People are creatures of habit. Dog trainers more so.
We insist as a rule that there’s nothing new under the sun and everything worth doing has been done before (and quite a few things not worth doing as well). To the professional dog trainer, success is our bread and butter, and we all understand on some level the danger of messing with a winning team.
We might talk about lifelong learning and extoll the benefits of getting out of our comfort zones and trying new things. But the truth is, most of us stretch only as far as we are comfortable and then return to our safe space with a few adjustments or variations on a theme we know well already.
This is progress in our eyes.
Sometimes, we become “crossover trainers” which means we leave one school of thought entirely for another one. Positive trainers become “balanced” and balanced trainers become “positive” trainers.
But for the most part, trainers seems to start with a handful of strong ideas about how dog training should be done, and those themes remain intact for the bulk of their careers. We all get stuck in time a bit.
I speak from experience.
When I first encountered what I now refer to as motivational dog training, I hated it! It made me angry because I knew it looked amazing, but I also “knew” it was a lie.
It was a gimmick.
Dogs were only working for a reward. I “knew” good dog training didn’t need a bunch of rewards. It wasn’t based on a gimmick. It was about building the dog’s “character.” The well-trained dog should do what he was told without the promise of a reward. This was common-sense and as far as I was concerned, it was indisputable.
In my eyes, those prancing “happy” dogs weren’t happy, they were neurotic. The training used on them was designed to bring them into an unnatural state of excitement. It seemed unfair to me. I would argue that the dog was actually being harmed by this unnecessary and unrealistic expectation of nervous excitement masquerading as “joy” to the untrained eye.
Like I said, it looked amazing and I was worried that people were going to start insisting that I give them that with their dogs. So I prepared my arguments and talking points for when the question was asked. I built complex explanations of the difference between character and working for pay. I was prepared to point out that dogs don’t feed each other tidbits in order to get each other to do what was expected. I had prepped to explain that in the natural world, dogs used punishment far more than reward and was ready to ask any potential client if we thought we knew more about teaching dogs than dogs do (BTW, the answer to that is so obviously “Yes” that I’m ashamed that I didn’t see that at the time!).
I cut my teeth in model of obedience that was focused on building the dog’s character. It was believed that the properly trained dog was working for us, out of respect, not out of self-interest. That was what I told my clients, and that was what I believed.
I believed that leash corrections for failure to comply with a known command were analogous with a nip a dog might give another for being rude or insubordinate. I believed very much in the pack hierarchy theory of training. So to be clear, I had no problem using a sharp leash correction to adjust a dog’s behavior, but offering them more than a scratch or a pet for the right thing would be detrimental to the dog’s character. My definition of “character” included working to avoid a punishment, but not working to earn something.
Working to avoid something = character.
Working to earn something = self interest.
This wasn’t even a thing to be questioned. The truth of it was so obvious that it needed no further explanation.
Of course I didn’t quite see it that clearly. What I would have said if someone had asked me is the dog is not really working to avoid the correction most of the time. Most of the time, he was working because of the relationship we had built, but if he flaked for any reason, the realities of behavior required that I correct it right away.
So, every once in awhile, his character needed a bit of help. And surprisingly, the more often I gave corrections the less often I needed them. But somehow, I would still tell my clients (sincerely) that those corrections were simply building the dog’s character more than teaching him avoidance.
The dog who was treat trained, on the other hand, was just engaged in self-serving behavior. Seeking a reward was somehow less noble than avoiding punishment.
I also would point out, “A dog should be serious when he’s working.”
Work isn’t supposed to be fun, and it shouldn’t need to be fun. School doesn’t need to be fun. Character is often about doing things that aren’t fun.
So a dog trained in a way that made obedience fun didn’t have the character necessary to work when the exercises weren’t fun. That sort of training was a joke really. Dogs trained this way couldn’t be counted on. It might look great in the controlled environment and familiar routine of an obedience competition, but it wouldn’t translate to a 45 minute walk in the park. I was certain of this.
As it turned out, no client ever asked for this style of obedience, so all the conversations became discussions with other trainers who shared my dislike of this “new” style of obedience.
The problem wasn’t with the training at all. It was with my own assumptions about what I thought I knew.
The fact is, I had a thick lens of ideas through which all ideas, concepts and techniques were viewed. Part of this was, “This is what a well-trained dog looks like when working.” Since I was equipped with that piece of information, it was easy for me to see that those enthusiastic dogs didn’t look like the dogs I trained, therefore, they obviously weren’t well-trained.
This all made sense to me back then. In fact, now 10 or more years later, I’m only beginning to understand how wrong I was.
Over the past six-months I have been incorporating more and more “motivational” methods into my work and the results have been far more encouraging than I had hoped. It’s not that I’ve replaced what I did before, but I’m layering more games, more high energy engagement and yes, more exciting rewards. The dogs are having more fun, I’m having more fun, and my clients are having more fun, and most importantly, they are getting the job done. The only downside is that I’m still learning this model and I often make rookie mistakes. But those mistakes rarely present any thing more than a short-lived detour and an opportunity to learn.
Apparently, I’m not the only one making this switch lately.
The whole reason I’m writing this is that I’ve been seeing a number of comments and discussions on social media using many of the arguments I used to make about this style of training. There seem to be more every week. Trainers who are my friends, and who have been instrumental in my own development are making waves among old school trainers with statements that can only be perceived as being anti-enjoyment. They dismiss the use of food. They dismiss the use of games, and they even dismiss a happy demeanor. The “seriousness” argument is in full swing.
Most interesting is they say it doesn’t matter if the dog is happy or not, as long as it’s obedient. This is a very interesting position, and a handy bit of rhetoric, and on a certain level, it is ultimately the truth.
If I had to chose between a dog who unhappily stays put instead of running in front of a train happily, I’d certainly choose the former.
But the problem is that you are literally comparing two worst case scenarios. If you can only get the dog to stay put (and therefore safe) grudgingly, then by God, get him to stay grudgingly. I have no problem with that. But “better than dead” doesn’t equate to great or even good training. It is literally the bare minimum to keep the dog safe.
If we can choose between a dog who stays put grudgingly and a dog who enjoys playing the “stay” game and does so joyfully, why wouldn’t we shoot for the latter?
It turns out, for many, the question is not so simple.
Many of my friends and respected colleagues would argue that the former is more reliable.
They would suggest that the dog who stays reluctantly is more reliable because he’s well-versed at denying his own ideas for the sake of obedience. They would say he’s proofed better against distractions because he’s not working for a reward but rather doing what is expected of him even when he doesn’t want to.
The problem with that reasoning is that he actually does want to.
We’ve put him in a game of “would you rather” and he’s choosing the least noxious of two options.
He’s choosing the path he believes will be most rewarding … Just like every other being on the planet, dogs will always do the thing they believe will be most rewarding. There is no escaping this fundamental law of behavior. Of course that statement becomes very complex very quickly because rewards come in many forms and layers, but it still remains an undeniable truth. The expectation of rewards ultimately drives behavior.
The dog who is doing the obedience grudgingly is ultimately doing it to avoid a negative outcome of some sort. It may be minor, but it is present. If the dog is particularly in tune with his handler, a disapproving look may be“punishment” enough. But the dog is still primarily motivated to avoid a consequence.
The dog who is joyfully obedient isn’t thinking about the consequence of failure, but is focused on the joy of doing.
Dr. Robert Sapolsky, at Stanford University, demonstrated that the opportunity to earn a reward can produce more dopamine than the actual reward. In fact, in his research the dopamine associated with the reward went away once the reward was produced. So the joyful dog is actually enjoying not the reward, but the opportunity to earn the reward.
As nice as this is, it might bring about the suggestion that this dog’s obedience will last only so long as he still believes that reward is likely. This is a valid concern but thankfully one that has been dealt with by trainers using this model. There are good defenses against it.
If you take the time to build the dog’s “tolerance for non-reinforcement” you can create a very stable dog in the absence of the reward. (By the way, the tolerance for non-reinforcement is the “character” I used to refer to when I built it through teaching avoidance).
The other problem with the assertion that this dog isn’t reliable is the assumption that the dog is working only for the external reward. I don’t believe this is the case. If a good relationship is built, then the dog will also not want to disappoint us, the fear of disappointing his handler is real, but unlike the dog in the first case, it isn’t at the forefront. It’s a secondary or even tertiary concern. But it still affects the dog’s idea about what behavior is most rewarding.
Motivational training is not devoid of corrections. The key difference between it and what I used to do is that it seeks to build a love of the work in the dog. I want the dog doing it primarily because he enjoys it, rather than because “I said so.”
So,in one case, have motivated the dog fully, we have a dog who actually enjoys what he’s doing, and character has been built through a systematic process of delayed and even denied rewards, all of this buttressed against the dog’s desire to work with us and for us because we matter to him.
The other scenario is we have a dog who doesn’t like doing what he’s doing but will do it because he doesn’t want to disappoint us and (maybe) he’s afraid of a correction.
It seems absolutely clear that the dog with no love for the work has fewer elements pushing him towards compliance than the joyful one. There is no reason to conclude that the motivated dog is less reliable except that it doesn’t match with what we might believe “good training” looks like.
But old ideas die hard, and the concept of the noble dog sacrificing his joy to be obedient has a certain degree of romance to it. And I don’t think there’s anything necessarily wrong with it, but it’s not the only valid idea of what a well-trained dog looks like. Too many times we let our theories get in the way of seeing something that’s quite beautiful because we are looking for something else. And this, I think, has been the primary obstacle to my learning over the years. When I think of all the dogs I could have helped more if I’d only explored this idea further when I first encountered it, it makes me a bit sad. But coming late to the party is better than not showing up at all, I’m here now, and hopefully, I’ll have something useful to add to the conversation down the road. Until then, I’ll be working and learning.