Monday, July 16, 2007

Climbing

The most arresting thing the great Jean Donaldson said in all seven hours of speaking at a seminar I attended yesterday at the Albany Obedience Club was not about Nelly and her ilk. it was about me.

In a fascinating overview of evolution, genetics, and how they affect behavior in general and canine behavior in particular, she told us about the Medawar effect: what happens to an organism after its reproductive period is over is invisible to evolution. It just doesn't give a hoot what illnesses you gather unto your bosom. You don't matter anymore.

Oh, the tragedy. I am about to vanish!

We were a group of about forty (though maybe it was seventy, or thirty: I can't estimate to save my life), sitting in our folding chairs in the large space that on other days echoes back the bark of dogs in agility practice, or the bark of people giving obedience commands--or so my prejudice against competitive obedience imagines it. (When I was a girl, and dreaming of having my first dog, I knew I wouldn't TRAIN it. I would have a NATURAL dog, not a robot who existed to do my bidding. That was the dichotomy, as my fervid little brain had it. Ha-ha-ha. Oh-ho.)

Needless to say, the audience was almost entirely female. This has been the case at every dog conference and seminar I've attended. Anyone want to hazard a guess why? Maybe because women are charged with--and wired for--nurturing and educating offspring? Thus we would illustrate another example of animal biology that Donaldson put forth: a slight "misfire" of an instinctual behavior, which happens to all sorts of creatures. In this case, our instinct to mother is triggered by the wrong species, by a dependent who is not genetically ours.

In person, Donaldson was far less prickly than she is in print. She was extremely generous in not offending (though her message was always clear, if you knew how to hear between the lines). The only time she allowed any righteous anger to boil over was in talking about breeders who allow or encourage the reproduction of spooky, fearful, or aloof traits, the kind of thing that's described as a breed characteristic, as in, say, "cautious" or "not easily socialized to strangers." Think Akitas, for instance. This she viewed as nothing less than criminal. She said, about those breeders, "I'm gunning for you," her voice tight with barely suppressed rage. "I've been cleaning up your messes for thirty years." The deliberate breeding of such an animal (sixty or more pounds of reactivity, armed with tearing teeth) is akin to selling a Beretta to any member of the general public who has a notion to buy one.

At the same time, given the fact that there are probably some 40 million dogs in this country, there are at most 12 to 20 killings by dogs per year. The incidence of bites is in fact decreasing, even as the dog population keeps rising and we live in ever closer proximity to them. So why the hysteria about a "dog bite epidemic"?

This is what gets dogs killed by the thousand, even though parrots and horses bite people all the time and are never euthanized for it. I suddenly saw where Donaldson was going when she asked why. What a mind. What an answer. A fear this unfounded, this primitive, she believes, could only be inborn: a residual fear of wolves, fanged predators, left over from the last evolutionary bottleneck for humans 100,000 years ago. We have not changed essentially in that long. It's too bad we can probably not count on a time that distant in the future, when we will have established a more reasonable fear of things with wheels or of bathtubs, since both of them kill exponentially more people than do dogs.

I love going to these seminars. I love seeing women (because that is what most of them are) who are barreling into science and the truth, armed with big questions. That's because the answers to them are required by the well-being of the creatures we care for. Who says we're not animals?

I drove home and picked up some Mexican takeout on the way. We ate out on the stone patio on a lovely summer evening. We wandered down to the garden to see if anything had escaped the cutworms and the teeth of the deer. I glanced up to see Nelly taking a little stroll on top of the dining table. She had just eaten an entire package of sweets. Pistachio-sesame-toffee-white-chocolate. They had looked quite good from the picture on front.

11 comments:

Anonymous said...

I do not agree with you about women and nurturing, men can also nuture. My way of "training is by bonding with my dogs. It has always worked for me, not counting Daisy. I have no problems with Fannie, and Jupiter is getting better because of bonding with me and my friend Dhia. Dhia hs Jupiter in her lap and cuddles him every day in the park. After awhile she give Jupiter back to me and I cuddle him. In a few weaks Jupiter has become more affectionate and usually stays close to me a lot of the time, and he comes whenI call him most of the time. This Rat Terrier is no saint, but considering the dog he is, he is doing well. I have seen so many people spend so much money on training with mixed results. All of the dogs that I have had in my life I never went beyond sit/stay and house breaking. As I stated above I had problems withonly one dog, Daisy. I am not my dog's Father", I'm one of the members of the pack. I do not humanize my dogs, that would be a insult to the canine species. I respect who they are, dogs, not people, and I allow them to be dogs and not what I think they should be. A dog's behavior is alien to us, so why second guess and interfere with ther canine life with theroies on what we think about dog life.

Melissa Holbrook Pierson said...

I never meant to imply that men can't nurture: far from it. I have a husband who does plenty of it to our son and our dog. There are single fathers, too, who are "just as good" (or as bad) as mothers. But take a look at the PTAs, the playgrounds, the dog seminars. Ninety percent female. Why? Culture, certainly. But also possibly biological. Anyway, a "misfire" can occur with a male caregiver just as easily as a female.
As for training, it's true that some dogs are easy. But other dog owners are propelled to it by necessity: their dogs are going to be killed if they're not trained. Or miserable: a naturally fearful dog lives in an unhappy state, and if an owner can use desensitization and counter-conditioning to improve her dog's quality of life, I think it would be criminal if she didn't. Ethology, learning theory, and animal behavior and neurology are not just people's flights of fancy--they are facts, Tony. I'm so grateful to my first trainer Polly for opening the door onto this world, which has enriched my life and, I have every reason to believe, that of my dogs.

Deirdre said...

I was out with the dogs yesterday and a man in a pickup pulled over to over me what he claimed was "friendly advice." This seemed more like a veiled threat to me when he said that if my dog chased deer someone would shoot him. And he added, "I wouldn't want to see you crying." I felt almost as if he WOULD want to see me crying and all because I have a dog who seemed a threat to deer. He ignored the little dog of course and focused instead upon the big one. I asked him why anyone would shoot a dog for chasing deer and he answered with that chestnut about "getting the taste for blood." I happen to know that Tiki ate a squirrel once and threw it up. That he killed a baby rat but only because he played with it for an hour and he then left it lying on my neighbor's lawn (for me to pick up). So much for the taste of blood. I wonder then if it is that fear of the wolf that really is at work here and that his man would protect himself (rather than the deer that overrun the area like enormous rats). I was feeling vulnerable as if a threat had been issued directly at my beloved and also as if the dog can no longer run free. I cried.

Melissa Holbrook Pierson said...

Deirdre, that's a terrible story. I think that man *was* threatening you--why, I don't know. Probably his own unconfronted fear. It's my belief that the people who yell the loudest at you to "get that dog on a leash" are simply afraid of dogs.
I don't really understand that law against running deer. (Apparently, it *is* legal to shoot a dog who's running deer.) Granted, I do not condone dogs chasing or killing wildlife; it's unnecessary, since we allow others to kill animals for them to eat. ;-< But is it, as I said before, because we want to save the deer for US to kill? Then that's just stupid.
Some sadistic men like to make women cry. Sorry to sound sexist, but I've experienced it plenty. And I've never gone out of my way to make a man cry.

Anonymous said...

I loved the end of that blog.

Anonymous said...

I only agree that the desire to manipulate one's environment or protect one's life are "inherent" in dogs (and humans). How we go about accomplishing those tasks is learned, though. A baby, for example, doesn't know how to clench its fist and threaten a punch when it wants to get its way. A newborn puppy can't retaliate to save its life...literally. (I've watched video footage of even several-week-old puppies attacked by other dogs or other animals, without so much as an attempt to retaliate. They usually just cry and try to flee. The ones with some experience using their teeth may try to bite, but they're so poorly practiced at it, the threat is practically moot to the attacker.)

(I watched as one breeder demonstrated how she "tests" for her progeny's "inherent aggressiveness", when what she was really doing was demonstrating how she teaches those puppies to respond by using their teeth.

What clearer evidence could there be, than the fact not one of the puppies she used for this "demonstration" tried to retaliate against her the first, second, even third times she pestered them?

Only once they were old enough to have some practice using their teeth with littermates, and had tried and failed with every non-violent coping strategy at their disposal, did they finally resort to using their teeth.

The breeder then halts the torment, and announces it proves how "inherently aggressive" her puppies are. Well...they aren't. She's teaching them that their teeth can be beneficial tools in getting their way with humans. Aside from the irresponsibility of her actions, they certainly don't come anywhere near proof the aggressive behaviours her dogs exhibit are innate. If anything, it was a text book case of how to teach a dog to behave aggressively.)

Just as with humans, there may be some dogs with impulse control issues, or more "sharp" personality types. This, too, does not necessarily conclude that aggressive behaviours will develop. A person with impulse control issues, for example, is as likely to take up extreme sports (a positive activity), as violence or criminality (a negative activity). Psychologists agree the difference lies in how those individuals are raised (i.e. what kinds of coping strategies are rewarded).

Very high drive dogs are known to develop obsessive behaviours, such as chasing in circles, jumping, frustated barking, repeated fetching of an item, and on and on. There are countless benigh behaviours high drive dogs can learn will garner their owners' attention, and other forms of reward.

On the other hand, high drive dogs can develop destructive behaviours, such as chewing, annoying/excessive barking, digging, and even aggression. When owners "reward" those behaviours, or otherwise allow them to escalate, they're actively encouraging them.

It's not the fact that the dog is high drive that's the problem. It's what the owner rewards that dictactes which behaviours will be continued.

As always, it's not about whether the dog is dominant or submissive, high drive or low drive, or high energy, or low energy. It's what the owner does with that personality type that determines if it will become a good canine citizen, a nuisance, or a terror.

The last thing that jumped out at me was theory of an "inborn" fear of anything...much less wolves.

Most people who've looked at this issue understand and accept that we learn fear. An infant is no more afraid of a wolf than a chair. Put a porcupine in front of a young child or puppy, and both are equally likely to reach out and get "stuck". Having had that experience, they'll certainly learn to fear porcupines. But it's not an innate quality we're born with.

Put a wolf in front of an infant, and it will be the anxiety and stress of the caregiver(s) that helps shape the infant's response. Having seen infants exposed to all kinds of strange animals, most show extreme curiosity in the absence of parental anxiety.

Oh, and one final thing. When I first read a Donaldson column about it being unacceptable to breed fearful dogs, I meant to respond to it directly. But I declined, at the time, because I certainly wouldn't want to imply that I'm promoting the breeding of fearful dogs.

But there's an awfully big area in between.

I absolutely respect Ms. Donaldson and am totally on-board with probabaly >95% of what she has to say. But I don't agree with everything, and the premise behind her position on this is one of those areas.

As an experienced dog trainer, it should come as no surprise that my own dog is one with an extreme personality type most dog owners wouldn't be able to cope with. (This particular dog was scheduled for euthanasia because of her out of control behaviour.) She's possibly the most fearful dog in existence. She's afraid of pretty much everything. With training and socialization, she's a bit better. More importantly, with all that training and bonding, she trusts me implicitly. As afraid as she might be, if I ask her to do something, she's going to do it.

There are few who'd disagree that this dog is one of the best-behaved dogs one is likely ever to meet. I teach extremely high levels of obedience and socialisation. I don't reward any kind of aggressive behaviour in dogs. (I specialized in re-training aggressive dogs for a decade.) This dog is now eight-years-old and has never growled (or seemingly felt the need to growl). Not one of my own dogs has ever bitten a living creature.

If she were a good representative of her breed, I can't imagine any theory behind which she shouldn't produce offspring. She's an excellent dog, by every account. "She's the best dog in the world," attests my neighbour's twelve-year-old son.

Can an inexperienced or malevolent dog owner encourage a fearful dog to develop aggressive behaviours? I believe so. I mean, it seems pretty common. Same goes with dominant dogs, submissive dogs, and dogs of every other kind of personality type.

(By the way, in one study, one of the most common side effects of spaying female dogs was generalised fearful behaviour.)

But as with all inappropriate dog behaviours allowed/encouraged by bad owners, the blame rests with the humans involved, not the dog...now matter how fearful or confident those dogs may be.

(For the record, even poorly-bred Akitas - or any other individual breed - are not responsible for the majority of dog attacks. Dogs of every size, shape and original breed purpose make up the dog bite, attack, and fatality statistics.)

I tend to be a bit prickly when we focus on the dog, for bad behaviour, rather than the cause: the owner. I don't fault people for being ignorant about dog training and behaviour (we all have to start somewhere), but there are countless books, magazines, web sites, and videos available to any conscientious dog owner who wants to prevent the development of inappropriate behaviours in their dogs. The really bad dog owners, the kind who ignore or encourage their dogs' bad behaviours, are (of course) wholly to blame for their negligence. No matter how you slice it, focusing on the dog, instead of the owner, does nothing about preventing or correcting undesirable dog behaviour. So whether a dog is fearful or confident is irrelevant in terms of which behaviours will ultimately be allowed to develop in, and/or be perfected by, any given dog.

Melissa Holbrook Pierson said...

Marjorie, what a fascinating comment--I know the nature/nurture debate has long been a heated one; I suspect we know a lot more facts now than we did just a few years ago. Recently I got to listen to Carolyn Wilki lecture about neurology (the other side of the coin to Donaldson's evolution lecture) and I know they all fit neatly together somewhere, but the science is deep & I'm a beginner swimmer.

Some people get easy dogs. Some people get tough ones. And when the latter go to "normal" people, not people with your discipline and commitment, those dogs end up miserable, or hurting someone, or dead. I have friends who are utterly *devoted* to their dogs, but they cannot or will not train, or even learn basic canine body language or the like. I'm afraid that's most dog owners. And the reason why *that* is will bring us back into arguments about humans that resemble the nature/nurture debate. Sigh.

Anonymous said...

Hey...thanks for the reply. ;-)

I will say I often hear the argument that some temperament types are "more difficult" to train, but I tend to qualify that kind of statement...a lot.

Undeniably, every dog is perfectly trainable.

But they don't all respond to the "one trick pony"-type techniques so many utilise. (If someone only knows one style of training, he/she may, indeed, find him/herself in difficulty with some dogs. Another person likely wouldn't have significant difficulty with that same dog, nor might the first person, if he/she just tried another strategy.)

And to complicate things even further, different canine temperament types often require different training methods. (Use the same methods on both a very dominant and a very submissive dog, and the dominant one may walk all over you, or you may destroy what little confidence the submissive dog has.)

And to make matters worse, individual people have their own affinities, as far as training styles go. (What may work for Cesar Millan or Stanley Coren, may not work for me...at least...verbatim.)

All these variables make a less-than-committed dog training effort anything but a guarantee of a well-mannered dog.

Still, it is up to the humans how much effort they're going to put into it. The dog doesn't really have much say. So it's really the owner who bears the responsibility when they haven't worked hard enough to ensure a well-behaved dog.

Every dog is perfectly trainable, if the person uses a training method to which the dog will respond. (If after a few days of attempted training, there is no improvement, it behooves the person to try something different. That's all.)

And that is why it is always down to the owner, as to whether or not the dog will turn out a good canine citizen, a nuisance, or a terror. (I know there are some very clever and/or high energy/high drive dogs that can test anyone's patience, at times. Still, that is just the way those dogs are, and it's up to humans to teach them accordingly.)

Humans choose which techniques they're going to employ; how much effort they're going to put in; and how consistent and patient they'll be with dog, as well. Again, the dog has little say in how effective the chosen training methods will be, or how correctly they'll be employed.

I liken responsible dog rearing to gaining weight, in some respects.

There are all kinds of arguments suggesting everything from genetics to a virus causes weight gain. While I wouldn't want to dismiss the possibility that some people gain weight more easily than others (in fact, I believe that is pretty well accepted), the fact remains there is no magic way to put on fat. Stored fat is the result of consuming more calories than one burns. It's not any sexier or more complicated than that, when you get right down to it. Eat fewer calories than you burn and, over time, you will have less fat.

And here's where I make the analogy. Once you gain that first 5 or 10 pounds, you have a choice to either address the weight gain (and do things like see a doctor; eat less; exercise more), or you can dismiss it, and hope it'll either go away, or won't get any worse.

(Quick story: My husband's colleague brought his family and their new puppy to my home for a visit a couple of days ago. In my house, the cat and dog food dishes are next to each other, on the floor - albeit the dog food is on a stand. The young puppy made a bee line for the cat food. The owner was about to put the cat food up on a table when I stepped in and merely taught the puppy that she wasn't allowed to eat the cat food...leaving it where it belongs. You see...I prefer teaching dogs what is and isn't acceptable behaviour, rather than rearranging everything to accommodate their undesirable behaviour. The puppy didn't learn this lesson immediately, but she did learn - about 10 minutes of trial and error. Pretty good, I'd say. Not too much effort, either. And while this dog would likely never be in my home again, it demonstrates how I taught my dogs not to eat my cat's food. Far too many people don't even try to teach their dogs to do what they'd like them to do in various situations.)

With dogs, once we see bad behaviour, we can either address it (and correct it), or we can choose to minimise or ignore it, and hope it'll go away or won't get any worse.

The thing is, when dogs are allowed to continue an undesirable behaviour, it usually escalates, over time. If it's situational, we can try to avoid those situations. But that's not a very good plan. We can't anticipate everything. And that way of thinking tends to just create more anxiety, anyway. (The owner is always worried about "what if..?")

Some people may try one strategy to deal with a developing problem and, if that doesn't solve it, they quit, and blame the dog or themselves. Others stay committed to fixing the problem, and try something new or seek out help from others. But, no matter what, they aren't going to let the bad behaviour go on one more day, if they can help it.

I think it has more to do with personal responsibility, than skill. I've known too many dog owners who were perfectly intelligent and well-meaning people; people who could readily "get" the relatively straightforward concepts behind dog training; who just don't want to face the reality of their dogs' bad behaviour. They'd rather ignore it or minimise it, because then they don't have to actually "do" anything.

And, long story still too long, we're right back to my point: hapless dogs are either taught how to behave appropriately in human society, or they're not. And whether they're dominant or submissive, high energy or low energy, big or small, that's always down to the owner.

My suggestion for any dog owner who, despite working hard, isn't seeing results with their training, is to seek out an alternative method, and see if that helps. All dogs are trainable. So if the owner is being consistent and putting in the effort, it's just a matter of finding a training style which compliments that individual dog.

Anonymous said...

Whoa Nelly! ;-)

I'm back to add something to the discussion about the theory that people have an innate fear of animals with "big teeth".

I was talking about this very issue, recently, when I pointed out that (in addition to there being little, if any, evidence to suggest infants are born with any kind of fear at all) children are the most common dog bite victims BECAUSE they just aren't afraid of dogs...even in circumstances where we, as adults with more experience, understand they should be.

Young children are not afraid of dogs, unless others create a lot of anxiety in the presence of dogs. Then the children may learn to fear dogs. As they grow older, children may learn that when a dog growls, raises its lips, etc., it is warning it will bite, if the person doesn't stop doing whatever it is that is bothering the dog.

But even when a dog is openly showing signs it will bite (such as growling, raised lips, staring, etc.) many children are so oblivious, they continue interacting with the dog, and are bitten.

Children are the most common dog bite victims precisely because they aren't afraid of dogs...even those dogs that are clearly dangerous. If anything, that fact alone suggests there's little merit to the concept of some kind of innate fear of dogs or "animals with big teeth".

Melissa Holbrook Pierson said...

Hey, Marjorie: Please tell how you taught the puppy not to eat the cat food, OK? That is one technique I'd like to know!

I've just posted (in "Babes") on subjects touching on this thread. So let me know what you think.

Anonymous said...

The puppy was loose, and trotted right over to my cat's food dish (which is next to my dog's dish...but the dog dish is on a raised platform and, thus, too tall for the puppy to easily reach).

The puppy's owner was about to pick up the cat's food dish, but I stepped in and explained that it would be a much better idea to start training the puppy how to behave, rather than trying to "manage" every environment she comes across. (Don't get me wrong. If it had been a temporary teething or housesoiling issue, that might be different. Those problems will go away as the dog matures. Eating everything within reach likely won't disappear with age.)

I simply halted the puppy before it could reach the food a second time and said, "Uh, uh." (It's not my dog, so I wasn't about to start training it actual commands like "out" or "leave it", etc.) I held my flat palm facing the puppy, just in front of the cat's dish. When the puppy didn't simply push past my hand, I squealed and praised, "VERY good!" And did my best to entice its attention to me, and what I was doing.

At the same time, I asked my husband to fetch a treat or toy for the puppy. He handed me one of the chew treats the owners brought with them.

Naturally, as soon as I removed my hand, the puppy went for the food again. (I know this is going to happen, so I'm ready for it.)

Again, I use my flat palm to halt the puppy's access to the cat's food, and again say, "Uh, uh" with authority, but not in a menacing way. (I try to always smile when I'm training dogs. With puppies, it's easy!)

It's the "authority" that's part of the puzzle, I find. Many dog owners just don't have the authority needed to direct a dog verbally. They can learn it, but it's hard for some people. Men, in particular, often confuse "authority" with "hostility". We're never trying to frighten dogs into doing what we say. We simply need to let them know we mean it, in a way that's understandable to a dog. And that's it. No matter how many times they try to go back to the cat dish, or break a sit, or whatever, we're going to notice it and correct it. Every time. That's all. There's nothing at all evil about that. In fact, that's the environment in which dogs flourish most. Dogs L-O-V-E to know exactly where they stand, and how to get the pack leader's attention. When we reinforce to them that they will be rewarded for doing what we ask, they will do that over and over and over again.

As predicted, the puppy stops, and looks at me, and I praise. "What a good girl," I say.

Then, I redirect the puppy. "Look what I have in my hand." "Is it a treat?" I use the treat to lure the puppy away from the cat's dish (just to get her focus off it), and maybe ask for a sit. Then I give the treat with a lot of praise. (This happened to be a pretty big - time consuming - treat, for a puppy. It's best to use a treat that can be consumed quickly, so you can move on to the next reinforcement exercise.)

(On the flip side, I don't use a lot of treats in training dogs, myself. I'd prefer luring the dog off to play or for a fun training exercise, or just a change of scenery, at the point I got its attention away from the cat's dish. The treat was just the easiest options with someone else's puppy that just arrived in my a few moments earlier.)

Typically, if you're there, watching, the puppy will consume the treat, and go back to the cat's dish. That's when I would repeat the above procedure. Stop the puppy from actually reaching the cat's food (which, by eating it, is positive reinforcement), so it again finds no benefit in going to the cat's dish. Then I would praise the puppy for obeying my "uh, uh" 'command' (if it did stop when I said, "Uh, uh."), and would redirect the puppy to a more desirable action, like chewing on an acceptable item, or following me to play a game in another room.

In this case, it was a social situation, and I wasn't about to spend a lot of time teaching their puppy not to eat my cat's food. So, after a couple of successful attempts at stopping the puppy with my palm and the words "uh, uh," but knowing the puppy wasn't yet ready to be on its own, I had them put a leash on the puppy, so the humans could be social, instead of having to eyeball the puppy constantly.

Knowing she would, again, try to go back to the cat's dish (because I specifically asked her to be kept within leash range of it), I asked the owner to give an ever so slight tug on the leash at the point the puppy clearly intended to go for the cat food.

Seeing in transpire, we both said, "Uh, uh." The owner was a bit confused about the leash tug, so I took it over for the next round.

(You never, ever want to have more than a second's worth of contact with a dog's neck. It's not effective, and tends to create an undesirable response from the dog.

One of the most basic rules in physics is that every action has an equal and opposite reaction. In the case of a constant, tight leash, one is teaching the dog to pull and resist, in opposition. ...So much so that I have to actually teach my dogs to accept those retractable leash things. I rarely have contact with my dogs' necks, so even the slightest contact has them stopping in their tracks. The first time I use a retractable leash on them, they invariably just stop and stand there, wondering why the contact. It takes some coaxing for them to learn that the slight pressure from a retractable leash is not some indication I want them to stop.) (Ah, the joys of a well-trained dog.) ;-)

Anyway, the puppy was again unable to reach the food (and, thus, wasn't rewarded), and turned around and payed attention to us when we called her name. That, of course, is where we all squealed and praised. We're rewarding the puppy for not just obeying us, but for paying attention to us when we call her name, insteady of remaining focused on the cat's dish.

The next time the puppy went for the dish, I did a proper (very gentle) leash correction along with the words, "Uh, uh." The puppy turned around on her own, and trotted right back over to me. I squealed and praised, of course. We may have given her another chew treat to keep her occupied. I can't recall.

That was the last time she attempted to eat out of the cat's dish, for the rest of the afternoon. (She was still on-leash for the next little while - maybe 20-30 minutes - where she layed near our feet. When we all went outside to play, that was the perfect opportunity to let her off-leash, but SCRUPULOUSLY focused on going outside to play (i.e. not just assuming she's doing to dutifully follow us outdoors)...right past the cat's dish. Sure, she could have failed, and made a bee line for the food...and that would've been understandable and perfectly predictable. But she didn't, which further reinforced the "training". While outside, she ate her own food, had a treat or two, and played until tired. By the time she came back inside, I think she'd forgotten all about the cat's food.)

We didn't do this, but if it had been a dog I was training, I'd have been sure to keep the leash on the dog when it came back inside, just so that if it did go back to the cat's dish, I could prevent it from being rewarded (by eating some food), and continue more training.

Like all training, it's mostly practice. Teach the dog there is no benefit for going to the cat's dish, but reward the dog for not going to the cat's dish or otherwise make it so the cat's dish isn't as rewarding as doing what you ask. (i.e. playing, having fun, feeding adequate amounts of dog food, giving treats, and getting enough exercise and mental stimulation throught the day, so that all the dog wants to do indoors is rest or play with "acceptable" toys.)

I realize I'm not explaining every thought, and every action, in detail. But you should get the idea. If there's no reward, then dogs (and people) won't continue that action. If there's a reward for NOT doing something or, more correctly, there are more interesting things to do in place of that action, then dogs and people will choose the more rewarding one.

There are some additional reinforcement strategies people can do, if they're trying to teach this to their own dogs, at home...once they've taught the dog the essence of what's acceptable and unacceptable behaviour. For instance, you can mark the off-limits item with a strong scent, like mint (if, in this case, it won't bother the cat too much). That way, there's a double reinforcement. You consistently disallow the dog access to the cat's food, and the area smells unique, so the dog comes to associate the smell with an off-limits area. It can work.