The Inverse Correlation

I recently posted a statement on Facebook that received quite a few comments, and started a few interesting conversations when people shared it. Here it is: “There is an inverse correlation between trainer skill and intensity of corrections: the better the trainer, the fewer and less harsh the corrections. Trainers who use harsh physical corrections simply do not have enough experience or knowledge to do better.”

The vast majority of trainers who have access to the page I posted the statement on are what would be categorized as “positive” trainers; a title that could benefit from an actual definition if you ask me, but you get my drift. Most comments were in agreement, but the topic of corrections is a touchy one, and where the line gets drawn is a never-ending controversy. Most of us agree that there is never a need to use hanging, helicoptering, or any of the other inhumane, over-the-top types of corrections that persist to this day. And then there are many of us who refuse to use harsh jerks on choke chains or similar collars. Some trainers believe in LIMA, which stands for least invasive, minimally aversive. If a correction is absolutely necessary, that sounds about right to me. Still other trainers believe that even verbal corrections are unnecessary and too harsh. That’s fine, but even though we do our best to set things up so that no corrections are necessary, life happens, and in my opinion, dogs need information at both ends of the spectrum, just as children do. We need a way to let them know it’s okay if you do this, but it’s not okay to do that.

The other day when it was raining, I finished working and went to return my laptop to its case. The soft envelope case, which had been lying on the carpet, was wet. For a moment I was baffled. Had water come up under the carpeting somehow? I wondered as I patted my palms all over the case. Then I spied a more vertical object nearby that seemed to have a splatter of water on it. Oh! Bodhi had apparently decided he didn’t want to go outside to pee. Imagine my joy. Now, I hadn’t been there when it happened, so I couldn’t very well correct him; but had I seen him in action, you bet I wouldn’t have just ignored it. There would have been a sharp verbal “Eh-eh!” to let him know that I didn’t appreciate his leaving P-mail, even if it was, appropriately, on my laptop case. Administered properly, a verbal correction is not traumatic to the dog. It’s information, nothing more. Of course, even a verbal correction can be taken differently by dogs who are more or less sensitive, so care must be taken, depending on the individual.

I’ve meandered a bit off point. My original statement wasn’t aimed at verbal corrections, or even the type of corrections many trainers use when teaching leashwork. We can debate definitions of “punishment” and “corrections” all day long, and trainers often do. I’m talking here about harsh physical corrections. I stand by my statement that, barring a physical attack on a person or some similar emergency, harsh physical coercion is never necessary. Again, the more you know, and the more experience you gain, the less need there is to ever resort to violence.

25 Responses to The Inverse Correlation

  1. Dave Hala says:

    I prefer the term “Modern Trainer”, over “Positive Trainer”

  2. sitstaylearnblog says:

    I find it frustrating when people think because I call myself a Positive Reinforcement trainer, that my dogs never receive any type of punishment, just get cookies and can do whatever they like
    . My dogs are punished when necessary, in an appropriate manner. But never in a harsh, physically painful manner.

  3. It frustrates me that these ‘conversations’ rarely distinguish between the use of corrections during formal training, and corrections that occur during living. Of course, the dog is always learning, as are we, but a clear distinction should always be made, imo, between formal training sessions and day in, day out living and management.

  4. LisaH says:

    I agree 100 percent but have never been able to convince a trainer who feels harshness is necessary that it is not. I have been told though, more than once, that I’m too nice, and need to be firmer if I want to get things done. My dogs are well-trained, have a lot of freedom because they are, and are trained to work with me, voluntarily, not cuz they are scared. I also don’t expect precision responses – if it sometimes takes my dogs 2-3 seconds to comply, that’s fine, they aren’t in the military. If something is urgent or we are in formal training mode, they hear it in my voice and respond immediately.

  5. me says:

    I agree for the most part which is why I try to avoid labels which constrain and convolute things. So if I use a label, now I say corporal punishment and intimidation free, which is the most apt description I’ve been able to come up with or heard yet. Anything else is pretty much game for me. Including non-threatening verbal corrections. I’d like to see someone allow the pee to soak through the carpet and into the bottom layers where it can’t be cleaned out without stripping the carpet. Are they going to stand by and not say anything? Maybe they’d rather just grab the dog, but how is that less invasive than an “ah ah!”? Especially if “ah ah” is just to interrupt and carries no physical reprimand?

    I still don’t agree with the better trainers using softer corrections. For a correction to work, the dog has to care enough to work to avoid it. If the correction is soft, then why would most dogs mind it? So yeah, they don’t need or probably use over the top corrections, but I don’t think they’re starting off with gentle corrections either. I see a lot of owners issue corrections that fail, a main factor is that they don’t want to hurt their dogs and are too afraid of giving really meaningful ones so the dog just tunes it out.

  6. Oddly enough, my first response when I read this was that if I spied my dog lifting his leg on something inappropriate, I would not shout eh-eh (which interrupts behavior, for sure, but doesn’t give the dog information about the situation other than that). I think my response would have been “Leave it!” because that is a cue the dog, in his elementary basic manners lessons, would know to mean “don’t go near that thing, person, or area.” The idea that a dog needs information is very true, but it needs to be information, not just noise.

    • wildewmn says:

      Actually, it’s not at all, and you’re absolutely right. “Leave it” is instructive, while “Eh-eh!” is an interruptor. Not being a perfect being at all times, I have to admit that I probably would have shouted “Eh-eh!” 😉

      • I have less problem with momentary lapses, which all humans (and dogs, though it seems they aren’t as easily excused) exhibit from time to time, than I do with intentional punishers;-))

    • Kim says:

      I agree with that. If I see my dogs doing something inappropriate, I try to say their name. Since I have conditioned their names as focus cues, they usually stop and look at me which gives me the opportunity to ask them to do something more appropriate. So I see a dog starting to squat or lift, I try say “Daisy/Demon/Angel/etc”, they stop, I say “Let’s go outside”.

      Notice I say try. I am still a work in progress.

      Even so, there is a difference between a cue (tells the dog what to do), an interrupter (momentarily stops the behavior) and a punisher (reduces the frequency of the behavior).

    • Evelyn Haskins says:

      As far as the dog is concerened “eh-eh!” means whatever we have taught ti dog that it means. Similarly with “leave it!”. For us verbal, literally-minded humans WE might see the words as different but I’m sure that to a dog the meanings are EXACTLY the same.

      Actually the way I use the terms ‘uh-uh!’ is a very mild word simply meaning “don’t do that now” whereas “leave iit” is something I use only in dire circumstances which means :Your life depends in NOT DOING THAT!” (such as eating a dead fish with a hook in its mouth 😦

      • So, how do you know your dog realizes the meaning of “eh-eh” if you use it as “don’t do that now” – did you actually train the behavior of stopping, waiting, or leaving an object alone, in response to that cue? If not, it may still just be an interrupter that gets the dog’s attention, but doesn’t really tell him what you’d like him to do instead. Therefore, it’s a bit unpleasant, and that’s the only reason it temporarily works, whereas if it was trained, there’s definite meaning behind it because the dog would have been reinforced for an alternate behavior that was actually cued.

  7. KellyK says:

    I really like the concept of LIMA. I think that a huge part of how aversive something is varies from dog to dog. I’ve had good luck with “timeouts” with our foster dogs, but I can’t do it with my own dog because trying to take her by the harness and move her somewhere scares her (and I really don’t want to poison a recall by associating it with punishment). So, if she’s getting into something, I usually just try to redirect with “leave it.”

    I also think it’s probably important to distinguish between a correction and a cue. Anything that just conveys the information “I want you to stop that and go do something else,” whether “something else” is specified or left up to the dog is really more of a cue than a correction.

    My personal yardstick for punishments is that they should be annoying or boring without being painful or scary. Enough to make it worth the dog’s while to learn how to avoid it, but not enough to damage the relationship or the dog. But finding that balance is tricky.

  8. Marilyn Marks says:

    As KellyK said, there is a difference between a correction and a cue. When a positive trainer gets an undesired response to a cue, they sit back and think why and try to help dog learn what they intended, where a coercion trainer would correct until desired results. It’s one thing to get mad and punitive upon disobedience to a command and another thing to provide info to the dog. I know a coercion based trainer would say they were providing info, but I think it’s the mindset that is the determining factor of whether I’m going to like/approve or not of a trainer’s use of some aversive.

    I have to be fair and say that I have seen shock collar trainers who do set ups to gradually expose dogs to desired decision-making (after training in the cue) and set one criteria for the session that will be tweaked (corrected). These trainers seem to be minimal or softer in the zap level they need to use and also in attitude. They seem to be shaping behavior more than setting up for failure and punishing. They also seem to really know dogs, etheologically, and take responsibility for developing a great dog, whereas the traditional, coercion ones seem to just take everything personally and get meaner if their dog disobeys.

  9. Marilyn Marks says:

    I’m sure we’ve all seen dogs use corrections to deter other dogs. We’ve also seen dogs that lack self-esteem be overly demonstrative to downright reactive to get their point across. So +p and -r can work or not work even within the species. Same goes for humans – there are parents who can go “eh” and the kid stops and others that the kid goes right on about his/her behavior and it becomes a power struggle. Staying out of a power struggle is important so that it is not personalized and one doesn’t have to ramp up the corrections until the subjugated one can’t respond emotionalIy or physically. When it’s not a power struggle, corrections can be given and received as info, like the cold in the game of hot and cold.

    So easy to say, so hard to get right in real life, especially when we were all raised in a power struggle!

  10. Mel says:

    I have a question. I am not a trainer. I have two rescued Shelties with very different personalities, both of whom i have taken to obedience classes. One of them behaved (and behaves at home) beautifully under almost all circumstances. The other is a well-meaning goober. He wants to please, but there are other things he wants more. From little background I was provided, he was taken from a situation where he was pretty much left to do whatever he wanted. He is a car chaser. When he’s on leash being taken for walks he will do well until a car goes by. Then he begins to growl and bark and he tries to run after the car. On-leash, he also is aggressive toward other dogs. He jumps and barks at them. At the training class–a positive training methods class–I was told to divert his attention, turn him away and “treat, treat, treat” so his attention is focused on me. But Connor is too fast for me. Turning away only works for a second and a treat is more likely to be snatched hastily away from me (fingers included if I’m not fast enough!) so he can refocus on the car or other dog. What can I do that doesn’t involve violence if he’s more interested in attempting to attack cars/dogs than in getting a treat? How can I get his attention to stay on me?

    • wildewmn says:

      Mel, when Connor goes after the cars (or the other dogs), he’s over threshold. You need to work him in situations where he can be successful before putting him into the types of situations that are happening now. Working with a professional trainer who can set up situations with their own (bomb-proof) dogs, and show you how to work around cars, etc. would be helpful. If you go to the APDT website ( and click on Trainer Search, you can enter your zip code and find a trainer near you. 🙂

  11. I am of a mind that “corrections” should not be physical or painful, but need to be “dramatic.” The mouthy dog who puts his teeth on me in greeting or during play gets a sharp “AH-ACK” verbal signal and body language (horrified facial expression, cessation of play activity, probably followed by a Time Out) that says he has just crossed the acceptable behavior line.

    I think we all agree that we will make leash-jerk exceptions to keep our dogs out of the face of pissed-off dogs, out from under the wheels of passing cars etc. I have an Australian Shepherd with extremely high motion sensitivity whose excellent “Leave it!” (she can perform this behavior as cars, buzzy motorcycles, bikes etc. zoom past us 98% of the time for as long as necessary) is trumped when Speed racer suddenly & silently appears from behind us on his bazillion gear bicycle @ 30 mph, or the approaching bicycle has a dog racing alongside it. I have no problem with bellowing “Leave it!” and jerking her back to avoid disaster. Her general reaction appears to be “THAT was EXCITING!”

    Dogs are not delicate china teacups (except, perhaps, the ones with “teacup” in their name). Watch how they behave toward one another. I tend to think that the “don’t make a ripple” tendency of trainers who suggest we NEVER raise our voices, block or physically intervene is unrealistic and…undoglike. A recent example of this is a panel discussion I listened to by a number of very canny Famous Dog People on whether Nothing In Life Is Free (NILIF) is too aversive. Kathy Sadao’s “Plenty In Life Is Free” (PILIF) has since appeared, arguing that we are being too oppressive in our expectations of our dogs. The argument just seems ridiculous to me. In the first place, nobody is so on-the-ball that NILIF actually works out to anything but PILIF in any practical sense. In my own home NILIF breaks down to: Sit & wait for food, to get the leash on, to go out the gate or hop out of the back seat, to ask to get on the bed or couch.

    I’d much rather hear ideas that err on the side of caution & kindness than those emanating from the Dark Side, but theories & solutions should also be required to WORK.

  12. Evelyn Haskins says:

    Years ago my daughter told me (that she had read) that Anger is a symptom of frustration/incompetence.

    So I suppose you could say that “good trainers’ don’t use ‘harsh corrections’ ” is a truism 🙂

  13. Matthew says:

    I always come back to….”how do you correct/punish a killer whale if it doesn’t perform”. The answer to my knowledge is you don’t. It is my understanding that whale trainers learn to read their animals to the point they know when not to ask for something if there is any doubt of the whale responding. So the whale doesn’t learn it has a choice to not respond.

    in terms of formal training, I am not sure corrections/punishments of any kind is appropriate. lack of “compliance” in my opinion, exposes an opportunity to improve as a trainer and thus the training of your dog.

    In my experience so far, it seems those that cling to corrections/punishments also often take it personally, like their dog is giving them the middle finger “oh, can’t have that” followed by punishment. They often do NOT take into account, proofing, or is there something in the environment that is affecting the dog to the point they can’t respond, or even that maybe, just maybe they didn’t train the cue they thought they did. the assumption is always seems to be defiance.

    I also find it bizarre that if we can take a wild animal, such as a killer whale and teach it to say jump through a ring of fire and it never occurs to anyone that we need to figure a way to punish lack of “performance”. BUT we feel punishment/corrections are required with a domesticated, biddable, docile animal bred to respond and work with us.

    I know this will be a bold statement, particularly as doing so at this time is beyond my experience and skill. BUT, IF the “first step” was to learn to train without force/fear/pain, then it seem to follow that the next “evolutionary” step in progressing dog training to question “do we really need Ah-Ah aka interrupters or light corrections etc.” is the fact we use these things still an indication we have much to learn and room to grow?

    in other areas of interest that I have, the “top names” are people that never accept the status quo. Telling a dog “no” or “ah-ah” is NOT remotely the same as shocking a dog with electricity in the name of training. But that doesn’t mean there is still not a better, more effective way out there just waiting to be discovered/figured out.

    • Mel says:

      Your observation is a thought-provoking one, Matthew. However, the whales do seem to know they have the option to refuse. I was once at the Sea World in Orlando when it happened. The show was delayed by some 45 minutes because the whales didn’t want to come out, and no amount of coaxing could get them to. We had to wait until they were “ready.”

      • Matthew says:

        but my point is the trainers look to avoid that. they work to learn their animal, read them, understand them. many people by contract make little to no effort to read their dog. so, the 45 min show delay happens, but if it happened often there wouldn’t be a show or shows at all.

        The whale/dog comparison does have it’s limits. owning and living with a dog raise challenges that whale trainers don’t face or have to deal with.

        BUT there are valuable lessons to be learned from them none the less

    • A couple of things come to mind here: We must take into account the innate ability of an animal to understand the parameters of what we are asking them to do, or expect of them. “The Misbehavior of Organisms” by Breland & Bailey speaks to this. I’m talking off the top of my head here, but I would guess that one thing whales don’t/can’t learn is not to pee in the pool. Arboreal species like birds & monkeys whose evolutionary niche depends upon gravity to dispose of their waste have never been successfully housetrained.

      Modern dog training replaces physical “corrections” with interruption & redirection. I can’t just tell my dog “Ah-Ah” or “Leave it!” without letting her know what to do instead of the undesirable behavior – “Leave it, let’s go” for instance.

      I use a No Reward Mark (“Too bad”) during operant training to tell a dog “that didn’t buy you anything” rather than a jerk on a spiked collar if the dog doesn’t deliver a behavior we are learning. There was a flurry of discussion a while ago about whether NRMs are aversive and therefore have no place in “Positive Reinforcement” training (whatever that is). If they were accompanied by a slap on the nose this would be worthy of Real World discussion, but otherwise it is just so much academic noise – interesting in the virtual world but pointless in the world dogs actually inhabit, where they frequently resemble remarkably strong & well-coordinated pre-schoolers armed with box cutters.

      The motto of my business is “Wake Up & Smell the Dogs!” When I consider many of the discussions that take place in forums & discussion groups peopled primarily by those of us in the Science-based training community (some of whom consider themselves “balanced” trainers & justify use of P+ techniques because they employ “one of the quadrants”) I ask myself “do I smell a dog in this discussion?” If the answer is “no” then I can deduce that we have wobbled into “virtual dog” territory, where analysis of theory can trump analysis of reality because virtual speeding cars, for instance, do so little damage.

  14. Paul Cottman says:

    The problem with comparing whale training to dog training is that whales are not expected to assimilate into our society in nearly the capacity that dogs are. The whales we train live in a very controlled environment. We have substantially more control over the level of distraction as well as the available motivators. While I don’t condone the use of harsh corrections or punishment of improperly calibrated intensity, whales are not expected to live in our homes, go with us to the whale park where there are dozens of other strange whales, walk through the neighborhood, sit outside at outdoor cafe’s, ride in our cars, sleep in our beds, play with our children, etc. A friend of mine who trained whales at sea world for two years and is now a professional dog trainer said to me ” we don’t need to train whales in the same context as we do dogs. After all, we take dogs to the park and expect them to listen to us, off leash, come when called, sit, stay and then approach the other dog only when we’ve asked the owner’s permission. Even then, we want them to approach the dog in a friendly way. Could we ask the same of thing of a killer whale in the ocean? Probably not. Thankfully, we also don’t need to.”

%d bloggers like this: