In previous articles, we’ve discussed the four elements that must be present for learning to take place, and the four stages of learning a new behavior. By understanding these concepts, you’re on the path to learning to become a dog trainer. The next piece of the puzzle is how to actually get the behavior you want and the best dog behavior training methods.
Again, there are four ways to get behavior in dog behavior training (four seems to be a very handy number):
Prompts cover a lot of territory. To get a little wonky (and we’ll go much deeper into this in later articles), a prompt is an antecedent stimulus. An antecedent stimulus is simply something that sets the occasion for an animal to perform a behavior. In dog behavior training, we strive to find antecedent stimuli which will help the animal achieve the behavior we’re wanting it to learn, and prompts can be very useful in this regard.
Generally, prompts help get the behavior initially – they give the animal enough extra information to head them in the right direction. However, before a behavior can be considered learned, the prompt must be faded. Some human examples of prompts might be:
Prompts can be all sorts of things: lures, targets, noises (verbal or physical such as finger snapping), etc.
Capturing is when you reinforce an existing behavior. So, if your dog has a tendency to howl, if you reinforce him when he howls, he’s likely to howl more often, creating the opportunity for more reinforcement. So, as someone wanting to become a professional dog trainer, here’s a handy tip: If you have a dog that is really hard to get into a “down” position (which is not at all unusual, because that position puts the dog in a very vulnerable state), you can simply capture the behavior when the dog is relaxed and decides to lay down. That reinforcement will often get the dog to start laying down more often, and once it’s predictable, you can add in a cue. At that point, you have a dog that lays down on cue, and it happened through capturing.
Physical manipulation means you actually place the animal into position. This is a technique that should be used with consideration. Physical manipulation is an aversive technique, which means that the animal is trying to avoid the pressure being put on him by the trainer. It can be fairly benign, such as lifting a dog’s paw to teach it to shake hands, or placing a barrier next to the dog to teach him to walk in a straight line. Or it can be quite painful and/or unpleasant, such as jerking on the leash and causing a slip collar to tighten around the dog’s neck, or rubbing the dog’s nose in its urine when the dog has had a house training accident.
Shaping is not actually a way to get behavior, but it takes an existing behavior (capturing) and through successive approximations, moves the original behavior toward the trainer’s ultimate goal behavior. Shaping is used in conjunction with one or more of the above techniques: prompting, capturing, or physical manipulation.
In producing my animal training courses and webinars, I try to include techniques that will produce the best online dog training course overall. As trainers, most of the behaviors we train will be shaped to some extent, so it is important to understand this concept in dog behavior training.
That’s it for this article. This is a brief overview of the four ways to get behavior. We’ll go into more detail on each technique in future articles. If you want to learn more, here’s a good course: How to Get BehaviorPosted by Susan Smith on November 18, 2018
When learning how to become a dog trainer, remember that these principles apply to all animals, including humans. We’re specifically talking about dogs, but you can use them with any animal and should use them with owners to help them efficiently build their skill set.
Now that we’re prepared with good timing, adequate motivation, achievable criteria and solid rate of reinforcement (see article: Dog Obedience Training – The 4 Elements That Must be Present for Learning to Take Place), we’re ready to go to the next phase and make sure you do know how to properly train a dog. Let’s discuss actually getting the behavior. There are four phases in learning a new behavior and they are:
Acquisition is when the dog is actually learning the new behavior, i.e., making the associations between the cue and the behavior and the behavior and the consequence. If you really want to know how to become a dog trainer – a really good dog trainer – knowing how to get behavior is crucial. When you first start training a new behavior to a dog he has no idea what you want, so it’s trial and error on their part. A savvy dog who’s been trained in positive reinforcement methods will usually start trying different things until they hit on something close enough to be reinforced. From there, we shape it into the behavior we want. If the dog isn’t so savvy, we may need to help them out a little. We’ll discuss the four ways you can get behavior from a dog (including shaping) in the next article.
Once the dog understands the behavior that is wanted – i.e., they’ve made an association between the cue and the behavior, and another association between the behavior and the consequence – we start working on fluency. Fluency is a combination of speed and accuracy. When the dog understands what the cue means and performs it quickly and accurately, then we can consider him to be fluent in that behavior. This is an important point – many trainers train for accuracy and consider the behavior fluent; however, fluency requires that the behavior also be performed speedily.
In the process of learning to become a dog trainer, it’s helpful to examine some common examples of human behaviors which most adults will be extremely fluent in are things like driving, using eating utensils, tying our shoes, or just walking. All of these behaviors had to be learned, but we learned them when we were young, we’ve used them regularly for years, and today we’re very good at them – we’re fluent. We don’t have to think about stepping on the brake when a car pulls out in front of us, we just do it – we do it accurately and we do it quickly; that’s fluency.
Generalization is the ability to take a new behavior and apply it under other contexts. Usually, generalization is going to deal with distance, duration, distraction and handler orientation (i.e., where is the handler in relation to the dog when he gives a cue).
When an aspiring trainer finally manages to become a dog trainer, we often do generalization as part of the training process. For instance, if we’re teaching loose leash walking we address duration and distraction as part of the criteria – i.e., we don’t think of it as generalization, we think of it as criteria. If we’re teaching a formal stay, we address duration, distraction, and distance as part of the criteria. The goal is to get the dog to stay no matter what, so we think of these elements as criteria rather than generalization. And this is fine, as long as we understand what we’re doing and why.
A great example of a behavior that often isn’t generalized, yet the owner thinks the behavior is well-learned and reliable, is house training. We are often fanatic about house training in our home, but if the dog goes to someone else’s house, they may eliminate. In fact, this probably happens more often than not, and that’s because the behavior hasn’t been generalized.
Once a behavior is learned and fluent, it must be maintained. Practice makes perfect. An interesting exercise for those of you teaching classes and private basic obedience would be to periodically assess your clients after they’ve finished their lessons with you. I’d assess at the following times:
I don’t know this for a fact, but I’ll bet they only use two to three of the behaviors they learned from you and the rest have been lost through disuse. If you find this to be true, it might be worth considering what behaviors you really need to teach. Perhaps fewer behaviors to higher fluency, better generalization, and more functionality (i.e., name recognition for recalls, leave it, loose leash walking, etc.) would make your clients happier over the long haul.
An example of a behavior we might have once been fluent in but are no longer, could be something like handwriting. Although I do occasionally write by hand, in the computer age I don’t do it nearly as often as I did when I was younger, and I’m not nearly as good at it – not that I was ever very good! Any sport we played in high school and no longer play on a regular basis, is going to be less fluent than it once was.
Okay, that’s it for the four stages of learning. In the next article on ‘how to become a dog trainer’ we’ll talk about the four ways to actually get behavior during the acquisition phase of learning. Here’s an in-depth course: Understanding Learning TheoryPosted by Susan Smith on October 31, 2018
I’m compelled to tell this wonderful story about my pal Guinness. Guinness is about 11 years old and is a re-homed Cocker Spaniel who lives with my good friend Karen at Tails by the Bay in Homer, Alaska. Karen took Guinness into her home several years ago, and has lots of experience training a rescue dog, (or more accurately, re-training) as she has been involved in rescue for many years.
We don’t know a lot about Guinness’s background, but we do know that he had some type of accident that injured his spine. He has had mobility issues since Karen got him, and they have been getting worse over the years.
The first time I met Guinness was in the fall of 2016 when I went to Alaska to watch Karen’s dogs while she was out of town. At that time, Guinness was doing pretty well – he was walking and getting around fine, if a little bowlegged. I went up again this past summer and Guinness’s mobility had deteriorated to the point where Karen had purchased a cart to support his back legs. At this time, Guinness could no longer walk. He was able to scoot around on the floor by pulling himself with his front legs, but his back legs were pretty much useless.
Well, Guinness didn’t need professional training for this – he took to the cart like a fish to water! What he really needed was some off leash dog training instruction! We had to keep a close eye on him, because he’d take off down the road like a sprinter, not realizing his limitations. He did take a couple of spills when rounding corners, but quickly learned not to panic and to wait to be righted. The cart has greatly improved Guinness’s quality of life – he’s now able to go out with the other dogs and able to go to the beach, again.
UPDATE!! I just got this video from Karen.
The cart (among some other things) has strengthened Guinness’s core to the point he can walk, again! Let’s hope he keeps improving and maybe won’t need the cart at all, someday.
It’s so heartwarming to see his improvement. He’s a real trooper – at 9 ½, when most dogs are gearing down, Guinness is taking it to a higher level. And, not to be self-serving, but it’s really helped with the house training, as well. Not that Guinness wasn’t house trained, but now he can take himself outside, rather than having to wait for someone to put him in his cart and help him out.
This is the first in a series of articles on the science and art of animal training and dog obedience training. Although the discussion is about animal training, don’t forget that these principles apply equally to humans, so if you’re a professional animal trainer, don’t forget to apply them to your clients, as well as their pets.
For canine obedience training, we have two problems we generally deal with: basic dog obedience training and dog behavior training. When dealing with behavior problems, we’re usually trying to reduce an unwanted behavior such as jumping up or separation anxiety. With obedience training, we’re teaching a dog to increase a desired behavior. For this series of articles, I’m going to focus on obedience training.
There are some basic dog obedience training methods that most trainers know – but it never hurts to review them. So, let’s start at the beginning with the four pieces of the training puzzle that must be in place before a new behavior can be learned:
If any one of these pieces are missing, the dog is less likely to learn the behavior.
Let’s start with timing. The “dog psychology” term for timing is contiguity – which is the proximity of a stimulus and response so that an association can be made between the two. In other words, the two events (stimulus and response – in this case, behavior and consequence) must happen close enough in time that the dog is able to make an association between the two.
There are different “rules” about how much time can pass between the behavior and the consequence, but there’s no doubt that sooner is better. Probably the most common recommendation is that it should happen between one and three seconds; personally, I would be aiming for between immediately and one second. AND – this is very important – if you’re a clicker or marker trainer, the marker can give you some extra time, but I’d still be going for less than one second between both the behavior and the click and less than one second from the time of the click to delivery of the consequence. The clicker buys you time, but the food is the consequence.
So what this means is, be prepared! Have treats in your hand, ready to be delivered. Don’t have them in your pocket, a baggie, or even your bait bag because it takes too much time to get to.
We all know what motivation is – it’s the reason we do something. Motivation can be to acquire something or avoid something. In the science-based/positive reinforcement dog training community, we generally use something the dog wants such as food, toys, etc. For purposes of this article, we’ll use food as the consequence. Food is easy to work with, highly motivating, and helps achieve a high rate of reinforcement.
You should use the smallest, least valuable reinforcer the dog will work for. Use smaller treats because you’ll be dispensing a lot of them and you don’t want to satiate the dog before you’ve completed the training session. As to the least reinforcing, well – save the big guns for more difficult challenges. Another thing to consider is when to break out the higher-value reinforcers. A lot of trainers jump to higher value treats as soon as the dog displays any indifference to the training session, or doesn’t appear to be learning. If the dog will eat the food if given freely, then it’s probably motivating enough for the training session. So, the problem is probably a function of criteria rather than motivation.
Most trainers understand timing and motivation. Where they often go wrong is with criteria. Criteria is the behavior you want the animal to do now. This can include many things such as position, duration, distance, speed, level of distraction, and so on.
The most important thing about criteria is that it should be achievable. If it isn’t achievable, all the motivation in the world isn’t going to help. In human terms, think about asking a 2nd grade school child to do a calculus problem. You can offer him $1,000.00, his favorite ice cream every day for the next year, whatever he wants. If he doesn’t know how to do it, the motivation isn’t going to help him.
Additionally, criteria should be specifically defined. If the requirements change from moment to moment, you have very fuzzy criteria and you’re going to get fuzzy behavior. In a later article, we’ll discuss specifically how to set criteria – when to raise it and how much to raise it.
Rate of Reinforcement
Rate is about how often something happens in a given period of time. Rate of frequency refers to the number of times a behavior occurs within a specified period of time. Rate of frequency is how we measure learning – is the frequency of the behavior increasing or decreasing? If not, then the animal is not learning what we’re attempting to train. Reinforcement refers to the consequence that increases a specified behavior. As we know, behavior is consequence driven – i.e., we are likely to repeat or not repeat a behavior, depending on the consequence of that behavior. So, rate of reinforcement is the number of times a reinforcer is delivered for performance of a specified behavior within a designated period of time.
Timing, motivation, criteria, and rate of reinforcement are the four elements that must be present for learning to take place.
In fact, if behavior is changing these four elements are in place – the trainer may or may not be aware that they’re in place, but they are. A good trainer knows they are necessary and arranges their training around them.
These same principles also apply to puppy obedience as well as when training an older dog. For that matter, they apply when teaching horses, cats, cockroaches and humans!
The next article will discuss the four stages of learning. Meanwhile, I will be presenting a webinar on this topic in January, so if you’d like to learn more, check it out!
NOTE: Since this is the first article in this series, I thought I’d bring this up now. Early in your career, you should know what to call yourself. A lot of dog trainers call themselves “dog behaviorists.” However the term “behaviorist” is an earned title that takes a lot of work, so out of respect for our affiliated professionals, we should not refer to ourselves as dog behaviorists unless we have earned that title. I suggest the term “dog behavior consultant.”
Susan Smith is a Certified Professional Dog Trainer as well as a Certified Dog Behavior Training Consultant. She has co-authored the book “Positive Gun Dogs: Clicker Training for Sporting Dogs and has authored or sponsored hundreds of online dog training courses.Posted by SBConsulting on October 6, 2018
If you would like to receive events updates please click here.