Cesar Milan

[Lynne B]

I disagree with many of Cesar Milan's methods, but I know he has quite a following...here's a website that Dr. Sophia Yin (a vet/trainer that I have much respect for) linked on her facebook page:
http://beyondcesarmillan.weebly.com/index.html

[Lynne B]

I have my own theory, it's called the 'slot machine theory' I live in a state where there are alot of casinos and so I have been able to observe lots of people playing slot machines. It's quite interesting and it closely parallels how dogs learn.

A person puts a coin into a slot machine and presses a button with the expectation that they will receive a 'reward' (money). A new player will even be rewarded by management by being given 'free play' (fake money to ensure they will get a reward 100% of the time when they first start playing). One 'hooked' they will continue putting money in a machine without winning but with the expectation each time that they might win. If, for some reason, they no longer sense that they have a chance at winning they will go to another machine, or quit gambling. At least in Northern Nevada, there are people who will drive through a snowy, icy pass in the winter just to get here and put their money in these machines- they are that strongly motivated.

A dog whose behavior is shaped with rewards acts exactly the same way. If you give them a reward (food treat, toy, belly scratch) when they perform a certain act that you want them to repeat, and you give them the reward everytime they repeat the behavior, they will associate the reward with the behavior and continue repeating it in anticipation of the reward. When you have suckered the dog into thinking that the reward always follows the behavior, you start withholding it once in awhile, and the dog will now repeat the behavior just because there is a chance it will get the reward. In time (if you do a good job at fading the reward, it will perform the behavior without even seeking a reward.

the nice thing is this method works, in fact it works 100% of the time. It works for training for obedience, learning tricks and desensitizing fearful dogs. If the dog misbehaves the slot machine (you) malfunctions for a few minutes and the lights and bells turn off..you just ignore the dog completely for a very short time and it will respond by trying to get the slot machine (you) to start working again. *ever see anyone in a casino pound a machine that malfunctions because they want to be able to put more money in it? It happens all the time.

Why do people claim it fails and they need to coerce their dog or punish him? For two reasons:

1. Human EGO: People start to ascribe all sorts of human emotion and feelings to their dogs. I hear people say they say they want the dog to 'respect' them, or they want the dog to know who is "boss", or they want the dog to perform without a reward 'just because the dog loves them' Well, to that I say horse puckey.. My dogs listen to me because I am the human slot machine. I keep treats in my pocket constantly, I train my dogs throughout the day for very short periods of time. I reward good behavior spontaneously. The love that I have for my dogs is based upon the pure joy I feel when I am around them- not on the fact that they respect me, fear me, or see me as pack leader (concepts I don't think dogs really can fathom anyway)

2. If the dog will not perform a behavior for a reward, the concept didn't fail, it's just that your reward sucks. Dog's seek rewards 100% of the time, all of their behavior is based upon it. They evaluate two alternatives by deciding which yields the best reward. Your dog might 'sit' for a piece of kibble, or for no reward at all, but if you want him to stop straining at the lead when he sees another dog, a piece of kibble probably won't work because the 'potential reward' of being able to get to the other dog is more valuable to him than the kibble, but, if you replace the kibble with a piece of fish, liver, or chicken- the dog will stop tugging at the lead because the 'jackpot' you offer is more enticing than the one it perceives it would receive if it could get to the other dog.

Cesar Milan's failure (in my opinion) is that when he tires a dog to the point of exhaustion, and then uses pressure on it's collar so that it is oxygen deprived, or jabs it in the neck. Of course the dog behaves, it will do whatever he wants- but did the dog learn anything at all from that? Is the dog now 'cured', or will the same negative behavior manifest itself again because he simply suppressed it, he did not replace a bad behavior with a good one.

I think Cesar Milan would be a better dog trainer if he went to a casino and learned about slot machines!

[Adina P]

I think the slot machine analogy falls apart with average pet owners...or perhaps average pet owners don't use it to its full capacity. But the difficulty I have (and I do know purely reward folks who have done great things with their dogs) is that there are TONS of environmental cues bombarding dogs all the time. Rewards are not always under our control and dogs are not seeking attention as their number one goal. The reality of training a dog to a high level of reliability is MUCH harder than it sounds...otherwise so many dog owners wouldn't say 'He's great as long as I have a treat...but forgets in XYZ situations'. More people would have obedient dogs who didn't pounce on guests, counter surf, pull on leash, etc. And they'd have dogs that obeyed regardless of whether a hot dog is in their pocket. But that's NOT the case...more dogs are unruly than trained. And to completely remove positive punishment as a choice (when it does work and has its place) I think diminishes opportunities for a speedier, more effective success.

I'm all for being positive, for teaching carefully and praising/rewarding the dog. I do NOT think my dogs obey me because of love and only love. But because they are making a choice between consequences based on a lot of practice and understanding of which choices yield which consequences. It's not 'fear' but self-interest. I stop at red lights not out of fear..Im not stressed or nervous when I stop. But I do know that going through a red light could at best get me a ticket (created consequence) and at worst end up with me in a crash (natural consequence). I'm not a fan of either...so via practice and education I've developed the habit of obeying traffic law and stopping at red lights...now there is a reward of safety that created and it started with knowing the unpleasant consequences (albeit merely in my head).

[Lynne B]

I guess that difference of opinion is why this is a valid discussion. If anyone who uses rewards as a training tool ends up having to have a reward in their pocket forever has failed to understand the casino theory. All you ever have to do is associate the reward with the behavior until the dog learns it, then you start withholding it.

I too stop at red lights out of fear, fear of getting a ticket or fear of getting in a wreck- but can our dogs really put together abstract pieces of information into some cogent model that allows them to make those decisions?

The biggest problem with punishment is that it is usually delayed and thus counterproductive. People do all their poking and yelling and jerking a second or two later than they should have and the dog isn't even doing the behavior at that time that it is being punished for, thus you just taught the dog that you might get disciplined no matter what you are doing...They simply cannot equate either reward or punishment with some behavior that is not currently occurring. If your training is positive, and reward based and is delayed, you at least have not taught your dog that humans are unpredictably cruel.

[Adina P]

Do you honestly feel FEAR while you stop at red lights? I find that hard to believe...I just stop based on a learned response. I DO feel fear on those rare moments when I go through the stop light (havent we all done that? Or caught ourselves speeding?) because then I HAVE done something to warrant a possible punishment. But I feel calm and safe BY stopping (avoiding the wrong behavior). I agree dogs don't think that abstractly, but then when I train my dogs I don't train with ideas...I train concretely. So it is just an analogy to relate a similar idea in our lives.

I also agree that most people don't have good timing...I didn't to start. But not being good at something NOW is no excuse to not improve. At the same time...based on my experience and that of my husband's and trainer friends...good timing with a correction is NOT all that illusive (ellusive?) and can be accomplished. I saw my dog progress REGULARLY when I trained him (Rosco) and there WAS evidence of learning at every step. If the chances of him being utterly confused were that great because I wasn't an experienced trainer...progress would have been thwarted for long periods and we would have gotten nowhere. But that is NOT what happened. Additionally, I believe the dog MUST receive reward/praise and that too is critical. It was the moments Rosco got it RIGHT in which he learned lots. But the moments he got it wrong were critical too...so he could form the contrast in his mind and be clear. There's A HUGE diff to me between catching a dog in a naughty act and applying a correction or punishment...AND the correction applied after extensive TEACHING of what is expected so the dog knows the dif between the correct behavior and the incorrect behavior. One is about trying to extinguish a behavior (counter surfing, digging, etc) and the other is about clarifying a behavior that has been taught and showing there is a consequence for a choice made based on the command given. Thanks for a good conversation :)

[Lynne B]

When you said this: "But I do know that going through a red light could at best get me a ticket (created consequence) and at worst end up with me in a crash (natural consequence)" I simply used the word 'fear' to explain it- maybe not the best word, and there are different levels of fear, I don't shudder and shake when I think about the consequences of going through a red light, but I do 'fear' the consequences- sorry if it was a bad choice of words, but I think you probably know what I meant

[Adina P]

Thank you for clarifying. I also agree there are probably degrees of fear. And while realizing my choice of words may have better terms in psychology than what I'm using...from my perspective there is active fear where one is actually experiencing the stress or FEELING of fear (at whatever level) and there is just knowledge of consequences where there is NOT really any stress or feeling of fear being actively felt.

My guess is that when we talk about not wanting a dog to be afraid or do things out of fear...it's because we don't want to subject a dog to the feeling/experience/stress of fear because it's not a NICE feeling or one we'd want a dog to feel on a regular basis as its motivation for doing what we want it to do (hope that sentence made sense). And I would agree with that. If every time I told my dog to sit or come or heel...it felt FEAR...I would NOT be happy with that. Early in teaching/training there might be some level of nervousness of trying to do it right..but I think that's the case when anyone is learning something new...a little stage fright. A little "I hope I do this right" kind of stress---which I find to be normal and healthy and goes away as the person or dog gets practice in the behavior or skill. They learn..oh I CAN do this right and then that initial normal, healthy stress associated with learning a new skill goes away. Learning to drive was stressful for me but not in an overwhelming way or to an extent that inhibited learning. And it made me focus more attentively and eventually with practice it became second nature to me so there is not stress associated with the driving task. Except when I visit a big city and have to deal with their horrendous traffic! ;-)

Anyone attempting to train their dog MUST pay attention to their dog's response to training, keep in touch with their trainer, and tailor their rewards and corrections to fit their dog. If a dog is trained under an exceptionally high level of stress...it WON'T learn and that will be evident in the lack of results gained from training. You just WON'T get a well trained dog if you're doing things to make the dog afraid of you or of other things...definitely not a dog that looks content working and obeying. I think BOTH the good positive-reward-only trainers and the good ones that use corrections agree on this. If a dog is completely freaking out or under undue stress...it won't learn. Ya gotta pay attention to your dog and be realistic about what you're setting it up for.

Also when a dog or person has a choice between right and wrong (and I'm not necessarily talking about 'moral' choices...just choices with associated consequences) and they choose RIGHT...there is no experience of a feeling of fear. Not from my experience or observation anyway. So when I stop at a red light...it's just habit based on my understanding and a choice I make as often as I possibly can. The feeling of fear only sets in when I make the wrong choice of running a red light (accidentally or on purpose)--the choice with potentially undesired consequences. So in this way I think corrections used in training of behaviors or skills is also not the same thing as punishment used to STOP unwanted behaviors that a dog owner NEVER wants. The dog will learn in context of training that getting up from a sit stay is not okay...but there is no global suppression of 'getting up' -- it's only learned as a 'wrong' in the context of SIT-STAY. The dog then learns that only a specific behavior will yield good results when he's told to "sit stay" and he gets good at it and doesn't have a feeling of fear while sit staying. Because while he's sit staying there's NOTHING to fear. So there is some avoidance training involved. So I guess my point (after all this long winded stuff) is that good correction training does not create a dog full of fear response but a dog with a choice to obtain desired results rather than undesired consequences and with enough opportunities to get it RIGHT, the dog finds the RIGHT choice more desirable and ends up choosing it of his own free will.

[F, Calla & Luca]

I agree completely. I think too that we go out of our way to avoid making people and dogs feel any disagreeable emotions. A dog can be displeased, uncomfortable or even a bit fearful of a consequence without ruining a relationship. My trainer advises us to be unemotional when giving commands etc. Yet I find my dogs recognize when I'm displeased and respond by trying, sometimes, to avoid this. Maybe it's not out of "love" but because they see me as a protector and supplier of good things, like food, but that's OK. For sure they know no rewards are forthcoming unless I'm pleased with their behavior. I think emotions such as shame in people have their place. People can learn the consequences of their behaviors in more subtle ways than animals and some actions should be regretted, make people feel ashamed etc. I think you can raise children to be generally proud of themselves and to have good self esteem without making them think everything they do is good, that they always win, etc. We love children despite their errors. Our dogs can bond with us despite being corrected at times.

[Lynne B]

Rewarding positive behavior does not mean that you have to make the dog think that everything it does is good. It Involves ignoring, or withholding rewards for negative behavior. There is also nothing wrong with a sharp well timed reprimand, But I won't and I never have used force or inflicted pain on a dog to get it to listen. I refuse to do it under any circumstances and my dogs are well balanced, happy, eager to please and very obedient.

[Karen, Jasper and Jackdoodle]

I think that a dog needs someone to be the leader, just like a child does. The wolf-pack thing really doesn't have to enter into it; lions live in organized groups with an "alpha" male head, and I'm sure lots of other animals live that way, whether they are related genetically or not. Humans don't live in "packs" with "alpha males", yet we do organize our lives in groups that naturally have "leaders". Every company has an authority structure, with a "boss" and subordinates proceeding from there. The government does, private organizations do, even rescue groups do. Society is organized in this way.You couldn't run a company, or anything else, with everyone being equal and free to make their own rules and decisions. So it really doesn't matter if dogs are wolves or not; they are like children in that they are our responsibility, they are dependent on us, and they are not capable of making choices on their own in the society in which they live, namely ours, without leadership, guidance, direction, whatever you want to call it. You don't have to follow an "alpha male" pack leader paradigm to still act as your dog's "boss", leader, or whatever you want to call it. I prefer to think of it as being a benevolent dictator.
The relationship that I think works best with a dog is one in which the dog naturally looks to you for direction as his leader; not out of fear, not in a manipulative way for his own benefit, but because a mutually beneficial relationship has developed between you in which the dog just naturally regards you as the source of everything he needs and wants. Kids and dogs are happier when they know the rules, know the boundaries, and know what to expect. Fairness and consistency is important. So while I don't think dogs should be treated like children, I do think that in this regard, the relationship is similar. The only way to help children and dogs live successfully in the world is to teach, guide and lead them. And part of that process involves corrections and consequences of some kind.

[Adina P]

I agree. Leadership is important...regardless of species =)

[1xar0x2rcx0li]

Agree!! with a close second to love and respect.

[F, Calla & Luca]

Your last paragraph says it all. I have always thought the world might be better off with benevolent dictators but unfortunately there are few if any around.