Why You Don’t Have an “Operant Dog”

8 | Posted:

By Dr. Sophia Yin

As I watched the video my friend sent me, I saw a reactive dog, named Frankie, lying down in front of his human. The owner was trying to reward quiet, calm behavior while other dogs were far away. Frankie fidgets and keeps moving, gets up and then lies back down, but he’s focused on his owner much of the time. The problem?  The owner’s timing of rewards is way too slow to effectively reward a calm down-stay. It’s like the owner is moving at the speed of The Lawrence Welk Show when the speed the dog wants is along the lines of MTV. As a result, the owner sometimes rewards a down and sometimes rewards Frankie as he starts to get up or even after he’s stood up. Ideally, the owner would be fast enough to reward him for lying down every 3 seconds or so at first and while he’s still lying down. Once Frankie starts relaxing at that rate, the owner would slow down to, say, 5-second intervals before increasing to longer intervals. Or the owner could, of course, just get a MannersMinder where the intervals have been determined through scientific study and the rewards are delivered in a location that’s perfect for a dog that is lying down.

In the midst of this dog’s fidgety performance, the owner states, “Frankie is hard to train because he’s very operant.” 

“Hmm…,” I think. “What does that mean? By “very operant” does the owner mean the dog is so fast that she can’t get her timing right? Or just that the dog tries a bunch of different things?” If I hadn’t seen the video I would be completely clueless as to what the owner meant. Why? Because using this term to describe the dog actually makes no sense. It’s like calling a dog dominant as a personality trait when what one means is the dog is aggressive, unruly, energetic, or the dog misbehaves, since dominance is a relationship, not an individual trait.

Am I The Only One With This Pet Peeve?

You might think the inappropriate use of the word “operant” to describe a dog is my pet peeve, but it turns out that it drives others behavior scientists crazy, too.

Here’s what Bob Bailey, former General Manager of Animal Behavior Enterprises—a company that showcased the science of operant conditioning by training over 14,000 animals of 140 species for television, amusement parks, sales conferences, military operations, and behavior research—says:

Let's see – a sleeping dog is probably not exhibiting lots of operant behavior. Most certainly a dead dog does little that would be considered operant. But, the rest of the time, a live and awake dog will exhibit lots of operant behavior- that is, behavior that has been influenced by consequence, whether reinforcing or punishing. I think those using the term “operant dog” simply don't know the real meaning of the word. Perhaps the word “operant” has a certain market value now and it is being exploited.

Dr. Eduardo J. Fernandez, a professor in the Psychology Department (Animal Behavior Program) at University of Washington and one of my collaborators on the research testing the operant conditioning protocol for the MannersMinder®, agrees.

This use of “operant” to describe a dog is odd. Skinner and most behaviorists would point out one very clear problem with this use of “operant” as an animal-describing adjective; operant terms, such as “reinforcement”, “punishment”, and “operant” itself, describe behaviors, not organisms!

It seems like what people are trying to describe, both good and bad, is an organism that will cycle through a large repertoire of behaviors in the hopes of being rewarded for something. In both cases, it would probably be more behaviorally accurate to simply say, “My dog offers a lot of behaviors!” Calling the dog “operant” is about as unclear as saying “my dog is very voluntary.” And really, when it comes to trying to describe any organism, it's always best to use clear, observable terms that describe the animal's behavior, rather than some non-observable quality that the animal supposedly has!

Fernandez knows the importance of describing animal behavior in clear cut terms because he spends a large part of his time developing programs to measure behavior in animals in zoos in order to determine where changes to care may be needed.

A Third Behaviorist Weighs In

Another well known behaviorist, Dr. Susan Friedman, a psychology professor at Utah State University and owner of www.behaviorworks.org, states:

“That phrase is irksome to me, too.” She also points out another irksome phrase. She states, “Another version I've heard too often among dog trainers, ‘He can't learn now, he's not operant.’”

Dr. Friedman, who teaches a highly rated online course for animal behavior professionals, continues,

I think it is just a new example of the explanatory fiction that an animal's behavior 1) resides inside the animal, and 2) is independent of the conditions in which it behaves. Both of these ideas are not to the goal of behavior change because they alleviate the trainer of responsibility for what the animal does (which is exactly the opposite of what we want).

In other words, Friedman feels this inappropriate descriptor serves as an excuse for people to explain why they have trouble training their dogs. Blame the dog for moving too fast or offering too many behaviors, instead of the owners blaming their bad timing or rate of reinforcement, inappropriate shaping steps, or other technique-related factors.

It’s Unanimous!

The vote is in. Scientists and others who value accurate definitions and descriptors can’t stand the misuse of the word “operant” to describe how a dog behaves. No doubt Skinner would be irked, too.

So what terms should you use instead? Just describe the behavior you see. If you mean the dog moves quickly and cycles through a lot of behaviors voluntarily you might use Fernandez’s, ”My dog offers a lot of behaviors.” If you mean something different, describe what you’re talking about.

Tags: , , , , , , ,

8 responses to “Why You Don’t Have an “Operant Dog”

  1. I have never heard the term “operant dog” or the term “operant” used in this manner before and I agree that it just sounds odd.

    Regardless, when I am dealing with a client, I try to stay away from jargon that would not mean that much to a layperson. The term “operant” would fall into that category. Instead, I try to simplify things and keep it to descriptions of what the dog is doing and why he/she is doing the behaviors. That is what clients care about. In this case, it sounds like the owner’s timing is slow so the dog is trying to figure out how to get the owner to click/respond/reward by trying out different behaviors to see wht works.

    D. Sakurai
    http://www.pawsitivefeedback.com
    Los Angeles, CA

  2. I really loved your previous post about treat timing, and I’m delighted to see your MTV/Lawrence Welk metaphor recycled here.

    I’m not a behaviorist or professional trainer. I train my own service dogs. (Currently training my third.) I use this term. I even have a post called “My Operant Dog!” in which I explain what I mean by that term. The other clicker trainers I know who use this term (some of whom are applied animal behaviorists — so I question your “unanimous” conclusion) use it as slang. We use it amongst ourselves as shorthand for “a dog who has been trained with shaping and as a result knows how to offer new and novel behaviors.”

    When I say, “Hey! I’ve got an operant dog now!” I know that most people won’t have any idea what I mean; but most people don’t know what I mean when I say “shape” or “capture” either. I use these terms with those who do know what I mean. When I say, “My dog is operant now,” I’m saying, “Yay! We’ve done enough free shaping that now my dog is eagerly offering a variety of behaviors, including new behaviors that I haven’t taught, and also trying out behaviors in situations where he hasn’t learned them before.” In a tweet or on a listserv, it’s easier and faster to use shorthand. English lends itself to neologisms, and the internet has speeded up this process a good deal. (Try to find “blog” in a dictionary published fifteen years ago.)

    From your description of Frankie, my guess is that he is not really “operant” in the way that I mean it. It doesn’t sound like he’s engaged and thinking and testing things out because he’s enthusiastic. It sounds like he’s nervous and confused. But just because Frankie’s trainer has bad timing and isn’t aware of the effect she’s having on her dog doesn’t mean that the term she’s using is flawed.

    I certainly respect your distaste for the term “operant” used in the slang way I use it. I’m an editor, and I can’t stand it when people use “impact” as a verb (except in reference to teeth). However, by now even dictionaries and style guides are allowing “impact” to be used as a synonym for “affect.” I think “an operant dog” is on its way to becoming accepted slang, even if its origins are painful to behaviorists.

  3. Sharon:

    I’m curious which certified applied animal behaviorists you’re talking about as I would suggest they make up a different term that still reflects the meaning of the original terms. (see below when I discuss the term BLOG).
    By unanimous, I meant amongst the behaviorists with a strong science background and who treat training as a science (e.g. take data, constantly improve based on what the results……, try to be precise about what they are describing) and remember every day that it is a science.

    I don’t know if you know the people who were quoted, but I asked the colleagues I most respect in terms of their understanding of operant conditioning as a technology (vs just the popularized version used with widely varying degrees of proficiency). These are people who understand the history of the words and the importance of where they came from and what they actually mean. And who also apply the principles to a degree that any trainer with knowledge has to respect.

    I have no problems with the words shaping and capturing because they are actually science terms that even beginners SHOULD quickly learn. They are part of a basic set of definitions that anyone who trains using clicker should know. Shaping is one of the first definitions explained in almost any psychology class on learning theory—after the categories of operant conditioning. I also don’t have problems with coming up with definitions that are well thought out given the understanding of what the current word means (vs out of a vague familiarity). For instance BLOG comes from short-hand of Web log. you are retaining both meaning in the new term. That is the perfect way to come up with a new word.

    Careless use of terms in science lead to problems such as people calling dogs dominant when dominance is not a personality trait (and as a result these dogs are particularly trained in a forceful way because people have assumed they have a certain personality that requires it). Or loose use of the word dominant or dominate to mean many different incorrect things—all of which end up in the use of force on the dog. Most likely if people use the word “operant dog” loosely as most people do, it will end up being used to describe many different behaviors which ends up actually muddying the waters.

    In science when we develop new terms we always define them every time they appear in an article for the first time. So perhaps if you guys want to use it and make it a definition you should ALWAYS use it with the definition. I bet you’d find there are about 10 definitions out there. And what one person says is an operant dog another persons says is just due to bad training.

    If behavior weren’t a science, if it was just some popular game people played, it wouldn’t matter. But if we are promoting that training be science-based (vs emotional-based such as can happen on either ends of the spectrum–“positive-only” vs dominance-force-based training) then all aspects of what we do should reflect the care taken in carrying out science.

    Sophia

  4. Hi Sophia, this is a very funny and accurate article. Here at I have heard people and clients using “operant” incorrectly before but never confronted or addressed it, I just went on with the issue or behavior at hand and focused on the timing and delivery. In order to help my clients with their timing and accuracy, especially with undertrained or energetic dogs, I split up the exercise into smaller, simpler, singular jobs. I will click/or mark the behavior and they will treat or vice versa. Giving them one aspect to focus on at a time typically puts them at ease and alleviates the timing issues. Russell Hartstein CPDT-KA

  5. Hi Sophia.

    To answer your questions, I know Bob Bailey’s work. (I don’t know him personally.) I’ve never heard of the other two. I try to read or watch work by many others (Overall, McConnell, Chong, etc.). I have no idea what their opinion is on using “operant” as an adjective/descriptor of a dog, although my guess is that it would be similar to yours.

    I don’t feel comfortable giving names or quoting opinions of others who haven’t published them publicly. (It is considered rude to publish or cross-post content without the poster’s permission.) I don’t know if the behaviorist I have in mind has used the term operant as an adjective publicly or just on clicker lists, but I remember a post in which she said, “[Crossover dog] is now operant” and went on to explain what he’s doing. That was actually the first time I’d come across the word used that way.

    Peace.

  6. P.S. This has me wondering if you’ve written a post — or might be interested in writing one — about the term, “reactive.” I see “reactive” now used to describe so many different behavior issues that I think it’s become kind of vague and euphemistic. I understand it’s an attempt to be more accurate and less blaming than the terms people used to use, but sometimes I wish I got more information. For example, is the dog reactive to cars? To strange dogs? IOW, does it make sense to say, “The dog is reactive,” period? Are there other words that could be used that are more specific? I’m curious as to your opinion on this. Thanks!

  7. This is a very valuable post, I found it looking through Google. I believe most readers will agree with your views. Finally – a person with common sense!
    PS I quite like the template you are using – where did you find it?

  8. Hi Sharon:

    Thx. This is actually all custom built in Expression Engine! More expensive than a template but also more flexible.

    sophia

Leave a Reply

Your email address will not be published. Required fields are marked *