Posted Date: August 4, 2014
Denise Fenzi explains these two different ways to interrupt an error in behavior in training.
A very common way that positive reinforcement trainers interrupt an incorrect behavior chain is through the use of a non-reward marker (NRM). The goal of a NRM is to interrupt an unwanted behavior and give feedback in a neutral manner, free from disapproval. A NRM is typically a word or sound (although it could also be a hand signal or body movement) that tells the dog that whatever he just did was wrong and that there will be no reinforcement as a result.
Ideally, if you give a NRM, your dog will cheerfully and immediately return to try again. Of course, this does not always happen. You can think of a NRM like a reverse click. It marks a moment of time, but instead of saying that a reinforcer is coming, it takes the possibility of one away. This suggests that NRMs are neutral communication, right? I would say no. In addition to marking a moment in time, a NRM also interrupts the dog's behavior, which communicates that you didn't like what happened. You have an opinion and it's not neutral at all!
There's really no way around this. A NRM is a marker of both the dog's specific behavior and the opinion of the person giving it. No matter how neutral your tone is, you are clearly saying, "That is wrong." I believe that dogs care about whether or not their people are happy with them; after all, they evolved to work cooperatively with us. This means that using a NRM comes with some risk.
Some dogs will stop the behavior you didn't like, but then display a stress reaction; this often happens when the dog doesn't truly understand what you wanted. When a NRM communicates disapproval, but provides no information about how to be right, it will likely lead to a lack of motivation, shutting down, or uninspired work. Then there are the dogs who ignore the NRM entirely. These dogs just don't care enough about the trainer's opinion or reinforcers to change his course of action, making the NRM ineffective.
Either way, you have a problem, and it's tempting to fix it by using a punisher, such as marching in (physical intimidation) or raising your voice (verbal and emotional intimidation). Look at it this way: with a reward marker, such as a clicker, if you mistakenly click, your dog gets a free cookie. While unfortunate for your training goals, it does not erode your dog's conditioned emotional response towards working with you. It's just poor training. But with a NRM, it's not just about your training goals - it's also about you and your dog and your relationship. It's more than just poor training. It's your foundation on the line.
It's not that NRMs should never be used; there is a place for them in training. But because NRMs tend to depress behavior, they should never be used during the initial teaching phase. The last thing you want to do with a dog that that is learning is to shut down his desire to try. The appropriate use of NRMs requires your dog to be trained to fluency on the individual behaviors before you pull those pieces together into chains. Don't spend energy pointing out what's wrong until the learner is clear on what's right.
Basically, a NRM should only be used when your dog is trial-ready for the exercise, and possibly learning to perform under unusual circumstances where attractive alternatives exist, or where the overall level of reinforcement may be minimal. A NRM in this circumstance is used to communicate to your dog that he is now 100% responsible for correct performance and that you will not help him. Be aware, however, that excessive use of a NRM will erode your working relationship with your dog. If this technique doesn't work very quickly, you need to consider the strong possibility that your dog is not ready for the responsibility you have given him.
If you're interested in a more in depth discussion on NRMs within the context of motivation, you might want to read Dog Sports Skills, Book 2: Motivation
, which I co-wrote with Deb Jones. (This book is expected to be available for pre-order
the first week in August with delivery mid-August.)
This dog is staying cheerful on course! Photo courtesy of Karen Moureaux.
There is a much safer option: the cheerful interrupter. A variation on the NRM, the cheerful interrupter marks the moment of the error by totally interrupting the dog in a way that maintains the dog's enthusiasm and gives him the support he needs to be successful. Used correctly, the cheerful interrupter is an excellent option for dogs in the learning phase. The difference between a NRM and a cheerful interrupter can be tricky to understand, so let's compare them using the [obedience] retrieve over high jump as an example.
First, the NRM: You send your dog to fetch, but as your dog gets to the dumbbell, he stops to sniff the ground, so you give him a NRM. From here, one of two things can happen. Either your dog understands the exercise and therefore stops and returns to you for another try, or he doesn't. If he doesn't, it could be that he didn't understand what you meant, or that your timing is off, or that he just didn't care. Regardless, you're stuck, and you'll likely have to enter the territory of punishment to reset the exercise.
By contrast, here's a working example of a cheerful interrupter: When you see your dog stop to sniff the ground, you immediately move towards him, speaking and moving in a cheerful manner that both interrupts the behavior and reorients your dog towards you. Now pick up the retrieve object and repeat the request as fast as possible. This combination of movement, cheerful chatter, and possibly showing the dog the reinforcer (that he won't get) almost always interrupts whatever the dog was doing wrong. In a short period of time, it becomes clear to the dog that there will be no reinforcement for that effort. I prefer this approach because I believe it maintains the relationship you have with your dog more effectively than a simple NRM. An error doesn't have to mean that you're upset or angry it simply means that the flow of training needs to be restarted. Being cheerful maintains engagement even though there's no reinforcement for that attempt. This approach can also be used in the learning phases; for a green dog, you might consider making the next repetition easier in some way.
Most of us use cheerful interrupters with human children, we just dont call it that. For example, if your child gets a math problem wrong, you wouldn't unemotionally say "no" (the traditional NRM). You're probably going to smile and say something like, "Close! Try again!" (the cheerful interrupter). Whether you're working with a child or a dog, the cheerful interrupter allows you to help your learner work through an error in an upbeat way, preventing a possible negative emotional response as a result. It also keeps your relationship intact, and maintains the joy of work.
Denise Fenzi is a long-time, successful dog trainer and dog behavior instructor. Denise has been showing and training dogs for 30 years and runs Fenzi Dog Sports Academy. Denise and Deb Jones are the authors of Dog Sports Skills Book 1: Building Engagement and Relationship, which can be purchased at www.thedogathlete.com.