Do dogs generalise badly?

Re: Do dogs generalise badly?

[quote author=Oberon link=topic=7888.msg111930#msg111930 date=1411386617]
See, I think training is all science. It's based on established principles. The principles behind it are quite straightforward. Yes, some skill is needed - mainly about good timing and observation, and some people are a bit better at that. But it's not an art. It's a science and it's done by systematically following known rules and steps.
[/quote]

I agree it is all science but historically viewed as art and today I would still say there are people for whom it is the unconscious application of that science that is supported and proven by the things they've observed in dogs for many years. They would not perhaps consider it science themselves and to speak in purely scientific terms potentially alienates both long term practitioners and newbies

 

Re: Do dogs generalise badly?

Wow I just want to train my dogs. ;D

I think sometimes this training lark can be made to sound hard and complicated. The more you learn a subject the easier it should be but it seems with dog training the more you look into it the more complicated it gets.

I'm not on Facebook so I've only read Rachael's version and I could've got the wrong end of the stick but it sounds a bit high brow, cliquey and possibly off putting for some. Maybe the 'experts' have difficulty generalising as well. Too much terminology but all meaning basically the same thing. :

 

Re: Do dogs generalise badly?

The people on the Clicker Revolution facebook site do, I think, have a different style. There doesn’t seem to be a consensus about whether they are talking about a different way of doing things, or just using different language – so it’s a bit odd that people disagree with them, really (you have to understand before you can disagree, surely).

I think they are just discussing, and exploring, how they should understand things. This is a useful and helpful thing to do. Well, if you enjoy that kind of thing, and not everyone does – some people are quite content that we have a perfectly adequate (for most things) explanation for gravity and find it intensely annoying, and time wasting, that others find it fascinating to discuss why we don’t all fall off (and all the more so if those people are without a PhD in physics). ;D ;D ;D

[quote author=Jen link=topic=7888.msg111968#msg111968 date=1411400309]
it sounds a bit high brow, cliquey and possibly off putting for some. Maybe the 'experts' have difficulty generalising as well. Too much terminology but all meaning basically the same thing. :
[/quote]

No, I don't think it's like that at all - they just have a different style and are enthusiastic.

[quote author=Joy link=topic=7888.msg111932#msg111932 date=1411387386]

1. They seem to suggest varying the trainer's body-language and surroundings from the word go rather than starting in your garden and then working in other places.

I don't believe that the clicker trainers on the facebook group would really expect a puppy to do a sit/stay for the first time for 5 minutes in a busy park. I bet they do gradually lengthen the time they ask for this behaviour - the D for duration.
[/quote]

I missed anyone saying either of those two things.

 

Re: Do dogs generalise badly?

You're right. I've read it all again and realise I misunderstood some remarks. In fact one poster specifically says that the behaviour is taught in one environment before moving to another and to make very small changes to start with.
Still don't quite see the problem with the idea of 3 Ds though. I think they're saying that a dog should be so sure of what the cue is that it will automatically understand it in a new environment and that you teach the dog to be sure of the cue by varying everything else. But surely 'everything else' includes distance, duration and things going on around the dog, ie distractions.

I think they're an interesting bunch of people who obviously care about their dogs.

 

Re: Do dogs generalise badly?

I think Kay's answer to Jan makes it a bit clearer - that it isn't a group for novices, it is a group for those who have already read Kay's book. A place to bat about ideas and concepts, rather than a place to help newcomers learn.

I also agree that it is important to discus and explore these issues.

However, I think the promotion of better dog training methods is a really important cause, and those of us involved in doing so need to present a clear and agreed understanding of the principles and techniques that will get the job done.

I'm not quite sure how we balance these two aims (exploring v teaching)

 

Re: Do dogs generalise badly?

You don't need a phd to discuss gravity you certainly shouldn't need a phd to discuss and train your dog. ;D

I'm all for discussion and ways to improve on what is already known but I don't see the point in making things more complicated. Rachael's explanation on DDD was simple, easy to remember and works. From what she has said the Facebook group are discussing a similar concept, proofing, but using different terminology and it sounds more complicated for the novice. Catchphrases may not be the ideal but they make what could be a complicated concept available to all which surely is what's important. However if the group isn't for novices or a place to find help then what does it matter. :

Sorry it was trained in to me as a primary school teacher to believe all learning should be inclusive not exclusive. :

 

Re: Do dogs generalise badly?

I know that Pippa has said this before and I completely agree - trainers (people teaching people to teach their dogs) need to use a common language so the learner aren't confused. If you are trying to learn a new discipline nothing is worse than everyone using different terms (and then not being able to explain those terms...).

The reason I don't really like the term 'art' when applied to animal training is that 'art' can imply that there is a mystery to it, that some people are born with the magic touch, that it's all about feeling not thinking, that you just have to 'know how to do it'. That makes newcomers despair. They wonder how will they ever succeed at something so elusive. The truth is - it is not elusive at all. It's really simple - give the dog something it wants when it does something you want. Repeat.

Most people don't want to do anything fancy with their dog. They want it to walk on the lead without pulling their arm off. They want it to come when it's called instead of running away. They want it to play nicely with other dogs. They don't want it to knock over their Grandma. They might want it to wait to be told to eat. That's about it. None of that requires anything special to train. Newcomers just need to understand that there are several basic steps that can be applied over and over to train any behaviour.

 

Re: Do dogs generalise badly?

[quote author=editor link=topic=7888.msg112016#msg112016 date=1411408783]
I think the promotion of better dog training methods is a really important cause, and those of us involved in doing so need to present a clear and agreed understanding of the principles and techniques that will get the job done.
[/quote]

[quote author=Oberon link=topic=7888.msg112075#msg112075 date=1411424245]
Newcomers just need to understand that there are several basic steps that can be applied over and over to train any behaviour.
[/quote]

I wonder, what might the implication of this be? That this thread, or the discussion on the facebook page, was inappropriate because newcomers might be confused?

I have to say, I very, very, much consider myself a newcomer, an absolute novice, yet feel able to cope with - and enjoy, with interest - a discussion that uses unfamiliar terms and may be a little unusual. I can always go back to following "recipe" type manuals if (perhaps when) I get confused.

Or is it protecting people like me from myself? ;D ;D ;D Which may be a fair concern. ;D ;D ;D

 

Re: Do dogs generalise badly?

You are definitely far, far from a novice

When I say newcomers I mean people who have never had a dog before and have not read anything at all about how to train it before getting the dog. That describes a fair proportion of the people starting out at the dog club I go to. They need it clear and simple.

That doesn't mean that we can't have in-depth discussions about principles, terms, approaches But I think that the fundamentals of how learning happens are known, and there is no need to go inventing a whole new jargon for it.

 

Re: Do dogs generalise badly?

I totally agree with you about the whole art, innate skiills side of things and the best trainers I've been to are good teachers of people. I've seen some great and successful dog trainers who don't have good people skills. I do think that we can learn from them though if you can understand how dogs learn as a student you can bridge that gap.

I'm afraid I enjoy the mental gymnastics of semantics and communication of ideas so, a bit like Julie, get pleasure out of these kind of discussions. It helps me to think about things and refine my understanding. More often than not the answer (in our case) is that the detail doesn't really matter as long as we're having fun ;D but I still like the discussions!

 

Re: Do dogs generalise badly?

[quote author=Jen link=topic=7888.msg112066#msg112066 date=1411418135]
You don't need a phd to discuss gravity you certainly shouldn't need a phd to discuss and train your dog. ;D

I'm all for discussion and ways to improve on what is already known but I don't see the point in making things more complicated. Rachael's explanation on DDD was simple, easy to remember and works. From what she has said the Facebook group are discussing a similar concept, proofing, but using different terminology and it sounds more complicated for the novice. Catchphrases may not be the ideal but they make what could be a complicated concept available to all which surely is what's important. However if the group isn't for novices or a place to find help then what does it matter. :

Sorry it was trained in to me as a primary school teacher to believe all learning should be inclusive not exclusive. :
[/quote]

I totally agree Jen, I have been reading this thread hoping to learn something but I just find it confusing, over complicated and exclusive. Why can't training just be simple and to the point, all this going round the houses is unecessary? Heidrun said it goes over her head so what hope do the rest of us have : x

 

Re: Do dogs generalise badly?

[quote author=bbrown link=topic=7888.msg112095#msg112095 date=1411453888]
I'm afraid I enjoy the mental gymnastics of semantics and communication of ideas so, a bit like Julie, get pleasure out of these kind of discussions. It helps me to think about things and refine my understanding. More often than not the answer (in our case) is that the detail doesn't really matter as long as we're having fun ;D but I still like the discussions!
[/quote]

Absolutely I enjoy it too. I'm hugely in favour of this kind of discussion.

My only point really is that on the FB thread there seemed to be a bit of a move towards a new set of terms and a rejection of established principles, when actually in practice what the posters seemed to be doing was quite consistent with those established principles.....i.e. keep your cues consistent and clear etc etc. It was just a bit confusing and jargony and generally seemed to be purporting to emanate from some kind of higher plane, that is all However, it was a private group and I asked to joint it and was very kindly allowed to, so I should not complain!

 

Re: Do dogs generalise badly?

Just a suggestion, as this is a private group on FB maybe it should remain there to be discussed amongst the group and not on the forum so that some of us don't feel quite so confused :-\ x

 

Re: Do dogs generalise badly?

[quote author=Oberon link=topic=7888.msg112102#msg112102 date=1411456290]
My only point really is that on the FB thread there seemed to be a bit of a move towards a new set of terms and a rejection of established principles, when actually in practice what the posters seemed to be doing was quite consistent with those established principles.....
[/quote]

Yes, I think so. I think it's deliberate though - I think it's a way of inviting people to think through the principles again. It's a bit of a theme in this collection of material.

 

Re: Do dogs generalise badly?

Just to drag this thread right back to the beginning ..and apologies for not reading this the whole way through! I have just realised that I have just applied this assumption..ie: dogs generalise badly, to Benson.

We started out by clicker training the retrieve using a tennis ball in our lounge initially as a distraction technique,(please do not de stuff that cushion, this is a much better game... ;D ) and the plan somehow formulated from there, which was: can I really train a dog to retrieve AND deliver to hand with a clicker? So using the same language, the same tone, making it very easy to succeed...and adding a new cue for when it is not right..."oh dear". My gut feeling is the challenge is not the throwing and bringing back..(he is a retriever after all), but where this happens, again reflecting back on some points about "noise" around him, meaning the actual environment is too stimulating/interesting. I liken this too when you have those "spot the difference" pictures...the less information in the photos..the easier it is, I guess with dogs this becomes 3 dimensional, as someone pointed out dogs are "observers". This overload of information translates into excitement, so we work at reducing the excitement and the dog starts to learn.

Benson is not only retrieving in a new off lead place...but blow me..he hunted too, and found that ball! I shouted "lost" and he went head down and found it, brought it back to my hand, not a very complex retrieve but for him it was a lot to ask, bearing in mind all the new and exciting information he was sifting through at the time. Now I will stop waffling there because I have remembered something from Julie post elsewhere re: Whipit training and time to mark..

 

Re: Do dogs generalise badly?

I’m not sure what the answer is. And I don’t want you to feel that I am criticizing you for starting this thread Julie, because I am not and I love talking about stuff like this.

But, and it is a big but, I feel that we are only just beginning to get the average dog owner on board with some of the concepts of behavioural science. And I do worry that more ‘public philosophising’ over concepts that we have only just got people on board with, may be counter-productive in this respect.

If there were a problem with teaching people the concept of generalization, if it was being misused, or leading people up the wrong path, if dogs were being trained less successfully or being harmed through this concept, then definitely – we should be hauling it over the coals and diving in to fix it.

But it isn’t broken.

Teaching people how to proof their dog’s recall, and other skills, transforms their lives. It really does. And to undermine the concepts on which proofing is founded – the understanding that dogs don’t generalize easily, and that we therefore need to do a certain amount of re-training in order to help them learn how to behave in different situations – well I just hate to think that could all be undone, by shaking people’s confidence in what they are doing.

Maybe we should have a section of the forum, especially to discuss behavioural science and the philosophy of dog training. A place to chuck around some of the more esoteric ideas behind what we are all doing with our dogs. I certainly don’t want to drive this kind of discussion away, it’s interesting to some of us, including me.

Or maybe I am just making too much of a big deal out of it, and should assume most people simply won't read this kind of stuff and therefore won't be affected by it.

I don't actually have the answer, just thinking out loud.

 

Re: Do dogs generalise badly?

That is really great Benson sounds like a natural!

How much 'proofing' is needed to get a dog to retrieve in different environments is a reflection of the value the dog places on the retrieve. For example, I have hardly had to proof Rachel's retrieve at all. There is virtually nothing in the world worth more to her than her precious dummy With other dogs, I have had to proof much more carefully initially, from indoors to out, and especially in the presence of other dogs and the greatest challenge of all, retrieving a dummy in the pheasant pens or a rabbit pen.

 

Re: Do dogs generalise badly?

I agree that training dogs is a science not an art. Follow the simple rules and it works for anyone. However I do think, like with a lot of things, some people have a natural aptitude towards it and find it easier than others. A reason I think to find the simplest, most effective method of training.

Something that occurred to me about looking at dog training as a science and that's the ever changing variable. The Dog.

It can be made as scientific as you like with scientific jargon and terminology and a long list of rules to get you to what is essentially the same result a trained dog but all dogs are different. Dogs learn at different rates, have different ideas about what's rewarding, different skills. Dogs can't read the list of rules they work on instinct. This science needs to be adaptable so ALL owners can adapt it to suit the ability and individuality of their dog. Another reason I think to find the simplest, most effective method of training.

I think it's a very interesting topic to discuss and only by discussing can new methods (if there are any) be found. My concern is [quote author=Oberon link=topic=7888.msg112102#msg112102 date=1411456290]
It was just a bit confusing and jargony and generally seemed to be purporting to emanate from some kind of higher plane, that is all
[/quote]

 

Re: Do dogs generalise badly?

Jen, you nailed it brilliantly. Thank you xx

 

Re: Do dogs generalise badly?

Jen, Helen - not sure quite what to say! Perhaps nothing is best in the circs.