The ethics of positive reinforcement training

[Joy]

I had already watched the Kamal/ Nando video and intensely disliked it. I've actually watched it again and still feel the same.

Nando and Jo-Rosie sounded as pleased as punch to be 'not afraid to say no' when their dog went to take food from their child's hand. Well I think as trainers who charge people a lot of money for advice they should be ashamed of not realising their own dog would be likely to do this and putting the dog somewhere else while the child ate. Of course I have occasionally shouted at dogs (though I have to say rarely at Molly because the more unacceptable you find it the less likely you are to do it - and of course I'm retired so less stressed!) but it's not a way of training or improving the dog's behaviour, it's just an involuntary human response.

Training with +R ( and ok I occasionally use -P such as removing attention) doesn't mean that the dog is allowed to behave in antisocial ways. Nando claims that he is seeing more 'brattish' behaviour from dogs - I'm not, most dogs I meet are happy and well adjusted.

I'm not convinced that using punishment as a means of training is either more effective or quicker. Anecdotally, the Flatcoat that I worked with for a year on not jumping up had spent 4 years being pushed down and shouted at / told no to no effect. My method wasn't very quick but it worked. My first dog as an adult was a rescue Springer who had had 3 homes before me, her only sin being dreadful food stealing. I knew very little about dog training and gave it almost no thought and I shouted and (to my shame) even smacked her once. It did no good. Thankfully I'd had a kind upbringing where being kind to our dogs was built in so I simply managed it by not leaving food out/ putting the dog in another room and when my son was born and started solid food I put him in a high chair to eat. Subsequent dogs, as puppies, I trained not to steal food.

I'm not even convinced about using force as a last resort, as with the dog forced out of its kennel. In that case they say it worked with no unwanted consequences. I think they were lucky and in many cases the dog might have bitten them or become totally shut-down.

I haven't yet watched the Jane Ardern video.

 

[Joy]

most people I know do NOT understand reward based training and need to be taught in a way that’s easy to understand.

I think that Kikopup is still one of the best online resources for clear easy-to-understand reward- based training advice.

I too felt irritated by John McGuigan's recent fb pages. I can see he was trying to stir up debate/ discussion but it was alienating a lot of people. And the 'sit' thing was daft. He was just trying to make people think about when you need to cue a sit, but actually I doubt that most pet dog owners overdo it in real life (I don't ask for a sit at a kerb for example and see very few dogs that do) but I think it's useful to be able to get a sit if you need it.

My dogs have benefited from me being able to learn HOW to teach them in a rewarding way, it is the only way I feel comfortable teaching them.

I'm glad!

 

[Cupcase]

OK I've checked and could someone please clarify that Kamal Fernandez is a force free trainer? I listened to his comment on how he got his adolescent male dog who was dog/dog aggressive to change from being reactive to other dogs. I really cannot believe this claim.
This comment is made from 37 minutes in.

 

[snowbunny]

OK I've checked and could someone please clarify that Kamal Fernandez is a force free trainer? I listened to his comment on how he got his adolescent male dog who was dog/dog aggressive to change from being reactive to other dogs. I really cannot believe this claim.
This comment is made from 37 minutes in.

Kamal is a crossover trainer - meaning he used to use aversive-based training and has moved over to positive-reinforcement based training.

I'm not even convinced about using force as a last resort, as with the dog forced out of its kennel. In that case they say it worked with no unwanted consequences. I think they were lucky and in many cases the dog might have bitten them or become totally shut-down.

Well, we all force our dogs to do things, there's simply no getting around that. We have to; unless we're incredibly lucky and we are able to get in all the cooperative care they need to be able to consent to any procedure before it's necessary. But can we do that before the puppy has his first round of vaccinations? Or before he develops an ear or eye infection at a young age? Even nail trimming needs to be done with some dogs before you get to the point where they will consent to it. If they happen to develop a fear of something that is incompatible with reasonable life; such as if your dog is afraid of going out the door, yet you have to take them out to toilet. If they're afraid of the car, but you have no option but to take them places in it. We can all work on these things to help the dog through them, and put cooperative care plans in place, but in the meantime, there are times where the dog has no choice, and we have to impose procedures on them. That's force.

I don't think you can say they were lucky. We shouldn't reduce the story in such a way; we don't know the experience of the handler in reading this dog's body language. There was a history with the dog, so they would have known the likelihood of it biting. It sounds like it was already shut down. If they know the dog - and the process - well enough to believe that flooding has a chance of working and saving the dog from extended suffering, then I think it would be unethical to NOT try it.

If we totally close ourself off to the idea of using methods that we wouldn't normally resort to, then we are falling into a trap that aversive trainers will throw at us - the story of the dog who is euthanised because his owners don't have the skill to fix his reactivity with positive methods, so he becomes walked less and less frequently, until he is locked away. Behavioural issues worsen, unskilled trainers come in and try to fix him with positive methods, but it's beyond their skill level. The world is FULL of crappy trainers (on both sides of the divide) who simply don't have the skills to address these challenging problems. In fact, I'd challenge that most trainers "on the street" are pretty limited. As the dog's behaviour worsens, they enter a downwards spiral, until the dog bites someone and is euthanised. The platitudes are "you did everything you could". But that's simply not true.
Again, I will reiterate that I do not condone using punishment-based techniques and I don't use them with my own dogs. But, I think it's totally narrow-minded to say that R+ alone has all the answers for every dog/handler combination in every situation. Because of the skill, the time and the energy that needs to go into it.

Let's face it, if I had told Shadow off for reacting to other dogs on lead, we'd have sorted that issue almost immediately. Would it have fixed his emotional state? Nope. But I also don't think it would have had any serious fallout. Because I don't let him meet other dogs on lead, it would have made our walks down the street a lot easier straight away. Is it fair to expect an inexperienced handler to put in the many, many hours (weeks/months/years) that are necessary to change the dog's emotional state, when so much of that is out of your control, and all the time the walks are becoming more stressful for both parties?
Again, to be perfectly clear, I am NOT saying that telling your dog off for reactivity is the answer. I am simply saying that for this one single dog that I know intimately, it would have been a quick fix for his leash reactivity. The fact I'm still not prepared to use it should speak for more about my ethics than the fact I am open to discuss the alternatives, although that's probably not what most people will pick up on
Not only that, I do wonder whether my choice to use only R+ in this situation has been the most ethical. It's easy to say "of course", but let's think on it. Shadow reacts on leash for a couple of reasons, but one of them is fear. He is telling the other dog to stay the hell back. This behaviour is negatively reinforced by the other dog retreating. If I told him off, he would find that aversive enough to suppress his behaviour. We know the behaviour is driven by emotion, and that hasn't changed - and this is where the idea of escalation comes in. The thing is, I am experienced enough to understand that his emotional state is still the same, so I would still be sympathetic and not pushing him too far. Yet, every time he walks past another dog, doesn't react, and the other dog ignores him, it's making him realise that the shouting isn't necessary. He gets the same negative reinforcement (the dog retreats) without the reactive behaviour. In fact, he's being reinforced for expressing his calm behaviours. The more this happens, the more his emotional state will change to one of indifference because these dogs are no longer a threat.
So, which is the more ethical path? The one I have chosen, or the "sliding doors" one? I'm undecided.

The danger in these discussions is that people believe they "know their dog" when they don't have the understanding of body language or psychology to back it up. So, someone reading this might say "I know my dog, I'm gunna try it!" when they just don't know what they don't know.

I don't have the answers, far from it, but I feel there is something missing from R+ at the moment. It's all far too left-brain.

 

[Joy]

I just watched Jane Ardern but I didn't think she said a lot: we need to have well-behaved dogs. Well, yes, of course.

 

[Cupcase]

( I need to learn how to insert specific quotes, so I hope you understand this)

@snowbunny , thank goodness someone else who has asked the same question I have asked myself, " I wonder what would have happened if I had put a negative into how I handled Cupar the first time he reacted to a dog', I witnessed a rescue dog with an inexperienced owner carry a bottle with stones in it to shake at the dog every time she reacted to another dog and two years later this dog goes everywhere with her and plays nicely with other dogs.
So listening to Kamal and his use of an aversive and changing the dogs behaviour for the better, makes me question myself once again.
Disclaimer, I felt nothing but relief when I found R+ training,I had never felt comfortable teaching my dogs by yanking on leads, rattling bottles at them, this is why I had such badly behaved dogs, as I couldn't bring myself to yank on the lead.

My opinion on the failure of the training of a lot of dogs, is not the fault of the type of training but the lack of it.
For example, couple get a spaniel puppy, they go to training classes once a week, they're using a clicker (wrongly). All the dog hears is dogs name and added to the end no. I don't believe they were taught that at training. The dog at 18 months old never gets walked and has a ball thrown in the garden every single time she's out. The dog has not been trained in anything. She has zero recall, it's scary witnessing this, she barks at everything, the odd occasion she's been 'walked' up the road, she's pulling like a steam train.

My point is, can we blame the dog trainer for that? Did the trainer make it clear that the training she's teaching must be carried out at home?
I wonder how many professional dog trainers, are just that 'dog' trainers. The owner needs to be the 'dog' trainer, the teacher needs to teach the owner on how to teach the dog. How many trainers are actually good teachers to the owners?
To train positively is a long slow process for most of us and unless the owners put in the time training there own dogs, yes R+training is in for a harder time, as more and more people switch over to it.

 

[Joy]

Well, we all force our dogs to do things, there's simply no getting around that.

Yes, you're right. And yes, maybe they did know enough about the dog forced outside to predict a good outcome. What I disliked was their apparently happy adoption of these methods and their appearing to advocate their use.

 

[snowbunny]

How many trainers are actually good teachers to the owners?

This is the absolute crux of the matter, and has been a bugbear of mine for years. I always hark back to skiing, because what I learn through my ski teaching is so applicable to the dog training community, too. “Dog trainers”, by which I mean the professionals, need to think of themselves more as teachers. They need to work not just on their training skills, but on their teaching skills, too. To be honest, that should be more important than their training skills! Most of what your average dog trainer sees is the same old stuff (same with ski teaching), and the great “dog trainers” are the ones who have a range of ways to teach that stuff, who can adapt what they are doing, and how and what they are doing sand saying, to cater for the individuals in their class. Not just “this is how we will be teaching xyz”, but getting into a dialogue about it.
The problem with this is that it takes a lot of learning on the teacher’s side. Learning to read the students and notice how they learn best (because asking them doesn’t work, interestingly). Then adapting your class to make it a class full of individuals that you can teach differently. It’s really not easy, but is what “dog trainers” should be trying to achieve.
If I taught classes, I would provide hand-outs before a class on a certain topic before that class, and expect my students to read it before class, because then they’re not getting the theory and the practice all thrown at them in one go - anyone can understand how that could be terribly confusing and frustrating. And I would have a FB group where I would encourage them to ask questions between classes, and share videos if they want.
But this is all time consuming and when you start doing things like this, your hourly rate goes through the floor.

The other thing is that many, many “dog trainers” aren’t people-centric at all. That’s why they spend so much time training dogs
But I think they need to start recognising that their ability to communicate to their clients is far more important than the fact they can train their dog to weave backwards through their legs, while whistling a tune and smoking a cigarette. They should be putting at least the same amount of professional education into that part of the equation as they do the understanding and training parts.

 

[Selina27]

@Cupcase , I completely agree with your point about how good or not the "trainers" can teach the dog owners. I think that some people, who maybe are good dog trainers, do not have the skills to educate the ordinary pet owner. I went to a "force free" class with Cassie at 7 months. I didn't like the woman at all, although I did in the main like her methods. But she made me feel very inadequate, I was after all learning a new skill and very keen to do so, but I found her very impatient. Luckily I had found a certain Forum by then and had plenty of support from there.
The trainer I go to now appears to like the people who come with the dogs, and seems to genuinely get pleasure from seeing dogs and their owners work towards their potential. If she doesn't she makes a very good show of it!

 

[Selina27]

Haha @snowbunny , double posted.! You absolutely have that right.

 

[Selina27]

I just watched Jane Ardern but I didn't think she said a lot: we need to have well-behaved dogs. Well, yes, of course.

Pretty much sums up what I thought too.

 

[snowbunny]

Haha @snowbunny , double posted.! You absolutely have that right.

You were slightly more succinct than I, though. Quelle surprise

 

[Selina27]

You were slightly more succinct than I, though. Quelle surprise

I don't know about that, I just sometimes can't get all my thoughts sorted out in my head well enough to type them out, I think your post is more thorough!

 

[snowbunny]

I’ve been thinking more on this. This topic is good for me, because it’s not something I’ve ever considered before, and I want to make sure I’m going about things in the right way. As some of you know, I’m in the process of writing a book. Whether or not it ever will ever get published, who knows, but this thought process has made me think very hard about whether I’m being hypocritical in writing this.

If I were in the position of teaching pet owners to train their dogs, I would not be happy with them using aversives, even mild ones. This is especially the case in a group class, where people might see something being explained to one owner and take it on board for themselves, when it’s entirely inappropriate for their situation. I see this all the time when teaching ski groups, so you have to be very clear when you’re giving one person advice that it is for that individual and could be counterproductive with others, otherwise everyone starts trying it.

So, if positive reinforcement can be difficult to get right, what can be done? Well, I decided early on that my book won’t be technical in the slightest. People don’t need to know that stuff in the early stages. I don’t even mention positive reinforcement, force free training or anything like that. What I focus on is quick wins alongside longer term tactics to bring about more robust behaviour change.
So maybe that’s where pet-focussed dog professionals should be looking, too. Give those owners quick wins and describe how they can be used in a variety of situations. Things like the reorientation pattern game is super easy technically, gives the handler something achievable they can do to get the dog focussed before they start a walk, when they arrive at training class, if there’s another dog around etc etc etc. That’s one part of the puzzle that seems missing, explaining a range of applications for a tactic and giving the handlers the task of going away and practicing it in those different scenarios.

So, how about we use this thread to think of some solutions for people who are trying to teach clients R+ methods to train their dogs. What tactics in classes have helped you the most in the real world, what did your teachers do or say that was really useful, and if you could give them some advice to get more people buying in, what would that be?

 

[Cupcase]

See I can't add to advice a dog trainer has given me in a positive way. The last dog trainer asked me not to bring my dog back.
I do have opinions on what I would now require in a dog trainer. None of the trainers in this area meet the criteria.

 

[Selina27]

Food for thought Fiona @snowbunny , I'll get my thinking cap on and get back to you

 

[Boogie]

Some of our trainers talk too much. I think it would be good if they could send info with the explanations on in advance. Then the sessions wouldn’t involve so much waiting and listening for us and the pups.

The best thing I’ve ever been taught is loose lead walking - she demonstrated first then let me practice while she watched, showing me exactly where to feed the pup and how much. She said to do this for three weeks before giving less food. It worked brilliantly. Now I need to remember it for the next pup.

Recall has always been 100% with my pups too - the ‘hiding as often as possible when they are tiny’ method has worked with every pup so far.

I am useless at teaching mine to settle in cafes - I’m going to ask for special tuition in that area for the next pup. Spencer plays me like a fiddle for attention in any situation where he’s meant to be settled

 

[kateincornwall]

Pretty obvious really but I would always advise not to set expectations too high , allow mistakes because in order to learn, our dogs will make mistakes . Never set the bar too high ( which has always been my problem ) try not to set your dog up to fail by putting him/her in uncomfortable situations before they are ready and be happy to make the grade slowly but surely x

 

[Natalie]

try not to set your dog up to fail

I think this is really difficult for complete beginners to do, and I include myself in that category. It sounds obvious, but when you are starting to shape a behaviour, how do you know what your dog can do in the first session without them failing? I cannot get Monty to walk backwards using positive methods, I've lured and shaped but he either lies down or turns in a circle. So each time I'm setting him up to fail and I have to change to something new to reward him for a behaviour he can do. So if I had a trainer, I'd want them to explain to me what this really means in practice and how to avoid 'failing'

 

[HAH]

I’ve got quite a few of these but will have to trawl them up from the depths so it might be a bit piecemeal as they come to mind

The main thing which has taken me a long time to properly ingrain is ‘deal with the dog in front of you’. This gets over so many new owner/new to training hurdles of guilt, ‘he did it in the garden’ type feelings. It goes hand in hand with ‘dogs aren’t robots’. Those are both philosophies rather than tactics, but perhaps useful for getting owners in the right frame of mind.

I’d say games like ping pong, and having a strong boundary are really useful to build from. And simple things like foot targeting and nose touches are a neat way to test engagement/arousal levels and develop that training mode - sorry to quote Absolute Dogs and compatriots but they seem to constantly refer to ‘conversation starters’ which now makes sense to me (seemed another slightly naff label when I first heard it).

Bouncing back to useful ideas, I think learning about arousal to describe excitement and other forms like fear was really helpful for me, in terms of understanding how that was the thing to deal with in the moment regardless of the cause (to some degree).