Pavlov’s Theory of Correction

Pavlov discovered that the best time to CORRECT a dog is when it is performing a task CORRECTLY.


The theory of Natural Dog Training (wherein emotion runs to ground through physiological states that then produce behaviors in conformance with the prey drive) can also be expressed in terms of Pavlov.  Note that the prey drive elicits the most intensely positive physiological state a dog can experience.

Pavlov’s process of conditioning whereby a tone is paired with the taste of food until the physiological state of salivation becomes associated with the sound of the tone alone is depicted in the diagram below.

Now, if during the process of training we were to follow the above schema but rather than a tone we were to  substitute a correction (jerk on lead) for a tone and the state of salivation with an obedience behavior, for example, the physiological state of wanting to lie down (wherein the dog wants to lie down because food/toy has been used to elicit the down/prey pounce position) then by pairing the jerk on the lead with the behavior resulting from the desire to lay down, the dog thereby associates the correction with the emotional state of wanting to lay down. Therefore, later in a real world situation if the dog finds itself distracted to the point that it doesn’t want to lie down, the conditioned stimulus (correction/jerk on lead) is applied and the dog wants to lie down and is not in a state of conflict because the physical/visceral/emotional states affiliated with prey drive, the physical memories of which are triggered by the jerk on the lead, are the most intensely positive that a dog can experience.

In contrast, in current methodology, a dog is corrected for making “mistakes” which are really states of conflict. Therefore the correction is being associated with a state of conflict and this violates Pavlov’s theory of correction because the dog is associating a correction with NOT WANTING TO LIE DOWN rather than wanting to lie down. Some might say the dog could be made to want to lie down in order to avoid a correction, but the dog is still associating the correction with a state of conflict because otherwise it would not be distracted. In other words, the distraction is a distraction because it elicits a stronger physiological/visceral response than the cue that had been associated with the obedience behavior from which the dog has been distracted. The correction is being associated with this stronger emotional/physiological state and the incorrect performance that the distraction elicits, even if ultimately the dog’s capacity to resist is overwhelmed by the handler and it is forced to lie down. Pavlov discovered that the worst time to correct a dog is when it is performing incorrectly.

Want to Learn More about Natural Dog Training?

Join the exclusive and interactive group that will allow you to ask questions and take part in discussions with the founder of the Natural Dog Training method, Kevin Behan.

Join over 65 Natural Dog trainers and owners, discussing hundreds of dog training topics with photos and videos!

We will cover such topics as natural puppy rearing, and how to properly develop your dog's drive and use it to create an emotional bond and achieve obedience as a result.

Create Your Account Today!

Published June 30, 2011 by Kevin Behan
Tags: , , ,

11 responses to “Pavlov’s Theory of Correction”

  1. Christine says:

    I much prefer the idea of eliciting a desired behavior vs overwhelming. Thanks for this insight, Kevin.

  2. Sorry, I’m not seeing how hearing a tone followed by being fed constitutes a correction.

    Could you explain?

    I know from my own experiences (following the principles and exercises in Natural Dog Training) that when a dog is in a high-drive state, a physical “correction,” if applied with precision, can have the opposite effect of what a correction is usually thought to do, or is “supposed” to do.

    But I don’t see the connection to Pavlov.


  3. kbehan says:

    Sorry if I wasn’t being clear. What I’m saying is that NDT methodology which focuses on drive is the only training 100% consistent with Pavlovian logic. So in Pavlov’s experiment dog associates food with positive emotional state, therefore, dog can associate correction with positive emotional state if that positive emotional state is strong enough. (Think wolves being kicked by deer after securing a full mouth grip on haunches.) So the worst time to correct a dog is when it is performing incorrectly because this is a state of conflict. Whereas in the training buildup, we can condition the dog to perceive a correction as a heightening of drive (just like the sound of a tone) so that when later out in the real world and the possibility exists that a dog could become distracted, a correction then reminds dog of a positive emotional state. So the negative now equals a positive (just like the tone equals the food), or, negative-as-access-to-the-positive and this is important because it fulfills the parameters by which the animal mind perceives its world. And therefore, other training systems which do not focus on drive are in violation of the most basic Pavlovian precept.

  4. Okay, I kind of see where you’re going.

    And I agree that training systems that don’t focus on the dog’s drive probably aren’t utilizing the Pavlovian model. I also agree that when you’re training while the dog is in a high-drive state, and you’re applying physical corrections, they don’t necessarily cause the dog to have a negative experience (depending on how high the dog’s level of drive is, how steady its temperament, and how forcibly the correction is applied).

    This relates back to something defensive linemen in the NFL talk about; most of them say they can’t wait to get that first hit from the opposing players. It amps them up for the job they have to do.

    The problem is, dogs aren’t getting paid millions of dollars to get knocked around by us. And not everyone who’s working with dogs has the right sensitivity to how much an individual dog can handle, or how to gauge how high the dog’s drive is at a critical moment.

    On a certain all stimuli are “perturbations” of an organism’s emotional homeostasis (or so Freud claimed, and I think he’s right). So an unusual tone is, at least hypothetically, not much different from a correction on a prong collar. It’s just that the calculus on the latter has to be right on the money or you’ve got some emotional clean-up/trust issues to take care of with that particular dog.

    I’m saying this primarily because I’m someone who’s pretty well immersed in the NDT paradigm, and even I’ve made mistakes in this area. I’ve found that I personally have to be very careful to do whatever corrections I’m going to do in such a way that the dog feels it’s part of a game. Dogs can take a lot of physical punishment when they’re in the throes of playing ecstatically with their pals.

    So I’m not disagreeing with the underlying thesis, re Pavlov and a dog’s drive states, etc. It’s just that I personally would rather use play as my model for “necessary roughness,” rather than the wolf vs. deer model. That’s life and death. Play isn’t.


  5. kbehan says:

    Those are good points but putting all the more complex distinctions and the matter of drive aside, the gist of what I’m saying is that if someone feels that a correction is ever going to be necessary at some point in a dog’s training, then the dog should associate it with doing the behavior correctly, rather than as things now stand, incorrectly. That’s the simple Pavlovian principle which is not in the mainstream training paradigm.

  6. Annie says:

    When the complex distinctions are removed, it is much easier for me to understand! I notice that most people either aren’t comfortable -or they’re downright outraged- when they observe me in the park engaging in a tug of war with Luke….they think that I”m teaching him how to be aggressive. “Why can’t you just let him play with the other dogs?”

    Until I started reading your work, and watching you train other dogs, I probably thought the same way. But I see the difference in my dog’s energy, returning home.from the park..his tail is higher, his stride is relaxed, in pace with mine, and he actually has a sparkle in his eyes. I want to learn more about the method of correction, so that if ever I do need to break his focus on an unwanted subject, and redirect it, I can do so without panic.

  7. Annie, that’s just how I’ve been feeling lately!

    I have a golden retriever who finally plays tug with me (yea!) after many many sessions trying to un-teach her the “drop it” command – which, I had been told in the past, was an indicator of a well-trained dog.

    The other day I was sitting with friends on the outdoor patio of a coffee shop where they allow dogs. Bella was off-leash, just napping under someone’s chair in the shade. A dog & his owner approached us and Bella grabbed her stuffed lion toy and trotted over towards them. As I expected, she pretty much ignored the dog and stuck her face up to the human, wiggling her butt. “Look at me and my cool toy!” While happily being petted by the other owner, Bella allowed the dog to sniff her, but when it became rude (pushy, lifting B’s leg for a better whiff), B flopped over onto her side to try & get the dog to stop. Wouldn’t stop, so B walked away. Still wouldn’t stop, so B stopped, turned & growled.

    My friend commented, “Yeah, Bella is definitely anti-social. She doesn’t like to play.” It was such an insulting tone of voice. As if there were something wrong with her preferring humans over other dogs. B plays with my friend’s dog, but cuts the play short if it becomes even slightly rough. But a game of tug with me? She’s like an energizer bunny, goes and goes and goes, and then we throw in some fetch/find it, and then we run away, me commenting, “Oh, wow, what an amazing toy you’ve got, B!” And then, “Ooh, look at me and this amazing stick I just picked up, guess you better come back over here so you can try & take it away…”

    And she’s running up and down the trails with that toy, head held high, so pleased with herself. How in the world is that something to criticize?

    *(in my head I went, “well, my dog is happier than your dog who has to be leashed up to your chair because you don’t trust her recall. Your poor dog is straining against her leash to join the party but she can’t because it’s not safe to let her off-leash to say hi to the other dog, as in a moment of excitement she couldn’t care less about you and instead would rush off. Neener, neener, neener.”)


  8. Lacey says:

    Apparently, dopamine rises in the pursuit of pleasure, not the gratification of it. The scientist states that this explains goal-driven behavior, but I’m not so sure. This looks to me like the work is the pleasure. Is this drive?

    Does this explain “frustration is better than gratification?”

  9. kbehan says:

    The number one motive for behavior is potential energy; this is what Sapolosky mischaracterizes as the “anticipation” of an impending pleasure. But it isn’t anticipation; rather it’s the activation of a physical memory of something positive, immediately juxtaposed to the absence of that positive in material actuality. So the animal has the physical memory of something positive kindled, but not the actual state of grounding by having it in “hand,” and this is potential energy. {Aesop had it wrong, in evolutionary logic the bird in the bush is worth ten in the hand.}
    The existence of potential energy is revealed to a dog via a state of conflict that is being processed via the hunger circuitry, and this is also why there is conditioning involved in all these experiments because first a physical memory via hunger has to be imprinted via Pavlovian conditioning. So if a dog is in a state of conflict (projects its physical center of gravity into an object) then even a 100% rate of reward will not diminish its desire to do work (as evidenced in bite work/search and rescue/drug and bomb detection/hunting training regimes ). When in Drive a dog’s level of arousal will always be heightened by whatever happens, and yes to your point, a state of conflict being processed via hunger reflects Drive. Whereas if you’re doing rote conditioning for basic obedience exercises that don’t incite a state of conflict (as hunting always does) then you need the intermittent reward schedule in order to incite a state of Drive. And to excite a dog into a state of Drive, frustration is necessary in order to induce a state of conflict and which by definition is simultaneous to it’s being processed via hunger circuitry otherwise the dog would be exhibiting avoidance or apathy to the circumstances of conflict rather than frustration. We must keep in mind that neurochemicals and hormones are merely the mechanics of implementing a behavior; they are not the actual dynamo of behavior but rather its symptoms. And you can see how thin such an understanding of behavior that is predicated on a neurochemical/hormonal model in the video linked below wherin Sapolosky is asked how do dogs know what humans are feeling.

    Sapolsky’s answer that dogs are this way because we bred them to be this way and/or they had to be this way in order to live with us for the last twenty thousand years is not a meaningful answer.

  10. Ben says:


    So when is the accurate time to start adding in corrections when the dog is doing it correctly? I assume this would by while working on the bite. So maybe it would look like shocks on a pinch collar literally while the dog is biting?

    I am also curious about why the dog goes down after the shock. Can being in down also be in a state of drive? I viewed it as shock into something like a push to conduct more drive.

    Thanks, Ben

  11. Kevin Behan says:

    A very good time is to lunge a dog in a circle with bite toy in its grip, and then make smooth turns and add flicks so that these are incentives to follow, also in the more active phase of bite work, also when the dog volunteers the down upon getting on the box. Yes Down when associated with “collecting” is a drive behavior as it allows the dog to gather force and collect itself for the next phase of forward motion (projection of force).

Leave a Reply

Books about Natural Dog Training by Kevin Behan

In Your Dog Is Your Mirror, dog trainer Kevin Behan proposes a radical new model for understanding canine behavior: a dog’s behavior and emotion, indeed its very cognition, are driven by our emotion. The dog doesn’t respond to what the owner thinks, says, or does; it responds to what the owner feels. And in this way, dogs can actually put people back in touch with their own emotions. Behan demonstrates that dogs and humans are connected more profoundly than has ever been imagined — by heart — and that this approach to dog cognition can help us understand many of dogs’ most inscrutable behaviors. This groundbreaking, provocative book opens the door to a whole new understanding between species, and perhaps a whole new understanding of ourselves.
  Natural Dog Training is about how dogs see the world and what this means in regards to training. The first part of this book presents a new theory for the social behavior of canines, featuring the drive to hunt, not the pack instincts, as seminal to canine behavior. The second part reinterprets how dogs actually learn. The third section presents exercises and handling techniques to put this theory into practice with a puppy. The final section sets forth a training program with a special emphasis on coming when called.
%d bloggers like this: