How interesting. We had a very protracted period of getting the dog used to our resident cat starting with weeks of them having to be segregated completely as the dog was obsessed with her and at one point had her in his jaws.
A bottle with stones was recommended, but Gymdog is not great with unexpected noise as it makes him hyper and barky - ie very similar to his behaviour around the cat - so I rejected that. We needed calm, not more chaos so I also rejected splashing water at him etc.
As ever, food was our friend. We progressed very slowly to Gymdog being able to be in the same room as the cat with her out of reach but he was still fixated, drooling, glazed eyes, legs quivering etc. Our big breakthrough happened by accident - I put a bowl of vile delicious smelling cat food in the same room but out of Gymdog's reach and away from the cat. Suddenly having two powerful stimuli in the same room, but in two different places, broke the fixation. Up until this point, so great was the fixation I could tuck a treat under his lip and he would not register it was there. With the fixation broken, he became responsive and I could work with him. I spent boxing day sat between cat and dog alternately feeding them bits of turkey and praising Gymdog for taking turns and eating turkey rather than cat.
So, distraction and positive reinforcement but not because of a conscious choice on my part between two models, but because that's what worked! Which is the best recommendation I guess. The majority of dog owners are unlikely to read up extensively on the different approaches, they just want techniques that work.
I feel using fear to train in as inhumane as a physical punishment. Why would you want to scare your own dog? Plus the points poached makes about it not working very well and having unintended consequences.
IMO becky no, that's not ok. Aversives like those are difficult to use "correctly" as inappropriate timing may lead the dog to associate the unpleasant stimulus with something other than the cat, for example the owner! Furthermore they heighten stress and anxiety which is counterproductive, and we also know that fear inhibits learning so they are actually ineffective.
In addition they are unnecessary. You can teach a dog to live with cats without any aversives. Or you can install a stairgate
Poached - I have tried that. I have politely sent links including that one and very politely mentioned the weight of scientific opinion. I can be very polite when necessary. The Chairman of the charity just said v patronisingly "well, we know you have a bee in your bonnet about this dog welfare stuff" and that they didn't want to offend Jan Fennell who at some distant point had donated a book or some equally tenuous link. He also said that their behaviour team liked using a dominance model as they "have experience of it working".
Sadly, when you have an all voluntary organisation there isn't much to be done in that case. All I can do is to review what if any voluntary work I do for them, and put my efforts into charities that do use positive methods.
Wow thanks D0oing and Spicy for taking the time to answer my questions at such length. Really appreciated. Until I realised I must do some work rather than MN, I was also going to make Redwing's point about being the boss being attractive. I think she's right and that's a big factor.
I read Milan's (first?) book over Christmas as my mum bought it in a jumble sale. A lot of his stuff in the early part of the book didn't seem so different from what I knew of the positive approaches - as Redwing says, calm energy etc - but having read this thread I might re-read the beginning where he describes growing up in Mexico and observing semi-feral dog packs. It's all very convincing but I guess that's the point at which the divergence starts. My memory is that it's very very different from D0oing's description! Be interesting to reread it now with a different perspective.
You explain and sumarise really well D0oing, not sure what course you're doing but you deserve to do really well
A few thoughts. One is that many smaller rescues, generally run by volunteers, STILL espouse this bloody nonsense. I'm struggling at the moment as I discovered inadvertently earlier this week that one of the greyhound rescues I volunteer for has several key people (including the main foster co-ordinator and the behaviour advisors) think dominance is the way to go and ADORE CM. Head. Desk. Sigh. So in turn they are advising new adopters etc. and so it continues.
Another is that there are so many dog trainers out there who are still doing well out of it, especially the ex police dog handler types. When I found our current trainer I googled trainers in our area, and was really depressed by the number (with smart, superficially convincing websites) who are still on this path. How are novice dog owners supposed to know this is wrong?
Thirdly, and this is more of an observation than a theory, but I've noticed that nearly all the dominance based trainers are men - CM obviously, but again a lot of the more militaristic types love this stuff. It's hierarchical, macho, alpha male, testosterone fuelled stuff. Overwhelmingly positive trainers are female. My current trainer is a bloke (very positive) and put a pic up recently on his FB page after attending a Gwen Bailey training course - he was the only man, and if you go to so many dog training events, it is overwhelmingly a female thing. Not sure why this is but there is definitely a gender split going on. Obviously some exceptions such as Jan Fennell in the dominance school but certainly a factor.
Totally agree re Cesar Milan. Some of his stuff starts as quite helpful, for example his ideas about calm energy and ignoring nervous or fearful dogs and letting them approach in their own time. Then he says something ridiculous about "the pack" and does an alpha roll or puts a dog on a treadmill and I want to punch the TV.
Negative and positive punishment did my head in for ages on my last module
Now I understand it I can't see how it was so confusing to me
Yes, most positive trainers use punishment, normally negative punishment, which is the removal of something to reduce the frequency of a behavior eg. removing attention from a dog who jumps up. It's not all treats all the time.
Positive punishment is normally the baddy, it's the addition of something to reduce a behavior eg. the addition of pain via a prong collar to reduce pulling on the lead
The reason most positive trainers don't mention punishments is because novices who only know that positive training is best, will tend to shy away when punishment is mentioned, people automatically assume that the punishment will be physically or mentally aversive, when that is not always the case.
If Cesar Milan would just study and understand the science behind it all and hold his hands up and admit he got it wrong, he could do so much good for owners and their dogs, he does have a natural affinity with dogs and he's very charismatic. It's just a shame he has to ruin it by talking
There's load about reinforcement schedules too, it's not as simple as just giving your dog a treat every time you want it to behave, that's not practical or healthy. The idea is your start training a behavior on a constant reinforcement schedule i.e the dog gets a treat every time and then you switch to a different reinforcement schedule, so the dog knows it might get a treat, so it still performs but your not constantly shoveling hotdogs into them. I think a lot of anti positive training people believe us positive folk are constantly dishing out treats every time our dog behaves well. There are also different kinds of reinforcers too, as Spicy mentioned earlier, some dogs find a smile reinforcing, some like play or affection, a word such as "Good dog" can be reinforcing, very few trainers use all treats, all the time.
And yeah, Whippy is a better cat than the actual cat
I think unfortunately pack theory fits in with some people's ideals about themselves and leadership more generally. Who doesn't want to be leader of the pack? Or the alpha? It sounds good to a lot of people. And it's much simpler to describe whole groups of behaviours as 'dominance' rather than look at a whole set of complicated reasons for each one.
The really good positive reinforcement trainers will have studied a lot of behavioural science so they'll understand about negative punishment and premack and so on. Some of this is complicated, and not so easy to explain. It would be good to see some of that on TV though instead of Cesar Millan.
My cat fails at being a cat and is too stupid and lazy to plan world domination
He can't even catch mice, he can only manage to catch insects and even then he doesn't kill them the bastard thing just brings them in, lets them go and looks confused when they fly off/scurry under the sofa
That's not my strong point but I prefer to think of dog-dog interactions in terms of competition for resources rather than "dominance". For example when SpicyDog takes a chew from near Pup and runs away with it, my perception is that she is claiming ownership of the resource rather than dominating Pup for the sake of it. From a human perspective it does appear that she is being a total cowbag though! And most humping I've seen looks more like an attempt at play than dominance.
Think it's worth mentioning that most positive trainers advocate some negative methods, but that are gentle and humane e.g. stepping away from a dog that jumps up combined with click treating for staying in a sit. That's a negative consequence but twinned with a positive reinforcement and not likely to cause undue stress to the dog
Tbh that is the part that we just starting to cover on my course, but from previous knowledge and little bits of info from my course, I will do my best to try and explain...
So basically the idea of pack theory and dominance came from studying wolves, which makes sense, since dogs are descended from wolves. What went wrong is that the wolves were not studied in their natural environment or within their natural family.
Wild wolves were captured and placed into an enclosure with other wild wolves. These wolves were not related to each other and had no clear leader, as they did in their own family groups, so disputes broke out between them, this was seen as dominance. The researchers believed that each wolf wanted to be dominant and that is how it would work with dogs and wild wolves. They'd all want to be dominant and would gain dominance through ritualistic displays and aggression.
What we now know is that wolf packs are actually family groups, a breeding pair and their off spring. The breeding pair lead by default, the older off spring help raise the younger, so they also lead the younger pack members, but not the older ones. Kind of like a flow chart based on age and experience. The younger pack members do not try to dominate the older ones. Eventually the cubs will grow up, find a mate and begin their own and thus become 'alpha' so each surviving wolf gets a chance to be pack leader in time. There are no ritualistic shows of dominance or aggression. The elders eat first, normally, because they hunted and caught the food, although in times of illness or when food is sparse they will allow the younger pack members to eat, to ensure their survival. It's a relationship based on mutual co-operation, the same as our family is.
When left to go feral dogs don't behave in the way same as wolves do, they don't have the same rigid family groups, the male dog has very little interest in sticking around to help raise the pups, the female is on her own. They'll mate with more than one spouse, even mating with members of other social groups.
They are each out for their own but will co-operate with other dogs if it is for their own gain. They catch and eat (or scavenge normally) their own meals, one leader does not feed the pack as wolves do. A dog who might show aggression with one particular other dog, in one situation, might not show aggression to another dog in the same situation, all based on their previous experience of that dog and the value of the resource they are fighting for. They are well aware injury may lead to death, so will only risk fighting if what they have is deemed worth it and they have a good chance of winning. They prefer to diffuse tension using calming signals and submissive behaviors which are normally copied by the aggressor.
They often leave one social group to join another, rather than staying in the same group like wolves. There is never a clear leader, although some of the group may naturally lead this is changeable due to the constant comings and goings of group members, it's all very flexible.
What this means to us when we are training is that dogs are adaptable and like to gain resources via co-operation rather than aggression. They don't understand pack rules or alpha rolls or eating first etc.
Also because the pup has already imprinted upon it's mother before meeting you, it is well aware that you are not the same species and you don't have the same rules as their species do.
Thanks both, a number of gaps filled there! I do find this whole area fascinating, one of the very unanticipated areas of dog ownership has been the time spent reading about it.
The relationship between transactional and emotional comes up a lot and I have seen it (I think) working with Gymdog and his reaction to other dogs. We've trained the transactional through repetition, ie see dog, look at me, eat sausage and that (I like to believe) has lead gradually to a more 'emotional' level. I do think Gymdog has developed a level of trust that by transacting with me in this way, things will remain calm and manageable. That in turn is just beginning to lead to him exploring fleeting hellos with certain dogs which is lovely to see. Is that right? Does that make sense?
Just further on why the pack theory continues to have a hold - I think some people (myself included) are confused about whether newer thinking applies just to dog/humans or to dog/dog as well. I have no problems understanding that pack theory doesn't apply to humans but I'm still confused over other dogs. Despite being on a thread that talked a lot about dogs being dominant over other dogs recently, I still wasn't sure by the end of it as to whether this was current thinking. I think a lot of people would find it hard to believe that dogs don't act in this way with each other as so much of their behaviour appears to bear it out. However, dealing with a reactive dog has shown me that a lot of aggression is actually fear and defence and that I had an awful lot of preconceptions that were wrong, so more than happy to be shown the light! Is it what you mentioned earlier in the week D0oing?, ie that there's a big difference between a dog living in a dog pack and a dog living on my sofa?
Oh but D0oin I love the idea of my pair getting together in the night like Pinky and the Brain to plot to overturn the pack leaders by such sophisticated methods as rushing through doors first, jumping on furniture and humping my arm. I almost wish it were true
I don't understand why dominance theory got so popular to begin with, even if you don't fully understand dogs and how they learn and their attachment with us etc. surely it is preferable to believe that your best friend will blindly follow you because you feed them hotdogs than to believe they spend every waking minute planning their next move to overturn you?
I'd rather not have an animal in my house who is constantly plotting my downfall
I really dislike the description of positive reinforcement as bribery. There are basically two reasons why a dog will do something for a human. Either because something good will happen when they do (positive reinforcement) or because something bad will happen if they don't (aversive). Dogs doing things out of love for and trust in their handlers is just a type of positive reinforcement. SpicyDog finds praise and a fuss rewarding. She therefore behaves (mostly!) in a way she knows will please us and make us happy. But she is getting something out of it, it isn't blind adoration because I am "pack leader". So basically, the emotianal is also transactional. It's a human conceit to believe otherwise.
You have identified another issue with a positive approach, which is that it attempts to be based on a scientific understanding of canine behaviour. This is currently incomplete as we do not fully understand every behaviour. Positive trainers and behaviourists are important in translating that knowledge into practical dog training approaches. I like to know the theory behind it but it's not necessary if people are using the right humane methods anyway.
No, no, no that is too simle a way of looking at it, Gymmummy
Although I did point it out as being that simple on the other thread. Dogs do have some level of cognition and a bond with us. All animals, including humans, do what is called imprinting, as in they form strong social bonds and identify with members of their own species. The awesome thing about dogs is they are one of only a few species who are able to imprint on more than one species at a time. That is why our relationships with them are so special. At a certain stage in a puppies development they actually prefer and seek out human contact over that of their litter mates and mother.
What they don't have is the level of empathy that allows them to understand that their behavior will impact upon other animals. They love us and have an innate desire to be near us and even with some dogs to protect us, what they don't have is the cognitive ability to understand that eating our shoes might piss us off or the ability to forward think, dogs live in the here and now, they don't plan forward, which would kinda put the kibosh on any plans of world domination, wouldn't it?
Behaviorally they are very simple and quite selfish, if it works for them and it's rewarding they'll continue doing it. They certainly don't have the ability to plan a coup de gras and they don't plan on upsetting or annoying us, they have no concept of their actions effect us or their standing in the 'pack'
i think sometimes the positive camp make it sound too simple though?
I was reading D0oing's description of the In-Out, repeat what works model earlier this week and it got me wondering where the human-type qualities we attribute to our dogs like loyalty and love and picking up on moods fit into that.
i think a lot of people (in the UK at least) imagine their connection with their dog to be based to some degree on an emotional relationship rather than just a transactional one. Dogs just doing stuff that works based on their previous experience doesn't tend to make me think of having a relationship with my dog with any mutual trust or respect. Maybe it shouldn't, maybe those are inappropriate concepts. They're pretty common ones though. I think the treat basis of the positive methods can also create the impression that the relationship is just based on bribery.
Before I got a dog last summer, I didn't even know there was a debate to be had and I have struggled on occasions to find the positive model (or my understanding of it at least) has enough detail to test all my experiences against. Cesar et al have a model that allows pretty much all behaviours to be fitted in and 'explained'. The model has a hold because it has a neat fit and gives an all-round picture. I've not really read a description of the positive model that has that all-round view. For example, I questioned my dog continually marking earlier this week. Cuebill set me straight and said it was a stress relief measure (so not 'marking' at all I guess), but I'd have struggled to reach that conclusion myself. OTOH I have little trouble guessing that Cesar et al would have called it territory marking/maybe dominance. So perhaps his explanations are more accessible and therefore have more of a hold?
I know what you mean. The biggest irony is that pack theory advocates often accuse positive folks of treating dogs like people when they are attributing much more of a human level of understanding to dogs. For example believing that they are actively intentionally "misbehaving" when they are just being dogs and doing what works.