using a clicker (provided the dog isn't sound sensitive) is whether or not I get more of the response I clicked.

Well said, sandgrubber

To you, yes. Not to everyone though. Your idea of training is not a specialised area of training, but it's not necessary to know all the details to be able to train a dog. For those of us who like to get at the "truth" of the subject (so far as we can), I think it's better to watch and then translate for your own benefit. To me, that's what the theory is all about - it puts stuff into a framework that we can use, and allows us to get rid of the stuff that never really worked and extend the stuff that does really work. I'd say some basic scientific competence was possibly more important than learning theory in this regard, but the more time I spend around academics the more I realise that people don't always generalise stuff like that very well.

I like them to be able to explain the difference between operant and classical conditioning, using the Rescorla-Wagner model and Herrnstein's Matching Law to illustrate the difference. If they can do that, I let them have their car keys back. Failing that, if they can show me they know when to click and treat, and how to satisfy themselves that the response is actually increasing and how to back up if it isn't, I'm happy. Good question, btw, worth thinking about.

Yes, Best Friends. I was referring to the television show because that is what most people would be familiar with.

The argument was that dogs specifically don't have "episodic memory" which is a term psychologists use to describe memories of specific events (or "episodes" in their life). I'm referring to episodic memory as well. To remember where a ball was last kept, a dog would need to remember where he played with it last, and thus remember the last time he played with it. Possibly, but other forms of memory could also explain this so the hypothesis presented by William Roberts could be valid without episodic memory or the ability to "mentally time travel". Spatial memory, for e.g, could just as easily explain how dogs know where objects are without them needing to remember that they placed a toy (or found prey) at some point in time, only that they placed it at some point in space. It's a very abstract concept and it's not how most of us experience our own thought, so I hope what I said made sense

The argument was that dogs specifically don't have "episodic memory" which is a term psychologists use to describe memories of specific events (or "episodes" in their life).

I've never seen CM do anything similar, although he is very good at getting people to remain calm and demonstrate that they are calm, almost certainly better at that than me. I like my clients to laugh, so we need them to feel safe otherwise it's just nervous laughter As for what I do, it's the absolute basics of learning theory but the problem here is that interpretations vary somewhat. First I establish operations, then I break the behaviour down into tiny little pieces and start with the first piece. When that is reliable, I move to the second piece. The rate of reinforcement is very high because we're talking about tiny durations of behaviour at first, but this is OK because it builds two things - a giant reinforcement history, and associations through classical conditioning. A good analogy is lifting weights. I don't know if it's something you've ever been involved with, but when most of us start we're usually pretty weak. So we have to start with a light bar, otherwise we can't do it. We lift that light bar over and over, then we get a little bit stronger. Then we can lift a little bit more, so that's what we do. Over and over. If we want to get anywhere with it, we don't say "OK, I think this weight x reps will keep making me stronger, I'll just stick with this" (actually, most people DO do that... ) nor do we say "OK, I'm going to put 4 more plates on the bar and just lift it" when we know we can't (actually a lot of people do this too, then they "cheat" by limiting range of movement, or enlisting a spotter to lift half of it for them while yelling "it's all you bro!"...) I would challenge that idea. How do we define what a "red zone" dog is? Think about that seriously. Cesar tells us the dog is "red zone", so do we just believe that because he is the expert? That is circular reasoning and does not include substantiated evidence. Also, do you think that the dogs Cesar works with are any different to the dog any of the rest of us work with? Why would that be? Is he breeding them specifically for his show? Realise that we can make almost any reactive dog look like a "red zone" dog if we put them into situations that they cannot handle.

It would be interesting to see these "red zone" dogs that CM works with given the same unbiased evaluation that the professionals dealing with the Vick dogs used. I have no idea what criteria is used to define a "red zone" dog, other than Cesar's opinion.

You should watch the Dogtown special on the Vick dogs then, I think you'll get a surprise. The worst dog I ever had foamed at the mouth when a plastic bag blew across the other end of a football field sized area. Her owner had lived in a few states and the dog was seven years old, so they'd been to a few trainers. His dog couldn't be walked anywhere where another dog was likely to be seen. Medication hadn't helped, and the dog was born that way. We had this dog taking food around other dogs within a couple of weeks, the owner was astounded. Then we very quickly got her down to sits, drops, looking at other dogs, and walking on a loose leash within 5m of another dog within another couple of weeks. She was never brilliant, but at least she could be walked without having a fit and taking days to come down.

Maybe you should watch Dogtown, also on National Geographic. They took the worst of the Michael Vick fighting pitbulls and rehabilitated all but one (I think) using positive reinforcement. Worth a look, although it comes with a Kleenex warning for the softies out there. Yes, but the Vick PitBulls were not reported to be highly aggressive, bar the one female sho was PTS after evaluation. The remainder of those dogs varied between shut down pancake dogs and highly aroused dogs ( and who bluddy knows how much of that was kennel stress! ). They were champion fighting dogs. That means they were successful in the fighting ring, fighting for their life with other dogs. Vick didn't remove her teeth because they were rotten, he removed them so that he could put a stud over her without her killing them. I'm not sure what criteria we need to define a "highly aggressive" dog, but killing other champion fighting dogs in fights would seem a reasonable definition! They were also tortured, abused and neglected by people. Thrown in the ring with their mouths taped shut. Hooked up to car batteries when they lost. By humans. Excuse me if they weren't scared shitless of everyone and everything. Can we not accept that Dogtown did an amazing job rehabilitating seriously aggressive, selectively bred for aggression, ex-fighting dogs, rescued from some of the most appalling conditions imaginable - because they used positive methods and therefore render the argument that "some dogs need more than rewards" invalid?

Well I rest my case then How many episodes have you seen? Maybe you should watch Dogtown, also on National Geographic. They took the worst of the Michael Vick fighting pitbulls and rehabilitated all but one (I think) using positive reinforcement. Worth a look, although it comes with a Kleenex warning for the softies out there.

And in some cases how do we teach the dog that if it does react in that manner there is a consequence to it, hence it should stop reacting like that? You cannot simply work through totally avoiding the dog reacting forever. Sometimes we have to push the envelope and get the dog to react then teach it when it does. Sorry, I don't agree that we need to get the dog to react so that we can use a punishing consequence to teach it not to react. We can teach a dog what to do in those situations without a punisher. I'm not saying that it isn't useful to use punishers, but I disagree that it's a necessary condition. There is nothing necessary in what Cesar does with Shadow, the Malamute, for e.g. He deliberately loads the dog up, kicks him, then when the dog predictably goes for him he chokes him, doesn't do that properly either, and gets bitten. That isn't necessary, and my understanding is that Shadow was not successfully rehabilitated. Everything I do is about pushing the envelope and teaching the dog what to do when that happens. If you aren't progressing steadily, you are doing something wrong in my opinion. No, but a dog trainer is not a dog either. And we know that treats and prey toys work, if used competently. I don't think we're all frightened of touching dogs, although some are. There are lots of different ways you can touch a dog, and a lot of different contexts. I don't think Cesar gets it right all the time, and I think he crosses an unacceptable boundary some of those times. Healthy, stable dogs do not do a lot of touching as discipline (although lots for affection and play). They have pretty good body language, not always great though, often poor submission. Really dominant dogs are masters of avoiding physical violence and aggression. Large packs of free-ranging domestic dogs do fairly poorly, they end up with an unstable hierarchy and they do damage to each other. Cesar is the master at controlling a large pack of unrelated dogs, a very unnatural situation even for domestic dogs.