The Dog Training Industry Cockfight
Research, Dichotomy, and a Path Forward
I started by writing an article to accompany a recent podcast conversation. The more I wrote, the more I realized the following was becoming a full-on exploratory story piece. It started with a conversation about a research paper. It evolved into the human tendency for dichotomy, for the “gap instinct” that divides - a term so beautifully coined by the late Swedish scientist Hans Rosling. Rather than editing it down to two or three shorter articles, I decided to let things flow naturally and take you along the ride of my thoughts and contemplations as they unfolded about a topic that is on the minds of dog trainers and behavior professionals around the globe.
This little ride may be bumpy with potentially controversial ebbs and flows, but I hope you settle in with your favorite tea (or tequila), give it an open-minded read, and if you feel inclined, add to the conversation.
The Podcast
I met with Jason Toy of Canine Scholars to discuss a recent research paper titled "The Efficacy and Welfare of Different Training Methods in Stopping Chasing Behavior in Dogs." I specifically asked Jason to join me because he has extensive experience with e-collar training and serves a different purpose in the industry, given that he focuses primarily on training. I focus primarily on behavior modification, giving us two different lenses to view and discuss the same information.
Most importantly, I invited Jason because he’s a refreshing human to converse with, and I respect him. Especially amid industry mudslinging ramped up by influencer propaganda and muddled with emotion-driven opinions, I wasn’t concerned that either of us would walk away personally offended or down a colleague.
Our conversation began addressing the research study and evolved into a discussion about the dog training industry's divide between force-free training and balanced training. You can find the paper HERE if you’d like to review it.
I’ve had a lot on my mind on the subject of the dog training divide and how research is intertwined with it, which I’ll expand on in this article. We’re going to get deep. But first, I’d like to point out Jason’s three most important takeaways from our conversation and two insights he shared before diving in with my professional and personal take on the state of the animal behavior and training industry and how it affects people and their pets.
Jason’s Three Top Level Items:
There’s no universal way to train or teach dogs.
The study's goal was not to show the conditioning of the (electronic) collar but to determine the fallout from the use of the e-collar to stop a specific behavior (in this scenario, chasing behaviors in high-drive dogs).
Studies should be used as another way to analyze what you are doing and what may be working or not working when training instead of laying the roadmap for how we teach.
Two additional points Jason brought to my attention during our conversation that I had yet to consider prior had to do with one of the professional trainers behind this study, Ivan Balabanov.
The first insight was the language barrier, given that English is not Ivan’s first language. This came up because I pointed out Ivan’s post announcing the study’s results, where he misuses words like “facts” and subjective phrases like “dogs trained with electric collars.” I found most of the words and phrases Ivan used in his post both undermining the actual scientific process and misleading. Jason mentioned that Ivan has a history of not always saying what he tries to convey in English. I thought this was very insightful and was somewhat embarrassed that I didn’t consider it prior. I take things very literally in English, which can sometimes impede intent in language. That said, while I agree this should be taken into account, I also don’t believe everything Ivan states fails to translate the way he intends - I think some of what he says may have more to do with Jason’s second point raised. And that is, Ivan has been repeatedly attacked online by force-free trainers poised to publicly shame him. A celebrity trainer is actively trying to get him banned from entering the UK. So he’s biting back.
I began losing respect for Ivan when he started mocking positive reinforcement-based methods by posting videos of parodies of training dogs with treats and (obviously) failing. Then he made fun of trainers who use affiliate links to make (very little) extra money. Dog trainers are not well-paid and, like veterinarians and technicians, don’t enter the field for the cash kickbacks. I found it arrogant and entitled to look down on those who need every financial lift they can get - especially coming from someone likely in the top 1% of earners in the dog training industry. Not to say he didn’t earn it - just to say it’s a dick move to mock people who need to be creative to survive while sitting on your proverbial cash pile. There are dog trainers who could run circles around Ivan; income is not an indicator of skill.
My respect clicked down several notches further after reading this research paper he touted as “clear findings” in what I considered a poorly conducted study with unreliable results. Perhaps a start to something, but unreliable.
I slid into the belief system that Ivan was turning out to be just another egotistical dog trainer with tunnel vision who only knew and accepted one lane. He knew nothing about companion animals, behavior modification, or using reinforcement effectively for positive reinforcement-based training. Some of that may still be true (and let’s be real - Ivan has an ego) - I don’t know, but my point is he’s a human with his own individual human response system, and I can’t really determine what he knows or doesn’t based on the small handful of videos I’ve seen. What I can determine is that he’s not a moron.
Given Ivan’s broad audience from years of success, he became an easy target and retaliated. I don’t agree with it, but it’s also not my place to say how another human being responds to threats and targeted hatred. I find it reminiscent of when Cesar Milan tried to take his own life. When a human feels so low that death appears to be the best option, things have gone way too far. I don’t agree with how Cesar Milan trained dogs, so I don’t train dogs like he did, but I certainly don’t support gathering the pitchforks and burning the witches down. With Ivan, I don’t have to agree with his approach either, but I’d much rather see a few mocking videos across my social feed that make my eyes roll than a post that he’s taken his own life. Maybe that’s an extreme way to look at things, but also, perhaps I’ve seen too much suicide as a result of hatred for one lifetime to err on the side of caution.
I am troubled by what this dog training divide has created between the force-free community and the balanced training community. The spotlight on harmful, abusive dog training practices was overdue and sparked needed and meaningful conversations about how we approach our relationships with animals. However, I fear that more emphasis is now being placed on what “side” everyone is on. We’ve lost the forest through the trees, and just as we would adjust the stress baseline of a traumatized dog, I think it’s time we adjust the stress baseline for the animal behavior and training industry.
The Study
The following is my professional opinion, with the occasional inclusion of thoughts brought up by Jason or myself during the podcast conversation (which you can access HERE).
Efficacy
“In this study, we compared e-collars to methods relying entirely on food rewards in order to stop dogs running after a lure.”
In operant conditioning, a reinforcer causes a behavior to repeat. It does not stop a behavior. Reinforcers vary in desirability from dog to dog and can include anything the dog really likes, such as playing tug, their favorite ball, access to a person, access to another animal, smelling something in the environment, free choice, treats or food, chase, and the list goes on. When choosing what is reinforcing to a dog in a given situation, we ask ourselves, is it the right reinforcer for the right dog at the right time in the right situation. Using food as reinforcement to avoid a lure, especially when selecting high prey drive dogs (dogs that find chasing extremely reinforcing) is setting the scene for failure. I can’t imagine any surprise from any dog pro that offering the choice between food or chase to high prey drive dogs, especially in close proximity, would lead to the dog choosing chase - I’d have bet my annual salary on that outcome. Food was a poorly selected reinforcer and utilized poorly in this study, which focused on punishment - a means to stop behavior.
In terms of the punishment chosen, stimulus from an electronic collar can be both novel, as Jason points out in our conversation (still potentially scary in my opinion), and also painful - as demonstrated in this study by delivering shocks at a level 6 out of 10 to start, and up to 20 times at once. Whether novelty or my theory, pain, the collar is punishing to the point the dogs yelp out in discomfort and further, to the point that two dogs were removed from the study because when the maximum of 20 shocks didn’t work, the trainer kept shocking the dogs surpassing the allowed number of shocks. That’s a welfare red flag for me when an animal continuously cries out in pain or discomfort, and the response from the human is to cause more pain or discomfort to the point of disqualification from the study. We’ll table welfare momentarily but dive in further shortly.
In dog training, teaching the dog not to do something also involves teaching alternative behaviors. You cannot simply suppress instinctual behaviors without a high likelihood of unintended consequence, rather, you provide appropriate outlets for the dog’s behavior (in this case, appropriate outlets to chasing and/or capturing prey) and draw a line between when those behaviors are appropriate and encouraged, and when those behaviors are not.
Learning is more complex than choosing between subpar reinforcement or the threat of punishment. We can not adequately judge the efficacy of training methods that are not effectively implemented to begin with. We can account for things such as an addictive adrenaline cycle that some dogs learn to self-perpetuate when too much stimulation is added.
We know punishment stops behavior. We know the chase is more exciting than food for high-prey-drive dogs. We know this is different from how learning or behavior modification occurs in real life, meaning we knew these were ineffective approaches to learning before the study started. So, what did we actually gain from this study?
Welfare
Looking at welfare poses the question: While we know punishment effectively stops a behavior, is inflicting pain in training and behavior modification necessary and ethical?
To quote the paper, “Aside from presumably pain-induced yelps in the dogs with e-collars when they received shocks, none of the dogs in any groups showed any signs of stress or distress. This sentence, to me, was a blatant dismissal of the dogs exhibiting overt signs of stress.
Depending on the individual, a dog yelping in pain is undoubtedly an overt sign of stress or distress. Given that we’re exploring welfare, and this was directly observed in the study, it is not simply anecdotal but rather recordable data relevant to the outcome. Even if, according to Jason’s open-minded insertion, it’s due to being startled by a novel stimulus.
In addition, pain tolerances, while subjectively calculated, are not taken into account despite the likelihood of generating different results from dog to dog. This study looks at inflicting pain or punishment to make observations and collect data, but it doesn’t account for varying sensitivities. Pain tolerance varies from dog to dog just as it does from person to person and sometimes from breed to breed. Most huskies respond to pain significantly differently than American Pit Bull Terriers, not just from a vocalization standpoint but physical and physiological differences that are measurable.
As part of this study's welfare component, fecal samples were collected to measure cortisol levels. This can be a helpful, noninvasive way to gauge stress in dogs—when it’s done with a specific individual over some time to develop a working baseline with well-journaled and analyzed behavior events day to day. It’s not as simple as saying cortisol levels are high, so the dog is experiencing distress, or cortisol is low, so the dog is not stressed.
Cortisol levels are not just related to distress; they are also associated with eustress, which is good stress. The difference between the two is how long the increased cortisol levels remain present over a period of time, which helps us make more determination, and even then, it’s still somewhat hypothesizing. Both good and bad stressors can occur within the same day and be completely unrelated to the study activities, making this exceedingly challenging to put any stake in. The dogs went from one stressful event to the next, such as being punished and then having fun, and we have no data on what happened that would affect their stress levels between each experiment throughout the study.
Furthermore, only nine dogs out of the nineteen provided fecal samples regularly, meaning there’s just not nearly enough data to put any stock in conclusiveness in this portion of the study.
Aside from this study's lack of a reliable measure of the welfare of the animals in the experiment, I keep coming back to the same pressing question: even if we know that using pain or punishment will stop a behavior, is it considered an ethical starting point for training another sentient being?
I relate learning in dogs to learning in children. With caution not to anthropomorphize, there are many similarities regarding their comprehension capabilities (outside of language reasoning) and even measurable emotional responses to physical punishment compared to dogs. In our podcast conversation, Jason mentioned that part of the purpose of this study was to look at the potential fallout of using the electronic collar as a physical punishment. Did the dog go and hide in the corner and cry, so to speak, or go have fun with his friends after receiving physical punishment? In the study, the dog had free choice and chose to have fun, seemingly unaffected, according to observations. Though this may give us a short-term answer, it doesn’t illuminate the complete picture.
Like physical punishment in children, the unintended consequences are not always immediate, which was another reason I was troubled by the short length of this study. An increase in aggressive behaviors in developing children and adults has been repeatedly studied and well-documented over the past 30 years. There’s a striking likeness between the results gathered from more extensive and compiled studies regarding the long-term unintended consequences of physical pain and punishment that don’t necessarily present until months or years later. The difference is that this has been studied more thoroughly and for longer in children than canines.
Subjectively, I also believe in the “does the punishment fit the crime” approach to consequences with children and dogs for several reasons, which we don’t have page space to discuss in the article. Though an example is to explain the consequence of an action, such as “If you don’t spend twenty minutes on your homework, you don’t get twenty minutes on your iPad today.” However, even if my child doesn’t spend twenty minutes on homework, the consequence is never to inflict physical pain or discomfort. It’s my belief system that harming a child with the hand that is also to comfort them is confounding. I’m even more lenient with dogs because they lack comprehension of language definition and reasoning. For example, they cannot comprehend deferred gratification from enduring momentary pain, as some may argue when using analogies like - “the tens unit hurts when I use it, but my muscles will feel better later.” That’s human comprehension, not canine. You cannot explain to the dog - “You may not like how the e-collar feels now, but if I do, I will feel comfortable letting you run off-leash through the woods, which you’ll love.”
In my many years of experience, I have worked time and again with dogs who were trained using fairly significant physical punishment and force who seemed “fine” and cooperative for a couple of years until they “all of a sudden” started attacking the owner, attacking other dogs, biting kids in the home or visiting strangers. There are two commonalities I’ve consistently found in these dogs, one of which is excessive physical punishment, with the most common form being shock delivered through an electric collar (or sometimes two collars at once).
I need to pause from my study remarks and come back to the dog training divide momentarily. The many clients and dogs I have worked with in the example I just stated were not trained by who I consider the average balanced dog trainer. This is a critical part of the conversation because the dogs I mentioned above were abused, in my opinion, by trainers who hide behind their “years of experience” and impressive party tricks to excuse their behavior and continue practicing without persecution. While they would be considered on the same “side” as every other balanced trainer due to the current labeling system, it’s unfair and unreasonable to lump them in the same category. There is a sliding scale, and it matters to the lives of dogs and the people that love them.
Jason, for one, has referred many clients to me over the years for behavior modification. Some he has trained on electronic collars, and some he has not. Not a single dog that he’s referred to me has shown signs of being abused or shut down during my work with them, and not a single client has ever shared with me something about what Jason has said or done that has caused me pause - not even for even a moment.
On the flip side of the divide, it’s unreasonable and unfair to assume all positive-reinforcement-based trainers (force-free trainers) and consultants are incompetent because you’ve witnessed failure with food or you’ve personally failed with food. It takes experience, education, competency, and often guidance to fully understand how to utilize the right reinforcement for the right dog at the right time in the right situation. There’s also something to be said for being well practiced in interruption, understanding neutral operants, redirection, and, most importantly, what only experience can lend you - adapting your strategies based on the individual in front of you. It’s an education that online courses, YouTube, or textbooks can’t replace.
I also find a critical common misconception among people not immersed and well-seasoned in the effective use of positive-reinforcement-based strategies: all stress is bad. Whether you choose to believe it or not, growth does not happen in the comfort zone, and almost every organism on this planet grows stronger under the right amount of stress. An example is when people stop taking antibiotics before the full course because they “feel better.” The infecting bacteria in your system are left weakened under stress but not destroyed. The bacteria make a mighty comeback and then mutate, becoming resistant to the very antibiotics you took to kill them, rendering the antibiotics useless the next time you are infected.
There’s something to be said about the right amount of stress, which is necessary in my field - behavior modification - to help animals overcome fears, conflicts, or anxieties that can rule their lives and those of their pet parents. Resiliency, the capacity to withstand or recover - is critical. But there’s a line. When too much stress is applied (including something that really hurts or overexposure to a trigger, for example) in dogs dealing with complex emotions, it can quickly get dangerous. The dog's mental state can’t be ignored when your practice is to improve the overall mental state of the dog. Dog training is slightly more straightforward (not as in dog trainers are not as skilled - to be very clear) because dogs are often neutral or have fun in dog training from the start. With behavior modification, we sometimes have to start at the not-so-fun parts and work our way back to things being fun again.
I don’t believe the underlying argument between force-free and balanced trainers (at least for the middle gray, not the outliers) is about the difference between a zero-sum stress environment and a highly stressful environment when training. I believe it’s about where the in-between line exists. Much like liberals who are Pro-Choice but believe in a line at which abortions should no longer occur. Or conservatives who are Pro-Life but believe there is a line and abortions are necessary when the mother’s life is in danger. Jason and I discussed the parallels during our conversation about politics and the dog training industry. No matter your individual beliefs, you get lumped into the same camp as (often extreme) red or blue, even if you’re unmistakably purple. You also get judged by both sides if you're too purple, which I find rather ironic.
Inconsistent Variables - Back to the Study
I recently was a substitute teacher for a private school's fifth and sixth-grade science class. Ironically, the topic was the Scientific Method. I already have a biology background and have gone through the experimental process more times than I could count in college, but teaching these methods to young minds is even more fun than our college experiments.
This is relevant because these students learn the importance of an independent variable and a constant even at a fifth and sixth-grade level. The parameters in this study muddy the process by using too many factors or variables. Some of these include:
The age of the dogs. The study enrolled puppies who were eight months old up through adult dogs who were five months old. Eight-month-old puppies are different behaviorally, socially, sexually, and maturely than fourteen-month-old dogs and two-year-old dogs and five-year-old dogs, and so on. More valuable data could be collected and compared by independently testing age groups and then combining the results for analysis.
Bidability - how easy a dog is to train varies from dog to dog, just as it does from person to person. What were the biddability ratings for the dogs selected for the study?
Small Population Size—The results used data from seventeen out of thirty dogs. More than half of the study population couldn’t be included in an already small population.
Only nineteen of the thirty dogs selected due to having a high prey drive (desirability to chase) completed the full study (aside from consistently providing fecal samples for cortisol levels) due to not being “consistently motivated to chase.” Why were high prey drive dogs not consistently motivated to chase if they were specially selected due to their high chase drive? Is this where more selective dismissal of data occurred as well?
Being offered a food reinforcer - or chasing (a reinforcer): While I’ve already touched on this, more variables come into play. If they chased, they were punished. If they continued to chase, they continued to receive punishment. If they stopped chasing after punishment, they were allowed free choice, which consisted of many options, such as interactions with other dogs, free choice to play independently, etc.
The dogs “had zero to very limited experience with the e-collar.” “At most, some owners had introduced e-collar training once or twice.” Additional variables rather than keeping the study population constant with experience, conditioning, and exposure prior.
Thoughts between Jason and I converged in the conversation on the simplicity of this study and the need not to give it too much weight when deciding how to train dogs. Further research and meta-analysis of many studies are needed to extract usable information to consider application in training methodologies.
A Path Forward
No side really wins when only one side wins.
We are societal beings that rely on cohesive coexistence literally for our survival. When we are only willing to see our “side,” we ignore the fact that we have things in common with other humans, and that dismissal turns them into the “others,” the enemies. Then, they get treated as such. Dog professionals with different opinions and different methodologies are not enemies.
An excellent parallel is between vegans and carnivore-diet proponents. Vegans have varying reasons for their choices, often intertwined with their belief systems, as do those who believe in the carnivore diet. Some beliefs are based on morality, some on biology, and some on health. When you read those three bases for beliefs, did veganism or carnivorism specifically come to mind? You may believe each point is specific to one side. Still, the reality is that the moral compass, biological processes, and health arguments are fought from both a vegan perspective and a carnivore perspective.
Neither vegan nor carnivore diet proponents will convince each other to make a drastic nutritional change and alter their belief system by shaming the other or, even more so, expecting an overnight change. Opening the door to conversation with the possibility of exploring any change starts with common ground. It starts in the middle—the grey—the purple.
It starts with talking about how pesticides sprayed on crops kill more than bugs - they kill the animals that eat the bugs. It starts with discussing what animals and plants are bountiful and which are threatened due to over-farming or over-fishing. It starts with talking about how factory farming has changed over the years but how far removed the practice is from treating animals as the sentient beings they are. It starts with being open to talking about the pros and cons - because they exist on both sides of the equation. Once these conversations are started and carried out respectfully, a beautiful thing happens: people walk away from them and make their own educated decisions about how to move forward personally. You don’t have to agree with them, though you just may!
It starts with conversing to listen and learn, not just talking to be heard, with dropping the self-righteousness that feeds our biases and closes our listening ears.
Where to begin
That said, I encourage you to have the conversations and ask the questions openly. I encourage you to avoid using those conversations against each other for clicks and likes and building an army of yes men.
I encourage you to find common ground. As it pertains to the animal training and behavior industry, I believe most of us can start by agreeing that abuse is not okay. Negligence is not okay. Being a registered business that carries insurance and is transparent about education, experience, and limitations is a basic requirement to operating a legitimate practice. Can you start here if you can’t bridge the gap anywhere else?
I encourage you to fight the urge to make everyone see things “your way,” which, if you found some magical way to bend everyone to that will, how the result would be a total loss of agency and the beautiful diversity that is the human species. Agency and diversity are two cherished qualities we fight to protect in the dogs we serve - why not the people, too?
I encourage you to be aware of the pendulum swing. When conversations about change arise, they can sometimes overcorrect - swing the opposite direction just a little too far before finally finding the right rhythm from center side to center side like a steady grandfather clock.
I encourage you not to twist information like research, statistics, or other findings to fit a personal agenda or feed your ego. Instead, take a (mostly - we’re all human) objective approach to the findings presented.
I encourage you to proudly but kindly express your beliefs and give the same grace to those you differ with as you would expect from them in return. I encourage you to do your best, not allowing those who may mock you, threaten you, or take advantage of your vulnerabilities to eat away at your core. You are not their words or opinions, just as you are not defined by one act. You are a collection of your behaviors, and those behaviors can change over time.
Putting a Bow on it All
This current and relevant conversation affects so many lives. Dogs, dog professionals, families, shelters, rescues, and anyone who comes into contact with dogs are affected, whether directly or inadvertently, by the dog training divide conversation.
Dogs, by law, are still considered personal property in many regions. By definition, personal property is moveable and not fixed to real estate, including tangible items like cars, furniture, jewelry, and electronics. Yet advancements in science have opened our eyes to the fact that dogs (and many other animals) are actually sentient beings that experience a vast array of emotions previously thought only to be indicative of our human species. For many, that drastically changed how we viewed relationships with dogs, which can be summarized by the language shift from “owning a dog” to being a “pet parent.”
For some, dogs are still treated like inanimate objects that are here to be shoved into our lifestyles whether they fit or not. Gone are the days of the white picket fence dream, and present are the days of the perfect brewery and dog park dog who loves everyone. It’s just not reality. I find that old-school trainers are still trying to sell that reality and end up ruining dogs, while new-school trainers try to reason with pet parents about what a happy medium is for both the dog and the person. Mostly. But not always - and that’s an important part of the conversation, too.
Aside from training practices evolving with the help of research and, anecdotal or not, good ‘ol hands-on experience, our companion animals are also evolving. This generation of companion dogs is not the same as ten generations ago. They aren’t raised the same, they aren’t eating the same foods, they aren’t bred the same, they aren’t the same dogs in shelters that were in shelters ten generations ago, and they are more and more being thrust into a lifestyle that they weren’t designed to be in. This is important because what we may have done twenty years ago that worked swimmingly, whether from a behavior, training, nutrition, or medicinal perspective, does not mean it can be applied the same to Bruno and Bella today.
The adage—the only way to get yourself out of a hole is to stop digging—isn’t just a wise financial analogy; it’s also a wise perspective on how to change the course of an otherwise futile path the training divide has paved. Regulation in the animal and training industry is likely coming. We can allow politicians without experience in the field to sponsor bills from the loudest voice or biggest check and play the wait-and-see game with our careers. Or we can shed some differences to come together on commonalities so that professionals at least have a dog in the fight that could significantly impact their income and business.
There’s a fine line between confidence and cockiness. I don’t remember who first said that to me (it may have been Stephen Belenky), but it rang true and still does. So with that, I say I’m a very confident handler, but I'm not arrogant enough to believe I know it all. I’m also a confident, introspective person, but I’m not cocky enough to withdraw from others' feelings or sentiments as though only mine are of importance. If there’s only one takeaway you walk away with, I hope it’s confidence in the notion that not everyone is as far from each other in this “divide” as it may seem.
Is it possible that without micromanaging and Monday-morning quarterbacking each other to death, we can find common ground, agree to disagree at times, and still be in collective agreement to stand against extremes as a starting point? Perhaps, for the love of dogs, we can start mending things by focusing on what we can control rather than what we cannot - our own practices and our leadership by example.