Alright, dog trainers. Today I'm responding to a video by Dr. Lisa Radosta. Her video was made as a response to a comment by Dr. Michael Bailey. Michael Bailey is the new president of the American Veterinary Medical Association. It was during a podcast where he was asked about the use of e-collars in dog training. And I guess it's important to be very clear about what Dr. Bailey actually said. He did not promote shock collars as a first line tool. He did not encourage casual or careless use. What he said was this: sometimes when the alternative is a dangerous dog or euthanasia, using an aversive tool - used humanely, intelligently and conditionally - may be the better option. That's it. That was the statement. And for that he received a massive backlash.

Now Dr. Radosta is a board certified behaviorist. She's also the president of the American College of Veterinary Behaviorist. She's one of the most influential voices in the force-free world today. That world holds a very clear position: No aversives ever, under any circumstances. That's not my interpretation. That's their published position. And anybody can go and look it up. Before I go any further, I want to say this clearly. Dr. Radosta is very intelligent. She's articulate, she's persuasive, and she's highly trained in what she does. I'm not afraid to say that openly. This isn't about intelligence. It's about the framework. Often very smart people can still operate inside rigid systems that do not allow them to see outcomes that fall outside that system.

And when framework matters more than outcomes, you tell me who pays the price.

What concerns me here is that Dr. Bailey's statement was measured, conditional and focused on safety and survival. However, it was met with an ideological response that allows zero wrong for nuances whatsoever. A lot of what is said in Dr. Radosta's video sounds authoritative. It sounds scientific, but it's driven far more by ideology than by serious dog training results. And the real world consequences of that matter. Dogs stay on SSRIs for years. Dogs get surrendered to shelters. Dogs get euthanized. Not because the owners or trainers didn't try hard enough, but because the positive only approaches - even when combined with the psychotropic medications - were just simply not enough.

So I'm going to go through this video as it plays claim by claim. This is not going to be a personal attack. However, it's a direct challenge to the force free ideology and to what that ideology does to dogs in the real world. This is not about what feels nice. It's about what actually works and what keeps dogs alive.

So, here is Dr. Radosta's first claim.

“Let's get at it. What does the peer-reviewed literature say about the use of shock? Whether it be an underground fence or a collar as it relates to training in dogs? The literature is very clear. Shock increases fear, stress and aggression.” Literature is very clear. Shock increases fear, stress and aggression.

Well, that statement sounds decisive, but it's not accurate. Acute stress responses do appear in some of the studies, particularly when aversive stimulation is unpredictable or poorly applied. There is a big difference, and no one would ever argue with that. What I disagree with is the claim that the literature is “very clear”, and that e-collars universally increase fear, stress, and aggression. 

So let's go back to science. Schalke et al 2007. The paper was published in Applied Animal Behavior Science and demonstrated something critical and important. Predictability and controllability determine welfare impact in dogs that could clearly associate the stimulation with their own behavior. So, in that study, they had a lure that was running. And if the dogs were to touch that prey, there was no significant rise in salivary cortisol and no persistent stress indicators. Even after four weeks, when they were retested.

Now, in contrast, dogs exposed to random or poorly timed stimulation did show elevated cortisol. Nothing surprising here. When ideology erases this distinction, it falsely portrays all e-collar use as inherently harmful. That is not a conclusion supported by the literature as a whole period.

Now let's go to claim number two. 

“Even when used correctly by a skilled trainer, it still changes the welfare of the animal, which is measurable in stress signals and also in physiologic stress measurements. And that is even true outside of the training situations. So in the home or in other situations” Even when used correctly by skilled trainers, it still changes welfare even outside the training situation.

Now this claim is repeatedly asserted but it is never ever demonstrated. Again if we go back to Schalke et al, it directly contradicts the idea of persistent welfare harm. When stimulation is predictable and controllable. Another frequently cited study Cooper et al 2014 published in PLOS One. Also does not support this sweeping conclusion. In that study, salivary and urinary cortisol showed no significant differences between dogs trained with e-collars and dogs trained with non aversive methods. Behavioral differences such as increased yawning and tension were observed, but no physiological evidence of chronic stress or lasting welfare impairment was found.

The only significant cortisol elevation reported in that paper occurred in a small preliminary study involving high intensity stimulation delivered without proper sensitivity testing or warning cues, a method that the authors themselves distinguish from best practice use. Declaring that dogs must be secretly stressed in daily life without individual evidence, it's not science, it's assumption.

Alright, so let's go to claim number three. 

“It's more stressful for animals, but it's not more effective. In studies that look at highly skilled trainers on both sides - shock and positive reinforcement - using shock is not more effective in training skills.” 

Now, this claim relies heavily on studies that never tested what they claim to evaluate. Going back to that most frequently cited example, Cooper 2014, that study is repeatedly referenced as evidence that reward based training is just as effective, if not better, than electric collars for stopping chasing behavior. However, there is a big problem. The study actually never tested predatory control. The dogs were trained on leash. The sheep were confined in small playpens. There was no free-running chase, no activation of the predatory motor pattern, no test of recall, and their genuine predatory arousal. In other words, the study avoided the very condition it claimed to evaluate. You cannot make this up really.

No one can demonstrate control of predation without allowing the predation to occur without that trigger activating that system. And that's not a minor limitation. It completely invalidates the study, and uses evidence for claims about real world chase behavior. Despite this, the paper continues to be cited in the force-free community is proof that reward based training is equally effective even though the central question was never tested.

By contrast, Johnson and Wynne 2024, published the paper in Animals, Did what Cooper et al did not dare to do. They allowed the dogs to actually chase. They tested real behavior, and the real motivation and the result was not subtle. The e-collar group rapidly inhibited chasing. The no-aversive groups did not. That difference matters because dogs are not injured or killed in laboratories. They are injured or killed outside, on the streets, on the fields.

Alright, let's go to claim number four. 

“Claims that shock collars are a last resort, and they must be used for animals who have severe behavior problems just isn't supported by the literature or my clinical experience. Right? Standard of care is to refer those animals to a board certified veterinary behaviorist, a certified applied animal behaviorist partnering with a force free dog trainer.”

This claim collapses when studies actually test high motivation behavior. Johnson and Wynne 2024 demonstrated that e-collars rapidly and reliably inhibited chase behavior, while non aversive methods failed under the same conditions. And that's very important. Also importantly, the authors report that no significant differences in urinary cortisol between groups and even more importantly, no evidence of ongoing welfare impairment beyond the brief discomfort during the learning stage.

The last resort is only doctrine on its own. It's also just a dangerous one. It allows undesired, dangerous behaviors to rehearse and strengthen while owners and trainers are told to keep trying methods that are clearly not working. Early, fair, skillful boundaries prevent escalation. It's very simple. Waiting until a dog becomes unmanageable, that's not humane. That's my opinion.

Claim number five. The class action lawsuit for 1.2 million, whatever dollars. It sounds really good. 

“A class action lawsuit was recently settled for $1.9 million. That was filed against a shock collar company for marketing their products as humane and safe. When the people using those products noticed that their dogs were more fearful, more aggressive, had skin damage or even burns on their skin.”

Well, let me tell you this: a lawsuit is not scientific evidence. The cited class action settlement concerns marketing claims and consumer complaints. The company settled without admitting wrongdoing. So just a classical routine business decision. There was no court that found that e-collars, when properly used, cause inherent behavioral harm. If a lawsuit determined the opposite then we would ban medicate. Like if lawsuits determine truth, we would ban medication, the harnesses, gentle leaders and countless other things. And the thing is, science is not conducted in courtrooms. It's conducted in data. It's conducted in real life training and evaluations. Okay. So there is nothing. This is just absolutely nothing in there to hold a straw. 

Claim number six.This one's a good one. 

“There is a reason why cities, counties, provinces, and entire countries have banned shock collar use. There's a reason why veterinary organizations across the globe and dog training organizations have come out to speak out against shock collars. The evidence is just overwhelming that these particular ways of training dogs increase stress, fear and aggression.” There is a reason cities, countries, counties, provinces have banned shock collars.

This argument relies on the implication that bans reflect overwhelming scientific consensus and they do not. Okay. The majority of countries in the world have not banned e-collars. Most countries, including the United States, still permit the use. Even when restrictions exist in some countries in Europe they are often narrow. And guess what? Filled with exemptions. “Well, this is bad, but this group can use it or this military unit can use it or whatever.” There is a reason for that. That is a good reason for that. These bans did not arise from new scientific discoveries. They came from political pressure, lobbying campaigns and emotional framing, not from evidence demonstrating unavoidable welfare harm under skilled use.

In the United States, attempts to ban electric collars are often introduced quietly and opportunistically attached to broader welfare bills, or pushed at the municipal level, where public scrutiny is low. These attempts fail every single time, and they fail because trainers, owners and working dogs’ handlers, conservation groups and actually many veterinarians push back with real world outcomes. Fewer bites, fewer livestock deaths, fewer dogs that are surrendered, fewer dogs are being euthanized. If the science was truly overwhelming, these bans would easily pass.

Well, they do not.

And when legislation fails again and again, it tells us something important. When the full picture is presented. Outcomes, welfare and risk. The case for prohibition collapses. 

“Positive reinforcement works, and it doesn't break down the human animal bond. Our obligation as veterinary professionals, as dog trained professionals, as pet parents, is to use the most humane and scientifically accurate training methods for the health, safety and welfare of everyone involved. The dog, the family and the public. ”

You're seeing these three dogs here. They are not an exception. There are many more of them. Zeke, Tina, Sammy, these dogs who are not theoretical. These are real dogs that were failing. They were running out of time. Balanced training saved them. Not because it's harsh, not because it ignores welfare, but because it allows boundaries to exist alongside reinforcement. Clarity alongside compassion. And again, these dogs are not exceptions.

Every single week, trainers and owners quietly saved dogs that were labeled too much, too dangerous, unfixable, and so on. They don't make headlines. They don't publish papers. They just go home with dogs that are alive.

So here is what I want to do next.

If you're a trainer or a pet owner and balanced training helped your dog when nothing else worked, I want to hear from you. Share your story. Share a photo. Share the dog that is alive today because someone was willing to step outside that ideology and do what actually worked. These stories matter not to win arguments, but to remind people that behind every policy, every position statement, and every restriction, there is also a dog whose life matters.

I'm going to close this with the postcard that I got for Christmas and Happy New Year from Zeke. I'm getting goose bumps as I'm showing you the card that I have read already numerous times. This dog is alive today. When the force-free community failed him, he was going to be euthanized. Think about that.

Sources

Cooper, Jonathan & Cracknell, Nina & Hardiman, Jessica & Wright, Hannah & Mills, Daniel. (2014). The Welfare Consequences and Efficacy of Training Pet Dogs with Remote Electronic Training Collars in Comparison to Reward Based Training. PloS one. 9. e102722. 10.1371/journal.pone.0102722.

E. Schalke, J. Stichnoth, S. Ott, R. Jones-Baade, Clinical signs caused by the use of electric training collars on dogs in everyday life situations, Applied Animal Behaviour Science, Volume 105, Issue 4, 2007, Pages 369-380, ISSN 0168-1591, https://doi.org/10.1016/j.applanim.2006.11.002.

Johnson, A.C.; Wynne, C.D.L. Comparison of the Efficacy and Welfare of Different Training Methods in Stopping Chasing Behavior in Dogs. Animals 2024, 14, 2632. https://doi.org/10.3390/ani14182632

Rodasta.“The ACVB’s Response to Recent Statements Made by AVMA (American Veterinary Medical Association) Leadership Regarding the Use of Shock Collars to Train Dogs”: Linkedin.com, 8 Dec. 2025, www.linkedin.com/posts/dacvb_the-acvbs-response-to-recent-statements-activity-7403840962688593920-HQHp. Accessed 27 Jan. 2026.