I’ve been watching a lot of World Dog Training Motivation Transparency Challenge videos this morning. If you haven’t heard of this, it’s an idea from the very wise Jean Donaldson. Dog training has picked up so many coy terms for punishment that it can be hard to tell when a trainer is using even heavy-handed aversive methods. So, Jean Donaldson created this set of questions (roughly paraphrased here) that should be asked to clarify any trainer’s position:
1) What happens if the dog gets it right?
2) What happens if the dog gets it wrong?
3) Is there a less invasive way to proceed?
Professional and amateur dog trainers have been posting videos explaining their stance on these three questions and challenging each other to do the same.
I’ll post the videos I watched this morning below, all of which are from trainers I respect:
First, Grisha Stewart. Her incredible sympathy for reactive/fearful dogs has been an inspiration for me, and her answer to the last question is really wonderful:
Second, Michael Baugh, whose video wins for beautiful production values + sound dog training:
Third, Thomas Mitchell, a great up-and-coming dog trainer from the UK. You may have heard of the Absolute Dogs Training Academy that he runs with Lauren Langman. Tom gets bonus points for fitting the Premack principle into his discussion of rewards (and for chopping off his head in the video, because I always do that, too.)
Finally, I also love Eileen’s video (from Eileen and Dogs), for general awesomeness and for acknowledging that we do stupid things and sometimes need professional help:
What I found most interesting in these videos was their answer to “What happens when the dog gets it wrong?” The answers vary slightly in their specificity, but the general theme is that if the dog makes a mistake it is your fault for asking too much, and we shouldn’t do this.
Let me just say: I agree with the first part of that. I’m not so sure about the second. In fairness, I absolutely believe these trainers would give more nuanced answers over a cup of coffee in a chat with an experienced dog person, and some of them do give more details. The whole point of the Dog Trainers’ Challenge is that it is intended for the general public. So, I’m using my luxury of space and audience here to ruminate.
Is constant success our goal?
If you really want a dog to be successful 100% of the time, you can arrange that. It’s absolutely important to remember both that we can help our dogs succeed and that we can inadvertently make them fail. The more carefully you think about the demands of your training environment, including things like distractions, the better off you’ll be.
But when was the last time you played a game you could always win? If you completed every level of Candy Crush on the first try, would you still be playing it? If every slot machine turned up a jackpot on every pull (and thus could only pay back exactly what you put in), would casinos even exist?
I’m not saying that you should constantly overwhelm your dog, but dog training with zero challenges is boring. Your dog is smarter than you think, and you’ll never know that if you’re obsessed with always setting the dog up to succeed.
As positive trainers, we can let failure be a big bogeyman. “I don’t know what to do if my dog makes a mistake! I can’t correct him?!” The key is to have a plan. Before you start training, ask yourself what you’re going to do if, within the context of this exact training moment, your dog makes the wrong choice. You’ll find that you have more options than you think.
My Imperfect Dog 
Great post – very interesting. Thanks for sharing.
LikeLike
I loved these three questions. I’ve had a hard time explaining to acquaintances how to tell if they’re using the right trainer. This is very simple.
Thanks for pointing out some of the challenge videos.
LikeLike
Aren’t they great?
I’ve developed a list of trainer website code words that mean “leash corrections/e-collars ahoy!” And they are, sadly, things like “obedience,” “attention,” and “listening.” AKA, things we all want to at least some extent and that sound completely innocent.
LikeLike
I feel weird commenting because I’m a blogless lurker, but I can’t resist, especially since I watched a bunch yesterday and have really had this on the brain lately.
So–this is definitely an interesting take on “What happens if the dog gets it wrong,” and I absolutely agree that challenges are important for keeping learners engaged. As you’ve phrased it here, though, this is really a criteria-raising problem, right?
So, I have mixed feelings on to what extent these trainers should possibly address this more fully. On the one hand, raising criteria is dreadfully difficult. Eileen of Eileenanddogs (of which I’m also a big fan!) openly admits that she finds raising criteria consistently extremely difficult. I’ve struggled with it, too, and I agree that most dogs will eventually grow bored if they know, 100% of the time, both that they will get it right and exactly what to expect for doing so. But it’s also crucial to balance that against remembering that if the dog isn’t getting it right often enough (my trainer says 50% or less success), the rate of reinforcement is too low, the game isn’t fun, and you’ll lose the dog that way too. For us, mixing in lots of jackpots and a messy adaptation of the CU “Gimme a Break” game works really well for resolving these issues.
I expect that, since these videos are aimed toward the general public rather than toward experienced dog people (as you say in your “cup of coffee” remark), the emphasis on trainer error is because this might be something of a revelation to many dog owners! I remember how groundbreaking it was for me when, reading something by Suzanne Clothier, I came across the (paraphrased) remark that the dog’s behavior is probably the dog’s best guess for what to do in a situation–mind blown, excuse me while I totally re-evaluate my relationship with this dog and kick a trainer to the curb.
*caveat: Grisha Stewart is clearly not addressing criteria raising, and shy and fearful dogs (as you know well!) kind of a whole different can of worms, even though these questions are just as or more important when finding a trainer to work with them.
LikeLike
Oh, goodness, I didn’t mean to suggest that anyone should be talking about this in their video. I hate it when I’m unclear. It’s not appropriate to the length or to the target audience. Rather, this is something that’s been nagging at me lately–we positive trainers are really uncomfortable talking about failure, and I do think that shows in the videos. Because nobody really wants to talk or think about what to do if they raise the criteria and it goes awry, people often just don’t raise the criteria. Hence “but he does it so well at home!” (Of which I, too, am guilty.)
On the other hand, you’re absolutely right about the risk you run of losing the dog if you make things too hard. I tried desperately to draw attention to Eileen’s passing mention of Bob Bailey’s 80% success rule, and I just couldn’t make it fit into a post of reasonable length. Susan Garrett says to only reward “average or better” performance, which I also think is a handy rule of thumb for raising criteria.
LikeLike
Oh no! I didn’t mean to imply that you were unclear–I’m definitely the one that read and wrote too hastily, leaving out important bits in her rush to get down her thoughts and put this train of thought aside for a few hours. Somehow I just rode right over the bit about taking space to talk about something they don’t have time for! That’s what I get for reading and writing on public transit.
You’re absolutely right that this is a major problem for [force-free, R+, whatever] trainers, and that this discomfort with failure affects both these videos and dog owners like us. In fact, when I watched the free material Susan Garrett offered before the last Recallers session, I found my ire going up over how much she talked about failure and its importance–a little silly, I quickly realized, and almost completely a product of how rarely we talk about failure in force-free circles. But it’s worth talking about, both because no one progresses when criteria is never raised and because the life of an over-managed dog is frequently not a great one.
For me, I guess it comes down to something really similar to what Eileen says in her video–I’m a dog behavior enthusiast, but I’m just an amateur (and way more so than Eileen, consider the analogy to her over now). My technical skills are mediocre, my dog and I have only known each other for about a month, and she’s a sensitive girl who was visibly surprised when she figured out how “target” worked–positive reinforcement based training appears to be completely new to her, and I’m working really hard not to not poison it for her, so I’ve been struggling with this a lot, too. Our trainer also gives her students a version of Bob Bailey’s rule–raise criteria if the dog is getting it right 80% of the time, lower for 50%, hold for 60-70%–for which I’m really grateful, because otherwise we’d never get anywhere. And even with this great rule, I still haven’t added the cue to heeling or moved our beginning loose leash work outside despite my dog’s 90% success rate, so…(I’m totally going to tomorrow, I swear.)
Speaking of unreasonably long posts–sorry for not whittling down this reply further. I tried, but I’m a wordy navel-gazer and I don’t get to talk about dogs in real life very often.
LikeLike
You are FINE. Wordy navel gazers are always welcome here.
And I believe we’re on the same page–talking around each other is a good way to flesh our ideas out.
I think some of what is going on in the conversation about failure is also about comfort with the other quadrants of learning. Allowing failure and then withholding a cookie is absolutely a kind of punishment, albeit a mild one. And, as your little brackets say, there’s some ambiguity about whether we’re okay with that. Does “positive” training mean “positive reinforcement only, all the time?” (Which is absolutely what some people mean, or at least think they mean.) Do we mean “I can use non-physical punishment once in a while?” Where is the line between withholding a cookie, saying “NO!”, and using a collar correction? Like a lot of good things in life, it’s less black and white than we would like.
LikeLike