This post is part of the series in response to Dunbar’s 2012 Australian seminars. See index.
Dunbar said: “We have lost words from training.”
Dunbar is saddened by the use of “buzzes and clicks” in dog training. He believes that words and language (verbal feedback) are efficient, descriptive, and ‘far better’ at gaining, maintaining, and rewarding behaviours. He thinks that language is the ‘feeling’ that is needed in dog training.
The Problem with Clicker and Shock Collar Trainers
Dunbar criticised clicker and shock trainers mostly for their inability to fade clicking and shocking. Clickers click and treat forever, zappers zap forever, and all the while the process is sterile and unrich (according to Dunbar).
To Ian, a click isn’t rich enough to convey levels of ‘good’ or ‘bad’. He described food and zaps as quantum (discrete) while verbal feedback is analog (continuous). According to Dunbar, both schools of thought (click vs zap) are inefficient at conveying degree of correctness. [Personal comment: I disagree. I think the rewards used following a click convey the degree of correctness to the dog. For example, the use of jackpots and mini-jackpots.]
Ian said he does use clickers, for things like shaping body positions (e.g. ears up, tail up, etc).
Verbal feedback as punishment
When Ian Dunbar talks about using punishments in dog training, he is referring to verbal feedback. This verbal feedback can be used to define the degree of incorrectness and urgency, and also redirect the dogs’ behaviour. (Read more about Dunbar’s thoughts on punishment.)
Further reading:
Patricia McConnell looked at two studies that found that clickers were better than verbal markers, but silence (no markers) were better than clickers! (Of course, Ian Dunbar would argue that teaching puppy owners about clickers is not an efficient use of training time – that is, dog owners struggle to use clickers effectively as experienced trainers.)
Steve DeBono on DogStar Daily goes into this with more detail. He talks about what daily life looks like, using frequent verbals, and how communicating with his dogs this way is natural.
Added 12/10/12: For further viewing, you may be interested in this interview between Zak George and Ian Dunbar on verbal feedback. Note, I haven’t watched it yet! But it was recommended in the lecture notes and seems relevant from the first five minutes.
I don’t think it has to be an either/or proposition, although I’ll admit I’m not a clicker purist. I use a mix of clicker and verbal feedback, both in the sense that I will talk after clicking, and that I don’t always use a clicker and use my voice instead. I started using a lot more verbal feedback after going to a Denise Fenzi seminar. It’s a style that works very well for me and my dog.
That said, I feel like the “degree of correctness” is a bit of a red herring. With clicker training, either the dog meets criteria or he doesn’t. It’s an either/or, black/white, wrong/right sort of thing. Ideally, if you’ve sliced your criteria thin enough, you won’t need to “cheer on” good tries.
Of course, I do cheer on good tries, and I do use my voice after the click if the dog has made an exceptionally good effort. Like I said- it doesn’t need to be either/or. There is a lot of flexibility in training.
Crystal, I agree with you. I don’t think that clicker training and verbal feedback are mutually exclusive. I think you can use bits and pieces depending on what you’re wanting, and still get pretty good results. And, yes, if you select your criteria fine enough, then you are shaping a correct behaviour – you don’t need to show a ‘degree of correctness’.
I normally use clickers in a pretty pure way when I do them, or otherwise use them for refining already taught behaviours (e.g. if the dog is being a bit imperfect with their heelwork, using C+T to pinpoint the place of ‘right’).
Thanks for introducing me to Denise Fenzi through your blog. I really like what she does and I follow her blog with interest.
“.. clickers is not an efficient use of training time – that is, dog owners struggle to use clickers effectively as experienced trainers.”
I agree with that! When teaching families it’s much better to let them use word as marker.
Jette, to me, this is one of the most important things I learnt from this seminar! In the nicest possible way (… mostly…) Dunbar explained that most dog owners are idiots, and we need to make things easy for them. I don’t think clicker training is the best tool for most dog owners – there’s too much learning for a dog owner to do, while it’s their pet that really needs to learning!
I guess on a novice’s (my) perspective, in the beginning, clicker training makes it easier to pick-up the timing rather than verbal cues then reward. I used to clicker train because it was easier to provide approval than saying “good dog” and then treat.
Eventually, I did away with the clicker since Ginger and Peanuts were still able to follow through the training without it. And it drove Coal crazy because he knew that every time the clicker clicked there was a treat to follow. Even when he was so far away from us, he could still hear it as if he were beside us. And would bark excessively after hearing the distinct sound.
Huggies and Cheese,
Haopee
Thanks for sharing your experiences, Haopee. Dunbar argues that most dog argues have trouble getting a behaviour, then clicking it, then rewarding it, and it all ends up in a big mess. I’m sure some trainers can work it out pretty well, but I think many probably get uninspired along the way (or decide that training dogs is ‘too hard’ and quit).
It’s nice to hear that you got what you wanted out of clicker training, though. I use clickers quite often for puppies, and then occasionally for adult dogs. I have no fear of breaking the clicker out, and knowing how to use a clicker is probably a good thing for pet people… Just probably not the first thing to teach them. 🙂
Pingback: Dog training doesn’t happen in a laboratory! | Some Thoughts About Dogs
I don’t typically follow a single type of approach when it comes to dog training. I use a combination of treats, positive reinforcement and traditional – especially if necessary. There are some dogs who are a little bit “difficult” so a little bit of force is necessary.
Dog training is very similar how we discipline our children. Some forms of more “rigid” training are necessary so they will obey while others respond a lot better with nurturing and calmer techniques. To me, it is best to get to know your dog and do what’s necessary.