Choosing an Approach to Training

The other day, someone asked a question that stuck with me: out of all the different trainers out there in the world, and all the different approaches, how do you choose what works for you?

It's no secret that I'm a staunch adherent of purely positive training. Why and how that happened, though, is not something I've given much conscious thought to in a while.

I'm a perpetual student. There is so much out there that I don't know, and so much that I want to be able to do with my dogs, that I am forever on the hunt for new ideas and resources. But how to decide which ones are credible?

I. Does It Work For the Dogs?

This is the most basic and important test. Do the dogs seem happy, attentive, and successful? Or are they stressed, shut down, and/or hopelessly confused? Putting my own ego and preconceptions aside, as best I can, I look to see whether the trainer's methods are working -- really working -- for the dogs. Because if the dog is not learning, then the training has failed.

Suzanne Clothier has described a happy, understanding dog as having a "light in its eyes," and I think that's a wonderful way of putting it. A dog who understands what is being asked of it, and who is joyous about doing that thing, has a brightness of spirit that I don't see in dogs who have been bullied into performing. The trainers I want to emulate are the ones whose dogs clearly love doing their work. I see that joy in Emily Latham (kikopup)'s dogs; I see it in Suzanne Clothier's, and Denise Fenzi's, and Julie Flanery's. I see it in the demo dogs that my instructors at Y2K9s show. I want my dogs to share that same happiness, so I emulate those trainers'  methods -- which happen, not coincidentally, to be purely positive methods. At the same time, I shy away from punitive methods, because I have seen the thousand-yard zombie stare in the eyes of a dog trained on a shock collar, and I never ever want my dogs to look like that.

As importantly, because we're training for competition, I want dogs who are precise and focused and reliable. A happy but undisciplined dog who can't execute a straight Front, or who runs amok on the course, is no good to me. I'm competitive: I want to win. I want ribbons and titles and championships. I want a canine partner who works with me seamlessly, with an almost telepathic connection on the course.

And again, it's no contest: purely positive methods carry the day. A dog who loves his work, and who above all loves working with his special person, will be laser focused on that work. A dog who's been clicker shaped with precision will have straight Fronts and sharp Heeling and (critically for freestyle) an enormous repertoire of flashy moves that simply cannot be obtained through compulsion. And because clicker training rewards the dog for paying extremely close attention to his handler, it develops that communicative bond to an incredible level.

So purely positive training passes the first test. It works for my dogs. It works for Dog Mob, it works for client dogs, and it's worked for all the Temporary Dogs without exception.

In the very beginning, and only with Pongu, I did briefly attempt force-based training in the form of leash corrections. That definitely did not work. I cringe every time I remember it: this four-month-old fearful puppy getting leash pops from me, the only person in the world he trusted, because he was pulling on lead in a desperate attempt to escape from a big scary world that terrified him. It didn't work, of course. It only damaged his trust in me. I am grateful that I saw the message in his eyes and stopped before any worse damage was done.

II. Does It Work For Me?

The second question is whether I can effectively use the methods being espoused. It's not kikopup or Pat Miller who's going to be training my dog. It's me. No matter how good they are at training their dogs (and they are very good!), if I can't understand what they're telling me to do, or can't carry out the programs clearly and consistently, then again, the training will fail.

Clicker work does require developing some new skills. Timing and observational skills are key to spotting desired behaviors (particularly in the early stages of shaping a new behavior, when the incremental steps will often be quite subtle). There are mechanical skills required in juggling the clicker, treats, target object(s), and/or leash, and often this has to be done while on the move. Eventually, treats have to be faded out if the behavior is to become truly reliable.

At this point, I'm comfortable using a clicker and treats. I know how to manipulate reinforcement rates and read the dog's body language and adjust my approach on the fly to accommodate the specific animal with whom I'm working. But if I did not know those things, and did not have the patience to learn them or apply them properly, then positive training might not work for me.

I've found that this is often the case with people who dismiss operant conditioning: they never learned how to apply the techniques properly and concluded that it was a fault of the training method, not of their implementation.

The flipside of that, of course, is that force-based training will never work for me because I don't know how to apply those techniques effectively. Of course I understand how to apply positive punishment in theory, but I've never done it in practice and I never will, because I know that I can get better results using more humane methods. I have no interest in learning how to ear-pinch a dog into retrieving or training a recall with a shock collar. I don't need to resort to those things to establish a reliable behavior. But for me to say those methods categorically "don't work" would be untrue, because they do work for some dogs and some trainers who are willing to inflict pain and fear, and accept a certain amount of collateral damage, in the name of instilling obedience. I'm not willing to do that, so those techniques don't work for me.

(This is, incidentally, why "balanced" or "crossover" trainers who tout their willingness to "use all tools" drive me batty. At first glance it might seem like a reasonable thing -- who wants to reject potentially useful tools out of hand? -- but then one has to reflect on what tools are being rejected, and why. We generally don't use switches to whip unruly children anymore, and we don't use public stocks to humiliate nagging spouses, and while there are no doubt people who think the world would be a better place if we still employed both of those barbaric "useful tools," I personally would not like to have them as parents or neighbors. Nor, I imagine, do their dogs enjoy having them as owners.)

III. Is It Scientifically Sound? Does It Make Sense?

There is a lot of misinformation circulating in the dog-training world. Metric craptons of it.

One quick filter to get rid of a lot of the garbage is to ask whether the explanations being offered for a particular technique stand up to basic scrutiny. Much of it, especially when it comes to "being the alpha" and dominance mythologies in training, is pseudoscientific hogwash and can safely be discarded.

Much more fails a basic common sense test, such as explanations of how prong collars work by mimicking the mother dog's teeth on her youngster's throat (uh, the pressure on a prong collar is at the front of the throat -- how many mother animals have you seen carrying their offspring around by their windpipes?), or admonitions against soothing fearful dogs because "it'll encourage the behavior" (as Debbie Jacobs is fond of asking, would a loved one patting your back make you eager to rush out and embrace something that scared you witless? Or might it just make you feel marginally safer?).

Sometimes the lack of scientific rigor is harmless. Canine homeopathy is getting to be a thing these days, and it is no more proven than homeopathic remedies are in humans, but as long as it doesn't substitute for appropriate veterinary care, I don't see how it hurts anybody to try. "Can't hurt/might help" is fine. But if there is potential to cause harm, then I don't think there's any excuse for uncritically accepting misinformation. And in general, I have found that positive trainers and writers tend to have more scientifically accurate explanations for their methods.

IV. Is the Source Credible?

I'm more inclined to accept someone as an authority if I know she is respected as such by other knowledgable persons in her field. Publication credits in independent journals or magazines carry weight, as do books published through external houses. The CPDT qualification matters to me -- if nothing else, it shows this person is familiar with the fundamentals of scientifically based training. Strong writing and clearly organized thoughts presented in a professional manner make a good impression too.

Also important is whether this person can do something with their dogs that I want to emulate. Have their dogs won titles that I aspire to get on my own dogs? Can their dogs do cool tricks that I want mine to learn? Have they successfully rehabilitated (not just shut down) difficult behavioral cases? Whatever the goal is, I want to learn from someone who has done it, and who has preferably done it repeatedly in a forum subject to knowledgeable, independent scrutiny.

All the most successful trainers in all the fields of endeavor that interest me espouse positive training methods. I take that as pretty good evidence that this is, in fact, the best way to get those things done.

V. What Are the Consequences of Getting It Wrong?

I will make mistakes. I make mistakes all the time. I drop treats, I mistime clicks, I reward in the wrong position and pull my dogs out of line. Often I misread my dogs' communications or fail to notice things that are distracting them and incorrectly assume they just don't want to pay attention. My handling cues are frequently sloppy and I miss opportunities to reward my dogs for correct behaviors and, all in all, I blunder so often that some days it's a wonder they know how to Sit.

So it is important to me that whatever training techniques I use be forgiving of those mistakes, because I'm going to make them and I'm going to make a lot of them. It's safe to assume that I will blunder every single new technique at least a dozen times while I'm learning how to do it.

If I make a mistake while clicker training, it's not a disaster. Worst case, I confuse my dog for a minute and have to spend another couple of repetitions clarifying that what I wanted was a backwards scoot, not a pop-up into Bow. Or maybe I drop a treat on the Rally course and we get knocked 3 points. It might cost us a placement ribbon (and it has!), but it's really not a big deal.

But if I make a mistake with a choke chain or a shock collar, I could do some real damage to my dog's psyche (and, less likely but possible, his body). I could inadvertently cause him to hate training, or fear approaching strangers, or lose his trust in me. These aren't hypotheticals; I have seen dogs who suffered these outcomes (and, as noted earlier, I almost did it to Pongu). To me, the risk is just not worth it. I don't want that to happen intentionally; I surely don't want it to happen by accident.


So that is pretty much how I landed where I landed, and how I evaluate new sources of information that I stumble across. I want happy dogs who are good at doing what they do, and this seemed like the best way of getting there. Now that I am there (although still with many greater heights to scale!), I look back and am satisfied with the road that I've chosen.