Wow, its the thirtieth blog today, and it's been a while since the last blog!
Today's topic is something I've noticed in almost every dog-human relationship - rewarding for an "almost". When teaching a dog something new, its is indeed valuable to mark
that they're going in the right direction, and communicate that yes,
that's what you're trying to get them to do. We're not talking about
this. You do need to mark and support your dog in new behaviours, but
they don't and shouldn't need a treat or the "this is perfect" marking
word. This leads to confusion. If you're using "Yes-treat" as your
marker for a job well done, you shouldn't be using this to mark a
correct direction. That note being aside, this isn't the instance we're
talking about. A new behaviour is complicated and difficult for your dog
to learn. There absolutely needs to be great communication, but no reward until they get it exactly right. Most people who do treat for almosts typically have a longer time training a behaviour, and it's often sloppy. While I totally understand the need to communicate that your dog is going the right direction, I personally disagree with this practice. The reason being, unless you're not doing a good job communicating to your dog, or teaching an incredibly complex behaviour, there is no need to reward the "almosts".
When a dog already knows a good sit, and can do it just fine, when you reward an almost, your dog learns that they don't have to do it exactly every time - and while that might not seem like a bad thing, dogs learn that they're allowed to be sloppy. This causes loads of behavioural issues, because if they can be sloppy on the little things and not take them seriously, this expands into other areas and causes a lot of handler frustration - especially when they're distracted, and go "oh yeah whatever, you don't mean come/sit/stay/stop. You didn't mean it before, so you don't mean it now." And because of that, many humans get mad at their dogs, when it's caused by their own lack of attention to detail. It might seem like a silly nit pick to have to ensure a sit is a flawless sit every time, but it really is necessary. When a dog is first learning a sit behaviour, and he is rewarded for an "I almost bummed the floor", they believe that first instance is a sit, and that's what it is - its harder to undo that behaviour than to reward a perfect sit.
When dogs learn that they can skimp on certain behaviours, such as waiting by a certain point when there is someone at the door and they're permitted to "break" that line without permission, they often believe they can do so with -all- behaviours, and consequently become the target of frustration. While this is caused by the handler, its ends up being perceived as defiance, or dominance, when it's something a simple as rewarding an "almost".
Most owners fall in to the "almost is fine" when they mistakenly believe that it's the same behaviour. For example, lets say its a rule that a dog is not allowed into a bedroom, and must wait at the door. You've trained your dog to sit at the door frame, and over time, your dog has started letting a paw just past the door, and of course since the dog is still behind the door, you ignore it and carry on. Except, you've just rewarded an almost. Remember, by ignoring a behaviour, you're communicating to your dog that what they're doing is acceptable, and that communicates just as much as a "yes" or a treat in a learned behaviour. So, your dog has learned "almost" is good enough at the door, and is likely to try "almost" with other things, such as sit for a treat, wait for the door, or even leash manners. With some canine personalities, this is a dangerous thing to teach, especially if your dog is a naturally assertive, intelligent type. Because most dogs learn fast, this is as much a good thing as it is a bad, but it means we must be careful what we let slide - and what else is going to slide.
Keep on training, everyone!
