Bangura (2026) — Aversive Dog Training Policy Library

A Position Paper

The Scientific Case Against

Aversive Dog Training Equipment and Methods

Convergent Welfare Evidence and Policy

Recommendations for the United States

Will Bangura, M.S.

CAB-ICB, CBCC-KA, CPDT-KA, FDM, FFCP

April 2026

Table of Contents

Foreword

I have spent years working as a force-free dog trainer and behavior consultant, and the longer I spend in this field, the clearer one observation becomes. The science is not the problem. The peer-reviewed welfare research on aversive dog training equipment and aversive handling techniques has been consistent for a long time, and the international veterinary profession has reached a formal consensus position. The problem is that the people who need to use the science, the practitioners, the guardians, the legislators, often do not have it pulled together in one place where they can see how thoroughly the evidence converges.

That is the gap this paper exists to close.

The peer-reviewed welfare research does not point in different directions depending on who reads it. Independent research groups, working in different countries, using different populations, different methodologies, and different outcome measures, have repeatedly found the same pattern. Aversive control is not necessary. It does not produce training outcomes superior to reward-based methods. It comes with welfare costs that reward-based methods do not. Some of those costs are behavioral. Some are physiological. Some, in the specific case of neck-pressure equipment, are mechanical, and they have been documented in peer-reviewed veterinary research. The welfare costs are real, the lack of necessity is established, and the international veterinary profession has reached a formal consensus that aversive equipment should not be used.

What I have rarely seen, and what I set out to provide here, is a single document that pulls all of that evidence together in one place. Behavior analysis, canine training outcome research, stress physiology, affective state assessment, clinical behavior medicine, nociception science, threat circuit neuroscience, peer-reviewed veterinary research on the physical effects of neck-pressure equipment, real-world survey data, and regulatory precedent from the dozens of jurisdictions that have already acted: each of these literatures, on its own, points the same direction. The agreement across them is what scientists call convergent evidence. It is the strongest kind of evidence a body of research can produce, and it is the kind of evidence we have on this question.

The reason there is still a public debate about aversive training equipment in 2026 is not that the evidence is in dispute. The evidence is settled. The debate persists because the evidence has not been consolidated, communicated, and operationalized at the level where guardians and policy makers actually encounter it. That is a problem the practitioner community can solve. It is a problem the practitioner community is best positioned to solve. Trainers, behavior consultants, certified applied animal behaviorists, and veterinary behaviorists are the bridge between the peer-reviewed welfare science and the pet guardians in the public. If that bridge is not built and walked across, the science stays in the journals, and the marketplace stays the way it is.

The kind of change I would like to see in the United States, the kind that is already in place in Wales, Switzerland, Germany, France, Australia, and a growing list of other jurisdictions, does not start in a legislative chamber. It starts with one person, in one location, talking to the guardians in their world. Each conversation moves the field a little. Enough conversations, multiplied across a country, change what the field looks like. That is how a public welfare consensus becomes a policy consensus, and that is how a policy consensus becomes law. The legislators who eventually carry these reforms forward will be responding to a constituency that already exists. Building that constituency is practitioner work.

This paper is written first for the practitioners who are doing that work. The science, the citations, the institutional consensus, the comparative jurisdictional record, and the policy logic are consolidated here in a form that is meant to be used. Read it. Cite it. Share it. Use the deployment toolkit in Section 7.6 for conversational versions of the core arguments. Use the proponent objections catalogued in Section 9 when you encounter them in real time. Use the comparative jurisdictional table in Section 7.4 when you talk with legislators or with other policy advocates. Use Section 10 when you are asked to recommend policy.

The paper can also serve as a reference for future legislative work, and I expect it will. That use is welcome. But the first step is education and organization. Anyone who is committed to force-free training and to ending the use of fear, pain, intimidation, and aversive equipment in dog training has a role to play. Get educated. Get organized. Get involved. Speak up. Advocate for change in the laws of your state. Silence is the only way the current consumer marketplace continues to operate. The convergent welfare science is on the side of the dogs. What the dogs need now is a practitioner community willing to carry that science out of the journals and into client conversations, into professional standards bodies, and into the legislative chambers where policy is actually written.

It is 2026. The peer-reviewed welfare science has been consistent for years. The international veterinary consensus has been formal since 2024. Multiple jurisdictions have operated under bans for more than a decade without producing any published evidence of harm from prohibition. The case for change is not novel, and it is not radical. It is the case the evidence has been making for a long time. What this paper offers is the synthesis, the language, and the references the practitioner community needs to make that case, in our own voices, to the people in our own communities. The dogs in those communities cannot speak for themselves. We can. We should.

That is why I wrote this paper.

Will Bangura, M.S., CAB-ICB, CBCC-KA, CPDT-KA, FDM, FFCP

Mesa, Arizona

How to Use This Paper

This is a long paper, and most readers will not need every section of it on a first read. The guide that follows points each kind of reader to the parts most likely to be useful for what they are trying to do. The paper works two ways at once. It can be read straight through as a sustained argument, and it can be returned to as a reference document. Table 1 (the jurisdictional summary), Table 2 (the quick-reference welfare evidence summary), the proponent objections in Section 9, and the terminology decoder in Section 9.12 are written for repeat reference, not single-pass reading.

Reading Paths by Audience

Pet guardians and clients new to this material. Start with the Executive Summary for the plain-language case, then read Section 2 (the functional definition of aversive control) for the operant conditioning vocabulary used throughout. Section 3.6 consolidates the welfare studies into one readable page, and Sections 10 and 11 state what the paper is asking for and why. The neuroscience in Section 4, the proponent objections in Section 9, and the device variability data in Section 8 are reference material best left for a second pass.

Trainers and behavior consultants newer to the underlying science. The working core is Sections 2 (the functional definition of aversive control), 3 (the convergent welfare evidence base), 3.6 (the quick-reference summary of those studies), and 9 (the anticipated proponent objections). Section 4 (nociception and mechanical injury) is study material for a second-pass read, once the operant framework in Section 2 is internalized. Section 9.12 (common terminology misuses and their behavior-science translation) is high-value preparation for client conversations and online engagement.

Experienced trainers, behavior consultants, and veterinary behaviorists. A full read is recommended, with particular attention to Section 4 (the nociception and mechanical-injury literature on neck-pressure equipment, including Carter et al. 2020, Pauli et al. 2006, Hunter et al. 2019, and the Grohmann et al. 2013 case report), Section 5 (the necessity claim and how it has been empirically tested), Section 7 (institutional consensus across veterinary, welfare, and professional bodies, including the comparative jurisdictional data in Table 1), and Section 8 (real-world use, device variability under Lines 2013, and the regulatory vacuum). The Pessoa and Knight personal communications in Sections 4.3 and 9.4 are particularly valuable for rebutting controllability and predictability arguments, and they are not available in any other current publication.

Policy advocates and legislative readers. The policy-focused materials are the Executive Summary, Section 7.4 (which presents Table 1, the comparative jurisdictional summary covering twenty-five enacted jurisdictions across Europe, Australia, Quebec, Latin America, and US states, alongside a survey of pending United States state-level legislative proposals in New York, New Jersey, Massachusetts, and Rhode Island), Section 10 (Recommended United States Policy, with five subsections covering equipment scope, standards of practice, confrontational handling, rationale, and supporting infrastructure), and Section 11 (Conclusion). The full evidence base supporting these recommendations is detailed in Sections 2 through 8 and is available for legislative referral, citation, and reference.

Using This Paper for Specific Practitioner Activities

Client conversations. Sections 2 (the functional definition of aversive control), 3.6 (the quick-reference summary of welfare studies), 4.1 (the distinction between nociception and tissue damage), and 9.12 (the terminology decoder) provide ready language and concise study summaries that translate the science into client-accessible explanations. The general principle to communicate to clients is in Section 2.1: aversive control is defined by what the procedure does to the dog's behavior, not by the label or vocabulary the trainer or manufacturer uses to describe it.

Online debate and social media engagement. Section 9 anticipates eleven specific arguments that balanced trainers commonly deploy and provides a direct rebuttal to each. Section 9.12 catalogs fourteen rhetorical and terminological moves grouped into four thematic categories, gives the accurate behavior-science translation for each, and provides response phrasing usable directly in conversation for the high-frequency items. The Pessoa and Knight personal communications, referenced throughout but particularly in Sections 4.3 and 9.4, are especially valuable for rebutting controllability and predictability claims, because they come directly from the senior authors of the neuroscience studies those proponent claims rely on.

Continuing education and professional speaking. Sections 3 (welfare evidence), 4 (nociception and mechanical injury), 5 (necessity claim), and 7 (institutional consensus) provide a structured argument suitable for course development, conference presentation, or professional webinar. Tables 1 and 2 are presentation-ready visual aids. Section 8 (real-world use) provides the consumer-protection framing that connects the welfare argument to public policy and is particularly useful for cross-disciplinary audiences in veterinary medicine, animal welfare advocacy, and consumer protection. The complete reference list at the end of the paper provides ninety-seven entries with DOI hyperlinks for citation and verification in derived materials.

This is a position paper, not a neutral literature review. It makes a case. It is written first for the practitioners who actually have to deploy the science: trainers, behavior consultants, certified applied animal behaviorists, and veterinary behaviorists who need a comprehensive reference they can learn, internalize, and use in client conversations, online debate, continuing education, and professional advocacy. Practitioners are welcome to read, cite, share, and build on the material in their own work.

Executive Summary

The United States should adopt animal welfare policy banning the sale and use of aversive training equipment for dogs. The equipment in question is electronic collars (remote, bark-activated, and containment variants), prong or pinch collars, choke collars (also called choke chains or slip collars), and spray collars (citronella, scentless air, and ultrasonic-or-audible-tone variants). Alongside that prohibition, the United States should adopt the Hierarchy of Dog Needs® and Best Force-Free Practices (Michaels, 2022) as the national standard of practice for commercial dog training and behavior modification. This standard is aligned with the position statements of the international veterinary profession, the largest national veterinary associations in the English-speaking world, the major animal welfare organizations, and the leading professional training and behavior bodies. The case for both is built on multiple independent lines of scientific evidence that all point in the same direction: behavior analysis, canine training outcome research, stress physiology, affective state assessment, clinical behavior medicine, the science of nociception (electrical and mechanical), threat circuit neuroscience, peer-reviewed veterinary research on the physical effects of neck-pressure equipment, survey data on real-world use, and regulatory precedent from multiple jurisdictions.

The core of the argument is a definition. When an aversive piece of equipment or an aversive handling technique reduces a behavior, it is functioning as positive punishment. When a behavior increases because performing it terminates, avoids, or prevents an unpleasant event, it is functioning as negative reinforcement. Both mechanisms require the stimulus to function as an aversive event for the dog. That requirement is the same whether the aversive is electrical stimulation from an electronic collar, mechanical nociceptive input from a prong collar, compressive airway and vascular restriction from a choke collar, or direct physical confrontation such as an alpha roll, scruff shake, leash jerk, or helicoptering. Aversive control comes with welfare costs. Those costs include stress-related behavior, conflict behaviors, suppressed body language, conditioned emotional responses to cues and context, negative affective bias, and, in some studies, elevated cortisol. For the equipment that applies physical force to the neck, the costs additionally include measurable mechanical effects documented in peer-reviewed research: elevated intraocular pressure during ordinary pulling, neck pressures in injury-relevant ranges, and, in at least one peer-reviewed case report, severe ischemic brain damage leading to euthanasia following a punitive choke-chain hanging technique. Those mechanical-injury findings apply specifically to prong and choke collars, not to electronic collars. The welfare case against electronic collars does not rest on mechanical injury, because electronic collars do not operate by mechanical force. The case against electronic collars rests on the functional properties of aversive control itself: nociceptive engagement, threat-system activation, stress physiology, and conditioned emotional response. None of these welfare costs, across any of these tool categories, are erased by lower stimulation settings, by a trainer self-testing the device on their own arm, by skilled application, by a fifteen-second video of an engaged dog, or by adding food rewards on top of the aversive contingency.

Proponents of aversive equipment have not demonstrated necessity. Cooper and colleagues (2014) and China, Mills, and Cooper (2020) both ran controlled studies showing that dogs trained without electronic collars achieve outcomes equal to or better than dogs trained with them. Head-to-head comparative studies pitting prong and choke collars against reward-based methods in everyday pet training contexts have likewise failed to produce evidence of necessity, and the international professional consensus treats all three categories as equally inappropriate. The one contemporary study frequently cited in favor of electronic collars for predatory chasing, Johnson and Wynne (2024), establishes narrow efficacy under specific experimental conditions, not necessity. Its protocol design has been challenged in the peer-reviewed literature (Bastos et al., 2025) and in the present author's separately published methodological critique (Bangura, 2025, SSRN). Herron, Shofer, and Reisner (2009) add a more specific finding about confrontational handling: alpha rolls, dominance downs, scruff shakes, and hit or kick corrections produced aggressive responses in a substantial fraction of the dogs on whom they were attempted. These methods are not recommended practice for any dog, and they are particularly contraindicated for dogs already presenting with aggression.

Two leading fear-conditioning researchers whose published work is frequently cited to justify electronic collar use, Dr. Luiz Pessoa (corresponding author of Limbachia et al., 2021) and Dr. David Knight (corresponding author of Wood et al., 2014), have each independently confirmed in written correspondence that their research does not support treating predictable, controllable aversive stimulation as neurologically neutral or benign. Both researchers clarify that controllability attenuates threat-related neural responses without eliminating them, and that a reduced response is not the absence of fear or stress (Pessoa, personal communication, 2026; Knight, personal communication, 2026). That clarification reinforces what the dog welfare research already shows: aversive-based training, whether delivered through electronic, mechanical, or confrontational means, comes with welfare costs that do not disappear with skilled application.

Professional consensus is explicit and international. In June 2024, four major international veterinary organizations, the Federation of Veterinarians of Europe, the Federation of European Companion Animal Veterinary Associations, the Federation of European Equine Veterinary Associations, and the World Small Animal Veterinary Association, unanimously adopted a joint position paper formally calling for a complete ban on the sale and use of electric pulse training devices including electric shock collars for dogs, and broadly stating that equipment and devices that cause pain or discomfort to modify behaviors should not be used and should be strongly discouraged by veterinarians and other allied professionals (FVE, FECAVA, FEEVA, and WSAVA, 2024). The American Veterinary Society of Animal Behavior explicitly names electronic collars, prong collars, choke chains, leash corrections, and other forms of physical or psychological punishment as aversive methods that should not be used under any circumstances (AVSAB, 2021). The American Animal Hospital Association Behavior Management Guidelines name electronic shock collars, prong or pinch collars, choke chains, alpha rolls, and physical punishment as techniques that can harm or destroy an animal's trust, negatively affect problem-solving ability, and increase anxiety, concluding that the only acceptable training techniques are non-aversive, positive techniques (AAHA, 2015). The American College of Veterinary Behaviorists, the American Society for the Prevention of Cruelty to Animals, the British Veterinary Association, the British Small Animal Veterinary Association, the Australian Veterinary Association, the Canadian Veterinary Medical Association, the New Zealand Veterinary Association, the European Society of Veterinary Clinical Ethology, the Royal Society for the Prevention of Cruelty to Animals, Humane World for Animals (formerly the Humane Society of the United States), Dogs Trust, the UK Kennel Club, Battersea, Blue Cross, Best Friends Animal Society, and a growing list of professional dog training and behavior organizations all independently recommend reward-based methods and oppose the use of aversive tools and aversive methods for training and behavior modification.

Jurisdictions around the world have already acted. Wales banned electronic collars under a Welsh Statutory Instrument made under the Animal Welfare Act 2006. Switzerland's Animal Protection Ordinance prohibits spiked, pinch, and electronic collars. Germany (under case-law interpretation of its Animal Welfare Act), Austria, France (in professional contexts), Spain, Sweden, Finland, Norway, Denmark, the Netherlands, Slovenia, Belgium-Flanders (full ban in force 2027), and Colombia have all enacted bans on various combinations of these devices. Quebec's Animal Welfare and Safety Regulation prohibits collars that interfere with breathing or cause pain or injury, language the Ministry of Agriculture has applied specifically to electronic and prong collars. In Australia, Victoria, Tasmania, and Queensland have banned pronged collars through their respective state legislation, and electronic collars are prohibited in the Australian Capital Territory, New South Wales, South Australia, and Queensland. The Australian federal government additionally bans the import of pronged collars. The United States is now out of step with international veterinary consensus and with international regulatory practice.

The empirical record on how aversive training equipment is actually used in the real world compounds the welfare and necessity arguments. Masson, Nigron, and Gaultier (2018) surveyed 1,251 respondents in France about electronic collar use. They found that 71.8 percent of users operated the equipment without professional advice, that 75 percent had tried two or fewer alternative methods before reaching for the collar, and that 7 percent of dogs on which collars had been used presented with physical wounds (Masson et al., 2018b). Lines, van Driel, and Cooper (2013) examined the electrical characteristics of thirteen commercially available electronic training collar models and documented an eighty-seven-fold range in stimulus energy at maximum settings across products marketed for the same use category, ranging from 3.3 millijoules to 287 millijoules. None of those products is required to disclose voltage, current, pulse width, or waveform at the point of sale (Lines, van Driel, and Cooper, 2013). The United States has no Food and Drug Administration regulation, no Consumer Product Safety Commission standard, and no state-level technical standard for these devices. That regulatory vacuum stands in stark contrast to the medical-device framework applied to therapeutic transcutaneous electrical nerve stimulation units for human use. United States public attitude data conducted by Edelman Intelligence and reported by Petco in October 2020 found that 70 percent of dog guardians believe shock collars negatively impact their pet's emotional or mental wellbeing, that 69 percent consider shock collars a cruel training method, and that 59 percent of pet guardians would prefer to shock themselves than their dog (Petco, 2020). On the available evidence, the substantial majority of United States guardians already perceive these devices as harmful and would not lose access to a product they value if aversive electronic training equipment were removed from the consumer marketplace. A prohibition is, in this framing, a consumer protection measure as much as an animal welfare measure. It removes from the marketplace a class of product whose advertised claims are not supported by peer-reviewed evidence, whose adverse effects are documented but not disclosed, and whose appropriate use requires professional gatekeeping that does not currently exist in the United States.

The recommended policy has two complementary components. The first is a prohibition on aversive training equipment, removing electronic, prong, choke, and spray collars from the consumer marketplace through legislation that regulates sale, import, and use. The second is the adoption of the Hierarchy of Dog Needs® and Best Force-Free Practices (Michaels, 2022) as the United States national standard of practice for commercial dog training and behavior modification, enforced through state licensure of trainers and behavior consultants. The first component is delivered through legislative architecture already established in Wales, Switzerland, Germany, France, and other jurisdictions that have enacted similar bans. The second is delivered through the same scope-of-practice mechanism by which welfare-affecting professions, including veterinary medicine, mental health counseling, and social work, are already regulated in the United States. Equipment prohibition without standards of practice leaves confrontational handling techniques (alpha rolls, dominance downs, scruff shakes, hanging or helicoptering) unaddressed. Standards of practice without equipment prohibition leaves the equipment available on the consumer market. Both components are necessary. Active legislative interest in the licensing component is already in motion at the United States state level, with pending proposals in New York (Assembly Bill A 6985 and Senate Bill S 7723) and New Jersey (Assembly Bills A 4206 and A 4207) that would establish dog-trainer licensing tied to non-aversive standards, and pending Massachusetts legislation (House Bill H 2342 and Senate Bill S 1459) that would exclude electric, prong, and choke collars from court-ordered dangerous-dog behavior modification plans, as documented in Section 7.4. Together, they accomplish what the international veterinary, professional, and welfare consensus has called for: the comprehensive removal of aversive control as the basis of canine training and behavior modification in the United States. A United States policy of this kind is not a radical proposal. It is a measured, conservative response to convergent scientific evidence, international veterinary consensus, and the welfare risks documented in the peer-reviewed literature.

1. Introduction: The Wrong Question and the Right One

Debates about aversive training equipment and aversive training methods almost always get trapped inside the wrong question. The wrong question is whether these tools and methods work. The right question is how they work, what welfare costs come with that mechanism, whether those costs are necessary, and whether broad public access to such devices and methods is justified when safer alternatives exist.

Electrical stimulation delivered contingently can decrease unwanted behavior. So can mechanical nociceptive stimulation. So can physical confrontation. Any sufficiently unpleasant consequence delivered contingently can decrease unwanted behavior. None of that is in dispute within learning theory. But efficacy alone does not establish welfare neutrality, necessity, or ethical justification. Many interventions can suppress behavior. The capacity to suppress behavior is not, on its own, a sufficient reason to leave an intervention on the consumer market or in the professional standards of practice.

A dog may stop chasing, lunging, barking, or failing to recall for several different reasons. The underlying emotional motivation may have changed. Another behavior may have become more reinforcing in that context. Or the original behavior may now predict an unpleasant consequence that the dog is working to avoid. Those are not equivalent outcomes. One reflects changed motivation or learned alternatives. The other reflects suppression, escape, avoidance, or threat prediction. Policy that treats suppression as equivalent to resolution is not evidence-based policy.

The argument that follows applies to all aversive training equipment and all aversive handling techniques as a single, unified category. That is not a stylistic shortcut. It is the way the international veterinary and welfare consensus already treats these tools and methods. The American Veterinary Society of Animal Behavior, the American Animal Hospital Association, the Australian Veterinary Association, the British Small Animal Veterinary Association, and the joint international veterinary position paper adopted by the Federation of Veterinarians of Europe and the World Small Animal Veterinary Association in 2024 all treat them this way. The reason is scientific, not rhetorical. The mechanism of action across electronic, mechanical, and confrontational modalities is the same: positive punishment or negative reinforcement via an aversive event. The welfare concerns that follow from that mechanism are the same. The convergent research documenting those welfare concerns spans all three modalities. Treating these tools and methods as separate debates has served mainly to let proponents pivot from one to the next when any one of them comes under scrutiny. A unified policy approach closes that escape route.

2. The Mechanism Is Aversive Control

2.1 Functional Definition, Not Descriptive Language

In behavior analysis, an aversive stimulus is not defined by what it looks like, what it is called, or what intensity it is set to. It is defined by its function. A stimulus is aversive when its presentation decreases the future probability of the behavior it follows (positive punishment) or when its removal, avoidance, or prevention increases the future probability of the behavior that produces that outcome (negative reinforcement). This is a functional definition, not a descriptive one. It applies identically to electrical stimulation, mechanical pinch, neck constriction, verbal threat, physical correction, or handling confrontation.

When an electronic collar reduces barking, the stimulation is functioning as a positive punisher. When a prong collar reduces lead pulling, the prong pressure is functioning as a positive punisher. When a choke chain correction decreases lunging, the neck compression is functioning as a positive punisher. When a dog stops jumping up because an alpha roll has been performed in that context, the alpha roll is functioning as a positive punisher. When a dog walks in heel position next to the handler to avoid a leash-jerk correction, the threat of the correction is functioning as a negative reinforcer. In every one of these cases, the behavior change mechanism requires the stimulus to function as an aversive event for the dog. No vocabulary choice eliminates that functional requirement. Calling the electronic pulse a "tap" or a "signal," calling the prong collar a "communication tool," calling the choke chain a "training collar," or calling the alpha roll a "boundary setting" does not change what the dog's nervous system is experiencing.

2.2 The Intensity or Force Dial Proves the Point

Every aversive training tool involves a dial, explicit or implicit. Electronic collars have stimulation intensity settings. Prong collars have prong sharpness, collar tightness, and the handler's choice of correction force. Choke chains have the handler's choice of jerk force, the dog's neck diameter, and the speed of the pop correction. Confrontational handling techniques vary in force of application, duration of hold, and degree of positional compromise. In every case, the handler is adjusting the magnitude of an aversive event.

The existence of that dial, and the clinical necessity of turning it up when the dog does not respond, is itself the evidence that the mechanism is aversive. If the stimulus were not functioning as an aversive event, turning the dial up would have no effect. The fact that behavior change depends on finding a threshold intensity or force that the dog works to avoid is definitional proof that the threshold is an aversive threshold. Proponents of these tools who argue that the tools operate below the threshold of aversiveness are making a claim that is logically inconsistent with the operation of the tools.

2.3 The "Barely Perceptible" or "Gentle Correction" Contradiction

A common proponent argument is that modern electronic collars, modern prong collars, or skilled use of choke chains operate at levels so low that they are effectively imperceptible, or are at worst a mild attention-getter, not a painful or frightening event. This argument runs into a logical contradiction. If the stimulus is behaviorally meaningful enough to change behavior through escape, avoidance, or suppression, it is by functional definition aversive. If it is not behaviorally meaningful, it is not doing the training work. There is no intermediate category where the stimulus is both strong enough to reliably shape behavior and mild enough to have no motivational salience.

The same contradiction applies to the argument that prong collars merely "self-correct" and do not cause pain, and to the argument that a properly fitted choke chain only "communicates" without discomfort. A prong collar that produces no unpleasant consequence when the dog pulls will not train the dog to stop pulling. A choke chain correction that produces no unpleasant consequence when performed will not train the dog to stop the behavior it follows. The behavior change mechanism is the aversive event. Denying the aversive event while claiming the behavior change requires a form of learning that does not exist.

2.4 Human Self-Testing and Trainer Demonstrations Are Not Welfare Assessments

Proponents of electronic collars sometimes demonstrate the tool on their own forearm to argue that it produces only a mild sensation. Proponents of prong collars sometimes show that the prongs do not puncture human skin when pressed lightly against a hand. Proponents of choke chains sometimes demonstrate that a skilled correction on another person's wrist feels like a minor tug rather than strangulation. Every one of these demonstrations shares the same category error. Human sensory, anatomical, and affective responses to aversive stimulation are not equivalent to canine responses. Human pain thresholds, skin thickness, neck anatomy, nociceptor density, and threat circuitry differ from the canine equivalents. Canine haired-skin epidermis is approximately three to five cell layers thick, considerably thinner than human epidermis (Affolter and Moore, 1994). Human cognition and context knowledge also differ in important ways from canine cognition and context knowledge. A demonstration that an electronic pulse, a prong pinch, or a chain correction feels tolerable on a human adult tells us essentially nothing about what the same stimulus does to a dog's nervous system.

2.5 Videos of Engaged Dogs Do Not Disprove the Contingency

Video demonstrations of dogs who appear engaged, happy, or relaxed while wearing aversive training equipment are routinely offered as evidence that the equipment does not harm welfare. The argument fails for several reasons. Visible engagement is not a physiological or affective readout. Threat and stress circuitry do not announce themselves through tail posture or eye expression alone. Once avoidance learning is well established, a dog may perform fluently and quickly precisely because the behavior prevents the aversive event. Fluency in the instrumental response is compatible with ongoing threat prediction. A dog who has learned to walk in heel to avoid a prong correction, or to come when called to avoid a shock, may perform the behavior smoothly and without visible distress while still maintaining the underlying threat association. Welfare is not reducible to what a dog looks like in a fifteen-second clip. Welfare is what is happening in the dog's nervous system, across time, in response to the entire training contingency.

3. The Welfare Evidence Is Convergent

3.1 Controlled Studies of Electronic Collar Use

Cooper, Cracknell, Hardiman, Wright, and Mills (2014) conducted the methodologically strongest controlled comparison of electronic collar training to reward-based training for pet dogs showing recall and chasing problems. Dogs trained with electronic collars showed more behavioral stress indicators than dogs trained without. Reward-based training achieved outcomes equal to or better than electronic collar training on the training objectives. The electronic collar conferred no necessity advantage. China, Mills, and Cooper (2020) re-analyzed the same dataset and reported on training efficacy in greater detail. The reward-based training group achieved superior outcomes on multiple measures, including faster latency to sit, fewer hand and lead signals, and faster general obedience progress. The electronic collar group did not show better learning outcomes than the matched non-e-collar control group operated by the same trainers, which means that whatever benefit professional skill brought to the training was not coming from the collar. The trainers in these studies were nominated by the Electronic Collar Manufacturers Association as representing best practice. They did not produce better outcomes with the tools than without them. The argument that e-collars work in expert hands has been tested directly using the experts the industry itself put forward, and it failed.

Schilder and van der Borg (2004) studied protection-trained working dogs in guard-dog training programs (the published abstract describes the sample as German Shepherd dogs) and found that dogs receiving shock collar applications displayed lower body postures, high-pitched yelps, avoidance behaviors, and other behavioral signs of fear and stress on shock application, and that dogs that had received shocks showed more behavioral signs of fear and stress in the training context relative to matched dogs trained without shock. Casey, Naj-Oleari, Campbell, Mendl, and Blackwell (2021) found that dogs whose guardians used two or more aversive methods were significantly more pessimistic on a judgment bias task than dogs trained only with reward-based methods, a direct measurement of negative affective state outside the training context itself.

Deldalle and Gaunet (2014) provided a direct observational comparison of two French training schools, one using primarily negative reinforcement and one using primarily positive reinforcement, with behavioral coding of dogs during training sessions. Dogs in the negative-reinforcement school showed significantly more stress-related behaviors during training and significantly less owner-directed gaze than dogs in the positive-reinforcement school. The reduction in owner-directed gaze is welfare-relevant in its own right because owner-directed gaze is one of the canonical behavioral indicators of secure attachment and comfortable engagement with the handler. The findings establish that within-session welfare differences track training method even when the comparison is between intact, ongoing programs rather than between experimentally manipulated conditions. The Deldalle and Gaunet finding directly counters the proponent claim that dogs in aversive-method programs look engaged and content during training.

3.2 Broader Research on Aversive Training and Canine Welfare

The welfare research on aversive training extends well beyond electronic collars specifically to cover aversive methods in general. Vieira de Castro and colleagues (2020) videotaped training sessions at seven schools and categorized them as reward-only, mixed, or aversive-based. Dogs in aversive-based schools showed significantly more stress-related behaviors during training, spent more time in tense and low behavioral states, panted more, and had higher post-training cortisol increases than dogs in reward-only schools. Dogs in mixed-method schools showed significantly more stress-related behaviors and panted more than dogs in reward-only schools, although the cortisol difference reached significance only for the aversive-based group. Dogs in aversive-based schools also performed more pessimistically on a cognitive bias task, indicating a more negative affective state extending beyond the training context. Vieira de Castro and colleagues (2019) separately showed that secure attachment patterns toward guardians were more consistently observed in dogs trained at reward-based schools than in dogs trained at aversive-based schools, when assessed through a Strange Situation Procedure. The Strange Situation Procedure is a standardized behavioral assessment, originally developed by Mary Ainsworth in the 1960s and 1970s for the study of human infant attachment to caregivers and subsequently validated for use with dogs, in which the subject is observed across a series of brief scripted episodes that systematically vary the presence of the guardian and a stranger in an unfamiliar room. Behaviors such as proximity-seeking, exploration, distress at separation, and quality of greeting on reunion are coded by trained observers and used to classify the subject's attachment pattern as secure or as one of several insecure patterns. Because the procedure relies on coded observable behavior under standardized conditions, it is methodologically more rigorous than guardian self-report and is now a standard tool in canine attachment research.

Blackwell, Twells, Seawright, and Casey (2008) found that guardians who used positive-punishment-based methods reported more undesirable behaviors in their dogs than guardians who used reward-based methods, with the highest aggression scores reported in dogs whose guardians combined positive reinforcement with positive punishment. Hiby, Rooney, and Bradshaw (2004) similarly found that the number of problem behaviors reported by guardians correlated with the number of training tasks taught using punishment, while obedience ratings correlated only with the number of tasks taught using rewards. Rooney and Cowan (2011), in a home-based observational study, found that dogs of guardians who reported using more physical punishment were less playful with their guardian and interacted less with the experimenter, while dogs of guardians who reported using more rewards performed better on a novel training task. Arhant, Bubna-Littitz, Bartels, Futschik, and Troxler (2010) found that high-frequency aversive training correlated with increased aggression and excitability in dogs, while reward-based training correlated with higher obedience without those side effects.

Herron, Shofer, and Reisner (2009) surveyed one hundred forty pet guardians presenting to the University of Pennsylvania veterinary behavior service. They found that confrontational training techniques produced aggressive responses in a substantial percentage of the dogs on whom they were attempted: hitting or kicking, forty-three percent; growling at the dog, forty-one percent; forcing the release of an item, thirty-nine percent; the alpha roll, thirty-one percent; staring the dog down, thirty percent; the dominance down, twenty-nine percent; grabbing the dog by the jowls or scruff and shaking, twenty-six percent; and choke or pinch collar use, eleven percent. Shock collars were used infrequently in this clinical sample, with ten percent of those dogs showing an aggressive response. Dogs presenting for aggression to familiar people were significantly more likely to respond aggressively to the alpha roll and to yelling "no" than dogs presenting for other behavior problems. The Herron study is a direct clinical demonstration that confrontational handling is not a benign or low-risk intervention. It is a clinically identified risk factor for guardian-directed aggression, with subsequent population-level multivariable analysis (Casey, Loftus, Bolster, Richards, and Blackwell, 2014) finding adjusted increased odds of family-member aggression in dogs whose guardians used aversive methods.

3.3 The Dissociation Between Behavioral and Physiological Stress Markers

An important methodological point cuts across the welfare literature. Behavioral stress markers (lip licking, yawning, low body posture, displacement behaviors, conflict behaviors, reduced approach, increased vigilance) and physiological stress markers (cortisol, heart rate, heart rate variability) often converge, but not always. Cooper et al. (2014) found significant behavioral stress markers in the electronic collar group, while the larger controlled study showed no significant cortisol difference between the e-collar group and either control group. The preliminary nine-dog phase did show elevated cortisol post-stimulation, but the larger controlled phase did not replicate this clearly. The behavioral findings have sometimes been dismissed by proponents on the basis of the unreplicated cortisol result. That dismissal is selective reading. Cortisol is one measure, and a blunt one. It can be suppressed, lagged, or buffered by context. Validated behavioral markers of stress in canine research are robust and meaningful on their own. A study that finds significant behavioral stress indicators in the electronic collar group has found welfare harm, regardless of whether a single hormone assay reached statistical significance.

This dissociation also applies across aversive modalities. A prong-collared dog may not show elevated cortisol during a short walk, but may show avoidance of the collar being put on, tension during leash handling, conflict behaviors around the handler, and reduced engagement in exploratory behavior. A choke-chained dog may not show elevated cortisol during a single correction, but may show whale eye, lip lick, yawn, and shoulder tension during the correction itself and in anticipation of subsequent corrections. Welfare is assessed through convergence of indicators, not through any single measure.

3.4 Conditioned Emotional Responses and Transfer of Aversiveness

Whenever an aversive event is paired with contextual stimuli, Pavlovian conditioning can produce a conditioned emotional response to those stimuli. A dog repeatedly corrected with a prong collar in the presence of other dogs can develop a conditioned negative association with other dogs. A dog repeatedly shocked in the presence of children can develop a conditioned negative association with children. A dog repeatedly alpha-rolled by the guardian can develop a conditioned negative association with the guardian. None of this is theoretical. Schilder and van der Borg (2004) identified exactly this kind of transfer in their shock collar work, noting that the dogs in their study appeared to associate the aversive event not only with their own behavior but also with the handler, with commands, or with the training context.

This mechanism is particularly dangerous when aversive equipment or aversive methods are used to address reactivity or aggression, because the triggers in those cases are, by definition, stimuli the dog already perceives as threatening. Adding aversive stimulation in the presence of such triggers can deepen the threat association rather than weaken it. That is the most foreseeable failure mode of aversive-based approaches to reactivity and aggression, and it is the failure mode that reward-based counter-conditioning and desensitization protocols are specifically designed to avoid.

3.5 Cumulative Exposure and Welfare Risk

Most aversive training equipment is not worn for a single correction and removed. A prong collar is often worn for every walk, for months or years. A choke chain is often worn continuously during handling. An electronic collar is often worn during daily exercise and off-leash time. Even if individual corrections produced only modest stress responses, the cumulative exposure over months and years is not welfare-neutral. Chronic stress exposure produces documented effects on the hypothalamic-pituitary-adrenal axis, the amygdala, and the prefrontal cortex in mammalian research (McEwen, 2012; Rosenkranz, Venheim, and Padival, 2010; Vyas, Mitra, Shankaranarayana Rao, and Chattarji, 2002; Arnsten, 2009). It produces documented changes in cognitive performance, stress reactivity, and affective state. A policy framework that assesses aversive training equipment only on the basis of a single-session stress response is assessing the wrong exposure window.

3.6 Quick Reference: Convergent Welfare and Training Outcome Evidence

The table below consolidates the principal peer-reviewed studies that anchor the welfare and training-outcome evidence base. It is meant to be used: in client conversations, in professional debate, on social media, in continuing education. Each row identifies the study, summarizes its design and sample, states the key finding, and notes the specific proponent argument the study addresses. The strength of the case lies in the agreement across these methods. Controlled experiments, direct observational studies, population-level surveys, affective state and cognitive bias measures, clinical referral data, and real-world containment data all point in the same direction. No single study has to carry the case alone. The agreement across methodologies, populations, countries, and outcome measures is what carries it.

Table 2. Quick-Reference Summary of the Convergent Welfare and Training-Outcome Evidence Base.

Several conclusions follow from this evidence base in aggregate. First, the welfare signal appears across every methodological approach (experimental, observational, survey, cognitive bias, clinical referral, and real-world containment), which is the structural feature that makes the case convergent rather than cherry-picked. Second, the population studied varies substantially across these works (pet dogs, working dogs, US samples, UK samples, French samples, Austrian samples, Portuguese samples), and the pattern persists across these populations. Third, the design diversity matters because no single methodology is decisive on its own, but the convergence across methodologies eliminates the possibility that the welfare signal is a methodological artifact. A practitioner facing a balanced trainer's dismissal of any single study can point to the convergence pattern documented in this table. The argument does not depend on Cooper alone, or on Vieira de Castro alone, or on Casey alone. It depends on agreement across all of them.

4. Nociception, Mechanical Injury, Aversion, and the Real Welfare Standard

4.1 What Aversive Equipment Actually Engages: Nociception and Threat Circuitry

The welfare case against aversive training equipment does not rest on a claim that the tools cause tissue injury. Modern electronic collar proponents commonly frame their defense around sensation severity rather than tissue damage. The argument runs roughly like this: modern e-collars operate at very low stimulation levels; the sensation is mild, imperceptible, or comparable to a transcutaneous electrical nerve stimulation unit; skilled trainers use the minimum effective setting; what the dog feels is no worse than a tap on the shoulder or a static-electricity shock. Those are claims about sensation severity, not about tissue damage. The counterargument has to meet them on their own terms.

The counterargument is nociception science. Nociceptors are specialized primary afferent neurons (sensory neurons that carry signals from the periphery into the central nervous system) that respond to stimuli capable of signalling actual or potential harm (Dubin and Patapoutian, 2010). They transduce mechanical, thermal, chemical, and electrical stimuli into neural signals that the central nervous system interprets as pain or as a noxious warning event. Critically, nociceptors fire at intensities well below the threshold of actual tissue injury. The system exists to warn the organism away from potentially damaging stimuli before the damage occurs. The International Association for the Study of Pain formally defines pain as an unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage (Raja et al., 2020). The phrase "potential tissue damage" appears in the definition precisely because pain is a warning signal, not an injury report.

Electrical stimulation from a training collar does not have to cause tissue damage to activate nociceptors and pain pathways. It simply has to be delivered at an intensity that the peripheral nervous system encodes as noxious. C-fiber and A-delta fiber nociceptors respond to electrical stimulation at intensities well below any injury threshold. Pressure applied to the dog's neck by a prong collar does not have to puncture skin or bruise deep tissue to engage mechanonociceptors. It simply has to exceed the threshold of noxious mechanical input. Compression applied by a choke chain does not have to cause vascular damage or tracheal rupture to engage mechanonociceptors and threat circuitry. It simply has to be delivered with enough force and duration to be noxious. In every case, the biological question is whether the stimulus crosses the nociceptive threshold, not whether it crosses the injury threshold.

The intensity dial on an electronic collar, the sharpness of the prong points, and the degree of neck compression produced by a choke chain are all functionally calibrated to deliver a stimulus the dog experiences as unpleasant enough to change its behavior. Unpleasant enough to change behavior through avoidance, escape, or suppression is, by definition, noxious. A stimulus that is not noxious will not drive avoidance learning. A stimulus that does drive avoidance learning is, by functional and by neurobiological definition, crossing the nociceptive threshold. The argument that aversive equipment operates in a zone above behavioral effectiveness but below nociceptive engagement is not consistent with how peripheral sensory neurons function.

This is why the welfare case does not require a showing of tissue damage. A dog's capacity to experience a stimulus as aversive, to undergo threat conditioning, to develop conditioned emotional responses, to experience stress, and to be worse off for the exposure does not begin at the threshold of visible injury. It begins at the threshold of nociceptive activation and threat-system engagement, which is a much lower threshold and the threshold the tools are built to cross.

Two further considerations strengthen the nociception argument and answer the proponent appeal to low-level or skilled application. The canine sensory anatomy that encounters these stimuli is not equivalent to the human anatomy proponents use in self-test demonstrations. As discussed in Section 2.4, Affolter and Moore (1994) document that canine haired-skin epidermis is approximately three to five cell layers thick, considerably thinner than human epidermis. A delivered electrical or mechanical stimulus that crosses the human nociceptive threshold at one intensity will reach deeper canine tissue at the same delivered energy. Self-testing on human skin systematically underestimates what the canine nervous system receives.

The equipment itself is also not standardized in ways that justify the proponent appeal to a low intensity setting. Lines, van Driel, and Cooper (2013) examined the electrical characteristics of thirteen commercially available electronic training collar models and reported an eighty-seven-fold range in stimulus energy at maximum settings, from 3.3 millijoules to 287 millijoules at a 50 kilohm resistive load representative of canine neck impedance. Within a single collar, the median ratio of maximum to minimum delivered energy across the available strength settings was 81, with individual collars ranging from 8 to 1,114. The authors reported that user-disclosed comparison data such as voltage, pulse parameters, and waveform are not available at the point of sale. Two of the thirteen new collars examined contained manufacturing faults, in one case capable of delivering a maximum-strength impulse regardless of the level chosen via the user dial. The authors concluded that a given strength setting cannot be assumed to deliver a similar stimulus across collar models or brands. From the canine nervous system perspective, what determines whether the nociceptive threshold is crossed is the actual electrical signal at the skin, not the user's intensity setting. The proponent appeal to a low intensity setting, even granting good faith user technique, is not informative about the welfare-relevant question of whether the stimulus is noxious to the dog.

Table 3. Nociception, Pain Neuroscience, and Sensory Engagement of Aversive Training Equipment.

4.2 Documented Physical Effects of Neck-Pressure Equipment (Prong and Choke Collars)

The nociception argument above applies equally to electronic, prong, and choke collars, because all three cross the nociceptive threshold by design. There is, however, a separate evidentiary question that arises specifically for equipment that applies mechanical force to the canine neck. Peer-reviewed veterinary research on prong collars and choke chains documents measurable physical effects on live anatomy during ordinary use, and, in one peer-reviewed case report, catastrophic injury during punitive use. This mechanical-injury evidence is distinct from the nociception argument and applies only to neck-pressure tools. Electronic collars, operated in their standard training modes, do not cause this kind of mechanical injury, and this paper does not claim otherwise. The mechanical-injury argument is a prong-and-choke-collar argument.

Carter, McNally, and Roshier (2020) tested seven collar types and a slip lead on a simulated canine neck model with a pressure sensor beneath the collar. Under force levels representing a firm pull (40 N), a strong pull (70 N), and a jerk (141 N average), collars produced pressures between 83 and 832 kilopascals on the model neck. Collar type and applied force each had significant effects on the pressure delivered to the neck. The authors concluded that no collar tested produced a pressure low enough to mitigate the risk of injury when the dog pulls on the lead (Carter, McNally, and Roshier, 2020). Hunter, Blake, and De Godoy (2019) measured force and pressure on the canine neck during ordinary on-leash walking and found peak contact pressure values reaching 44.61 newtons per square centimeter, with significant differences in how different collar constructions transmit force to the neck. Both studies establish that ordinary collar use transmits substantial pressure to the canine neck. The pressures produced by prong collars, which concentrate force at the prong points, and by choke chains, which apply force to a progressively narrowing circumference, are higher than what flat collars produce under identical pull conditions. This is an engineering reality of how those collars are designed to operate.

Pauli, Bentley, Diehl, and Miller (2006) measured intraocular pressure in fifty-one eyes of twenty-six dogs while the dogs pulled against a collar or a harness. Intraocular pressure rose significantly from baseline when pressure was applied via a collar, but not when equivalent pressure was applied via a harness. The authors concluded that dogs with weak or thin corneas, glaucoma, or any condition where elevated intraocular pressure could be harmful should wear a harness rather than a collar (Pauli et al., 2006). The proposed physiological mechanism is ventral neck pressure compressing the jugular veins and obstructing ocular aqueous outflow. Elevation produced by a flat collar under pull is a published and reproducible finding. Prong and choke collars concentrate or constrict force delivery in ways that flat collars do not, which on the underlying mechanism would be expected to produce equal or greater intraocular pressure elevation. The peer-reviewed literature has not yet directly tested prong or choke collars against the same protocol, and the burden of demonstrating that these tools are mechanically safer than the flat collars Pauli measured lies with the manufacturers and proponents who market them, not with the welfare science community.

Grohmann, Dickomeit, Schmidt, and Kramer (2013) published a detailed peer-reviewed case report in the Journal of Veterinary Behavior describing severe ischemic brain damage in a one-year-old German Shepherd Dog subjected to a punitive training technique in which the guardian lifted the dog off the ground by the choke chain collar. The dog initially appeared normal, then became progressively ataxic (uncoordinated and unable to walk normally), began circling to the left, and showed reduced consciousness. Magnetic resonance imaging showed multifocal T2 and diffusion-weighted changes consistent with severe cerebral edema from ischemia. The injury mechanism was carotid artery compression producing cerebral hypoxia. Because of the severity of the neurological findings, the dog was euthanized. The authors attribute the injury to the punitive hanging technique commonly referred to in the training community as helicoptering or hanging. This is a documented peer-reviewed case of fatal ischemic brain injury directly caused by a choke chain training technique (Grohmann et al., 2013).

The clinical veterinary literature additionally recognizes that repeated collar pressure is a clinical concern for tracheal collapse, and harnesses are commonly recommended in place of collars for dogs diagnosed with tracheal collapse (Rozanski, 2022). Cough induced by collar pressure is a recognized diagnostic feature of the condition in veterinary medicine.

Taken together, the peer-reviewed literature on neck-pressure equipment establishes that prong and choke collars, by their mechanical design, transmit pressure to the canine neck in ranges capable of producing measurable physical consequences, including elevated intraocular pressure under ordinary pull conditions and, under punitive use, documented fatal cerebral ischemia. None of this evidence applies to electronic collars, which operate by a different mechanism. The mechanical-injury case applies specifically to neck-pressure equipment and adds a further layer of evidence, beyond the nociception argument, that prong and choke collars should not remain on the consumer market.

Table 4. Mechanical Injury and Physiological Effects of Neck-Pressure Equipment.

The mechanical evidence base summarized in Table 4 is independent of the nociception argument set out earlier in this section. Even if a reader sets aside the question of whether prong and choke collars engage nociceptors during ordinary use, the engineering, physiological, and clinical evidence establishes documented physical effects on the canine neck under conditions that fall within the operating range of these tools as marketed and used. The case for prohibition of neck-pressure equipment does not depend on the nociception argument alone.

4.3 Threat Circuitry, Controllability, and Avoidance Learning

The neuroscience of fear and threat conditioning provides further evidence that aversive-based training engages welfare-relevant neural processes regardless of how skillfully the tool is applied. The amygdala and its connected circuits respond to predicted aversive events, encoding threat associations and driving avoidance learning (LeDoux, 2014). Limbachia et al. (2021) showed that when human participants had control over aversive stimulation, the magnitude of threat-related neural responding was attenuated compared to uncontrollable aversive conditions, but not eliminated. The senior author of that study, Dr. Luiz Pessoa, confirmed in written correspondence that the attenuation should not be interpreted as rendering controllable aversive stimulation neurologically neutral or welfare-benign. Wood et al. (2014) documented that the amygdala mediates the emotional modulation of threat-elicited responses, situating the amygdala as the key node through which aversive stimulation produces emotional response. Dr. David Knight, the senior author of that study and a fear-conditioning researcher whose broader research program (including work on the conditioned diminution of unconditioned responses) is sometimes cited as showing that predictable aversive stimulation is neurologically neutral, confirmed in written correspondence that his research cannot be used to support the proposition that predictable aversive stimulation is neutral or benign (Pessoa, personal communication, 2026; Knight, personal communication, 2026).

Methodological note on these personal communications. The two personal communications cited above (Pessoa, 2026; Knight, 2026) consist of written email correspondence between the present author and the named senior researchers, in which each researcher independently confirmed the substantive interpretation of their published work as represented in this paper. Both exchanges are documented and are held on file by the present author. Excerpts of the correspondence are available, on written request and subject to the correspondents’ permission, to qualifying parties including academic peer reviewers, legislative drafting staff, professional credentialing bodies, and accredited journalists. This disclosure framework follows standard academic practice for citing personal communications in policy contexts where the communications support a substantive interpretive position rather than introducing novel empirical findings.

This neuroscience has direct implications for aversive-based training. A dog trained with an electronic collar to avoid predatory chasing has learned to avoid an aversive event, not to become neutral to one. The same is true of a dog trained with a prong collar to avoid lead pulling, or a dog trained with a choke chain to avoid forging. In every case, the learned behavior is avoidance driven by negative reinforcement of an aversive event that the dog's brain continues to represent as aversive. Controllability and predictability do not dissolve the aversive. They modulate the intensity of the neural response to it.

The fear-and-avoidance literature has moved beyond the older framing of avoidance as a reflexive, negatively reinforced, fear-driven response. Cain (2019), in a review of contemporary active avoidance research, describes avoidance as a goal-directed instrumental behavior that the brain mounts in contexts where harm is anticipated and a behavioral solution is available. Importantly, the shift from a fear state to an anxiety state during effective avoidance does not eliminate the underlying threat representation. The warning stimulus retains its conditioned threat value; what changes is that the dog has acquired a behavioral option that controls exposure to the aversive event. When the avoidance response is blocked or fails, the fear state returns along with the inflexible defensive reactions characteristic of fear. This framing is directly relevant to the welfare evaluation of aversive-based training, because it locates the welfare cost not in observable freezing or panic during training, but in the underlying threat representation that the warning stimulus retains throughout.

The neurobiological literature on stressor controllability underscores the same point. Maier and Watkins (2005), in a review covering decades of stressor controllability research, set out the dorsal raphe nucleus, serotonergic, and corticotropin-releasing factor pathways through which controllability modulates the consequences of an aversive stressor. The work establishes a robust modulation finding: controllable aversive events produce a different downstream profile than uncontrollable aversive events. The work does not establish that controllable aversive events are stress-free or welfare-neutral. The aversive remains aversive. The animal still recruits stress-system machinery in response to it. What controllability does is attenuate certain downstream sequelae, including the spread of activation into prefrontal regions that produces the broader behavioral signature of learned helplessness. This is the same point Pessoa made in written correspondence about the Limbachia data, expressed at the level of system neurobiology rather than human imaging.

The empirical literature on canine remote shock collar use under conditions designed to maximize controllability and predictability is consistent with this picture. Christiansen, Bakken, and Braastad (2001) studied 114 hunting dogs (Norwegian elkhounds, English setters, and hare hunting dogs) across two consecutive years of pasture confrontation testing with sheep. Remote shock collar use was deployed under conditions where the dog could control the aversive event through behavioral compliance and where the aversive was predictable and contingent on a defined behavior. The authors’ own welfare measures were limited, relying largely on owner report and temperament tests, and the study did not detect a significant fear or anxiety effect using those measures. The methodological thinness of the welfare assessment, rather than a clean positive welfare conclusion, is what prevents the data set from supporting a welfare-benign reading. The Christiansen study is sometimes invoked by proponents to argue that controllable, predictable shock-collar use in field conditions is welfare-benign. The data set itself does not support that reading.

The most recent neurobiological work on active avoidance reinforces this convergence. Sears, Andrade, Samels, Laughlin, Moloney, Wilson, Alwood, Moscarello, and Cain (2026), in a study using a shuttlebox active avoidance paradigm with rats, demonstrated that response-produced feedback cues are transformed during training into safety signals that positively reinforce avoidance. Moderately trained avoidance was goal-directed and depended on the posterior dorsomedial striatum. Overtrained avoidance became habitual, depended on the dorsolateral striatum, and was insensitive to devaluation of the safety signal. The Sears and colleagues finding has three implications relevant to the present argument. First, the safety signals they identify acquire their value entirely from their inverse relationship with the aversive contingency. Without the aversive event, no warning stimulus acquires threat value, no feedback cue acquires safety value, and the avoidance response is not reinforced. To describe avoidance as positively reinforced by safety, in the technical sense Sears and colleagues use, is not to claim that the underlying training regime was not aversive. The aversive contingency is the precondition for the entire learning architecture. Second, the authors describe the shift from fear to anxiety that accompanies effective avoidance, in which the dog (or rat) is in an anxiety state where action is possible because safety is attainable, and they explicitly note that when the avoidance response is impaired, the fear state returns along with inflexible reactions like freezing. This is the mechanism behind the proponent observation that a successfully shock-collar-trained dog looks engaged and content during work. The dog is in an anxiety state mediated by an effective avoidance response, not in the absence of threat representation. Third, the same dorsolateral striatum-mediated habitual circuit that the Sears study isolates is the circuit Gillan, Morein-Zamir, Urcelay, Sule, Voon, Apergis-Schoute, Fineberg, Sahakian, and Robbins (2014) implicate in obsessive-compulsive disorder, and that Gordon, Patterson, and Knowlton (2020) implicate in the stronger avoidance habits observed in survivors of early life stress. The argument is not that aversive training causes obsessive-compulsive disorder in dogs. The argument is that the learning architecture aversive-based training depends on, particularly under prolonged or overtrained conditions, is the same architecture implicated in clinical populations as the substrate for persistent maladaptive avoidance.

Taken together, the neuroscience literature on threat circuitry, controllability, predictability, and active avoidance does not exempt aversive-based training from welfare scrutiny. It does the opposite. The modulating factors that proponents invoke (skilled application, predictable timing, controllable contingencies, low intensity) operate on top of an aversive contingency that the dog's nervous system continues to represent as such throughout. Table 5 summarizes the convergence.

Table 5. Threat Circuitry Research Under Controllability and Predictability Conditions.

4.4 The Compound Schedule Problem: Why Adding Food Does Not Subtract the Aversive

There is a more sophisticated proponent argument that holds that the problem with aversive equipment is not the aversive itself but the absence of positive reinforcement, and that combining aversive stimulation with high-rate food reinforcement eliminates the welfare concern. The argument does not hold up. Adding food reinforcement to an aversive-based contingency creates a compound schedule in which positive reinforcement and positive punishment, or positive reinforcement and negative reinforcement, operate in parallel. The presence of the positive component does not erase the aversive component. The dog's nervous system continues to register and respond to the aversive event.

Empirically, the controlled electronic collar studies that produced welfare concern used trainers nominated by the industry, trainers who did in fact add food reinforcement and praise to their electronic-collar work. The welfare indicators appeared anyway. The Vieira de Castro et al. (2020) finding of elevated stress behaviors and increased panting appeared in schools that used aversive methods in combination with reward-based methods (Group Mixed), not only in schools that used aversive methods alone, although the cognitive-bias finding was specific to the high-aversive group. Compound schedules that pair food with aversive events also produce their own set of welfare concerns, including approach-avoidance conflict and conditioned ambivalence toward the handler. The argument that adding food to a shock, pinch, or correction protocol eliminates the welfare concern has not been demonstrated in any controlled study and is not supported by the studies that have examined compound schedules empirically.

5. The Necessity Claim Fails

5.1 Cooper (2014) and China (2020): No Necessity Advantage Under Best-Practice Conditions

Cooper et al. (2014) and China et al. (2020) matter for the necessity claim because of how they were designed. The electronic collar trainers in both studies were nominated by the Electronic Collar Manufacturers Association as representing the trade's best practice. They used low-level stimulation. They followed manufacturer-recommended protocols. They were evaluated on the problem categories the industry most strongly claims as the e-collar's home territory: recall failure and chasing. Under those conditions, reward-based training produced outcomes equal to or better than electronic collar training, while the electronic collar group showed behavioral welfare indicators that the reward-based group did not.

The implication is decisive. If electronic collars are not necessary in the hands of trained, industry-nominated, best-practice trainers working on the problems that most favor the tool, then they are not necessary at all. The argument for electronic collars has already failed at the professional level. That failure cannot be rescued by pivoting to "use by professionals only" as a policy concession. The studies show that even under professional, industry-nominated, best-practice conditions, the tool produces welfare harm without an outcome benefit.

5.2 No Necessity Advantage for Prong and Choke Collars

The published comparative outcome literature for prong and choke collars against reward-based methods is thinner than the electronic collar literature, but the pattern is consistent. No peer-reviewed study has demonstrated that prong or choke collars produce long-term training outcomes superior to reward-based alternatives in everyday pet training contexts. The welfare literature, including Hiby et al. (2004), Blackwell et al. (2008), Arhant et al. (2010), Rooney and Cowan (2011), and Casey et al. (2021), consistently associates aversive methods with worse rather than better training outcomes and with elevated risk of problem behaviors, including aggression. The absence of any controlled study demonstrating long-term superiority of prong or choke collars over reward-based methods is itself meaningful. The burden of proof for a device whose mechanism engages nociception, and whose use has been associated in the peer-reviewed veterinary literature with measurable physical consequences on the canine neck, lies on the proponent to demonstrate necessity, not on the opponent to refute a never-demonstrated claim.

The veterinary behavior literature is explicit on this point. The AAHA 2015 Canine and Feline Behavior Management Guidelines conclude that the only acceptable training techniques are non-aversive positive techniques, and specifically name electronic shock collars, prong or pinch collars, choke chains, alpha rolls, cattle prods, entrapment, and physical punishment as techniques that can harm or destroy an animal's trust, negatively affect problem-solving ability, and increase anxiety (AAHA, 2015). The AVSAB 2021 position statement is similarly explicit that aversive methods including but not limited to electronic collars, prong collars, choke chains, leash corrections, and other forms of physical or psychological punishment should not be used under any circumstances, and that there is no evidence that aversive training is necessary for dog training or behavior modification (AVSAB, 2021). The BSAVA position statement explicitly names electric shock collars, prong collars, spray collars, choke chains, and electric containment fences as aversive devices that the BSAVA recommends against, and supports legislation banning their sale and use (BSAVA, 2024). The professional consensus has already assessed the necessity claim and rejected it.

5.3 Johnson and Wynne (2024): A Narrow Finding, Not a Necessity Proof

Johnson and Wynne (2024) is frequently cited by proponents as evidence that electronic collars are necessary for predatory chasing problems. It is neither as strong nor as generalizable as that citation suggests. The study examined a narrow problem profile under specific experimental conditions and reached a narrow efficacy conclusion, not a necessity conclusion. Its protocol design has been challenged in the peer-reviewed literature (Bastos, Warren, and Krupenye, 2025), with specific methodological concerns about the reward-based comparison condition, the duration of the training trial, and the baseline comparability of groups. Johnson and Wynne (2025) published a response to that critique, but the substantive methodological concerns about the original protocol design were not resolved by the response. The present author's separately published methodological critique (Bangura, 2025, SSRN) raises additional concerns about the study's internal validity and about the generalizability of its findings to general pet dog populations. A single contested efficacy finding under narrow experimental conditions is not a foundation for policy that grants broad consumer access to a device whose mechanism engages nociception and threat circuitry.

5.4 Practice-Based Evidence on Force-Free Alternatives

The claim that aversive equipment is necessary for difficult cases is contradicted by the clinical practice of board-certified veterinary behaviorists, who treat the most severe aggression, reactivity, anxiety, and predatory problems in canine medicine without relying on aversive equipment. The ACVB clinical standard of care is built on behavioral assessment, environmental management, and reward-based behavior modification, integrated with psychiatric medication when clinically indicated. ACVB-board-certified veterinary behaviorists treat the most severe canine aggression, reactivity, anxiety, and predatory cases using this integrated approach, without relying on aversive equipment. If aversive equipment were genuinely necessary for difficult behavior cases, the veterinary specialty that handles those cases would be using it. It is not. The specialty that sees the hardest cases has already determined that reward-based, force-free methods are the appropriate standard of care (ACVB, 2025).

6. Reactivity, Aggression, and Confrontational Handling: Where Aversive Approaches Compound Harm

6.1 Suppression Versus Resolution

Aversive-based training, whether delivered through equipment or through confrontational handling, is especially concerning when it is used with reactivity and aggression, because reactivity and aggression usually occur in the presence of stimuli the dog already perceives as threatening, frustrating, overwhelming, or unsafe. A reactive dog barking and lunging at another dog, a stranger, a child, a bicycle, or an unfamiliar object is not simply disobeying. The dog is already in a state of heightened arousal, threat appraisal, and defensive preparation. Adding an aversive event in that moment, whether electrical stimulation, prong correction, choke correction, leash jerk, alpha roll, or physical correction, lands on top of an already activated emotional and physiological background. The dog can associate the aversive not only with its own behavior but also with whatever else is present, including the trigger itself. This is the conditioned emotional response mechanism described in Section 3.4, applied in the exact circumstance where aversive Pavlovian pairings are most foreseeable and most dangerous.

6.2 The Herron Finding on Confrontational Handling

Herron, Shofer, and Reisner (2009) provide direct clinical evidence on confrontational handling. In their sample of dogs presenting to a university behavior service, confrontational techniques produced aggressive responses in a substantial percentage of cases: hitting or kicking the dog, forty-three percent; the alpha roll, thirty-one percent; the dominance down, twenty-nine percent; grabbing the jowls or scruff and shaking, twenty-six percent. Dogs presenting for aggression toward familiar people were significantly more likely to respond aggressively to the alpha roll and to yelling "no" than dogs presenting for other complaints (Herron, Shofer, and Reisner, 2009). The implication is direct. Confrontational handling is not merely aesthetically uncomfortable or philosophically disfavored. It is a clinically identified independent risk factor for guardian-directed aggression. It is an unsafe intervention on grounds of human safety, not only animal welfare.

This finding directly contradicts the popular claim that confrontational methods are needed to address aggressive dogs. The evidence shows the opposite. Confrontational methods are contraindicated for aggressive dogs and for many dogs without existing aggression, because the methods themselves elicit defensive and guardian-directed aggressive responses in a substantial minority of cases. A method that triggers aggressive responses in a quarter to a half of the dogs on whom it is attempted is not a safe, reasonable, or evidence-based intervention, regardless of how it is rationalized by dominance theory or by celebrity television demonstration.

6.3 The Mechanism of Compounded Harm

Schilder and van der Borg (2004) identified the compound harm mechanism directly in the electronic collar context, noting that dogs in their study appeared to associate shocks not only with their own behavior but also with the handler, commands, or training context. The same mechanism applies to prong collars in the presence of triggers, to choke chains in the presence of triggers, and to confrontational handling performed in the presence of triggers. Under these conditions, the visible reactive or aggressive behavior may decrease while the underlying emotional problem worsens. The dog may bark less, lunge less, or appear more controlled, but the trigger may now become even more predictive of discomfort, conflict, pressure, or threat. Suppression of visible behavior is not the same as resolution of fear, anxiety, frustration, or defensive motivation. A dog that looks more controlled on the outside while carrying a heavier emotional load on the inside can escalate or redirect when the suppression fails, and the failure often occurs at the worst possible moment.

7. Professional and Regulatory Consensus

7.1 Veterinary Behavior Specialists and Veterinary Organizations

On June 14, 2024, the Federation of Veterinarians of Europe, in conjunction with the Federation of European Companion Animal Veterinary Associations, the Federation of European Equine Veterinary Associations, and the World Small Animal Veterinary Association, unanimously adopted a joint position paper on animal behavior, training methods, and the welfare implications of equipment used to modify behavior. The paper's seventh formal recommendation is a direct and unambiguous call for a complete prohibition on the sale and use of electric pulse training devices, specifically including electric shock collars for dogs (FVE, FECAVA, FEEVA, and WSAVA, 2024). The paper states broadly that equipment and devices that cause pain or discomfort to modify behaviors, such as electric shock collars for dogs and cats, should not be used and should be strongly discouraged by veterinarians and other allied professionals. The signatories represent the veterinary profession across the European Union, across the European companion animal and equine sectors, and globally through WSAVA. This is as unambiguous an international veterinary consensus as exists in this literature. Four major veterinary organizations, representing tens of thousands of veterinarians across multiple continents, formally and unanimously recommend a ban.

The American Veterinary Society of Animal Behavior, in its 2021 position statement reaffirmed in 2025, states explicitly that aversive methods including but not limited to electronic collars, prong collars, choke chains, leash corrections, and other forms of physical or psychological punishment should not be used under any circumstances. AVSAB also states that there is no evidence that aversive training is necessary for dog training or behavior modification (AVSAB, 2021). The AVSAB position matters for policy because proponents often shift, under pressure, from "it works" to "it is necessary." Necessity is a stronger claim than efficacy. A procedure can be effective and still unnecessary. A procedure can suppress behavior and still be inappropriate if less intrusive methods can accomplish the same goal with lower welfare risk.

The American College of Veterinary Behaviorists, the AVMA-recognized specialty organization for board-certified veterinary behaviorists, took its position further in a formal letter to the American Veterinary Medical Association in December 2025. ACVB addressed a public statement by AVMA leadership that seemed to leave room for shock collars as a last-resort alternative to euthanasia. The ACVB response was explicit and emphatic. Electronic collars carry significant risks of fear, aggression, physical pain, and long-term welfare harm. There is no evidence that electronic collars reduce euthanasia risk. Shock collars are not medically necessary, are not evidence-based for preventing euthanasia, and are not aligned with the standard of care for veterinary behavior medicine. ACVB urged that cases involving serious or complex behavior concerns be referred to a board-certified veterinary behaviorist rather than escalated to shock, and aligned its position with AVMA's own published JAVMA guidance on humane training (ACVB, 2025).

The American Animal Hospital Association, representing the certifying and accreditation body for companion animal veterinary practices in the United States, opposes aversive training techniques. The AAHA 2015 Canine and Feline Behavior Management Guidelines, updated in subsequent editions, identify electronic shock collars, prong or pinch collars, choke chains, alpha rolls, dominance downs, cattle prods, lunge whips, starving or withholding food, entrapment, beating, and other forms of physical punishment as training techniques that can harm or destroy an animal's trust in its guardian, negatively affect problem-solving ability, and increase anxiety. The AAHA Guidelines conclude that the only acceptable training techniques are non-aversive, positive techniques that identify and reward desired behaviors (AAHA, 2015).

The European Society of Veterinary Clinical Ethology has similarly argued against the use of electronic collars, concluding after review that there is no evidence of superior efficacy compared to reward-based training and that risks associated with timing errors, lay use, and welfare harm are substantial (Masson et al., 2018a). The British Veterinary Association calls publicly and repeatedly for a complete ban on the sale and use of electronic shock collars for dogs and cats in the United Kingdom, describing electronic shock collars applied even at low intensity as causing physiological and behavioural responses associated with stress, pain, and fear (BVA, 2024). The British Small Animal Veterinary Association opposes aversive training methods broadly, stating explicitly that aversive methods and devices, including electric shock collars, prong collars, spray collars, choke chains, and electric containment fences, have the potential to cause physiological and psychological suffering and that the BSAVA supports legislation banning the sale and use of devices that enable aversive training (BSAVA, 2024). The Australian Veterinary Association holds that collars designed to inflict pain, discomfort, or fear to achieve behavioural change should not be used on dogs, naming electronic and prong collars specifically in that prohibition and adding that prong collars should be illegal in all Australian jurisdictions (AVA, 2022). The Canadian Veterinary Medical Association strongly discourages aversive training techniques and asserts that remote-controlled shock collars are not a necessary method of training or behaviour modification (CVMA, 2021). The New Zealand Veterinary Association's current position states that NZVA does not support the use of electronic behaviour-modifying collars that deliver aversive stimuli for the training or containment of dogs, and recommends that guardians use positive reinforcement methods instead (NZVA, n.d.).

The convergence across veterinary behavior medicine, small animal veterinary practice, international veterinary federations, and national veterinary associations on multiple continents is not selective or niche. It is the established professional consensus.

7.2 Animal Welfare and Humane Organizations

Animal welfare and humane organizations have aligned with the veterinary behavior consensus. In the United States, the American Society for the Prevention of Cruelty to Animals (ASPCA) states explicitly that it is opposed to any training equipment that causes a pet to experience physical discomfort or undue anxiety, and supports training methods that incorporate kindness and respect for both the pet and the guardian, making primary use of lures and rewards such as food, praise, petting, and play (ASPCA, n.d.). Humane World for Animals (formerly the Humane Society of the United States), Best Friends Animal Society, the San Francisco SPCA, and Michigan Humane have each publicly opposed the use of electronic shock collars, prong collars, choke chains, and aversive training methods, and have endorsed reward-based training as the appropriate standard. Best Friends Animal Society has stated publicly that it does not use aversive tools such as pinch collars or electronic collars, and does not endorse their general use, citing the potential for harm when such tools are used by the public (Best Friends Animal Society, 2025). In October 2020, Petco, one of the two largest pet specialty retailers in the United States with more than fifteen hundred locations, ended the retail sale of human- and bark-activated electronic shock collars, citing evidence that shock collars increase fear, anxiety, and stress in dogs, and announced an industry-wide #StopTheShock campaign calling on other retailers, manufacturers, and pet guardians to discontinue the sale and use of these devices (Petco, 2020).

Internationally, the same alignment is even more comprehensive. The Royal Society for the Prevention of Cruelty to Animals, RSPCA Australia, Dogs Trust, the UK Kennel Club, Battersea Dogs and Cats Home, Blue Cross, the People's Dispensary for Sick Animals, the British Columbia Society for the Prevention of Cruelty to Animals, the Scottish Society for the Prevention of Cruelty to Animals, and Cats Protection have each publicly opposed the use of electronic shock collars, prong collars, choke chains, and aversive methods, and have called for legislative bans or for exclusive use of reward-based methods. RSPCA Australia has explicitly stated that it is opposed to the import, sale, or use of equipment used to modify the behaviour of a companion animal that is inhumane, causes injury, pain, suffering, or distress, or can be used to abuse animals, including pronged or pinch collars (RSPCA Australia, n.d.). The British Columbia Society for the Prevention of Cruelty to Animals has stated that it does not support the use of devices and techniques that cause anxiety, fear, distress, pain, or injury, including choke chains, prong collars, and shock collars (BC SPCA, n.d.). In the United Kingdom, the British Veterinary Association, the Kennel Club, Dogs Trust, RSPCA, Battersea, and Blue Cross have acted jointly as a coalition advocating for a complete ban on the sale and use of these devices in England (Dogs Trust, 2024; BVA, 2024).

This is not a fringe coalition. These are the largest and most widely recognized animal welfare organizations in the United States, United Kingdom, Canada, and Australia, joined by a major US pet specialty retailer that decided to remove shock collars from its shelves. When the largest veterinary, welfare, shelter, and retail organizations in the anglophone world each independently reach the same conclusion about aversive training equipment and aversive methods, policy makers should take that convergence as what it is. A field consensus that these tools and methods should be off the market.

7.3 Professional Training and Behavior Organizations

The leading professional training and behavior organizations have adopted the most explicit positions of all. Their standards directly prohibit member use of electronic, prong, and choke collars and of other aversive equipment and methods. The Joint Standards of Practice, as updated in November 2025, are endorsed by the International Association of Animal Behavior Consultants, APDT International, the Karen Pryor Academy, Assistance Dogs International, the Grisha Stewart Academy, Science Matters, Understand Horses, the Victoria Stilwell Academy, and the IAABC Foundation (IAABC, 2025). These standards commit signatory organizations to reward-based methods and explicitly reject the deliberate use of pain, fear, or intimidation in training.

In February 2025, the International Association of Animal Behavior Consultants sunset its previous shock collar addendum and adopted a clarified position explicitly opposing the intentional use of aversive stimuli and specifically requiring members to refrain from using shock in any training or behavior modification context. Members with existing clients using shock devices are expected to help transition those clients away from shock. This is a substantive strengthening of the professional consensus, not a cosmetic update.

The direction of professional trajectory is also telling. Multiple leading education and certification organizations in the dog training field have actively disassociated themselves from broader industry frameworks that continued to permit aversive tools. In particular, the International Association of Animal Behavior Consultants, the Karen Pryor Academy, the Victoria Stilwell Academy for Dog Training and Behavior, the Academy for Dog Trainers (Jean Donaldson), the Pet Professional Guild, and APDT International have each formally distanced themselves from a broader certification framework that continues to permit the use of electronic, prong, and choke collars. Each organization cited alignment with contemporary welfare science and ethical standards as the basis for that departure. The professional consensus is not merely stable. It is tightening. Organizations that permit aversive tools are becoming smaller and more isolated within their own field.

Additional professional organizations that explicitly prohibit aversive equipment or that require reward-based practice of their members include the Pet Professional Guild, Pet Professional Guild Australia, AnimalKind, the Association of Professional Dog Trainers United Kingdom, the Association of Professional Dog Trainers New Zealand, the Canadian Association of Professional Dog Trainers, and the Animal Behaviour and Training Council in the United Kingdom. The Association of Pet Dog Trainers New Zealand has issued a position statement stating that the use of electronic training collars in the context of training is not only unnecessary but a form of cruelty toward dogs, and that shock collars should no longer be an accepted practice in dog training (APDTNZ, 2022). That kind of direct language is now the professional norm among organizations setting standards for modern reward-based training.

7.4 Jurisdictions That Have Enacted Prohibitions on Aversive Training Equipment

Multiple jurisdictions have enacted binding legal restrictions on aversive training equipment, providing regulatory precedent and evidence that bans can be implemented without producing peer-reviewed evidence of increased public safety risk. The pattern across the verified record is consistent. The earliest national prohibitions in this area are now between fifteen and twenty years old, the most recent are continuing to be enacted, and several of the jurisdictions that adopted bans well over a decade ago, including Wales, Switzerland, Austria, and Germany under the case-law interpretation of its Animal Welfare Act, have continued to operate under those prohibitions without any peer-reviewed evidence of harm from prohibition.

In Europe, Wales prohibited the use of electronic collars on dogs and cats under the Animal Welfare (Electronic Collars) (Wales) Regulations 2010 (S.I. 2010/943, W. 97), which came into force on 24 March 2010 under section 12 of the Animal Welfare Act 2006. Switzerland's Animal Protection Ordinance (Tierschutzverordnung) of 23 April 2008, in force 1 September 2008, prohibits training devices delivering electric shocks at Article 76 under the Animal Protection Act of 16 December 2005. Austria's Federal Animal Protection Act of 28 September 2004, in force 1 January 2005, prohibits at §5(2)(3)(a) spike collars, prong collars, and animal training devices using electricity or chemical substances. Germany operates a case-law prohibition: §3 No. 11 of the Animal Welfare Act (Tierschutzgesetz), originally 1972 and consolidated on 18 May 2006, was interpreted by the Federal Administrative Court (Bundesverwaltungsgericht) on 23 February 2006, in case BVerwG 3 C 14.05, to cover electronic training devices on the basis of their design and function, regardless of how an individual user might attempt to apply them.

Denmark's Bekendtgørelse nr. 607 of 25 June 2009 prohibits remote-controlled and automatically operating electric devices, sharp or pointed prong collars, and the advertising and sale of prohibited equipment, under the Animal Welfare Act of 6 June 1991. The Netherlands prohibited equipment delivering electric shocks to dogs through the Besluit of 26 April 2018 amending the Besluit houders van dieren, with the professional exception closed by Staatsblad 2021, 361, and a separate pinch collar ban in force from 1 July 2018. Norway's Animal Welfare Act of 19 June 2009, in force 1 January 2010, supports an implementing regulation prohibiting electric training devices, anti-bark electric collars, invisible electric fences, and prong collars; the predecessor 1974 Act also restricted training collars. Sweden's Animal Welfare Act 2018:1192, in force 1 April 2019, supplemented by the Animal Welfare Ordinance 2019 and Jordbruksverket regulations, prohibits equipment delivering electric shocks. Finland's Animal Welfare Act 693/2023, in force 1 January 2024, was the first Finnish statute to contain an explicit prohibition on electric and spike collars.

France adopted the Arrêté of 19 June 2025, which at Article 14 prohibits electric, prong, and strangling collars (without stopping buckle) in professional contexts; the prohibition applies to educators, breeders, kennels, refuges, and presenters but does not yet cover private use, with the broader Assembly proposition de loi (passed 16 January 2023) still pending in the Senate. Slovenia's Animal Protection Act (Zakon o zaščiti živali, ZZZiv) of 18 November 1999, with most recent amendment ZZZiv-G in force 1 August 2025, restricts electronic training collars under its general Animal Protection Act framework. Spain's Ley 7/2023, in force 29 September 2023, prohibits at Article 27(ñ) the use of electric, impulse, punishment, and choke collars, with hunting, herding, and guard dogs exempt and serious-infraction penalties of €10,001 to €50,000 under Article 76. The Flemish region of Belgium adopted a decree on 13 July 2018 establishing a principle prohibition under the federal Animal Welfare Act of 14 August 1986; a phase-out scenario set in 2021 brings the full ban into force on 1 January 2027, with no exception for military, police, or behaviour therapists, although invisible-fence collars remain permitted.

In Latin America, Colombia enacted Ley 2480 de 2025 (Ley Kiara), in force in 2025, which at Article 10 prohibits prong and electric collars in regulated pet care services including kennels, training centres, transport, grooming, and spas; private-use cases are addressed under the general anti-cruelty framework of Ley 84 de 1989 as updated by Ley 2455 de 2025 (Ley Ángel) of 18 April 2025. In North America, the Canadian province of Quebec adopted the Règlement on the welfare and safety of domestic companion animals and equines (chapter B-3.1, r. 0.1), which came into force on 10 February 2024 and replaces the earlier Règlement under chapter P-42, r. 10.1; the Quebec Ministry of Agriculture, Fisheries and Food has specifically identified prong-type collars and electric shock collars as collars that violate the requirement that an animal's collar must not interfere with breathing or cause pain or injury.

Australia operates a federal prohibition on the import of pronged collars under the Customs (Prohibited Imports) Regulations 1956 (Commonwealth), with sale and use addressed at the state and territory level. The Australian Capital Territory's Animal Welfare Act 1992 prohibits administering an electric shock to an animal except by a prescribed device; the Animal Welfare Regulation 2001 lists permitted electrical devices, and electronic training collars are not on that list, with the framework further strengthened by the Animal Welfare Legislation Amendment Bill passed 26 September 2019, which also recognised animal sentience. New South Wales prohibits the use, sale, and possession of electric collars at section 16 of the Prevention of Cruelty to Animals Act 1979, with containment systems permitted only inside a fence at least 1.5 metres high. Queensland's Animal Care and Protection Amendment Act 2022, passed 2 December 2022 and in force 12 December 2022, added section 37A to the Animal Care and Protection Act 2001, prohibiting the possession and use of pronged dog collars; electronic collars in Queensland are regulated rather than banned. South Australia's Prevention of Cruelty to Animals Regulations (No. 2) 2000, regulation 8(1)(a), prohibits placing on an animal a collar designed to impart an electric shock. Tasmania's Animal Welfare Act 1993 was amended by Act No. 36 of 2022 to insert section 8(2)(ja), prohibiting pronged collars and similar pinching collars in force from 30 November 2022. Victoria's Prevention of Cruelty to Animals Regulations 2019 prohibit pronged collars at regulation 11, while electronic collars are heavily regulated under regulations 23 to 29A with technical specifications set by Ministerial Approval Notice S 56 published in the Victorian Government Gazette on 6 February 2020.

At the United States state level, several jurisdictions have enacted partial restrictions in the tethering context. Hawaii Revised Statutes §711-1109(1)(j), as amended by Act 182 of the Session Laws of 2021, makes it a criminal offence of cruelty to animals in the second degree to tether or restrain a dog by means of a choke collar, pinch collar, or prong collar unless the dog is engaged in an activity supervised by its owner or an agent of the owner. Rhode Island General Laws §4-13-42, as substantially expanded by H 8095, Chapter 079 of 2024 (in force 12 June 2024), prohibits tethering a dog with a choke-type, head, or prong-type collar, restricts permanent tether area to no less than 113 square feet (or a six-foot trolley radius at ground level), prohibits tethering for more than ten hours in any twenty-four-hour period and between 10:00 p.m. and 6:00 a.m. (with a fifteen-minute exception), and incorporates the Tufts Animal Care and Condition Weather Safety Scale to limit outdoor confinement under adverse conditions. Connecticut General Statutes §22-350a, as amended by Public Act 10-100 with effect 1 October 2010 (and subsequently amended, current version under Public Act 22-59), prohibits tethering a dog by means of a coat hanger, choke collar, prong-type collar, head halter, or any other collar, halter, or device that is not specifically designed or properly fitted for the restraint of the dog. The Animal Legal and Historical Center records that twenty-three states and the District of Columbia have enacted laws regulating the tethering of dogs, with several states explicitly naming choke, prong, or pinch collars as prohibited tethering equipment. These statutes are partial restrictions in the tethering context rather than comprehensive prohibitions on the sale and use of aversive equipment, but they establish that United States state legislatures have already recognised the welfare concerns associated with these tools and have begun legislating accordingly (Animal Legal and Historical Center, 2022). No comprehensive sale-and-use prohibition has yet been enacted at the United States state level. Pending legislative activity in the 2024 to 2026 period, however, reflects active interest across multiple states and across multiple legislative-design models, as discussed below.

Table 1. Comparative Summary of Jurisdictions That Have Enacted Prohibitions or Restrictions on Aversive Dog Training Equipment.

The following table consolidates the jurisdictional record of legislative and regulatory action against aversive dog training equipment, organized by region. The table is not exhaustive but documents the principal jurisdictions cited throughout this paper. Statutory citations and effective dates are drawn from primary sources where available and from the Welsh Government's 2017 review and the FVE, FECAVA, FEEVA, and WSAVA 2024 joint position paper for jurisdictions where primary verification was conducted through those secondary references.

Pending United States legislation in the 2024 to 2026 period reflects three distinct legislative-design approaches to aversive equipment and aversive methods, beyond the enacted tethering statutes already discussed. The first approach is professional licensing of dog trainers tied to non-aversive standards. New York Assembly Bill A 6985 and Senate Bill S 7723 of the 2025-2026 session would have added Agriculture and Markets Law section 113-a, requiring licensing and educational standards for individuals providing canine training to non-service and non-police dogs, with the statutory language explicitly mandating non-aversive, evidence-based, positive reinforcement techniques as the basis of those standards. The Assembly version had its enacting clause stricken on 20 February 2026; the Senate version remains in the Senate Agriculture Committee. New Jersey has introduced parallel proposals under the same professional-licensing approach. Senate Bill S 3814 of 2024 would have established a Dog Training Licensing Board with an evidence-based humane training code precluding aversive methods (the bill was pulled after testimony from trainers in early 2025). Assembly Bills A 4206 and A 4207, both introduced 19 February 2026, would establish, respectively, a Board of Examiners of Dog Trainers under the Dog Trainer Licensing Act, and a New Jersey Dog Trainer Licensure Board under the Dog Training Licensure Act. A 4207 expressly ties licensure standards to professional codes of ethics that incorporate the Least Intrusive, Minimally Aversive Effective Behavior Intervention Policy adopted jointly by APDT, CCPDT, and IAABC. The second approach is restriction of aversive equipment within specified behavior-modification contexts. Massachusetts House Bill H 2342 and Senate Bill S 1459, in the 194th General Court, would require that any dangerous-dog behavior modification plan ordered under the proposed dangerous-dog statute use exclusively evidence-based training techniques that do not result in pain, discomfort, fear, or anxiety, and would explicitly exclude electric, prong, and choke collars from such plans, with required adherence to the principles of the American Veterinary Society of Animal Behavior and the American College of Veterinary Behaviorists. Both bills were reported favorably out of the Joint Committee on Municipalities and Regional Government in 2025 and remain pending. The third approach is enhancement of existing tethering and care statutes. Rhode Island House Bill H 7487 of 2026, introduced 4 February 2026 and referred to the House Judiciary Committee, would increase penalties for repeat violations of the existing dog care and tethering statute and would expand enforcement authority to include city and town animal control officers. None of these proposals had been enacted as of the date of this paper. Collectively, however, they establish that United States state-level legislative interest in regulating aversive training equipment, aversive methods, and the dog training profession itself is active across multiple states and across multiple legislative-design models.

The pattern across these jurisdictions is consistent. Where evidence-based welfare considerations have been weighed by national, regional, or state legislative bodies, the consistent direction of policy has been toward restriction or prohibition of aversive training equipment, never toward expansion of access or normalization of use. The United States is, at the federal level, an outlier in this comparative regulatory landscape.

7.5 The Burden of Proof and the Absence of Adverse Outcomes

The burden of proof in a public welfare policy debate lies with the party defending devices whose mechanism engages nociception and threat circuitry, and, in the specific case of neck-pressure equipment, whose use has been associated in peer-reviewed literature with measurable physical effects on the canine neck. That burden has not been met. No controlled study has demonstrated that aversive training equipment produces better training outcomes than reward-based training when the comparison is fair. No controlled study has demonstrated that aversive training equipment is necessary for any training or behavior modification problem that cannot be addressed by reward-based methods. No jurisdiction that has banned these devices, some of which have operated under bans for more than fifteen years, has produced peer-reviewed evidence of increased public safety risk attributable to the prohibition. Wales has banned electronic collars since 2010. Switzerland has banned spike, pinch, and electronic collars for years. Germany and Austria have operated under their respective bans for years. The predicted adverse consequences to canine safety, dog-guardian relationships, or public safety have not been documented in the published literature.

This absence of documented adverse outcome evidence is itself data. When a significant intervention is removed from public access across multiple jurisdictions over more than fifteen years without any observable peer-reviewed evidence of increased public safety risk, the claim that the intervention is necessary for public safety has been empirically tested and has failed.

7.6 Conversational Deployment of the Argument: Translating the Science Into Spoken Language

Sections 1 through 7.5 lay out the case for prohibition in technical, peer-reviewed terms. The practitioners this paper is written for, however, including trainers, behavior consultants, certified applied animal behaviorists, and veterinary behaviorists, will not always be in a position to deploy the argument that way. Client conversations, social media exchanges, podcast interviews, continuing education sessions for mixed audiences, and informal exchanges with colleagues all call for the practitioner to translate the academic argument into spoken language that stays accurate to the science but lands with the listener. The subsection that follows is a deployment toolkit. For each of ten core argumentative pillars of this paper, three formulations are offered: a technical version (the academic statement), a plain version (one to two sentences a practitioner can use in conversation), and a thirty-second version (a single tight sentence for situations where even the plain version is too long). Practitioners are welcome to adapt these formulations to their own voice and audience. The three-tier structure is a teaching scaffold, not a script.

Concept 1: The mechanism is aversive, regardless of label (Section 2.1).

Technical version: Aversive stimuli are defined functionally, not descriptively. A stimulus that drives avoidance, escape, or termination behavior is, by behavior-science definition, an aversive stimulus, regardless of the vocabulary used to describe it.

Plain version: Whatever the trainer or manufacturer calls the collar, the dog's nervous system processes the experience the same way. The label does not change what is actually happening to the dog.

Thirty-second version: If it changes behavior, it bothers the dog. That is what aversive means.

Concept 2: The intensity dial proves the mechanism (Section 2.2).

Technical version: The existence of an intensity adjustment on aversive equipment, and the clinical necessity of escalating that adjustment when the dog does not respond, is itself evidence that the mechanism is aversive control. If the stimulus were not functioning as an aversive, escalation would have no behavioral effect.

Plain version: Every shock collar has a dial that goes higher. That dial exists because sometimes the trainer has to turn it up to get the dog to respond. If the lower setting were already enough, nobody would need a higher setting. The dial is the proof that the collar works by being unpleasant.

Thirty-second version: Why does the collar have a dial? Because sometimes you have to turn it up. That is the answer.

Concept 3: The "barely perceptible" contradiction (Section 2.3).

Technical version: If the stimulus is behaviorally meaningful enough to change behavior through escape, avoidance, or suppression, it is by functional definition aversive. If it is not behaviorally meaningful, it is not doing the training work. There is no intermediate category.

Plain version: Some trainers say the modern e-collar is so mild that the dog barely notices it. But if the dog barely noticed, it would not change the behavior. If it changes the behavior, the dog is noticing. Both things cannot be true at the same time.

Thirty-second version: If the dog barely feels it, it would not work. That it works tells you the dog feels it.

Concept 4: Mechanism, not tissue damage, is the welfare question (Section 4.1).

Technical version: The welfare case does not depend on tissue injury. Nociceptive engagement and threat-system activation occur well below the threshold of visible tissue damage. The biological question is whether the stimulus crosses the nociceptive threshold, not whether it crosses the injury threshold.

Plain version: The welfare question is not whether the collar leaves a mark. The welfare question is whether the device activates the dog's pain and fear systems. Those systems do not need tissue damage to be triggered. They are designed to warn the body before damage happens. That warning system is what these collars activate.

Thirty-second version: You do not have to injure a dog to harm the dog. Pain and fear do not require visible damage.

Concept 5: The convergence argument (Section 3 and Table 2).

Technical version: The welfare evidence is convergent across multiple independent methodological approaches, populations, countries, and outcome measures. No single study carries the case alone, and the agreement across methodologies eliminates the possibility that the welfare signal is a methodological artifact.

Plain version: This is not based on one study. Fifteen peer-reviewed studies from multiple countries, using completely different methods, including controlled experiments, observational research, surveys of thousands of dog guardians, cognitive bias tests, and clinical referral data, all find the same welfare cost. When that many independent methods point in the same direction, the evidence is solid.

Thirty-second version: It is not one study. It is fifteen, in different countries, with different methods, all pointing the same direction.

Concept 6: The necessity claim has been empirically tested and has failed (Section 5 and 7.1).

Technical version: Board-certified veterinary behaviorists, the clinical specialty that treats the most severe canine aggression, anxiety, fear, and predatory cases, treat those cases as their standard of care without aversive equipment. The American College of Veterinary Behaviorists has formalized this position in their December 2025 letter to the American Veterinary Medical Association.

Plain version: The veterinary specialists who handle the worst aggression and anxiety cases in the country do not use shock or prong collars. The American College of Veterinary Behaviorists, the highest credential in this field, does not include aversive equipment in their standard of care. If the hardest cases can be solved without these tools, the everyday cases certainly can.

Thirty-second version: The veterinary specialists who treat the toughest cases do not use these tools. That tells you everything.

Concept 7: Adding food does not subtract the aversive (Section 4.4).

Technical version: Compound schedules pairing positive reinforcement with positive punishment or negative reinforcement do not eliminate the welfare cost of the aversive component. The dog's nervous system continues to register and respond to the aversive event regardless of whether food reinforcement is present in parallel.

Plain version: Some trainers say it is fine to use a shock or prong collar as long as you also give the dog treats. Adding treats does not erase the shock. The dog's brain still registers the unpleasant event. The peer-reviewed studies that have looked at this directly find welfare costs even when food rewards are also present.

Thirty-second version: Treats do not cancel out shocks. The dog's brain registers both.

Concept 8: The working level is the aversive level (Section 2.3 and 4.1).

Technical version: Behavior change driven by escape, avoidance, or suppression requires the stimulus to function as an aversive event. A stimulus below the dog's aversive threshold cannot drive avoidance learning. The setting at which the stimulus successfully changes behavior is, by functional definition, the setting at which the stimulus has crossed the dog's aversive threshold.

Plain version: Trainers using these tools talk about finding the dog's working level, meaning the lowest setting that gets the dog to respond. The working level and the aversive level are the same number. By definition, the setting that makes the dog change behavior is the setting the dog wants to avoid. There is no working level below the aversive threshold, because below that threshold, the device would not work.

Thirty-second version: The working level is the aversive level. Those are the same setting.

Concept 9: Predictability and controllability do not eliminate the welfare cost (Section 4.3).

Technical version: Controllability and predictability attenuate but do not eliminate threat-related neural responding. They do not render aversive stimulation neurologically neutral or welfare-benign. This has been confirmed in writing by Dr. Luiz Pessoa, the senior author of Limbachia et al. (2021), and Dr. David Knight, the senior author of Wood et al. (2014).

Plain version: Some trainers say a shock the dog can predict and control is welfare-neutral. The neuroscience says the opposite. Predictability and controllability reduce how strongly the brain responds, but they do not eliminate the threat response. The senior researchers whose work is most often cited to support the proponent claim, Dr. Pessoa and Dr. Knight, have confirmed in writing that their research does not support that claim.

Thirty-second version: Predictable shocks still hurt. The researchers being cited say so themselves.

Concept 10: International consensus and jurisdictional precedent (Section 7 and Table 1).

Technical version: Aversive training equipment is prohibited or restricted in multiple national and subnational jurisdictions across Europe, the United Kingdom, Australia, North America, and Latin America, and is opposed by every major international veterinary, welfare, and professional behavior organization that has issued a position statement on the question, including the FVE, FECAVA, FEEVA, and WSAVA joint position paper of June 2024.

Plain version: Wales has banned shock collars since 2010. Switzerland banned aversive collars years ago. Quebec banned them in 2024. Most of Australia, France, the Netherlands, and many other countries have restrictions or full bans. Every major veterinary organization that has reviewed the evidence has reached the same conclusion. The United States is increasingly the outlier on this issue.

Thirty-second version: Most of the developed world has banned or restricted these tools. The United States is the outlier.

Three notes on deployment. First, the formulations above are written for clarity, not for memorization. The most important step for a practitioner is to internalize the underlying conceptual structure of the argument from Sections 2 through 7 of this paper, after which the practitioner's own voice will produce conversational versions naturally. The formulations above are useful as study aids and as starting points, not as scripts to recite. Second, the deployment context matters more than the wording. A grieving guardian who has just lost a dog after a behavioral euthanasia following an aversive training failure does not need the thirty-second version. They need the practitioner to listen first and to deploy the argument only when invited. A balanced trainer in a public online debate is the opposite context, in which the thirty-second version may be all the practitioner has time to deliver before the discussion moves on. Read the room. Third, the practitioner's credibility is built over time, through consistent, calm, accurate communication, not through any single phrasing. The argument is correct. The practitioner who delivers it in their own voice, calmly, with care for the listener, is the practitioner the argument needs.

8. Real-World Use: Why Research Conditions Underestimate Risk

The welfare problems with aversive equipment are intrinsic to the mechanism by which the equipment operates. Aversive control engages nociception and threat circuitry whether the handler is a novice or a master. Skill does not eliminate the aversive event, because the aversive event is what makes the equipment work. The convergent evidence from controlled studies, neuroscience, ethology, and clinical veterinary behavior medicine establishes that no level of handler skill renders an aversive welfare-neutral. The international veterinary, professional, and welfare consensus reflects this. AVSAB, the ACVB, the AAHA, the FVE/WSAVA joint position paper, the BVA, the BSAVA, the AVA, the CVMA, the NZVA, and the leading professional training and behavior organizations all conclude that aversive equipment should not be used at all, by anyone, in any setting. The policy this paper recommends is therefore not a restriction directed at lay users while reserving expert use. It is a comprehensive prohibition on sale, import, and use, applying equally to professional trainers, behavior consultants, hobbyists, and pet guardians.

Real-world use data add a separate and additional welfare concern. Even setting aside the intrinsic mechanism-based welfare argument entirely, the empirical record on how aversive equipment is actually used in the population shows predictable harm at scale. The following subsections document five categories of real-world evidence: who uses these devices and how widespread that use is (Section 8.1), how users acquire and apply the equipment (Section 8.2), what the manufacturing and product-engineering record shows about the devices themselves (Section 8.3), what United States guardians actually believe about these devices (Section 8.4), and the regulatory and informational environment in which sale and use occur (Section 8.5).

8.1 Prevalence and User Profile

Aversive training equipment is widely available in the United States consumer market. Electronic collars, prong collars, choke chains, and remote training systems are sold through national pet specialty retailers (with the notable exception of Petco, which ended retail sale of human and bark-activated electronic shock collars in October 2020), through national online marketplaces, through breed-club networks, and through the personal sales channels of trainers who use these devices in their practice. There is no national US registry of training equipment sales, no required reporting of use or adverse events, and no requirement that purchasers receive any education or assessment before purchase. The published research consequently relies on guardian self-report surveys for prevalence estimates.

Blackwell et al. (2012), in a UK sample, found that owner attendance at training classes and owner gender were the strongest factors distinguishing electronic collar users from owners using reward-based methods, suggesting that source of training advice rather than dog characteristics drove tool selection. A significantly higher proportion of owners in the reward-based comparison group reported training success than those in the e-collar group for comparable recall and chasing problems (Blackwell et al., 2012). The same general profile, in which the source of advice (or absence of advice) is the principal driver of equipment selection, has been documented across multiple jurisdictions and survey populations.

Starinsky, Lord, and Herron (2017) examined the effect of various containment methods on escape rates and biting histories in dogs confined to their guardians' properties. Their findings did not support a clear protective effect of electronic containment systems. Escape rates were nearly twice as high for dogs confined by electronic fence (forty-four percent) as for dogs confined by physical fencing (twenty-three percent), with comparable rates for tethered dogs (twenty-seven percent) (Starinsky et al., 2017). The tool most often cited as necessary for suburban dog containment does not, in the available evidence, rescue the bite or escape profile of the population in which it is used. The user profile for electronic containment systems substantially overlaps with the user profile for remote training collars, and the two product categories are sold by overlapping manufacturer networks.

The user profile for prong collars and choke chains differs in important ways from the electronic collar user profile but converges on similar welfare implications. Prong collar use in the United States is concentrated in (a) sport dog and protection dog communities, (b) certain working-breed enthusiast subcultures, (c) lay guardians purchasing through retail channels for leash-pulling control, and (d) a subset of professional trainers operating in the balanced training tradition. Choke chain use is broader and includes a substantial cohort of guardians who purchased the equipment based on advice from breed clubs, older training manuals, or family tradition rather than contemporary professional behavioral guidance. Across all of these populations, the empirical record from Hiby et al. (2004), Arhant et al. (2010), Casey et al. (2021), and the broader survey literature is consistent: aversive method use correlates with worse rather than better behavioral outcomes, regardless of user sophistication.

8.2 Professional Guidance Patterns and User Behavior

Masson, Nigron, and Gaultier (2018) surveyed 1,251 respondents in France about electronic collar use. They found that 26 percent reported having used an electronic collar at some point. Among those users, 71.8 percent used the collar without professional advice, 75 percent had tried two or fewer other solutions before reaching for the collar, and 7 percent of dogs on which collars had been used presented with physical wounds. The authors concluded that real-life use is far from the idealized conditions of experimental studies and may put dog welfare at higher risk than the scientific literature suggests (Masson et al., 2018b).

These numbers are critical for policy because they describe the actual exposure profile in the population, not the experimental exposure profile in laboratory or controlled-trial conditions. A device used by a lay guardian, without professional supervision, after fewer than three attempts at any alternative method, on a dog whose underlying motivation for the target behavior has not been assessed, in environments where the behavioral context for application of the stimulus is not controlled, is not the same intervention that appears in published efficacy studies. It is a different intervention, with a different exposure profile, applied to a different population. Policy that addresses only the experimentally idealized version of aversive training equipment ignores the version that actually exists in the consumer marketplace.

The same real-world considerations apply to prong collars and choke chains. Consumer-purchased prong collars are commonly fitted incorrectly, used with excessive force, left on unsupervised dogs, and applied by lay handlers with no training in timing or body language. Choke chains are routinely used with force profiles that exceed what any clinician would consider safe. The Herron et al. (2009) data on aggressive responses to choke and pinch collar use in the general pet population, eleven percent, indicates a rate of confrontational outcomes that would be unacceptable for any consumer product with a safer available alternative. The same study documented aggressive responses in the high single to low double digits across the full set of confrontational handling techniques the authors examined, none of which are restricted in the United States consumer marketplace or regulated under any state or federal training standard.

When these guardian-survey patterns are combined with the experimental and observational welfare data summarized in Section 3, the result is convergent. Aversive training equipment, in the population in which it is actually used, produces measurable welfare harm and produces no measurable advantage in training outcome. The pattern holds across country, language, survey methodology, and outcome measure.

8.3 Device Variability and the Absence of Manufacturing Standards

There is also a body of real-world evidence about the devices themselves. Lines, van Driel, and Cooper (2013), in research published in Veterinary Record, examined the electrical characteristics of thirteen commercially available electronic training collar models under realistic conditions, including the electrical impedance of dogs' necks measured separately for wet and dry coats. The study found large differences between e-collar models in delivered energy, peak voltage, number of pulses, and pulse duration. Stimulus energy at the maximum strength setting at a 50 kΩ load ranged from 3.3 millijoules to 287 millijoules across the tested models, an eighty-seven-fold range across products marketed for the same use category (Lines, van Driel, and Cooper, 2013).

That variability has direct welfare implications. A guardian who buys an electronic training collar at maximum setting from one manufacturer is buying a meaningfully different stimulus than a guardian who buys an apparently equivalent product from another manufacturer. Comparison shopping, in any informed-consent sense, is impossible, because the relevant electrical characteristics (voltage, current, pulse width, waveform, and total energy delivery) are not disclosed by manufacturers in any standardized format on packaging, in marketing materials, or at the point of sale. The Electronic Collar Manufacturers Association maintains a voluntary technical standard, but adherence is self-reported and is not enforced by any government regulatory body in the United States.

The United States has no Food and Drug Administration regulation of these devices, no Consumer Product Safety Commission standard for them, no United States Department of Agriculture regulatory framework for their manufacture or sale, and no state-level technical standard. The regulatory vacuum does not extend to other consumer products that deliver electrical stimulation to a body. Therapeutic transcutaneous electrical nerve stimulation devices for human use are regulated as medical devices by the Food and Drug Administration, with required disclosure of pulse parameters, mandatory clinical evidence for marketing claims, and adverse event reporting requirements. Electronic dog training collars, which deliver substantially higher peak voltages than therapeutic TENS units (Section 9.7), are subject to no comparable framework. The asymmetry between the regulatory treatment of human-targeted electrical stimulation devices and animal-targeted electrical stimulation devices is hard to defend on the merits.

Prong collars and choke chains are not subject to United States federal product safety regulation either. There is no required tensile strength rating, no required disclosure of intended pressure load, no point-of-sale guidance on fit or use, and no adverse event reporting requirement. The same product category, sold under different brand names, varies in prong sharpness, prong spacing, link weight, and chain composition, with no standardization. A consumer who buys one of these products is buying a different functional intervention than a consumer who buys another. The lack of standardization on the manufacturing side compounds the lack of standardization on the user side documented in Section 8.2.

8.4 United States Public Attitude Data

The third strand of real-world evidence is what United States guardians actually believe about these devices. The most substantial available data point comes from October 2020, when Petco, then the second-largest pet specialty retailer in the United States, ended the retail sale of human and bark-activated electronic shock collars and announced its #StopTheShock campaign. As part of that announcement, Petco disclosed survey data conducted by the market research firm Edelman Intelligence (Petco, 2020). Seventy percent of dog guardians surveyed reported that they believed shock collars had a negative impact on their pet's emotional or mental wellbeing. Sixty-nine percent of dog guardians surveyed considered shock collars a cruel training method. Petco additionally reported separate consumer research finding that fifty-nine percent of pet guardians surveyed would prefer to shock themselves than their dog (Petco, 2020).

These figures, even discounted appropriately for the limitations of corporate-sponsored survey research, document something important. United States guardians are not generally enthusiastic users of aversive electronic equipment. The substantial majority of United States guardians, by Petco's own reported numbers, already perceive these devices as harmful to canine emotional wellbeing and consider them cruel. The constituency that would be inconvenienced by a sale and use prohibition is a minority of the consumer population, not a majority of it.

This is structurally important for policy. Public welfare regulation that prohibits a class of consumer product faces predictable opposition framed in terms of consumer choice and personal liberty. The empirical evidence available from Petco's 2020 corporate disclosure, however, suggests that the consumer choice argument has weak support in the population it claims to represent. A majority of dog guardians, on the basis of the available US survey data, would not lose access to a product they value if aversive electronic training equipment were removed from the consumer marketplace. They would lose access to a product they already perceive as harmful, and that they would, in significant numbers, prefer not to use on their dogs in the first place.

The visible public face of opposition to aversive equipment prohibition in the United States, comprised of certain trainer trade associations, certain breed-specific advocacy groups, and certain manufacturer-funded campaigns, does not represent the modal United States dog guardian. It represents a vocal subset whose professional or commercial interests align with continued availability of these devices. Public welfare policy is properly oriented to the empirically documented preferences of the affected population, not to the loudest organized voices in the policy conversation.

8.5 The Information Environment and the Regulatory Vacuum

Todd (2018) analyzed the barriers to adoption of humane training methods by the general public, identifying lack of knowledge of welfare risks, poor quality of information available to guardians, lack of regulation of dog trainers, and inconsistent positions among professional bodies as factors that maintain the continued use of aversive methods (Todd, 2018). Each of those factors operates with particular force in the United States consumer environment.

The first, lack of knowledge of welfare risks, is structural. Aversive training equipment carries no required warning labels, no point-of-sale disclosure of welfare research, and no regulatory disclosure framework comparable to the warning frameworks applied to other consumer products with documented adverse effects. Guardians purchasing these devices commonly do so on the basis of marketing claims (the device is gentle, the device is humane, the device is safe at low settings) that are inconsistent with the published welfare and neuroscience literature.

The second, the quality of available information, is inconsistent across sources and across professional channels. The veterinary, behavioral, and welfare-organization consensus opposing aversive equipment (Section 7) is not communicated to guardians at the point of purchase, in retail catalogs, or by manufacturers. Internet search results for training equipment commonly surface industry-funded sources that contradict the peer-reviewed welfare literature. Trainer marketing materials commonly endorse aversive equipment notwithstanding the contrary professional and welfare consensus.

The third is the regulatory vacuum around dog trainers themselves. Commercial dog training is essentially unregulated worldwide. Germany is the closest exception: under §11 of the Animal Welfare Act (Tierschutzgesetz), anyone who commercially trains dogs for third parties must obtain a permit from the local veterinary office demonstrating animal-related qualifications. The §11 permit applies to a range of animal businesses (dog training, grooming, daycare, breeding), and the granting decision is administrative rather than examination-based. It is a permit requirement, not a licensure framework comparable to those that govern veterinarians, mental health counselors, and social workers. No other developed country, including the United States, has even that level of regulatory gatekeeping for the canine training profession. Anyone may operate as a dog trainer in the United States without any required education, examination, supervised practice, or continuing education requirement. There is no state regulatory body to which guardians can complain when a trainer's methods cause harm. There is no professional malpractice framework comparable to the one operative in veterinary medicine, mental health counseling, or social work. New York Senate Bill S 7723 and Assembly Bill A 6985 (2025-2026 session) represent one of several serious United States state-level attempts to address this regulatory vacuum; parallel licensure proposals in New Jersey (Senate Bill S 3814 of 2024 and Assembly Bills A 4206 and A 4207 of 2026) are documented in Section 7.4. Until comparable measures pass, the United States consumer market for canine training and behavior modification operates without the professional gatekeeping that protects consumers in welfare-affecting professions generally.

Fourth, inconsistent positions among professional bodies have been substantially resolved in the past decade. The convergence documented in Section 7 (AVSAB, ACVB, AAHA, FVE/WSAVA, BVA, BSAVA, AVA, CVMA, NZVA, ESVCE, and the major welfare and training organizations) leaves no serious mainstream professional disagreement on the welfare implications of aversive equipment. The remaining inconsistency lies in the gap between what the veterinary and behavioral consensus has concluded and what consumers encounter at the retail level. This gap is structurally maintained by the absence of a regulatory framework. Closing it requires policy action.

The implication for policy is direct. Consumer-level availability of aversive training equipment, in a market where lay guardians have limited access to high-quality behavioral information, predictably produces welfare harm even before anyone considers the intrinsic properties of the devices. A prohibition on the sale and use of aversive training equipment is therefore a consumer protection measure, not just an animal welfare measure. It removes from the consumer marketplace a class of product whose advertised claims are not supported by the peer-reviewed evidence, whose adverse effects are documented but not disclosed, and whose appropriate use requires professional gatekeeping that does not currently exist in the United States and is not on the visible legislative horizon at the federal level.

Policy must be designed for the actual properties of the equipment and the actual conditions of its use. The intrinsic argument is that aversive equipment works by aversive control, which engages nociception and threat circuitry regardless of who is operating the device. The empirical argument is that real-world use compounds that intrinsic welfare cost through inconsistent timing, poor fit, prolonged wear, use in aggression and anxiety cases, use around triggers, and use by people without education in behavior analysis or canine body language. Both arguments lead to the same conclusion. The right policy response is a comprehensive prohibition on the sale, import, and use of these devices, not a tiered system that imagines safe expert use on one side and unsafe lay use on the other. The international veterinary, professional, and welfare consensus has already rejected the tiered framing. The United States should align with that consensus.

9. Anticipated Proponent Objections

9.1 "You Are Cherry-Picking Studies."

This objection can only be answered by convergence, and convergence is what this paper provides. The argument does not rest on any single study. It rests on multiple independent lines of evidence using different methods, populations, and outcome measures: welfare-focused experimental studies (Cooper, China, Vieira de Castro, Casey), direct observational studies (Deldalle and Gaunet, Rooney and Cowan), guardian survey data (Blackwell 2008, Blackwell 2012, Hiby, Masson), confrontational handling clinical data (Herron), peer-reviewed physical-effects research on neck-pressure equipment (Carter, Pauli, Hunter, Grohmann), threat-circuit and fear-conditioning neuroscience (Limbachia, Wood, LeDoux), stress neurobiology (McEwen, Rosenkranz, Vyas), and the gate-control theory of pain that underwrites the TENS-versus-shock distinction (Melzack and Wall). No single study in this list is perfect. Together, they point in the same direction.

A critic who wants to dismiss the convergence argument must explain why each of these independent methodological lines is wrong, why the same welfare signal would appear across such different methods by coincidence, and why every major veterinary, professional, and welfare organization that has reviewed this body of evidence has reached the same conclusion. That explanation has not been provided. Sections 7.1 through 7.5 of this paper detail the institutional consensus; this section addresses only the cherry-picking critique of the underlying scientific evidence itself.

9.2 "Cortisol Was Not Significant in Cooper 2014."

As discussed in Section 3.3, cortisol is one measure, and a blunt one. Behavioral stress indicators in Cooper et al. (2014) were significant in the electronic collar group, and these are validated welfare markers in canine research. Dismissing a study on the basis of one non-significant physiological measure while ignoring significant behavioral measures is not a valid methodological critique. It is selective reading.

9.3 "The Reward Groups Just Had More Reinforcement."

As discussed in Section 5.1, this argument either concedes that the aversive equipment added no outcome benefit, in which case it carried welfare cost for no training advantage, or reduces to a hypothetical about a better-designed protocol that has not been tested. The trainers in the aversive equipment groups of Cooper and China were industry-nominated, and the high-reinforcement protocol is available to any trainer who wishes to deliver it. They did not. The existing evidence is what policy must address.

9.4 "Controllability and Predictability Make Aversive Use Safe."

Controllability and predictability reduce the magnitude of threat-related neural responding. They do not convert aversive stimulation into a neutral stimulus (Limbachia et al., 2021; Wood et al., 2014; Pessoa, personal communication, 2026; Knight, personal communication, 2026). Controllability is precisely what negative reinforcement provides, and the aversive nature of the stimulus is precisely what makes the avoidance or escape response reinforcing. A claim that controllability neutralizes the aversive contradicts the definitions of the procedures under which controllability operates.

9.5 "Low-Level Stim or Gentle Corrections Are Benign."

As discussed in Section 2.3, the low-level argument runs into a logical contradiction across all aversive modalities. If the stimulation or correction is behaviorally meaningful enough to change behavior through escape, avoidance, or suppression, it is by functional definition aversive. If it is not behaviorally meaningful, it is not doing the training work. There is no intermediate category where the stimulus is both strong enough to reliably shape behavior and mild enough to have no motivational salience.

9.6 "It's Not Really Aversive, the Dog Just Finds It Weird or Tingly and Wants to Turn It Off."

When the aversive-control argument is pressed, a specific retreat position is common among proponents of electronic collars. The retreat runs like this. At the low levels that skilled trainers use, the stimulation is not painful and not aversive. It is merely unusual, odd, prickly, or tingly. The dog simply notices it, finds it strange, and works to make it stop. That is not the same as pain or fear, and therefore it is not aversive in any welfare-relevant sense.

There are four reasons the argument fails, each of them sufficient on its own.

The first is that the argument misuses the word aversive. In behavior analysis, which is the scientific field that defines these terms, the category "aversive" is functional, not subjective. A stimulus is aversive when its presentation decreases the future probability of the behavior it follows (positive punishment), or when its removal, termination, avoidance, or prevention increases the future probability of the behavior that produces that outcome (negative reinforcement). This is the standard textbook definition in applied behavior analysis (Cooper, Heron, and Heward, 2020). It is not a preference. It is a definition. If the dog works to terminate, avoid, or prevent the stimulus, the stimulus is functioning as an aversive by the only definition the science of behavior recognizes. The argument that the dog finds the stimulus "weird" or "odd" but not aversive substitutes a subjective characterization of the dog's private experience, which the dog cannot confirm or deny, for the functional definition that textbooks, credentialing bodies, and the peer-reviewed learning literature actually use. It is not a scientific argument. It is a relabeling exercise.

The second is that the argument cannot distinguish itself from the mechanism it claims not to be. The statement "the dog finds it weird and wants to make it stop" is a plain-language description of negative reinforcement. The dog's behavior is being maintained by termination of a stimulus. That is exactly what negative reinforcement means, and by definition any stimulus that functions as a negative reinforcer is an aversive stimulus. There is no category of stimulus called "weird but not aversive" in the experimental learning literature that supports avoidance learning. Such a category does not exist in the controlled behavioral studies that underlie the field.

The third is that the argument is contradicted by the neurobiology of attention and habituation. A stimulus that is merely novel, unusual, or odd, without any negative valence, produces an orienting response that is followed by habituation. The organism attends to the stimulus once, possibly twice, and then the brain filters it out as irrelevant. This is documented in hundreds of studies on habituation across species, including dogs (Thompson and Spencer, 1966, for the foundational characterization of habituation across taxa). What a merely novel stimulus does not produce is sustained, reliable behavioral avoidance that persists across hundreds of trials without habituation. The fact that electronic collar stimulation continues to drive behavioral modification after repeated exposure, without habituating away, is evidence that the dog's nervous system is not classifying the stimulus as merely novel. It is classifying it as something worth working to avoid, which is the functional signature of an aversive.

The fourth and most decisive is that the argument is empirically closed by the industry-nominated best-practice studies. Cooper et al. (2014) studied trainers nominated by the Electronic Collar Manufacturers Association as representing best practice. Those trainers used low stimulation settings, manufacturer-recommended protocols, and pre-warning cues. The dogs still showed stress-related behavioral indicators, including increased yawning, panting, yelping, and tense body posture. China, Mills, and Cooper (2020) re-analyzed the same dataset focusing on efficacy and confirmed that the sophisticated low-level protocol, in which stimulation intensity was matched to the dog's tolerance, a pre-warning vibration cue was used, and the stimulus was applied as negative reinforcement rather than positive punishment, did not produce superior training outcomes compared with the reward-based training group. That is the "it's just a tingle" protocol, operationalized as the industry itself describes it. The behavioral welfare indicators documented in Cooper (2014) appeared in the e-collar group anyway, and the reward-based comparison group achieved equal or better training outcomes without any stimulation. The "it's only weird at low levels" defense has been tested on the industry's own terms, by the industry's own nominated trainers, using the industry's own best-practice protocols. It failed. The proponent who retreats to this position is retreating to a defense that has already been empirically refuted.

9.7 "An Electronic Collar Is Just Like a TENS Unit."

A common proponent analogy compares electronic training collars to transcutaneous electrical nerve stimulation (TENS) units, which are used in human medicine for pain relief and musculoskeletal therapy. The argument runs that TENS is medical-grade, widely used in physical therapy, considered safe, and therefore an electronic training collar, which also delivers electrical stimulation through the skin, should be treated the same way. This analogy fails on the biology.

TENS and electronic training collars have opposite therapeutic purposes, and therefore use opposite electrical parameters. Therapeutic TENS is designed to produce analgesia. It works according to the gate control theory of pain proposed by Melzack and Wall (1965) and elaborated in the subsequent fifty years of pain neuroscience. The gate control theory holds that pain signals traveling from the body to the brain can be modulated, dampened, or blocked at the level of the spinal cord. Two kinds of nerve fibers feed into a so-called gate in the dorsal horn of the spinal cord. Large-diameter A-beta fibers carry harmless touch and vibration signals. Small-diameter A-delta and C fibers carry pain signals. When the large-diameter touch fibers are active, they close the gate, reducing how much pain signal gets through to the brain. This is the same mechanism by which rubbing a stubbed toe reduces the perceived pain. TENS exploits this mechanism deliberately. The pulse parameters of a therapeutic TENS unit, including low intensity, specific frequency ranges, and pulse durations typically in the fifty to two hundred fifty microsecond range, are calibrated to selectively activate the large-diameter A-beta touch fibers without crossing the nociceptive threshold. The therapeutic goal is reduction of pain perception by recruiting the non-painful sensory pathway to dampen pain transmission.

An electronic training collar is designed to do the opposite. It must deliver a stimulus the dog works to avoid, escape, or terminate. As established throughout this paper, that requirement means the stimulus must function as an aversive event, which by definition means crossing the nociceptive threshold and engaging nociception and threat circuitry. A stimulus calibrated to remain below the nociceptive threshold, like therapeutic TENS, could not drive avoidance learning, and therefore could not train anything. The two devices cannot share a biological mechanism because they have opposite biological goals.

The analogy also fails on electrical parameters. Training collar manufacturers disclose very little about the specific voltage, current, pulse width, and waveform characteristics of their devices, a finding documented by Lines, van Driel, and Cooper (2013) in Veterinary Record, who measured substantial variation in the characteristics of commercially available training collars. The fact that both TENS and training collars deliver electricity through electrodes on the skin is not sufficient biological similarity to support the analogy, any more than the fact that a flashlight and a surgical laser both emit light makes them biologically equivalent.

The analogy also involves a category error about the recipient. TENS is self-administered by a human who controls the intensity, can adjust it in real time, can verbalize the sensation, and can turn the device off instantly if it becomes unpleasant. An electronic training collar is administered to a non-consenting dog who cannot turn it off, who cannot verbalize the sensation, and who cannot negotiate the intensity. The ethical and welfare situations are categorically different, even if the electrical hardware were identical, which it is not.

Finally, the analogy fails on the internal logic of what the tool does. If a training collar really produced only a TENS-level A-beta sensation without nociceptive engagement, the dog would have no behavioral reason to work to avoid it. The fact that the collar functions as a training tool, at all, depends on the dog experiencing the stimulus as something worth terminating. That is not what TENS does. That is the opposite of what TENS does.

9.8 "The Prong Collar Just Gets the Dog's Attention."

The prong collar produces behavior change because the prong points apply concentrated force to the dog's neck when the leash tightens. That force is sufficient to make pulling unpleasant enough that the dog stops pulling. If the prong were only an attention-getting device without an aversive component, the dog could be trained to stop pulling with any other attention-getting device, including a tap, a verbal cue, or a sound. The specific efficacy of the prong collar is its mechanical delivery of an aversive event. This is functionally identical to the electronic collar operating at a level sufficient to stop behavior. The vocabulary of "attention" does not change the mechanism of action.

9.9 "The Choke Chain Mimics How a Mother Dog Corrects Her Pups."

This claim is biologically inaccurate. Mother dogs do not perform sustained neck constrictions on their puppies for training purposes. The alpha-wolf mythology on which this claim rests originated with Schenkel's 1947 captive-wolf studies, was popularized in L. David Mech's 1970 book on wolf ecology and behavior, and has since been extensively critiqued by wolf biologists and canine behavior scientists, including by Mech himself in the peer-reviewed literature (Mech, 1999) and in subsequent public retractions (Mech, 2008). The claim that choke chain corrections mimic maternal canine behavior is not supported by the ethology of mother-pup interactions and is not a rationale for applying choke chain corrections to adult dogs.

9.10 "My Dog Looks Happy in Training."

As discussed in Section 2.5, visible engagement is not a physiological readout. Threat and stress circuitry do not announce themselves through tail posture or eye expression alone. Once avoidance learning is well established, a dog may perform fluently and quickly precisely because the behavior prevents the aversive event. Fluency in the instrumental response is compatible with ongoing threat prediction. Welfare is not reducible to what a dog looks like in a fifteen-second clip.

9.11 The Rescue Device Problem: Unfalsifiability

A pattern that appears repeatedly in proponent argumentation deserves to be named. Every study that documents welfare harm is dismissed by pointing to some methodological limitation. Cooper had reinforcement rate differences. China had reinforcement rate differences. Vieira de Castro was not randomized. Schilder studied working dogs. Cortisol is complicated. Cognitive bias tasks are indirect. Neuroscience studies are in humans, not dogs. Herron relied on guardian reports. Carter used a model neck, not a live dog. Each of these critiques has some partial validity in isolation.

However, when every study showing harm is set aside for a different reason, a pattern emerges. The methodology critique is being used as a rescue device rather than as a genuine pursuit of better evidence. The appropriate diagnostic question is this: what would count as evidence against the tools? If the answer is that no existing study, no convergent finding across disciplines, no professional consensus, and no regulatory precedent would be sufficient, then the position is not scientifically falsifiable. It is a commitment defended by serial critique, not an evidence-based conclusion.

That pattern itself is relevant to policy. Tools whose safety is defended by an unfalsifiable argument structure cannot have their safety demonstrated, because no conceivable finding is permitted to count against them.

9.12 Common Terminology Misuses and Their Behavior-Science Translation

A category of proponent argumentation that deserves separate attention is the systematic misuse of behavior-science terminology to make aversive procedures sound benign or to reframe them outside the welfare evidence base. This is not pedantry. The integrity of the operant conditioning framework, of nociception and threat-circuit science, and of the welfare research summarized in Sections 3 through 8 of this paper depends on accurate terminology. When the language drifts, the welfare evidence gets rhetorically neutralized without ever being substantively contested. This subsection catalogs the most common terminological moves a practitioner will encounter and provides the accurate behavior-science translation for each. The list is illustrative rather than exhaustive.

Group 1: Euphemisms That Rename the Aversive Stimulus

"Stim," "stimulation," "e-touch," or "tap" instead of "shock." These terms are marketing language, not behavior-science language. The functional question is whether the electrical event the device delivers is sufficient to drive avoidance learning. If it is, it is, by definition, crossing the dog's nociceptive threshold and engaging threat circuitry (Section 4.1). The vocabulary chosen by the manufacturer or trainer does not change the neural processing. A useful response in conversation: "What we call it doesn't change what the dog's nervous system does with it. If the stimulus is strong enough to make the dog change its behavior to avoid it, that's an aversive event by any scientific definition."

"Correction" instead of "punishment." In operant conditioning, positive punishment is the application of a stimulus following a behavior that decreases the future probability of that behavior. Calling the application a correction does not move it out of the operant punishment quadrant. If the procedure decreases behavior, it is, by definition, functioning as positive punishment. A useful response in conversation: "In behavior science, the procedure is named by what it does, not by what we call it. If applying the leash pop reduces the behavior, it's positive punishment. That's not an opinion, it's the definition."

"Pressure" or "tap" instead of "pinch" or "compression." Used for prong collars and choke chains to suggest the mechanical event is mild. The mechanical force applied by a prong collar that successfully reduces lead pulling is, by functional necessity, exceeding the dog's mechanonociceptive threshold. If it were below that threshold, it would not change the behavior. The same applies to choke chain compression that reduces lunging or forging.

"Self-correction" for prong collars. This phrase is used to imply the dog, not the handler, is the cause of the aversive event, and therefore that no punishment is being applied. In operant terms, the procedure is unaffected by who is the proximate cause. The contingency defines the procedure. When the dog pulls and the prong applies concentrated pressure, the dog has experienced a behavior-contingent aversive event. Whether the handler activated it, or whether it was activated by the dog's own action, the operant function is identical. Procedures are defined by their contingencies, not by the identity of the agent who closes the contingency.

Group 2: Reframings of the Tool's Function

"Communication tool" instead of "aversive device." Communication does not require an aversive event. Verbal cues, visual cues, hand signals, and trained marker words all communicate without delivering an aversive consequence. If a device communicates by applying an unpleasant consequence the dog works to avoid, it is functioning as an aversive device, regardless of whether the handler frames the application as communication. The label does not change the function.

"Working level" or "minimum effective level." Used to imply that low-level electronic collar use is welfare-benign. The working level is, by definition, the level at which the stimulus is sufficient to change the dog's behavior. As established in Section 2.3, behavior change driven by escape, avoidance, or suppression requires the stimulus to function as an aversive event. A stimulus below the dog's aversive threshold cannot drive avoidance learning. "Working level" and "aversive level" are the same level. The phrase obscures this rather than communicating it.

"Pavlovian" or "classical conditioning" framing for electronic collar use. Sometimes invoked to suggest the e-collar functions through associative learning rather than punishment, with the implication that classical conditioning is welfare-neutral. The framing fails on two grounds. First, classical conditioning to an aversive unconditioned stimulus is the foundational paradigm of fear conditioning research, and decades of neuroscience document the threat-circuit engagement that paradigm produces (Section 4.3). Calling the procedure classical does not exempt it from welfare cost. Second, in actual e-collar training, the operant component (behavior-contingent stimulation that the dog learns to escape, avoid, or prevent) is the working mechanism. The Pavlovian frame is a relabeling that does not change what the device is doing.

Group 3: Methodological Frame Reframings

"Balanced" training as a moderate middle ground. The term implies that balanced training combines reward-based and aversive methods in a moderate, considered way, and that force-free training is the extreme position. From a behavior-science perspective, balanced training is defined by the inclusion of positive punishment and negative reinforcement, regardless of whether positive reinforcement is also present. The inclusion of aversive procedures, not the proportion, is the methodological signature. The convergent welfare evidence in Section 3 (and consolidated in Table 2) applies to any methodology that includes aversive procedures, including those that combine them with positive reinforcement (the Vieira de Castro et al. 2020 "mixed" group showed welfare cost as well, not only the aversive-only group).

"LIMA" (Least Intrusive, Minimally Aversive) invoked to defend aversive use. LIMA in its original formulation establishes a hierarchy in which less intrusive interventions must be demonstrably exhausted before more intrusive interventions are considered, and in which any aversive intervention requires documented justification by a qualified practitioner. As deployed in much of the balanced training community, LIMA is invoked at the level of label rather than at the level of practice, with no documented exhaustion of less intrusive methods. The framework, when properly applied, is a constraint against routine aversive use, not a defense of it. A useful response in conversation: "LIMA properly applied requires that you've documented that less intrusive methods don't work for this specific case before you escalate. Routine use of aversive equipment without that documented hierarchy is not LIMA, it's the label of LIMA on a different practice."

"All four quadrants" or "the four quadrants of operant conditioning are equally valid." Behavior science describes four operant procedures (positive reinforcement, negative reinforcement, positive punishment, negative punishment) functionally. This is a descriptive taxonomy of how behavior changes. It is not a normative endorsement of all four procedures for all training contexts. The welfare research has specifically established welfare costs associated with positive punishment and negative reinforcement that are not associated with positive reinforcement. Pointing to the existence of four quadrants in the textbook does not establish that all four should be used routinely on companion dogs in the consumer marketplace. Descriptive science does not equal prescriptive endorsement.

Group 4: Conceptual and Logical Deflections

"Pack leader," "alpha," or "dominance" framing for confrontational handling. The conceptual basis for this framing was Schenkel's 1947 captive-wolf studies, popularized in Mech's 1970 book on wolf ecology. Mech himself has subsequently retracted the alpha-pack-leader model as applied to wild wolves and to domestic dogs, in the peer-reviewed literature (Mech, 1999) and in subsequent public retractions (Mech, 2008). Contemporary canine ethology does not support the use of dominance-based confrontational handling techniques on domestic dogs. The Herron, Shofer, and Reisner (2009) clinical data establish that confrontational handling techniques justified on this basis (alpha rolls, dominance downs, scruff shakes, hit or kick corrections) elicit aggressive responses in a substantial percentage of the dogs on whom they are attempted. The framing has been retracted by the source. The techniques justified by the framing are clinically associated with elicited aggression.

"My dog tells me when to use it" or "the dog asks for the correction." Dogs cannot consent to aversive procedures and cannot communicate informed preference for them. Behavioral indicators sometimes interpreted as the dog asking for, requesting, or anticipating the correction are commonly displacement behaviors, conflict behaviors, or stress signals. Reading these as consent reverses the inferential direction. The behaviors are evidence of the welfare cost, not evidence of the dog's endorsement of the procedure that produced them.

"It worked for my client" or "I have hundreds of success stories." Anecdotal individual cases do not establish welfare neutrality, training equivalence, or absence of long-term cost. The convergent welfare research summarized in Section 3 already demonstrates that visible behavioral suppression and welfare cost can coexist. A dog whose problem behavior has been suppressed by an aversive intervention is not, by that fact alone, a dog whose welfare has been preserved. The visible outcome (the suppressed behavior) and the unmeasured outcome (the conditioned emotional response, the threat circuit engagement, the affective state, the relationship cost) are not the same thing. Population-level controlled and observational research is the appropriate evidence base for welfare claims, not individual anecdotes.

"Positive reinforcement doesn't work for serious cases." This claim is empirically falsified by the clinical practice of board-certified veterinary behaviorists. ACVB Diplomates treat the most severe canine aggression, anxiety, fear, and predatory cases in companion animal medicine, and they do so without aversive equipment, using behavioral assessment, environmental management, reward-based behavior modification, and psychiatric medication when clinically indicated. The clinical specialty that handles the hardest cases does not use aversive equipment as standard of care. The ACVB position has been formalized in their December 2025 letter to the AVMA (Section 7.1; ACVB, 2025). The necessity claim has been empirically tested in the highest-acuity clinical setting and has failed.

The categorization above is not exhaustive, but it covers the principal terminology moves a practitioner is likely to encounter. The general principle behind the corrective response is consistent across all of these examples. Behavior science defines procedures by their function, not by their label. Aversive equipment and confrontational handling techniques are defined as aversive by the contingencies under which they operate, not by the words a trainer uses to describe them. The welfare evidence applies to the function, not the label. A practitioner who keeps the functional definition in view can decode any new terminology variant a balanced trainer introduces, by asking the same diagnostic question: does this procedure depend on the dog working to avoid, escape, or terminate a stimulus, or does it not? If yes, the operant procedure is aversive, and the welfare research applies. If no, no aversive procedure is occurring, and there is nothing to defend.

9.13 Studies Frequently Cited by Proponents and How to Address Them

Beyond the rhetorical and terminological objections catalogued above, balanced trainers and aversive equipment proponents frequently cite specific studies in support of their position. A practitioner who engages in client conversation, professional debate, online exchange, or written correspondence with a proponent will encounter these citations. The subsection that follows identifies several of the most commonly invoked studies, summarizes the proponent reading, summarizes what the studies actually establish on careful reading, and provides a practitioner-deployable response. The list is illustrative rather than exhaustive, and practitioners may encounter additional studies in their specific contexts. Across the studies catalogued below, a consistent pattern emerges. The proponent reading typically depends on one of three moves: selective extraction of a single finding from a larger study whose overall conclusions point in a different direction; extension of a narrow methodological finding into broad claims the study does not support; or citation of older work that predates contemporary canine welfare science methodology.

Salgirli, Schalke, Boehm, and Hackbarth (2012). This study compared three training methods (electronic collar, pinch collar, and a conditioned quitting signal) for distraction-based training in a sample of 42 Belgian Malinois police dogs in Germany, with salivary cortisol and behavioral measures recorded across three training sessions. Proponents commonly cite this study to argue that electronic collar use produced less stress than pinch collar use under the trained-handler conditions of the study, and to position electronic collars or pinch collars as welfare-acceptable. What the study actually shows is more limited. The comparison was among three aversive interventions, conducted with experienced police dog handlers in a working-dog protocol, with the explicit acknowledgement that handler proficiency was not equally available across the three conditions. The study does not establish that electronic or pinch collars are welfare-neutral. It does not compare any of the aversive modalities to reward-based training. The senior author Esther Schalke has been an outspoken critic of electronic collar use in companion dog contexts, and her broader research program (including Schalke et al., 2007) supports the welfare case against aversive equipment. The response is to acknowledge the study but reframe its scope: a comparison among aversive modalities in working-dog contexts cannot establish welfare neutrality of any of them. The relevant comparison for companion dog training and consumer policy is aversive versus non-aversive, not e-collar versus pinch collar, and that comparison is addressed by the convergent welfare evidence summarized in Table 2.

Steiss, Schaffer, Ahmad, and Voith (2007). This study evaluated plasma cortisol levels and behavior in 24 kennel dogs (8 per group) wearing electronic bark collars, citronella spray bark collars, or inactivated control collars across a structured exposure protocol. Proponents commonly cite the study to argue that bark collars do not significantly elevate cortisol above baseline and are therefore welfare-neutral. What the study actually shows is that under the specific test conditions (kennel dogs, brief structured exposure to an unfamiliar dog as the bark stimulus, plasma cortisol sampled at scheduled intervals), cortisol elevation was modest. As discussed in Section 3.3 of this paper, cortisol is a single physiological measure of stress and not a comprehensive welfare assessment. Behavioral stress markers and cortisol do not always converge. The Cooper et al. (2014) study, despite a non-significant cortisol finding in its main study phase, found significant behavioral stress indicators in the electronic collar group. The Steiss study cannot support the claim that bark collars are welfare-neutral; it can only support the claim that under specific test conditions, the cortisol elevation was modest. The response is that absence of cortisol evidence is not evidence of absence of welfare cost, and that the welfare assessment of any aversive device requires the full set of behavioral, affective, and physiological measures, not cortisol alone.

Schalke, Stichnoth, Ott, and Jones-Baade (2007). This study examined dogs trained with electronic collars under three conditions varying in predictability and contingency of stimulus delivery. Proponents sometimes cite the study selectively for the finding that dogs receiving stimulation under conditions where the stimulus was contingent on a clearly identifiable target behavior (the rabbit-chasing condition) showed less stress than dogs in conditions where the contingency was less clear. The proponent reading is that controllable, predictable e-collar use is therefore welfare-acceptable. What the study actually shows is that all three groups experienced welfare cost, with the lower-contingency conditions showing the largest effects. The study's overall conclusion was not that e-collar use is welfare-acceptable when contingencies are clear; the conclusion was that the variability of welfare risk across use conditions, combined with the difficulty of guaranteeing optimal contingency conditions in real-world use, argues against approving the equipment for general use. The senior author Esther Schalke has been an outspoken critic of electronic collar use. As established in Section 4.3 and corroborated by the Pessoa personal communication (2026), controllability and predictability attenuate but do not eliminate threat-circuit engagement; they do not render aversive stimulation welfare-neutral. The response is that the study supports the welfare case, not the proponent case, and that selective citation of one sub-finding misrepresents the authors' overall conclusions.

Christiansen, Bakken, and Braastad (2001). This study examined remote shock collar use to prevent sheep predation in a sample of 114 hunting dogs (Norwegian elkhounds, English setters, and hare hunting dogs) tested across two consecutive years in Norwegian sheep pasture confrontation tests. Proponents sometimes cite this study to argue that shock collars are necessary and effective in predation prevention contexts, and that the welfare cost is justified by the alternative of livestock loss or the dog being shot by a farmer. What the study actually shows, on careful reading, is that the protocol involved structured exposure conditions in fenced enclosures, hunting-dog populations, and a sheep-predation-specific context, not companion dog training. The authors themselves recommend that the device be used only for the specific purpose of preventing livestock attack, in connection with positive reinforcement, and the Norwegian Council on Animal Ethics on which the study is partly based recommends a ban on public use of electronic devices outside that specific predation-prevention context. The Masson et al. (2018b) survey establishes that in actual real-world use in countries without regulation, electronic collars are predominantly used by lay guardians without professional supervision for routine companion-dog behavioral problems unrelated to livestock predation. The response is that even granting the Christiansen finding for its specific context, the finding does not transfer to consumer marketplace conditions. The relevant policy question is whether aversive electronic equipment should be available to lay guardians for routine training, not whether structured predation-prevention protocols can reduce sheep attacks under specialized professional handling.

Tortora (1983). This older study, published in the Journal of Experimental Psychology: General, used what the author called "safety training," a multi-stage avoidance-learning protocol involving electronic collar negative reinforcement combined with extensive positive reinforcement, to address what the author defined as avoidance-motivated aggression in 36 dogs. Proponents sometimes cite this study as establishing that electronic collar training resolves aggression and improves welfare. What the study actually shows is more nuanced and more limited than the proponent reading. The protocol was not simple aversive conditioning; it was a complex nine-stage process that began with positive reinforcement training of obedience behaviors, used both play and choke collars (no electronic stimulus) in early stages, introduced a conditioned safety signal as a negative reinforcer, and reserved electronic stimulation for later stages of the protocol. Tortora's own Experiment 3 within the same paper showed that simple aversive use of the electronic collar (full-intensity signaled shock used to punish aggression directly) produced only slight decreases in aggression, undermining the proponent reading that the study supports straightforward aversive conditioning of aggressive behavior. The study is also from 1983, predating contemporary canine welfare science methodology by decades, with no behavioral welfare measures, no physiological measures, no follow-up assessment of conditioned emotional responses, no comparison to modern reward-based aggression behavior modification, and methodological standards that would not meet contemporary peer-review expectations. The response is that the convergent welfare evidence base from 2004 forward, using contemporary welfare science methodology, supersedes a methodologically thin 1983 design. Citing Tortora as evidence against current force-free aggression behavior modification is asking practitioners to weigh a single 1983 multi-stage protocol study without modern welfare measures against forty years of subsequent peer-reviewed convergent welfare research.

Johnson and Wynne (2024). This recent study compared training methods for stopping predatory chasing behavior in dogs and is increasingly cited by proponents as decisive evidence that electronic collars are necessary or uniquely effective for the predatory chasing intervention. The full treatment of this study is in Section 5.3 of this paper. The summary position is that the study establishes narrow efficacy under specific protocol conditions, not necessity, and that its protocol design has been directly challenged in the peer-reviewed literature by Bastos, Warren, and Krupenye (2025) and in the present author's separately published methodological critique (Bangura, 2025, SSRN). The proponent reading treats the study as decisive. The accurate reading treats it as one experimental finding under specific conditions, contested in peer review, that does not establish necessity over reward-based alternatives applied with comparable intensity and follow-through. Practitioners encountering Johnson and Wynne in conversation can refer interlocutors to Section 5.3 of this paper for the full treatment, and to Bastos et al. (2025) as the peer-reviewed methodological critique.

Lindsay's Handbook of Applied Dog Behavior and Training (2005, Volume 3). Proponents sometimes invoke Steven Lindsay's three-volume textbook, particularly Volume 3 (Procedures and Protocols), in which Lindsay defends electronic collar use and characterizes low-level electronic stimulation as a "pulsing tingling or tickling sensation" rather than a noxious event. What this source actually is: a textbook chapter expressing the author's interpretation of training methodology, not a peer-reviewed welfare research finding. Lindsay's characterization of low-level electronic stimulation has not been substantiated by peer-reviewed nociception research. As established in Section 4.1 of the present paper, electrical stimulation strong enough to drive avoidance learning is, by functional definition, crossing the nociceptive threshold; the characterization of the same stimulation as a non-aversive sensation cannot be reconciled with the operant requirement that the stimulation function as an aversive event. The response is that a textbook chapter expressing one author's interpretation is not a peer-reviewed welfare finding, and that the convergent welfare research catalogued in Section 3 and Table 2 supersedes textbook interpretation when the two are in conflict.

The pattern across these proponent-cited studies is consistent. Each study, on careful reading, supports a narrower conclusion than the proponent reading attributes to it. In several cases, most notably Schalke (2007), the study's overall conclusions actually support the welfare case against aversive equipment, with the proponent reading depending on selective extraction of a sub-finding contrary to the authors' own conclusion. In other cases, most notably Tortora (1983), the cited source predates contemporary welfare science methodology by decades and does not include the welfare measures that contemporary peer review would require. In all cases, the convergent welfare evidence catalogued in Section 3 and Table 2, drawing on multiple independent methodological lines, populations, countries, and outcome measures, addresses the welfare question more directly and more comprehensively than any single proponent-cited study can. A practitioner who encounters a balanced trainer's citation of any of the above studies can deploy the same general response: acknowledge the study, summarize what it actually establishes within its specific protocol and context, and reframe the conversation toward the broader convergent evidence base and toward the contemporary clinical consensus of veterinary behavior specialists who treat the highest-acuity cases without aversive equipment.

10. Recommended United States Policy

The recommended policy has two complementary components. The first is prohibition of aversive training equipment, addressed in Section 10.1. The prohibition is delivered through legislation regulating sale, import, and use, modeled on the legislative architecture already in place in Wales, Switzerland, Germany, France, and the other jurisdictions cataloged in Section 7.4. The second is the adoption of the Hierarchy of Dog Needs® and Best Force-Free Practices (Michaels, 2022) as the United States national standard of practice for commercial dog training and behavior modification, addressed in Sections 10.2 and 10.3 and enforced through state licensure of trainers and behavior consultants, addressed in Section 10.5. Component one removes aversive equipment from the consumer marketplace. Component two ensures that commercial training and behavior modification services delivered to the public are conducted using non-aversive methods, including the regulation of confrontational handling techniques applied without equipment. Both components are necessary. Equipment prohibition without standards of practice leaves confrontational handling techniques unaddressed. Standards of practice without equipment prohibition leaves the equipment available on the consumer market. Together, the two components accomplish what the international veterinary, professional, and welfare consensus has called for: the comprehensive removal of aversive control as the basis of canine training and behavior modification in the United States.

10.1 Equipment Prohibition: Scope

A United States policy should prohibit the sale, import, and use of aversive training equipment for dogs. This includes, without limitation, electronic collars of all types (remote-controlled, bark-activated, and electronic containment), prong or pinch collars of all designs, choke chains, choke collars, and slip collars designed to constrict the neck under load, and spray collars of all types (including citronella, scentless air, and ultrasonic-or-audible-tone variants, whether remote-controlled or bark-activated). The prohibition should cover equipment used for training, behavior modification, punishment, negative reinforcement, containment, barking suppression, and behavior interruption. Properly fitted flat collars, body harnesses (front-clip and back-clip), head collars (such as the Halti and Gentle Leader), and limited-slip martingale collars used at non-tightening positions are not aversive equipment and are not within the scope of this prohibition.

10.2 Adoption of the Hierarchy of Dog Needs® as the National Standard of Practice

The second component of the recommended policy is adoption, as the United States national standard of practice for dog training and behavior modification, of the Hierarchy of Dog Needs® and Best Force-Free Practices, developed by Linda Michaels, M.A., Psychology, and set forth in The Do No Harm Dog Training and Behavior Handbook (Michaels, 2022). The Hierarchy of Dog Needs® (HDN) is a comprehensive welfare and training framework, embedded with an ethical code, that establishes the "First, Do No Harm" principle as the governing standard for canine training and behavior modification.

The HDN organizes care across five tiers of needs, addressed in sequence: biological needs (nutrition, water, exercise, shelter, sleep, gentle grooming, gentle veterinary care, freedom from pain); emotional needs (physical and emotional safety, trust, love, benevolent leadership, secure attachment); social needs (two-way non-threatening interaction with humans and conspecifics, age-appropriate socialization); cognitive needs (choice, problem-solving, novelty, opportunity for normal dog behaviors); and force-free training needs, addressed only after the preceding tiers have been assessed and supported.

The HDN's Best Force-Free Practices, set forth in Chapter 5 of the Handbook, comprise the eight permitted methods under this Standard: management, antecedent modification, positive reinforcement, differential reinforcement (including DRI, DRA, DRO, and DRL), classical and counterconditioning, desensitization, the Premack Principle, and social learning (Bandura, 1965, 1977, 1986). Per Michaels, these methods may be used in any order or combination ("Choose Any or All Methods"); they are non-hierarchical among themselves. The HDN specifically excludes positive punishment, negative reinforcement, negative punishment, and engineered extinction. Any tool or method whose mechanism of action depends on these processes, or on the elicitation of fear, pain, intimidation, or frustration, is outside the Standard.

The Hierarchy of Dog Needs® is a standards-of-practice framework, and that distinction matters for understanding why it is the appropriate model for a United States national standard rather than other widely cited frameworks in the field. Standards of practice, learning-theory frameworks, and procedural-priority principles are three different categories of document, and they do different work.

A learning-theory framework catalogs and ranks the full operant landscape because that is what learning theory is. Susan Friedman's Humane Hierarchy is a procedurally rigorous ladder that orders positive reinforcement first, then negative reinforcement, then extinction and negative punishment, then positive punishment (Friedman, 2009). The framework is methodologically sound as a description of behavior analysis, and it serves a useful pedagogical purpose for behavior analysts who need to understand the full operant landscape across applied contexts, including zoo enrichment, laboratory animal welfare, and applied behavior analysis with humans. It is a teaching framework. It is not, and was not designed to be, a standards-of-practice framework for pet dog training.

A procedural-priority principle, such as LIMA (Least Intrusive, Minimally Aversive), orders interventions from least to most intrusive but does not categorically exclude aversive procedures (Lindsay, 2005). LIMA allows aversive procedures at the bottom of the priority hierarchy under an "exhaust less intrusive alternatives first" principle. That is consistent with a procedural ranking framework, but it leaves the door open to procedures whose welfare costs the convergent science of the past two decades has clearly documented. A practitioner working under LIMA can, in principle, justify aversive equipment after exhausting other options. The convergent welfare science, however, does not support that conclusion in any application within commercial canine training and behavior modification.

A standards-of-practice framework operates differently. It is a normative document about what working practitioners should and should not do with the animals in their care, and it draws its lines from the welfare evidence rather than from the methodological inventory. The Hierarchy of Dog Needs® is built as a standards-of-practice framework. Its exclusion of positive punishment, negative reinforcement, negative punishment, and engineered extinction is not a methodological oversight or a simplification of learning theory. It is a deliberate operationalization of the welfare conclusion that those procedures, regardless of their position in any procedural priority hierarchy, are not appropriate in commercial canine training and behavior modification given the convergent welfare evidence catalogued in Sections 3 through 7 of this paper.

This is also the structure of the international veterinary consensus. The 2024 joint position paper of the Federation of Veterinarians of Europe, the Federation of European Companion Animal Veterinary Associations, the Federation of European Equine Veterinary Associations, and the World Small Animal Veterinary Association does not call for positive punishment to be used last (FVE, FECAVA, FEEVA, and WSAVA, 2024). It states that equipment causing pain or discomfort should not be used. That is a standards-of-practice claim, not a procedural-priority claim. It categorically excludes the equipment, which categorically excludes the procedures that depend on the equipment. The Hierarchy of Dog Needs® does the same thing on the procedural side.

The recommendation that the Hierarchy of Dog Needs® and Best Force-Free Practices (Michaels, 2022) be adopted as the United States national standard of practice is therefore not a claim that other force-free frameworks are deficient. Friedman's Humane Hierarchy is doing what a learning-theory framework is supposed to do. LIMA is doing what a procedural-priority principle is supposed to do. The Hierarchy of Dog Needs® is doing what a standards-of-practice framework is supposed to do, which is what a national standard of practice for commercial canine training and behavior modification specifically needs to be. The difference is one of category, not of quality, and the policy purpose at hand is one that calls for a standards-of-practice framework.

The HDN is aligned with the position statements of the American Veterinary Society of Animal Behavior, the American College of Veterinary Behaviorists, the American Animal Hospital Association, the Federation of Veterinarians of Europe and the World Small Animal Veterinary Association joint position paper, the British Veterinary Association, the British Small Animal Veterinary Association, the Australian Veterinary Association, the Canadian Veterinary Medical Association, and the New Zealand Veterinary Association, and with the standards of practice adopted by the International Association of Animal Behavior Consultants, the Karen Pryor Academy, the Victoria Stilwell Academy, the Academy for Dog Trainers, the Pet Professional Guild, and APDT International. The international veterinary, welfare, and professional consensus has converged on the methodology the HDN articulates. Adoption of the HDN as the United States national standard of practice operationalizes that consensus through a single citable framework with a clear ethical code and an explicit set of permitted methods.

Equipment prohibition (Section 10.1) addresses one mechanism of aversive practice. Adoption of the HDN addresses the broader category of aversive interventions, including confrontational handling techniques applied without equipment, which are addressed specifically in Section 10.3.

10.3 Confrontational Handling Techniques and Standards of Practice

The recommended policy reaches not only aversive equipment but also aversive techniques applied without equipment. Confrontational handling techniques operate by positive punishment or negative reinforcement, both of which are excluded under the Hierarchy of Dog Needs® standard adopted in Section 10.2. Confrontational handling techniques include, without limitation, the alpha roll (forcibly rolling a dog onto its back and pinning it), the dominance down (forcing a dog into a prolonged down position as punishment), the scruff shake (grabbing the loose skin at the nape of the dog's neck and shaking the dog), hanging or helicoptering (lifting a dog off the ground by a leash or collar), hitting or kicking the dog, forced retrieves involving ear pinches or toe pinches, finger jabs to the neck or ribs, and intentional yelling at the dog as an aversive intervention. These techniques operate by the same mechanism as aversive equipment. They engage nociception and threat circuitry. They produce avoidance learning rather than resolution of the underlying behavior. The clinical evidence summarized in Section 6.2 (the Herron findings) documents that confrontational handling techniques produce aggressive responses in a substantial percentage of dogs on whom they are attempted, including dogs who present without prior aggression history.

Aversive techniques applied without equipment cannot be reached through equipment prohibition, because no equipment is involved. They are reached, under the framework recommended by this paper, through the second component of the policy: state licensure of commercial dog training and behavior modification, with the Hierarchy of Dog Needs® and Best Force-Free Practices (Michaels, 2022) as the required standard of practice. Under that framework, a licensed trainer or behavior consultant who applies confrontational handling techniques to a client's dog would be in violation of professional standards of practice, with the same disciplinary consequences (license suspension, license revocation, exclusion from the regulated profession) that apply to a veterinarian, mental health counselor, or social worker who practices outside the standards of their profession.

The boundary between confrontational handling as a deliberate aversive intervention and ordinary verbal communication or supervisory contact with a dog is the same boundary that veterinary medicine and human behavioral health already manage. A frustrated tone of voice in a stressful moment is not aversive practice. A startled exclamation when a dog runs into the street is not aversive practice. A physical restraint applied during a medical emergency to prevent a dog from injuring itself or others is not aversive practice. The standard reaches techniques deliberately employed as aversive interventions in the course of training or behavior modification, performed in the commercial context where licensure applies. The framework is not novel. It is the standard scope-of-practice mechanism by which welfare-affecting professions are regulated in the United States.

10.4 Rationale

Six considerations support the recommendation.

The first is mechanism. Aversive equipment and aversive methods produce behavior change, when they work at all, through punishment or negative reinforcement, suppressing behavior or reinforcing escape and avoidance. That mechanism is not vocabulary-dependent. It is definitional in behavior analysis.

The second is welfare risk, supported by convergent evidence. Experimental, observational, survey, cognitive bias, attachment, and neurobiological research consistently associates aversive-based training with stress-related behavior, conflict behaviors, suppressed body language, pessimistic affective bias, and conditioned emotional responses. For neck-pressure equipment specifically (prong and choke collars), peer-reviewed veterinary research adds elevated intraocular pressure during ordinary pulling (Pauli et al., 2006), neck pressures in injury-relevant ranges (Carter et al., 2020; Hunter et al., 2019), and, in the peer-reviewed case report of Grohmann et al. (2013), fatal cerebral ischemia following a punitive choke-chain hanging technique.

The third is the failure of the necessity claim. Cooper et al. (2014) and China et al. (2020) do not support the necessity claim for electronic collars. No peer-reviewed controlled study has demonstrated necessity for prong or choke collars. Johnson and Wynne (2024) is a narrow efficacy finding that has been challenged in the peer-reviewed literature for its protocol design (Bastos et al., 2025) and in the present author's separately published methodological critique (Bangura, 2025, SSRN). Herron et al. (2009) identifies confrontational handling as a clinical risk factor for guardian-directed aggression, not a safe training modality.

The fourth is the empirical record on real-world use. Masson et al. (2018b), Blackwell et al. (2012), and Herron et al. (2009) show that most users of aversive equipment and aversive methods operate without professional guidance, and that guardian-reported success is not higher for aversive-based approaches. That is an additional layer of welfare concern, not the foundation of the case. The foundation of the case is that aversive control engages nociception and threat circuitry regardless of who is operating the device, which is why the international veterinary and professional consensus calls for these tools to be removed from use entirely rather than restricted to expert hands.

The fifth is the existence of less intrusive alternatives. The Hierarchy of Dog Needs® and Best Force-Free Practices (Michaels, 2022), endorsed by veterinary behavior organizations and aligned with the international professional consensus, addresses the full range of problems proponents claim as the strongest applications of aversive tools: recall, predatory chasing, reactivity, aggression, and barking. The ethical standard in behavior change intervention, applied widely in human behavior analysis and in veterinary medicine, is the least intrusive effective method. Aversive equipment and aversive methods are not the least intrusive effective option for any presenting problem in canine training or behavior modification, in anyone's hands, in any setting. The international veterinary, professional, and welfare consensus has reached this conclusion on the same evidence presented in this paper. The conclusion is not that aversive equipment is acceptable when applied skillfully and unacceptable otherwise. The conclusion is that aversive equipment is not the appropriate tool, regardless of who is holding it, because less intrusive effective alternatives exist.

The sixth is international veterinary consensus, which has already reached the policy conclusion. The June 2024 joint position paper unanimously adopted by the Federation of Veterinarians of Europe, the Federation of European Companion Animal Veterinary Associations, the Federation of European Equine Veterinary Associations, and the World Small Animal Veterinary Association formally calls for a complete ban on the sale and use of electric pulse training devices including electric shock collars, and broadly states that equipment causing pain or discomfort should not be used (FVE, FECAVA, FEEVA, and WSAVA, 2024). The United States can either align with this international veterinary consensus or remain an outlier. The evidence does not support the outlier position. The Hierarchy of Dog Needs® (Michaels, 2022) provides the citable, authored framework through which the United States can operationalize that international consensus as a national standard of practice.

10.5 Supporting Infrastructure

A United States ban should be paired with public education, professional standards for trainers, and accessible humane training alternatives. Legislation should be accompanied by funding or incentives for reward-based public education, veterinary referral pathways, accessible behavior consultation resources, and support for pet guardians dealing with serious behavior problems such as aggression, predatory chasing, separation-related issues, and reactivity. A prohibition without supporting infrastructure is less effective than a prohibition embedded in a broader public policy of humane companion animal care.

Trainer regulation is not within the traditional scope of federal companion animal legislation in the United States, but the states and professional bodies can support this infrastructure with certification requirements aligned with the force-free standards of practice, continuing education standards, and public information resources. The absence of dog trainer regulation has been identified, including by Todd (2018), as a structural barrier to widespread adoption of humane methods. Any serious welfare policy benefits from closing that gap.

States should adopt licensure requirements for any person who provides dog training or behavior modification services for pet dogs on a commercial, fee-for-service basis. Licensure should require, at minimum, demonstrated competency through certification by an independent credentialing organization whose standards align with the Hierarchy of Dog Needs® and Best Force-Free Practices (Michaels, 2022) described in Section 10.2; documented continuing education in evidence-based and science-based behavior practice, with periodic renewal requirements; adherence to a published professional code of conduct; and accountability through a state regulatory board with authority to investigate complaints and impose disciplinary action, including license suspension or revocation. The current US baseline, under which anyone can advertise as a dog trainer or behavior consultant with no education, no certification, no examination, no continuing education, and no professional accountability, is not consistent with the welfare interests of dogs or with the consumer-protection interests of the public. Comparable licensure requirements already exist for veterinarians, veterinary technicians, mental health counselors, social workers, and many other professions whose practice carries welfare or public-safety implications. Dog training and behavior modification, which directly affect canine welfare, the human-animal bond, and the public safety profile of dogs in the community, warrant the same regulatory seriousness.

The licensure architecture this section recommends is already under active legislative consideration in multiple United States jurisdictions, as documented in Section 7.4. New York Assembly Bill A 6985 and Senate Bill S 7723 of the 2025-2026 session would require licensing and educational standards for canine trainers under non-aversive, evidence-based, positive reinforcement principles. New Jersey Assembly Bill A 4206 (Dog Trainer Licensing Act) and Assembly Bill A 4207 (Dog Training Licensure Act), both introduced 19 February 2026, would establish state licensure boards and tie licensure standards to professional codes of ethics that incorporate the Least Intrusive, Minimally Aversive Effective Behavior Intervention Policy. New Jersey's earlier Senate Bill S 3814 of 2024 would have established a Dog Training Licensing Board with an evidence-based humane training code precluding aversive methods. The pattern across these proposals reflects an emerging United States legislative interest in implementing the licensure framework recommended here, even though no comprehensive licensure statute had been enacted at the United States state level as of the date of this paper.

11. Conclusion

Aversive training equipment and aversive training methods change behavior. That is not the disputed point. The questions that actually matter scientifically are how they change behavior, what welfare risks come with that mechanism, whether they are necessary, and whether public access to such devices and methods is justified when less intrusive alternatives exist. This paper has answered each of those questions using convergent evidence drawn from multiple independent scientific disciplines, peer-reviewed veterinary research on the physical effects of neck-pressure equipment, clinical behavior medicine, veterinary welfare consensus, and established regulatory practice.

When aversive equipment or an aversive handling technique reduces a behavior, the equipment or technique is functioning as positive punishment. When a behavior increases because the dog can escape, avoid, delay, or prevent the aversive event, the equipment or technique is functioning as negative reinforcement. In both cases, the stimulus has to function as an aversive event for the dog. That requirement is not eliminated by a softer vocabulary, by a lower intensity setting, by a trainer self-testing the device on their own arm, by a fifteen-second video of an engaged dog, by an appeal to controllability, or by an appeal to discredited dominance theory. None of those moves changes the underlying mechanism. The mechanism is what it is, and what follows from the mechanism follows. This is not a matter of preference or ideology. It follows from the definitions of the procedures by which these tools and methods operate.

The evidence does not support the claim that aversive training equipment is necessary for effective dog training or behavior modification. The welfare case against electronic collars does not rest on a claim that they cause tissue damage. The case rests on what the equipment must do to function. Modern electronic collar proponents commonly argue that the stimulation is mild, imperceptible, comparable to a tap on the shoulder, or therapeutically similar to a TENS unit. The nociception framework answers those sensation-severity arguments directly. Nociceptors and threat circuitry respond to stimuli well below the threshold of actual injury. The system exists to warn the organism away from potentially harmful events before damage occurs. A stimulus that is behaviorally effective because the dog works to avoid, escape, or terminate it is, by that fact alone, crossing the nociceptive and threat-system thresholds, regardless of whether the stimulus would ever leave a mark. The argument from absence-of-marks is not an argument about welfare. It is an argument about visibility. For prong and choke collars, a separate evidentiary question does arise, because these tools operate by applying mechanical force to the canine neck. Peer-reviewed veterinary research establishes that such force produces measurable physical effects: elevated intraocular pressure during ordinary pulling (Pauli et al., 2006), neck pressures in injury-relevant ranges (Carter et al., 2020; Hunter et al., 2019), and, in the peer-reviewed case report of Grohmann et al. (2013), fatal cerebral ischemia following a punitive choke-chain hanging technique. The mechanical-injury literature applies specifically to neck-pressure equipment, and it adds a second layer of welfare concern beyond the nociception argument for that category of tools.

Two of the fear-conditioning researchers whose work has been frequently cited by aversive-equipment proponents, Dr. Luiz Pessoa and Dr. David Knight, have each now confirmed in direct correspondence that their published findings on controllability and predictability cannot be used to support the claim that predictable or controllable aversive stimulation is neurologically neutral. The neuroscience of fear conditioning does not vindicate aversive equipment use. It undermines the rhetorical defense that has been built on its back.

The professional consensus is not a matter of opinion. It is documented, convergent, and international. In June 2024, the Federation of Veterinarians of Europe, the Federation of European Companion Animal Veterinary Associations, the Federation of European Equine Veterinary Associations, and the World Small Animal Veterinary Association unanimously adopted a joint position paper formally calling for a complete ban on the sale and use of electric pulse training devices including electric shock collars for dogs, and broadly stating that equipment causing pain or discomfort should not be used. The American Veterinary Society of Animal Behavior, the American College of Veterinary Behaviorists, the American Animal Hospital Association, the British Veterinary Association, the British Small Animal Veterinary Association, the Australian Veterinary Association, the Canadian Veterinary Medical Association, the New Zealand Veterinary Association, and the European Society of Veterinary Clinical Ethology have each independently reached the same position. The largest animal welfare organizations in the English-speaking world have reached it. The leading professional training and behavior organizations, including the International Association of Animal Behavior Consultants, the Karen Pryor Academy, the Victoria Stilwell Academy, the Academy for Dog Trainers, the Pet Professional Guild, and APDT International, have reached it. Multiple jurisdictions have enacted legislative bans, in some cases for more than fifteen years, without producing peer-reviewed evidence of increased public safety risk attributable to the prohibition.

The convergent welfare and consensus arguments are reinforced by what we know about how these devices are actually used in the population and what United States guardians actually believe about them. Real-world use data show that aversive equipment is predominantly purchased and operated by lay guardians without professional supervision, applied after few or no alternative methods have been attempted, and associated with physical wounds in a measurable percentage of cases (Masson et al., 2018b). The devices themselves vary by nearly two orders of magnitude in delivered electrical energy across products marketed for the same use category, with no manufacturer disclosure of pulse parameters and no United States regulatory framework comparable to the medical-device framework applied to therapeutic electrical stimulation units for human use (Lines, van Driel, and Cooper, 2013). Survey data from Petco's October 2020 corporate disclosure indicate that 70 percent of United States dog guardians believe shock collars harm their pet's emotional or mental wellbeing, that 69 percent consider them cruel, and that 59 percent would prefer to shock themselves than their dog (Petco, 2020). The constituency that would be inconvenienced by a sale and use prohibition is a minority of the consumer population, not a majority of it. A prohibition on the sale, import, and use of aversive training equipment is therefore not only an animal welfare measure but a consumer protection measure. It aligns the United States consumer marketplace with the welfare consensus the international veterinary profession has already reached and that, on the available data, a substantial majority of United States guardians already share.

The case for a United States ban on aversive training equipment, and for the adoption of the Hierarchy of Dog Needs® and Best Force-Free Practices (Michaels, 2022) as the standard of practice, is not radical. It is conservative in the technical sense. It follows established scientific evidence, established veterinary welfare consensus, established peer-reviewed veterinary research on the physical effects of neck-pressure equipment, and established international regulatory practice. It aligns the United States with the recommendations of the international veterinary profession. It protects dogs from nociceptive engagement, threat conditioning, conditioned emotional harm, and, in the case of prong and choke collars, the mechanical pressure effects documented in peer-reviewed veterinary research. It protects guardians from the clinically documented risk of guardian-directed aggression associated with confrontational handling. It supports public safety by prioritizing the methods that actually resolve, rather than suppress, problem behavior. And it removes from the consumer market a category of devices whose mechanism is aversive control and whose risks accumulate with every hour of exposure.

The policy conclusion is direct. Aversive training equipment works through aversive control. Aversive training methods work through aversive control. Aversive control comes with welfare risk. For the specific category of equipment that applies mechanical force to the canine neck, peer-reviewed veterinary research adds documented physical effects on top of that welfare risk. Less intrusive alternatives exist and are supported by the international veterinary profession. Real-world users overwhelmingly operate without professional guidance. Multiple jurisdictions have already restricted or banned these devices without producing peer-reviewed evidence of increased public safety risk. The United States should follow the evidence and the international veterinary consensus. The recommended policy has two parts. The first is the prohibition of the sale, import, and use of aversive training equipment, including electronic collars, prong collars, choke chains, and spray collars. The second is the adoption of the Hierarchy of Dog Needs® and Best Force-Free Practices (Michaels, 2022) as the national standard of practice for commercial dog training and behavior modification, enforced through state licensure of trainers and behavior consultants, with confrontational handling techniques excluded from licensed practice. Pending United States legislation in the 2024 to 2026 period in New York, New Jersey, Massachusetts, and Rhode Island, documented in Section 7.4, indicates that aversive-equipment regulation and trainer licensure are already under active legislative consideration in multiple states. Both parts are necessary. Together, they are a measured, evidence-based, and conservative response to convergent scientific evidence, to international veterinary consensus, and to the welfare risks the peer-reviewed literature has already documented.

Closing Note for the Practitioner Community

This paper is written first for the practitioner community: the trainers, behavior consultants, certified applied animal behaviorists, and veterinary behaviorists who carry the responsibility of explaining, defending, and operationalizing the science of canine welfare in their daily work. Legislative and broader policy use may follow. The immediate purpose, though, is to put a comprehensive, citable, science-based reference in the hands of the practitioners who are already doing the work of moving the field toward the Hierarchy of Dog Needs® and Best Force-Free Practices (Michaels, 2022) as the standard of practice and away from aversive training equipment and confrontational handling techniques.

The deployment contexts for this paper are concrete and immediate. In a client conversation, the paper supports the practitioner who needs to explain to a guardian why a training colleague's recommendation of an electronic or prong collar should be declined. Sections 2, 3.6, 4.1, and 9.12 provide ready language. In online debate and social media exchange, the paper supports the practitioner who encounters balanced trainer rhetoric in real time. Sections 9 and 9.12 catalog the most common arguments and terminology moves and pair them with direct rebuttals. In continuing education and professional speaking, the paper provides a structured, citable scaffold suitable for course development, conference presentation, professional webinars, and guest lectures, with Tables 1 and 2 functioning as ready visual aids. In professional advocacy, whether that means correspondence with veterinary practices, shelters, training schools, training certification organizations, or animal welfare organizations, the paper consolidates the convergent scientific evidence, the institutional consensus, the comparative international jurisdictional record, and the policy logic in a single citable document. In mentoring relationships with junior practitioners and force-free credential candidates, the paper offers an in-depth treatment of the operant, neuroscientific, and clinical foundations of the force-free position, suitable for study and discussion in mentoring conversations, peer study groups, and continuing education cohorts.

Practitioners are welcome to use, cite, share, and build on the material in this paper. Cite it in client handouts, in educational content, in social media posts and blog posts, in podcast episodes, in video scripts, in continuing education materials. Quote passages with attribution. Adapt the structured arguments for new contexts, including legislative testimony, public comment letters, professional position statements, and academic teaching materials. The paper exists to be used. The purpose is not to sit on a server. The purpose is to circulate through the practitioner community and into the professional and public conversation about canine welfare, training methodology, and consumer protection policy.

A final note about the role of the practitioner community in the welfare argument itself. The convergent welfare evidence catalogued in Section 3 and Table 2 has existed in the peer-reviewed literature for years, in some cases for two decades. The international veterinary consensus catalogued in Section 7 has been in place since the 2024 joint position paper of the Federation of Veterinarians of Europe, the Federation of European Companion Animal Veterinary Associations, the Federation of European Equine Veterinary Associations, and the World Small Animal Veterinary Association. The mechanical-injury literature on neck-pressure equipment has been published since 2006. None of this material is new. What is needed at this point is not more evidence. What is needed is for the practitioner community to translate the existing evidence into client conversations, into professional advocacy, into public education, and into policy pressure. Trainers, behavior consultants, certified applied animal behaviorists, and veterinary behaviorists are the professional bridge between the peer-reviewed welfare science and the dog-owning public. The dogs who will benefit from a future shift away from aversive training equipment will not benefit from the welfare research alone. They will benefit from the practitioners who carry that research into the daily contexts where dogs are actually trained, where behavior is actually modified, and where equipment is actually purchased and used. This paper exists to support that work.

Glossary

This glossary covers technical vocabulary used in the body of the paper that goes beyond the standard learning theory and behavior modification terminology with which the practitioner audience is presumed to be familiar. The basics of operant conditioning (positive and negative reinforcement, positive and negative punishment), the classical conditioning paradigm, and the standard force-free training procedures (counter-conditioning, desensitization, differential reinforcement) are not redefined here. The terms below are organized by domain: neuroscience and threat-circuit terminology, stress physiology, nociception and pain science, anatomy specific to neck-pressure injury, welfare science and assessment methodology, specialized learning theory beyond the basics, electronic device and electrical terminology, and regulatory and professional terms. Within each category, terms are listed alphabetically.

Neuroscience and threat-circuit terminology

Amygdala. A subcortical structure in the temporal lobe of the mammalian brain, central to processing threat-related stimuli, fear conditioning, and the assignment of negative affective valence to events. Activation of the basolateral and central amygdala is one of the most replicable findings in the fear-conditioning neuroscience literature on which Section 4.3 of this paper draws.

Anterior cingulate cortex. A cortical region involved in pain processing, conflict monitoring, and the integration of affective and cognitive information. Part of the broader threat-processing network that responds to aversive stimuli regardless of predictability or controllability.

Conditioned emotional response (CER). A learned emotional reaction, typically fear or anxiety, elicited by a previously neutral stimulus that has been paired with an aversive event. CERs are central to understanding why aversive training produces emotional sequelae beyond the immediate behavior change. A dog corrected with a prong collar in the presence of other dogs may develop a CER linking other dogs with aversive stimulation. Treated in detail in Section 3.4.

Conditioned freezing. A defensive immobility response to a stimulus that has been paired with an aversive unconditioned stimulus. One of the standard behavioral indicators in fear-conditioning research, used as a measure of threat-circuit activation in mammalian models.

Controllability. In aversive learning research, the technical property of an aversive stimulus referring to whether the subject can perform a behavior that terminates or prevents the stimulus. Pessoa, Knight, and others have shown that controllability attenuates threat-related neural responses without eliminating them. A dog who can turn off a shock by performing a behavior is in a more controllable condition than a dog who cannot, but both dogs are still in an aversive condition. Discussed in Section 4.3.

Defensive circuitry (also called survival circuits). Joseph LeDoux's preferred terminology for the neural networks (centered on the amygdala but extending through hypothalamus, periaqueductal gray, and locus coeruleus) that process threat-relevant stimuli and produce species-typical defensive behaviors. LeDoux distinguishes defensive-circuit activation from the conscious experience of fear, but does not characterize circuit activation as welfare-neutral.

Extinction learning. In Pavlovian terms, the process by which a conditioned response weakens when the conditioned stimulus is repeatedly presented without the unconditioned stimulus. In contemporary threat-circuit research, extinction is understood as new inhibitory learning rather than erasure of the original association. The original threat-related learning remains and can be reinstated. Relevant to claims that aversive training can be "faded out" once the dog has learned the association.

Fear conditioning (the experimental paradigm). A research protocol in which an organism learns to associate a neutral stimulus with an aversive unconditioned stimulus, typically electric shock in animal research. The paradigm has produced decades of neuroscience documenting the neural substrates of threat learning. References to fear conditioning in this paper are to the experimental paradigm and the science derived from it, not to colloquial usage.

Hippocampus (and hippocampal involvement). A medial temporal lobe structure central to spatial and contextual memory. In threat-conditioning research, hippocampal involvement is associated with contextual fear, where the threat is associated with a place or situation rather than a discrete cue. Relevant to predictions about generalization of fear from a specific aversive event to broader environmental contexts.

Hypothalamic-pituitary-adrenal (HPA) axis. The neuroendocrine system that produces the physiological stress response, releasing cortisol and other glucocorticoids in response to perceived threats. Repeated activation of the HPA axis over time produces documented changes in brain structure and function. See "chronic stress."

Learned helplessness. A behavioral and neural state, originally identified by Maier and Seligman, in which an organism exposed to inescapable aversive stimulation later fails to escape or avoid the same stimulus when escape becomes possible. The original work used uncontrollable shock. Subsequent research, particularly Maier and Watkins (2005), has refined the framework but reinforced the conclusion that lack of control over aversive events produces lasting behavioral and neural changes.

Locus coeruleus. A brainstem nucleus that is the primary source of noradrenergic projections in the mammalian brain. Activated by threat-related stimuli, contributing to arousal, vigilance, and the autonomic component of the stress response.

Periaqueductal gray (PAG). A midbrain region central to producing defensive behaviors (freezing, flight) and to descending pain modulation. Part of the canonical fear-conditioning circuit. PAG activation is a robust marker of threat-related neural processing.

Predictability. In aversive learning research, the property of an aversive stimulus referring to whether it is reliably preceded by a warning signal. Predictability attenuates some threat-related neural responses but, like controllability, does not eliminate them. The Pessoa and Knight personal communications referenced in Sections 4.3 and 9.4 specifically clarify that attenuated does not equal absent.

Prefrontal cortex. A set of cortical regions involved in executive function, decision-making, and top-down regulation of subcortical structures including the amygdala. Chronic stress produces documented atrophy of prefrontal dendrites and impairment of prefrontal function, which feeds back into reduced regulation of threat responses.

Safety signal. In aversive learning research, a stimulus that reliably predicts the absence of an aversive event. Safety signals are studied in the broader context of how predictability and controllability shape welfare during aversive procedures. Relevant to the proponent argument that structured, predictable aversive training is welfare-neutral, an argument the predictability and controllability literature does not support.

Threat circuit. A general term for the neural networks (centered on the amygdala, with contributions from hypothalamus, periaqueductal gray, locus coeruleus, and prefrontal cortex) that process aversive and threat-related stimuli and produce defensive responses. Contemporary fear-conditioning neuroscience characterizes aversive training, regardless of method or modality, as engaging this circuitry. Section 4.1 develops the implications.

Stress physiology

Allostatic load. The cumulative biological cost of chronic activation of the stress response systems, with documented effects on cardiovascular function, immune function, brain structure, and metabolic regulation. Relevant to the cumulative-exposure argument in Section 3.5: a single aversive correction may produce a modest stress response, but repeated exposure across months or years produces cumulative biological cost.

Catecholamines. Norepinephrine and epinephrine. Hormones and neurotransmitters released during the acute stress response, producing increased heart rate, blood pressure, and vigilance.

Chronic stress. Sustained activation of the stress response over extended time, producing documented changes in HPA axis regulation, hippocampal volume, prefrontal function, and amygdala reactivity. Distinct from the everyday colloquial use of "stress." References to chronic stress in this paper are to the technical research literature.

Cortisol. A glucocorticoid hormone released by the adrenal cortex as part of the HPA axis response. Often measured in welfare research as a proxy for stress, but with significant limitations. Cortisol can be suppressed, lagged, or buffered by context, and a non-significant cortisol finding does not establish welfare neutrality. Discussed in Section 3.3.

Glucocorticoids. A class of steroid hormones, of which cortisol is the primary one in dogs and humans. Produced by the adrenal cortex in response to HPA axis activation. Long-term elevation produces documented effects on brain, immune, and metabolic function.

Heart rate variability (HRV). The variability in time intervals between heartbeats, regulated by the autonomic nervous system. Reduced HRV is associated with sympathetic dominance and chronic stress states. Used as a non-invasive welfare marker in some canine research.

Stress reactivity. The magnitude and duration of the physiological response to a stressor. Chronic stress produces documented changes in stress reactivity, generally in the direction of heightened response to subsequent stressors.

Sympathetic nervous system (and autonomic arousal). The branch of the autonomic nervous system that produces the fight-or-flight response, including increased heart rate, blood pressure, and vigilance. Activated rapidly by threat-related stimuli, including aversive training events.

Nociception and pain science

A-delta fiber. A type of myelinated nerve fiber that conducts sharp, fast pain signals from the periphery to the spinal cord. Activated by the high-intensity stimuli used in aversive training equipment, particularly mechanical-pressure equipment.

C-fiber. A type of unmyelinated nerve fiber that conducts slow, dull, burning pain signals. Often activated alongside A-delta fibers during sustained or intense nociceptive stimulation.

Gate-control theory of pain. The Melzack and Wall theoretical framework explaining how non-painful sensory input can modulate the transmission of pain signals at the spinal cord level. Underlies the mechanism by which therapeutic transcutaneous electrical nerve stimulation (TENS) units produce analgesia. Critically, gate control operates only at sub-nociceptive intensities. Stimulation strong enough to drive avoidance learning is, by functional definition, above this range. Section 9.7 develops the comparison between TENS units and electronic dog training collars.

Mechanonociceptor. A nociceptor that responds preferentially to mechanical stimulation (pressure, pinching, stretching). Activated by the concentrated point pressure of prong collars and by the circumferential compression of choke chains.

Nociception. The neural process of detecting and transmitting potentially damaging stimuli. Distinct from pain, which is the conscious experience that may or may not result from nociceptive input. Aversive training equipment, by operating at intensities sufficient to drive avoidance learning, engages nociception by definition. Section 4.1 develops the distinction.

Nociceptive threshold. The minimum stimulus intensity that activates nociceptors. Stimuli below this threshold are not registered by the nociceptive system. The "working level" of an electronic collar, which by definition is sufficient to drive avoidance learning, is above the nociceptive threshold.

Nociceptor. A specialized sensory neuron that detects potentially damaging stimuli (mechanical, thermal, or chemical). The receptor end of the nociceptive system.

Tissue damage threshold. The stimulus intensity above which physical tissue injury occurs. Distinct from the nociceptive threshold: a stimulus can be nociceptive (engaging the pain system) without producing tissue damage. Aversive equipment proponents sometimes argue that a stimulus is harmless if it produces no visible tissue damage. That argument confuses these two thresholds.

Transcutaneous electrical nerve stimulation (TENS). A therapeutic medical device that delivers low-intensity electrical pulses through skin electrodes for pain management, operating by gate-control mechanisms at sub-nociceptive intensities. TENS units are FDA-regulated medical devices with required disclosure of pulse parameters and adverse event reporting. The asymmetric regulation between TENS units and electronic dog training collars (which deliver substantially higher peak voltages with no comparable regulatory framework) is examined in Sections 8.3 and 9.7.

Anatomy specific to neck-pressure injury

Carotid artery. The major artery supplying blood to the head, located in the neck. Compression of the carotid by neck-pressure equipment can produce reduced cerebral perfusion and acute increases in intraocular pressure.

Cervical spine. The vertebral column of the neck, comprising vertebrae C1 through C7. Vulnerable to compression and shear injury under the forces produced by neck-pressure equipment.

Hyoid apparatus. The set of bones and connective tissue suspending the larynx and pharynx, located in the throat. Vulnerable to compression and dislocation injury under choke chain compression.

Intraocular pressure. Fluid pressure within the eyeball. Documented in peer-reviewed canine research (Pauli et al., 2006) to increase under leash-and-collar pulling forces, with measurable elevations from neck-pressure equipment. Relevant to dogs with glaucoma or other ocular conditions, but elevation is documented in healthy dogs as well.

Jugular vein. The major venous return from the head, located in the neck. Compression of the jugular by neck-pressure equipment produces venous engorgement that contributes to the documented elevation of intraocular pressure.

Larynx (and laryngeal). The voice box, located in the upper neck. Contains the vocal folds and is the entry point of the airway. Vulnerable to compression injury, edema, and paralysis under neck-pressure equipment.

Recurrent laryngeal nerve. A branch of the vagus nerve that innervates the laryngeal muscles. Vulnerable to compression injury from neck-pressure equipment, with potential for laryngeal paralysis and airway compromise.

Thyroid gland. An endocrine gland located in the upper neck near the trachea. Anatomical proximity makes it vulnerable to compression from neck-pressure equipment, with case-report-level documentation of thyroid impact in some clinical literature.

Trachea (and tracheal). The windpipe. The primary airway between the larynx and the bronchi. Vulnerable to compression and structural injury under choke chain and prong collar forces, including documented tracheal collapse.

Welfare science and assessment methodology

Affective state. The internal emotional condition of the animal, comprising both valence (positive vs negative) and arousal. Contemporary welfare science treats affective state as a primary outcome of welfare assessment, alongside physical health.

Behavioral stress markers. The set of validated, observable behaviors used in canine welfare research as indicators of stress, including lip licking, yawning, low body posture, displacement behaviors, conflict behaviors, reduced approach, and increased vigilance. These markers are robust and meaningful as welfare indicators on their own and do not depend on convergent physiological measures. Section 3.3 develops this point.

Cognitive bias paradigm (also called judgment bias test). An experimental method for assessing affective state by measuring how an animal interprets ambiguous stimuli. Animals in negative affective states tend to interpret ambiguous cues pessimistically, anticipating a negative outcome. Animals in positive affective states interpret them optimistically. One of the few research paradigms that can directly assess affective state in non-verbal animals.

Convergent evidence. A research finding that the same conclusion is supported by multiple independent lines of evidence using different methods, populations, and outcome measures. The strongest form of evidence in any empirical field, because it cannot be undermined by methodological critique of any single study. The argument of this paper rests on convergent evidence across canine welfare science, neuroscience, stress physiology, nociception research, and clinical behavior medicine.

Five Domains model. A contemporary framework for welfare assessment, developed by Mellor and colleagues, that evaluates an animal's welfare across five domains: nutrition, environment, health, behavior, and mental state. Has largely replaced the older Five Freedoms framework in scientific welfare assessment.

Welfare indicator. Any measurable variable used to assess an animal's welfare state. Contemporary welfare science emphasizes that welfare assessment requires convergence across multiple indicators (behavioral, physiological, affective) rather than reliance on any single measure.

Specialized learning theory beyond the basics

Compound schedule of reinforcement. A schedule in which two or more contingencies operate simultaneously or in alternation. When food rewards and aversive corrections are both present in a training session, the dog is operating under a compound schedule. The apparent absence of stress in the food-presentation moments does not subtract the welfare cost of the aversive component. Section 4.4 develops this analysis.

Conditioned suppression. A reduction in ongoing operant behavior in the presence of a stimulus that has been paired with an aversive event. A measure used in some research paradigms. Relevant because it shows that aversive associations affect behavior beyond the immediate punished response, suppressing other ongoing behaviors as well.

Functional definition. A definition based on the effect a procedure has on behavior, not on the procedure's label, topography, or descriptive properties. The behavior science definition of "aversive" is functional. A stimulus is aversive if its delivery decreases behavior or its termination reinforces behavior. This is the framework used throughout the paper. Section 2.1 sets it out.

Humane Hierarchy. Susan Friedman's procedural ranking of behavior-change interventions, ordered from least intrusive to most intrusive: distant antecedents, immediate antecedents, positive reinforcement, differential reinforcement of alternative behaviors, negative reinforcement, extinction and negative punishment, and positive punishment (Friedman, 2009). Developed as a teaching framework within applied behavior analysis. The Humane Hierarchy is methodologically rigorous as a description of the operant landscape but, because it includes all four operant quadrants, it does not function as a standards-of-practice framework. Discussed in Section 10.2.

LIMA (Least Intrusive, Minimally Aversive). A procedural-priority principle, coined by Steven Lindsay (2005) in volume 3 of the Handbook of Applied Dog Behavior and Training, that directs trainers and behavior consultants to select the least intrusive, minimally aversive procedure reasonably expected to succeed. LIMA orders interventions by intrusiveness but does not categorically exclude aversive procedures, allowing them at the bottom of the priority hierarchy under an "exhaust less intrusive alternatives first" principle. Adopted in modified form by several professional training organizations including the International Association of Animal Behavior Consultants and the Certification Council for Professional Dog Trainers. Discussed in Section 10.2.

Operant function vs operant procedure. The function of an operant procedure is the contingency that defines it (positive reinforcement, negative reinforcement, positive punishment, negative punishment). The procedure is the specific topographic form (a leash correction, a click-and-treat, a shock, a praise word). The same procedure can serve different functions depending on contingency, and the same function can be served by many different procedures. The distinction matters because welfare arguments apply to function, not to procedure.

Standards-of-practice framework. A normative document that specifies what working professionals should and should not do within their scope of practice, with the lines drawn from welfare evidence rather than from a methodological inventory. Distinct from a learning-theory framework (which catalogs the full operant landscape) and from a procedural-priority principle (which orders interventions but does not categorically exclude any). The Hierarchy of Dog Needs® and Best Force-Free Practices (Michaels, 2022) is the standards-of-practice framework recommended in this paper for the United States. Section 10.2 develops the distinction.

Two-factor (or two-process) theory of avoidance. Mowrer's classic theoretical framework for avoidance learning. The organism first acquires a Pavlovian fear association to the warning signal, then operantly learns to perform a response that terminates the warning signal (negative reinforcement). The framework explains why avoidance behaviors, once learned, are remarkably persistent and resistant to extinction. Directly relevant to electronic collar training, where the dog is in a continuous Pavlovian-plus-operant compound contingency.

Electronic device and electrical terminology

Bark-activated mode. An electronic collar configuration that delivers stimulation automatically when the dog vocalizes. Removed from Petco's retail inventory in October 2020 along with handheld remote-controlled shock collars.

Continuous, momentary, and "nick" stimulation modes. Different temporal profiles of electrical stimulation delivered by electronic training collars. Continuous delivers a sustained stimulus. Momentary delivers a brief pulse. "Nick" is a marketing term for a very brief stimulus. The terms are not standardized across manufacturers and do not reliably predict the actual electrical characteristics of the delivered stimulus.

Electrical impedance. The resistance to electrical current flow through tissue, varying significantly based on coat type, skin moisture, and electrode contact quality. Lines, van Driel, and Cooper (2013) documented that impedance varies substantially across realistic conditions, contributing to large variations in delivered stimulus energy from the same nominal collar setting.

Peak voltage. The maximum instantaneous voltage delivered during a stimulation pulse. One of several electrical characteristics that determine the actual stimulus intensity experienced by the dog. Lines, van Driel, and Cooper (2013) documented that peak voltage varies dramatically across commercially available products at the same nominal setting.

Pulse duration (also called pulse width). The temporal length of an individual electrical pulse delivered by an electronic collar, typically measured in microseconds or milliseconds. Combined with peak voltage and pulse rate, determines the total energy delivered. Not standardized across manufacturers and not disclosed at the point of sale.

Remote-controlled mode. An electronic collar configuration in which the handler manually triggers stimulation. The most common configuration in use among lay guardians per Masson, Nigron, and Gaultier (2018).

Stimulus energy. The total energy delivered in a stimulation event, typically measured in joules or millijoules. The product of voltage, current, and time. Lines, van Driel, and Cooper (2013) documented an 87-fold range of stimulus energy across commercially available models at the same nominal maximum setting, ranging from 3.3 millijoules to 287 millijoules.

Regulatory and professional terms

Accreditation. Independent evaluation and certification of an organization's training programs or credentialing standards. Distinct from individual certification.

Board-certified veterinary behaviorist. A veterinarian who has completed a residency in veterinary behavior and passed the certification examination of the American College of Veterinary Behaviorists or its international equivalents. The highest credential in the canine behavior field.

Certified applied animal behaviorist (CAAB or ACAAB). A non-veterinary credential issued by the Animal Behavior Society in the United States, requiring graduate-level academic training in animal behavior and supervised clinical experience.

Credentialing. The formal evaluation and recognition of an individual's professional qualifications, typically through examination and continuing-education requirements.

Diplomate of the American College of Veterinary Behaviorists (DACVB). The credential held by board-certified veterinary behaviorists in the United States. Comparable credentials exist internationally.

Licensure. Government-issued authorization to practice a profession, with associated standards-of-practice requirements and disciplinary mechanisms. Required for veterinarians, mental health counselors, social workers, and many other welfare-affecting professions. Currently not required for dog trainers or behavior consultants in any United States jurisdiction. The recommended policy in Section 10 includes state licensure of commercial dog training and behavior modification.

Scope of practice. The set of activities that a licensed professional is authorized to perform under their license. Practicing outside scope is a basis for disciplinary action under most state licensing frameworks.

Standards of practice. The body of professional norms that define what constitutes competent practice within a field. The recommended policy in Section 10 includes adoption of the Hierarchy of Dog Needs® and Best Force-Free Practices (Michaels, 2022) as the United States national standard of practice for commercial dog training.

References

Affolter, V. K., and Moore, P. F. (1994). Histologic features of normal canine and feline skin. Clinics in Dermatology, 12(4), 491-497. https://doi.org/10.1016/0738-081X(94)90215-1

American Animal Hospital Association. (2015). 2015 AAHA canine and feline behavior management guidelines. Journal of the American Animal Hospital Association, 51(4), 205-221. https://doi.org/10.5326/JAAHA-MS-6527

American College of Veterinary Behaviorists. (2025). Letter to the American Veterinary Medical Association regarding public comments on shock collars in dog training, December 2, 2025. https://cdn.ymaws.com/www.dacvb.org/resource/resmgr/docs/ACVB_letter_to__AVMA_12-3-25.pdf

American Society for the Prevention of Cruelty to Animals. (n.d.). Position statement on training aids and methods. https://www.aspca.org/about-us/aspca-policy-and-position-statements/position-statement-training-aids-and-methods

American Veterinary Society of Animal Behavior. (2021). AVSAB position statement on humane dog training. American Veterinary Society of Animal Behavior. https://avsab.org/wp-content/uploads/2021/08/AVSAB-Humane-Dog-Training-Position-Statement-2021.pdf

Animal Welfare (Electronic Collars) (Wales) Regulations 2010. (Welsh Statutory Instrument 2010 No. 943 (W. 97)). Made under the Animal Welfare Act 2006, Section 12. https://www.legislation.gov.uk/wsi/2010/943/contents/made

Arhant, C., Bubna-Littitz, H., Bartels, A., Futschik, A., and Troxler, J. (2010). Behaviour of smaller and larger dogs: Effects of training methods, inconsistency of owner behaviour and level of engagement in activities with the dog. Applied Animal Behaviour Science, 123(3-4), 131-142. https://doi.org/10.1016/j.applanim.2010.01.003

Arnsten, A. F. T. (2009). Stress signalling pathways that impair prefrontal cortex structure and function. Nature Reviews Neuroscience, 10(6), 410-422. https://doi.org/10.1038/nrn2648

Association of Pet Dog Trainers New Zealand. (2022). Position statement: Use of electronic stimulation (shock) collars in training dogs. https://www.apdtnz.org.nz/position-statement-use-of-electronic-stimulation-shock-collars-in-training-dogs

Animal Legal and Historical Center. (2022). Table of state dog tether laws. Michigan State University College of Law. https://www.animallaw.info/topic/table-state-dog-tether-laws

Australian Veterinary Association. (2022). Policy: Use of behaviour-modifying collars on dogs. https://www.ava.com.au/policy-advocacy/policies/companion-animals-dog-behaviour/use-of-behaviour-modifying-collars-on-dogs/

Bandura, A. (1965). Influence of models' reinforcement contingencies on the acquisition of imitative responses. Journal of Personality and Social Psychology, 1(6), 589-595. https://doi.org/10.1037/h0022070

Bandura, A. (1977). Social Learning Theory. Englewood Cliffs, NJ: Prentice-Hall. ISBN 978-0138167516.

Bandura, A. (1986). Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice-Hall. ISBN 978-0138156145.

Bangura, W. (2025). A critical evaluation of Johnson and Wynne's (2024) methodology in comparison of the efficacy and welfare of different training methods in stopping chasing behavior in dogs. SSRN, posted February 24, 2025. https://doi.org/10.2139/ssrn.5154127

Bastos, A. P. M., Warren, E., and Krupenye, C. (2025). What evidence can validate a dog training method? Learning and Behavior, 53(2), 227-228. (Advance online publication, November 20, 2024.) https://doi.org/10.3758/s13420-024-00658-9

Best Friends Animal Society. (2025). Training for reactive, people-friendly dogs. https://bestfriends.org/stories/julie-castle-blog/best-friends-statement-re-dog-training

British Columbia Society for the Prevention of Cruelty to Animals. (n.d.). Position statement on animal training. BC SPCA. https://spca.bc.ca/programs-services/leaders-in-our-field/position-statements/position-statement-on-animal-training/

Blackwell, E. J., Bolster, C., Richards, G., Loftus, B. A., and Casey, R. A. (2012). The use of electronic collars for training domestic dogs: Estimated prevalence, reasons and risk factors for use, and owner perceived success as compared to other training methods. BMC Veterinary Research, 8, 93. https://doi.org/10.1186/1746-6148-8-93

Blackwell, E. J., Twells, C., Seawright, A., and Casey, R. A. (2008). The relationship between training methods and the occurrence of behavior problems, as reported by owners, in a population of domestic dogs. Journal of Veterinary Behavior: Clinical Applications and Research, 3(5), 207-217. https://doi.org/10.1016/j.jveb.2007.10.008

British Small Animal Veterinary Association. (2024). Position statement: Aversive training methods. https://www.bsava.com/position-statement/aversive-training-methods/

British Veterinary Association. (2024). BVA policy position: Electric shock collars and training aids. https://www.bva.co.uk/take-action/our-policies/electric-shock-collars-and-training-aids/

Canadian Veterinary Medical Association. (2021). Position statement: Humane training of dogs. https://www.canadianveterinarians.net/policy-and-outreach/position-statements/statements/humane-training-of-dogs/

Cain, C. K. (2019). Avoidance problems reconsidered. Current Opinion in Behavioral Sciences, 26, 9-17. https://doi.org/10.1016/j.cobeha.2018.09.002

Carter, A., McNally, D., and Roshier, A. (2020). Canine collars: An investigation of collar type and the forces applied to a simulated neck model. Veterinary Record, 187(7). https://doi.org/10.1136/vr.105681

Casey, R. A., Loftus, B., Bolster, C., Richards, G. J., and Blackwell, E. J. (2014). Human directed aggression in domestic dogs (Canis familiaris): Occurrence in different contexts and risk factors. Applied Animal Behaviour Science, 152, 52-63. https://doi.org/10.1016/j.applanim.2013.12.003

Casey, R. A., Naj-Oleari, M., Campbell, S., Mendl, M., and Blackwell, E. J. (2021). Dogs are more pessimistic if their owners use two or more aversive training methods. Scientific Reports, 11, 19023. https://doi.org/10.1038/s41598-021-97743-0

China, L., Mills, D. S., and Cooper, J. J. (2020). Efficacy of dog training with and without remote electronic collars vs. a focus on positive reinforcement. Frontiers in Veterinary Science, 7, 508. https://doi.org/10.3389/fvets.2020.00508

Christiansen, F. O., Bakken, M., and Braastad, B. O. (2001). Behavioural changes and aversive conditioning in hunting dogs by the second-year confrontation with domestic sheep. Applied Animal Behaviour Science, 72(2), 131-143. https://doi.org/10.1016/S0168-1591(00)00203-3

Connecticut General Statutes. § 22-350a. Tethering dog to stationary object or mobile device. Prohibited means. (As amended by Public Act 10-100 effective 1 October 2010, current version under Public Act 22-59 effective 1 October 2022.) https://www.cga.ct.gov/current/pub/chap_435.htm#sec_22-350a

Cooper, J. J., Cracknell, N., Hardiman, J., Wright, H., and Mills, D. (2014). The welfare consequences and efficacy of training pet dogs with remote electronic training collars in comparison to reward based training. PLOS ONE, 9(9), e102722. https://doi.org/10.1371/journal.pone.0102722

Cooper, J. O., Heron, T. E., and Heward, W. L. (2020). Applied Behavior Analysis (3rd ed.). Pearson. ISBN 978-0134752556.

Deldalle, S., and Gaunet, F. (2014). Effects of 2 training methods on stress-related behaviors of the dog (Canis familiaris) and on the dog-owner relationship. Journal of Veterinary Behavior: Clinical Applications and Research, 9(2), 58-65. https://doi.org/10.1016/j.jveb.2013.11.004

Dogs Trust. (2024). Government must deliver on its commitment to ban cruel electric shock collars, say leading veterinary and animal welfare organisations. (Joint statement with The Kennel Club, RSPCA, Battersea, British Veterinary Association, and Blue Cross.) https://www.dogstrust.org.uk/about-us/media-centre/shock-collars-commitment

Dubin, A. E., and Patapoutian, A. (2010). Nociceptors: The sensors of the pain pathway. Journal of Clinical Investigation, 120(11), 3760-3772. https://doi.org/10.1172/JCI42843

Federation of Veterinarians of Europe, Federation of European Companion Animal Veterinary Associations, Federation of European Equine Veterinary Associations, and World Small Animal Veterinary Association. (2024). Joint position paper on the animal welfare implications of animal behavioural modification, training methods, and ability to express species-specific behaviours. Adopted unanimously by the FVE General Assembly, June 14, 2024. https://wsava.org/wp-content/uploads/2024/11/FVE-FEEVA-FECAVA-WSAVA-Behaviour-and-training-position-ADOPTED.pdf

Friedman, S. G. (2009). What's wrong with this picture? Effectiveness is not enough. APDT Chronicle of the Dog, January/February 2009. Reprinted from Good Bird Magazine, 4(4), Winter 2008. https://www.behaviorworks.org/files/articles/APDT%20What's%20Wrong%20with%20this%20Picture%20-%20Dogs.pdf

Gillan, C. M., Morein-Zamir, S., Urcelay, G. P., Sule, A., Voon, V., Apergis-Schoute, A. M., Fineberg, N. A., Sahakian, B. J., and Robbins, T. W. (2014). Enhanced avoidance habits in obsessive-compulsive disorder. Biological Psychiatry, 75(8), 631-638. https://doi.org/10.1016/j.biopsych.2012.11.002

Gordon, A. L., Patterson, T. K., and Knowlton, B. J. (2020). Early-life stress is associated with a preponderance of habitual responding in a novel instrumental avoidance learning paradigm. Neurobiology of Learning and Memory, 175, 107316. https://doi.org/10.1016/j.nlm.2020.107316

Gouvernement du Québec. (2024). Règlement sur le bien-être et la sécurité des animaux domestiques de compagnie et des équidés. (Replaces Règlement sur la sécurité et le bien-être des chats et des chiens, P-42, r. 10.1.) In force 10 February 2024. https://www.quebec.ca/nouvelles/actualites/details/entree-en-vigueur-reglement-bien-etre-animaux-compagnie-equides

Grohmann, K., Dickomeit, M. J., Schmidt, M. J., and Kramer, M. (2013). Severe brain damage after punitive training technique with a choke chain collar in a German shepherd dog. Journal of Veterinary Behavior: Clinical Applications and Research, 8(3), 180-184. https://doi.org/10.1016/j.jveb.2013.01.002

Hawaii Revised Statutes. § 711-1109. Cruelty to animals in the second degree. (As amended by Act 182, Session Laws 2021.) https://www.capitol.hawaii.gov/hrscurrent/vol14_ch0701-0853/HRS0711/HRS_0711-1109.htm

Herron, M. E., Shofer, F. S., and Reisner, I. R. (2009). Survey of the use and outcome of confrontational and non-confrontational training methods in client-owned dogs showing undesired behaviors. Applied Animal Behaviour Science, 117(1-2), 47-54. https://doi.org/10.1016/j.applanim.2008.12.011

Hiby, E. F., Rooney, N. J., and Bradshaw, J. W. S. (2004). Dog training methods: Their use, effectiveness and interaction with behaviour and welfare. Animal Welfare, 13(1), 63-69. https://doi.org/10.1017/S0962728600026683

Humane World for Animals. (n.d.). Which type of dog collar is best for your dog? (Formerly Humane Society of the United States; position statement that choke, prong, and shock collars are inhumane and should not be used.) https://www.humaneworld.org/en/resources/dog-collars

Hunter, A., Blake, S., and De Godoy, R. F. (2019). Pressure and force on the canine neck when exercised using a collar and leash. Veterinary and Animal Science, 8, 100082. https://doi.org/10.1016/j.vas.2019.100082

International Association for the Study of Pain. (n.d.). IASP terminology. https://www.iasp-pain.org/resources/terminology/

International Association of Animal Behavior Consultants. (2025). Joint Standards of Practice for the Animal Behavior and Training Industry (November 2025 update). Endorsed by IAABC, APDT International, Karen Pryor Academy, Assistance Dog International, Grisha Stewart Academy, Science Matters, Understand Horses, Victoria Stilwell Academy, and IAABC Foundation. https://iaabc.org

Johnson, A. C., and Wynne, C. D. L. (2024). Comparison of the efficacy and welfare of different training methods in stopping chasing behavior in dogs. Animals, 14(18), 2632. https://doi.org/10.3390/ani14182632

Johnson, A. C., and Wynne, C. D. L. (2025). Chasing solutions: A response to Bastos et al. (2024). Learning and Behavior. Advance online publication. https://doi.org/10.3758/s13420-025-00672-5

Knight, D. C. (2026). Personal communication with Will Bangura regarding the interpretation of Wood et al. (2014) and the aversive character of predictable threat. [On file with author.]

LeDoux, J. E. (2014). Coming to terms with fear. Proceedings of the National Academy of Sciences, 111(8), 2871-2878. https://doi.org/10.1073/pnas.1400335111

Limbachia, C., Morrow, K., Khibovska, A., Meyer, C., Padmala, S., and Pessoa, L. (2021). Controllability over stressor decreases responses in key threat-related brain areas. Communications Biology, 4, 42. https://doi.org/10.1038/s42003-020-01537-5

Lindsay, S. R. (2005). Handbook of Applied Dog Behavior and Training, Volume 3: Procedures and Protocols. Blackwell Publishing. ISBN 978-0813807386.

Lines, J. A., van Driel, K., and Cooper, J. J. (2013). Characteristics of electronic training collars for dogs. Veterinary Record, 172(11), 288. https://doi.org/10.1136/vr.101144

Maier, S. F., and Watkins, L. R. (2005). Stressor controllability and learned helplessness: The roles of the dorsal raphe nucleus, serotonin, and corticotropin-releasing factor. Neuroscience and Biobehavioral Reviews, 29(4-5), 829-841. https://doi.org/10.1016/j.neubiorev.2005.03.021

Massachusetts House Bill H 2342. (194th General Court, 2025-2026 Session). An Act Relative to Dangerous Dogs. Reported favorably by the Joint Committee on Municipalities and Regional Government and referred to the House Committee on Ways and Means on 14 August 2025. https://malegislature.gov/Bills/194/H2342

Massachusetts Senate Bill S 1459. (194th General Court, 2025-2026 Session). An Act Relative to Dangerous Dogs. Reported favorably by the Joint Committee on Municipalities and Regional Government and referred to the Senate Committee on Rules on 19 November 2025. https://malegislature.gov/Bills/194/S1459

Masson, S., de la Vega, S., Gazzano, A., Mariti, C., Pereira, G. D. G., Halsberghe, C., Leyvraz, A. M., McPeake, K., and Schoening, B. (2018a). Electronic training devices: Discussion on the pros and cons of their use in dogs as a basis for the position statement of the European Society of Veterinary Clinical Ethology. Journal of Veterinary Behavior, 25, 71-75. https://doi.org/10.1016/j.jveb.2018.02.006

Masson, S., Nigron, I., and Gaultier, E. (2018b). Questionnaire survey on the use of different e-collar types in France in everyday life with a view to providing recommendations for possible future regulations. Journal of Veterinary Behavior, 26, 48-60. https://doi.org/10.1016/j.jveb.2018.05.004

McEwen, B. S. (2012). Brain on stress: How the social environment gets under the skin. Proceedings of the National Academy of Sciences, 109(Suppl 2), 17180-17185. https://doi.org/10.1073/pnas.1121254109

Mech, L. D. (1999). Alpha status, dominance, and division of labor in wolf packs. Canadian Journal of Zoology, 77(8), 1196-1203. https://doi.org/10.1139/z99-099

Mech, L. D. (2008). Whatever happened to the term alpha wolf? International Wolf, 18(4), 4-8. https://www.wolf.org/wp-content/uploads/2013/09/267alphalegend.pdf

Melzack, R., and Wall, P. D. (1965). Pain mechanisms: A new theory. Science, 150(3699), 971-979. https://doi.org/10.1126/science.150.3699.971

Michaels, L. (2022). The Do No Harm Dog Training and Behavior Handbook: Featuring the Hierarchy of Dog Needs®. Foreword by Marc Bekoff, PhD. ISBN 978-1732253704.

New Jersey Assembly Bill A 4206. (2026-2027 Session). Dog Trainer Licensing Act. Introduced 19 February 2026 by Assemblywoman Carol A. Murphy; referred to the Assembly Regulated Professions Committee. https://www.njleg.gov/bill-search/2026/A4206

New Jersey Assembly Bill A 4207. (2026-2027 Session). Dog Training Licensure Act (creating the New Jersey Dog Trainer Licensure Board). Introduced 19 February 2026; referred to the Assembly Regulated Professions Committee. https://www.njleg.gov/bill-search/2026/A4207

New Jersey Senate Bill S 3814. (2024-2025 Session). Establishes Dog Training Licensing Board to license and regulate dog trainers. Introduced 24 October 2024; held by sponsor following committee testimony in February 2025. https://pub.njleg.gov/Bills/2024/S4000/3814_I1.HTM

New Zealand Veterinary Association. (n.d.). Policy: Behaviour modifying collar use on dogs. https://nzva.org.nz/positions-advocacy/position-statements/collars/

New York State Senate Bill S 7723 / Assembly Bill A 6985. (2025-2026 Regular Session). An act to amend the agriculture and markets law, in relation to requiring the commissioner of agriculture and markets to establish licensing and educational standards for individuals providing canine training for non-service and non-police dogs. Assembly Bill A 6985 had its enacting clause stricken on 20 February 2026; Senate Bill S 7723 remains in the Senate Agriculture Committee. https://www.nysenate.gov/legislation/bills/2025/S7723

Pauli, A. M., Bentley, E., Diehl, K. A., and Miller, P. E. (2006). Effects of the application of neck pressure by a collar or harness on intraocular pressure in dogs. Journal of the American Animal Hospital Association, 42(3), 207-211. https://doi.org/10.5326/0420207

Petco. (2020, October 6). Stop the Shock: Petco ends the sale of electronic shock collars, firmly establishes itself as the health and wellness company for pets [Press release]. Petco Corporate. https://corporate.petco.com/2020-10-06-Stop-the-Shock-Petco-Ends-the-Sale-of-Electronic-Shock-Collars-Firmly-Establishes-itself-as-the-Health-and-Wellness-Company-for-Pets

Pessoa, L. (2026). Personal communication with Will Bangura regarding the interpretation of Limbachia et al. (2021) and the aversive character of controllable stressors. [On file with author.]

Prevention of Cruelty to Animals Regulations 2019 (Victoria, Australia). Made under the Prevention of Cruelty to Animals Act 1986. https://agriculture.vic.gov.au/livestock-and-animals/animal-welfare-victoria/pocta-act-1986/prevention-of-cruelty-to-animals-regulations

Raja, S. N., Carr, D. B., Cohen, M., Finnerup, N. B., Flor, H., Gibson, S., Keefe, F. J., Mogil, J. S., Ringkamp, M., Sluka, K. A., Song, X. J., Stevens, B., Sullivan, M. D., Tutelman, P. R., Ushida, T., and Vader, K. (2020). The revised International Association for the Study of Pain definition of pain: Concepts, challenges, and compromises. Pain, 161(9), 1976-1982. https://doi.org/10.1097/j.pain.0000000000001939

Rhode Island General Laws. § 4-13-42. Care of dogs. (As substantially expanded by H 8095, Chapter 079 of the 2024 Public Laws of Rhode Island, in force 12 June 2024.) http://webserver.rilegislature.gov/Statutes/TITLE4/4-13/4-13-42.htm

Rhode Island House Bill H 7487. (2026 January Session). Penalty escalation and expanded enforcement authority for the existing dog care and tethering statute. Introduced 4 February 2026; referred to the House Judiciary Committee. https://webserver.rilegislature.gov/BillText/BillText26/HouseText26/H7487.pdf

Rooney, N. J., and Cowan, S. (2011). Training methods and owner-dog interactions: Links with dog behaviour and learning ability. Applied Animal Behaviour Science, 132(3-4), 169-177. https://doi.org/10.1016/j.applanim.2011.03.007

Rosenkranz, J. A., Venheim, E. R., and Padival, M. (2010). Chronic stress causes amygdala hyperexcitability in rodents. Biological Psychiatry, 67(12), 1128-1136. https://doi.org/10.1016/j.biopsych.2010.02.008

Rozanski, E. A. (2022). Tracheal collapse. Today's Veterinary Practice, February 2022. https://todaysveterinarypractice.com/respiratory-medicine/tracheal-collapse/

RSPCA Australia. (n.d.). Are dog pronged collars illegal in Australia? RSPCA Knowledgebase. https://kb.rspca.org.au/categories/companion-animals/dogs/general/are-dog-pronged-collars-illegal-in-australia

Salgirli, Y., Schalke, E., Boehm, I., and Hackbarth, H. (2012). Comparison of learning effects and stress between 3 different training methods (electronic training collar, pinch collar and quitting signal) in Belgian Malinois police dogs. Revue de Médecine Vétérinaire, 163(11), 530-535.

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

Schenkel, R. (1947). Expression studies of wolves. Behaviour, 1(2), 81-129. https://doi.org/10.1163/156853948X00065

Schilder, M. B. H., and van der Borg, J. A. M. (2004). Training dogs with help of the shock collar: Short and long term behavioural effects. Applied Animal Behaviour Science, 85(3-4), 319-334. https://doi.org/10.1016/j.applanim.2003.10.004

Sears, R. M., Andrade, E. C., Samels, S. B., Laughlin, L. C., Moloney, D. M., Wilson, D. A., Alwood, M. R., Moscarello, J. M., and Cain, C. K. (2026). Devaluation of response-produced safety signals reveals circuits for goal-directed versus habitual avoidance in dorsal striatum. Nature Communications, 17, 2542. https://doi.org/10.1038/s41467-026-69119-3

Starinsky, N. S., Lord, L. K., and Herron, M. E. (2017). Escape rates and biting histories of dogs confined to their owner's property through the use of various containment methods. Journal of the American Veterinary Medical Association, 250(3), 297-302. https://doi.org/10.2460/javma.250.3.297

Steiss, J. E., Schaffer, C., Ahmad, H. A., and Voith, V. L. (2007). Evaluation of plasma cortisol levels and behavior in dogs wearing bark control collars. Applied Animal Behaviour Science, 106(1-3), 96-106. https://doi.org/10.1016/j.applanim.2006.06.018

Swiss Federal Council. (2008). Animal Protection Ordinance (Tierschutzverordnung, TSchV), 455.1, Article 76: Prohibited Devices. Entered into force 1 September 2008; current version in effect. https://www.fedlex.admin.ch/eli/cc/2008/416/en

Thompson, R. F., and Spencer, W. A. (1966). Habituation: A model phenomenon for the study of neuronal substrates of behavior. Psychological Review, 73(1), 16-43. https://doi.org/10.1037/h0022681

Todd, Z. (2018). Barriers to the adoption of humane dog training methods. Journal of Veterinary Behavior, 25, 28-34. https://doi.org/10.1016/j.jveb.2018.03.004

Tortora, D. F. (1983). Safety training: The elimination of avoidance-motivated aggression in dogs. Journal of Experimental Psychology: General, 112(2), 176-214. https://doi.org/10.1037/0096-3445.112.2.176

Vieira de Castro, A. C., Barrett, J., de Sousa, L., and Olsson, I. A. S. (2019). Carrots versus sticks: The relationship between training methods and dog-owner attachment. Applied Animal Behaviour Science, 219, 104831. https://doi.org/10.1016/j.applanim.2019.104831

Vieira de Castro, A. C., Fuchs, D., Morello, G. M., Pastur, S., de Sousa, L., and Olsson, I. A. S. (2020). Does training method matter? Evidence for the negative impact of aversive-based methods on companion dog welfare. PLOS ONE, 15(12), e0225023. https://doi.org/10.1371/journal.pone.0225023

Vyas, A., Mitra, R., Shankaranarayana Rao, B. S., and Chattarji, S. (2002). Chronic stress induces contrasting patterns of dendritic remodeling in hippocampal and amygdaloid neurons. The Journal of Neuroscience, 22(15), 6810-6818. https://doi.org/10.1523/JNEUROSCI.22-15-06810.2002

Welsh Government. (2017). Electronic collars in dogs and cats: Review of welfare implications. https://www.gov.wales/sites/default/files/publications/2017-12/electronic-collars-in-dogs-and-cats-review-of-welfare-implications.pdf

Wood, K. H., Ver Hoef, L. W., and Knight, D. C. (2014). The amygdala mediates the emotional modulation of threat-elicited skin conductance response. Emotion, 14(4), 693-700. https://doi.org/10.1037/a0036636

Debate Playbook

for Force-Free Practitioners Engaging Balanced and Aversive-Based Trainers

Companion to The Scientific Case Against Aversive Dog Training Equipment and Methods

Bangura, April 2026

Reference document. Read for study, deploy as needed.

How to Use This Playbook

This document is a deployment companion to the policy paper. The paper is the long-form citable case. This is the practitioner’s debate reference. Read it through once for the structure, then keep it accessible as a quick lookup during podcast appearances, online exchanges, in-person debates, client conversations, and continuing education sessions.

Section 1 sets the frame, the single most important move in any debate over aversive equipment and methods. Sections 2 through 4 are the substance: the nine pillars of the argument, the fifteen terminology and rhetorical moves to recognize and counter, and the twelve objections most likely to come up. Section 5 is the citation block to keep in pocket. Section 6 is closers. Section 7 names the practitioner mistakes that most often derail force-free debaters.

A one-page rip-card is provided as a separate file for quick on-the-spot reference. The rip-card is a stripped distillation; the playbook is the studied version.

Page references throughout point to The Scientific Case Against Aversive Dog Training Equipment and Methods (Bangura, 2026), the source for the underlying evidence and citations.

1. The Frame: Refuse the Wrong Question

Most debates over aversive equipment and methods get trapped inside the wrong question. The wrong question is whether these tools work. The right question is how they work, what welfare costs come with that mechanism, whether those costs are necessary, and whether broad public access is justified when safer alternatives exist.

Efficacy of suppression is not in dispute within learning theory. Anything sufficiently unpleasant delivered contingently can suppress behavior. Conceding this up front frees the debate to move to the question that actually matters: the welfare and necessity case.

Opening move: “I’m not going to argue with you about whether shock or prong collars work. They can change behavior. So can a lot of things. The question is how they change it, what that costs the dog, and whether that cost is necessary when alternatives exist that produce the same outcome without it.”

Once the frame is established, the proponent has to either accept the welfare and necessity question or refuse to engage it. Refusing to engage exposes the position. Accepting the question moves the debate onto evidentiary ground where the convergent welfare research overwhelmingly favors the force-free position.

2. The Nine Pillars

These are the nine conceptual moves that carry any debate over aversive equipment and methods. Internalize them. Three formulations are given for each: the technical version (the academic statement), the plain version (one to two sentences in conversation), and the thirty-second version (a single line for fast deployment).

Pillar 1. The Mechanism Is Aversive, Regardless of Label

Technical. Aversive stimuli are defined functionally, not descriptively. A stimulus that drives avoidance, escape, or termination behavior is, by behavior-science definition, aversive, regardless of what vocabulary describes it.

Plain. Whatever the trainer or manufacturer calls the device, the dog’s nervous system processes the experience the same way. The label does not change what is actually happening.

Thirty-second. If it changes behavior through avoidance, it bothers the dog. That is what aversive means.

Pillar 2. The Intensity Dial Proves the Mechanism

Technical. The existence of an intensity adjustment, and the clinical necessity of escalating it when the dog does not respond, is itself evidence that the mechanism is aversive control. If the stimulus were not functioning as an aversive, escalation would have no effect.

Plain. Every shock collar has a dial that goes higher. That dial exists because sometimes the trainer has to turn it up to get a response. If the lower setting were already enough, nobody would need a higher setting. The dial is the proof.

Thirty-second. Why does the collar have a dial? Because sometimes you have to turn it up. That is the answer.

Pillar 3. The “Barely Perceptible” Contradiction

Technical. If the stimulus is behaviorally meaningful enough to change behavior through escape, avoidance, or suppression, it is by functional definition aversive. If it is not behaviorally meaningful, it is not doing the training work. There is no intermediate category.

Plain. Some trainers say the modern e-collar is so mild the dog barely notices. But if the dog barely noticed, it would not change the behavior. If it changes the behavior, the dog is noticing. Both things cannot be true.

Thirty-second. If the dog barely feels it, it would not work. That it works tells you the dog feels it.

Pillar 4. Mechanism, Not Tissue Damage, Is the Welfare Question

Technical. Nociceptive engagement and threat-system activation occur well below the threshold of visible tissue damage. Nociceptors fire to warn the organism away from potentially damaging stimuli before damage occurs (Dubin and Patapoutian, 2010), and the International Association for the Study of Pain definition of pain explicitly includes potential tissue damage and applies to nonhuman animals (Raja et al., 2020). Canine haired-skin epidermis is approximately three to five cell layers thick, considerably thinner than human epidermis (Affolter and Moore, 1994), so a self-test on the human forearm or wrist systematically underestimates what the canine nervous system receives at the same delivered energy. Compounding this, the relationship between user setting and delivered stimulus across commercial e-collars is heterogeneous: an eighty-seven-fold range across products, an eighty-one-fold median ratio within collars, manufacturing faults in two of thirteen new collars, and no point-of-sale disclosure of stimulus parameters (Lines, van Driel, and Cooper, 2013). The biological question is whether the stimulus crosses the nociceptive threshold, not whether it crosses the injury threshold.

Plain. The welfare question is not whether the collar leaves a mark. The international scientific definition of pain explicitly includes potential damage. Nociceptors fire before injury, by design. Canine skin is much thinner than human skin, so self-tests on the human forearm underestimate what the dog feels. And the e-collar industry does not disclose stimulus parameters at point of sale. The “low setting” on the dial does not tell you what the dog actually feels.

Thirty-second. Pain does not require visible damage. The international pain definition says so. And canine skin is much thinner than ours.

Pillar 5. Convergent Evidence, Not One Study

Technical. The welfare evidence is sufficiently convergent across multiple independent methodological approaches, populations, countries, and outcome measures to support precautionary welfare policy. No single study carries the case alone, and the agreement across methodologies eliminates the possibility that the welfare signal is a methodological artifact.

Plain. This is not based on one study. More than twenty peer-reviewed studies from multiple countries, using different methods, including controlled experiments, observational research, surveys of thousands of guardians, cognitive bias tests, clinical referral data, pain neuroscience, threat-circuit imaging, and active avoidance neuroscience, all converge in the same direction. When that many independent methods point in the same direction, the evidence is sufficient for precautionary welfare policy.

Thirty-second. It is not one study. It is twenty-plus, in different countries, with different methods, all pointing the same direction.

Pillar 6. The Necessity Claim Has Not Been Supported Under Best-Practice Conditions

Technical. Where the necessity claim has been empirically tested under best-practice conditions, with industry-nominated trainers (Cooper et al., 2014; China, Mills, and Cooper, 2020), the tools produced welfare cost without producing better outcomes than reward-based methods. Independently, board-certified veterinary behaviorists, the clinical specialty that treats the most severe canine aggression, anxiety, fear, and predatory cases, treat those cases as their standard of care without aversive equipment. The American College of Veterinary Behaviorists has formalized this position in its December 2025 letter to the American Veterinary Medical Association.

Plain. The necessity claim has been tested directly. Industry-nominated trainers using e-collars in controlled studies produced welfare cost without producing better outcomes than reward-based trainers. Separately, the veterinary specialists who handle the worst aggression and anxiety cases in the country do not use shock or prong collars. The American College of Veterinary Behaviorists, the highest credential in this field, does not include aversive equipment in its standard of care.

Thirty-second. Tested with industry-nominated trainers, the tools added nothing. The veterinary specialists who treat the toughest cases do not use them.

Pillar 7. Adding Food Does Not Subtract the Aversive

Technical. Compound schedules pairing positive reinforcement with positive punishment or negative reinforcement do not eliminate the welfare cost of the aversive component. The dog’s nervous system continues to register and respond to the aversive event regardless of whether food reinforcement is present in parallel.

Plain. Some trainers say it is fine to use a shock or prong collar as long as you also give the dog treats. Adding treats does not erase the shock. The dog’s brain still registers the unpleasant event. The peer-reviewed studies that have looked at this directly find welfare costs even when food rewards are also present.

Thirty-second. Treats do not cancel out shocks. The dog’s brain registers both.

Pillar 8. International Consensus Is Already In

Technical. Aversive training equipment is prohibited or restricted in multiple national and subnational jurisdictions across Europe, the United Kingdom, Australia, North America, and Latin America, and is opposed by every major international veterinary, welfare, and professional behavior organization that has issued a position statement, including the FVE, FECAVA, FEEVA, and WSAVA joint position paper of June 2024.

Plain. Wales has banned shock collars since 2010. Switzerland since 2008. Germany under case law since 2006. Austria since 2005. The Netherlands restricted pinch collars in 2018 and electric stimulation devices comprehensively by 2021. Quebec in 2024. France in professional contexts since June 2025. Colombia since 2025. Belgium-Flanders has a full ban in force 1 January 2027. Most of Australia has prohibited various combinations of these tools. Every major veterinary organization that has reviewed the evidence has reached the same conclusion. The United States is increasingly the outlier on this issue.

Thirty-second. Most of the developed world has banned or restricted these tools. The United States is the outlier.

Pillar 9. Active Avoidance Is Anxiety, Not Welfare Neutrality

Technical. Active avoidance is goal-directed instrumental behavior under threat, not a welfare-neutral compliance state (Cain, 2019). The shift from a fear state to an anxiety state during effective avoidance does not eliminate the underlying threat representation; the warning stimulus retains its conditioned threat value. Controllability modulates downstream consequences but does not render the stressor benign (Maier and Watkins, 2005). Safety signals acquire their value entirely from their inverse relationship with the aversive contingency, and overtrained avoidance becomes habitual via the dorsolateral striatum, the same circuit implicated in obsessive-compulsive disorder (Gillan et al., 2014) and in stronger avoidance habits in survivors of early life stress (Gordon et al., 2020) (Sears et al., 2026).

Plain. Some sophisticated balanced trainers will say that successful e-collar training transitions the dog to an anxiety state where the dog is positively reinforced by the absence of shock, and that this exempts the training from welfare scrutiny. The contemporary neuroscience does not support that interpretation. Safety signals acquire their value entirely from their inverse relationship with the aversive event. Without the aversive event, no warning stimulus acquires threat value, no feedback cue acquires safety value, and the avoidance response is not reinforced. The aversive contingency is the precondition for the entire learning architecture. The calm-looking dog is in an anxiety state mediated by an effective avoidance response, not in the absence of threat representation. And the brain circuit that mediates overtrained avoidance is the same circuit implicated in obsessive-compulsive disorder and in survivors of early life stress.

Thirty-second. The safety signal is defined by the aversive event. Take away the aversive, the safety signal means nothing. The aversive is the precondition.

3. The Fifteen Moves: Terminology and Rhetorical Decoder

Recognize these moves when they appear. The general principle is the same in every case: behavior science defines procedures by their function, not by their label. Whatever vocabulary variant the proponent introduces, the diagnostic question is whether the procedure depends on the dog working to avoid, escape, or terminate a stimulus. If yes, it is aversive.

Group A. Euphemisms That Rename the Aversive Stimulus

Move 1. “Stim,” “stimulation,” “e-touch,” or “tap” instead of “shock”

Translation. Marketing language, not behavior-science language.

Response. What we call it does not change what the dog’s nervous system does with it. If the stimulus is strong enough to make the dog change its behavior to avoid it, that is an aversive event by any scientific definition.

Move 2. “Correction” instead of “punishment”

Translation. In operant conditioning, positive punishment is the application of a stimulus following a behavior that decreases its future probability. Calling it a correction does not move it out of the operant punishment quadrant.

Response. In behavior science, the procedure is named by what it does, not by what we call it. If applying the leash pop reduces the behavior, it is positive punishment. That is the definition.

Move 3. “Pressure” or “tap” instead of “pinch” or “compression”

Translation. Used for prong and choke collars to suggest the mechanical event is mild.

Response. Mechanical force that successfully reduces lead pulling is, by functional necessity, exceeding the dog’s mechanonociceptive threshold. If it were below that threshold, it would not change the behavior.

Move 4. “Self-correction” for prong collars

Translation. Used to imply the dog, not the handler, is the cause of the aversive event, and therefore that no punishment is being applied.

Response. Procedures are defined by their contingencies, not by the identity of the agent who closes the contingency. When the dog pulls and the prong applies concentrated pressure, the dog has experienced a behavior-contingent aversive event. The operant function is identical.

Group B. Reframings of the Tool’s Function

Move 5. “Communication tool” instead of “aversive device”

Translation. Communication does not require an aversive event. Verbal cues, visual cues, hand signals, and trained markers all communicate without aversive consequences.

Response. If the device communicates by applying an unpleasant consequence the dog works to avoid, it is functioning as an aversive device, regardless of how the handler frames it.

Move 6. “Working level” or “minimum effective level”

Translation. Used to imply that low-level e-collar use is welfare-benign.

Response. The working level is the level at which the stimulus is sufficient to change behavior. Behavior change driven by escape, avoidance, or suppression requires the stimulus to function as an aversive event. Working level and aversive level are the same level.

Move 7. “Pavlovian” or “classical conditioning” framing

Translation. Sometimes invoked to suggest the e-collar functions through associative learning rather than punishment.

Response. Classical conditioning to an aversive unconditioned stimulus is the foundational paradigm of fear-conditioning research. Calling the procedure classical does not exempt it from welfare cost. And in actual e-collar training, the operant component (behavior-contingent stimulation the dog learns to escape or avoid) is the working mechanism.

Group C. Methodological Frame Reframings

Move 8. “Balanced” training as a moderate middle ground

Translation. The term implies that balanced training combines reward-based and aversive methods in a moderate, considered way, and that force-free is the extreme position.

Response. Balanced training is defined by the inclusion of positive punishment and negative reinforcement, regardless of whether positive reinforcement is also present. The inclusion of aversive procedures, not the proportion, is the methodological signature. The convergent welfare evidence applies to any methodology that includes aversive procedures.

Move 9. “LIMA” invoked to defend aversive use

Translation. LIMA properly applied requires that less intrusive interventions be demonstrably exhausted before more intrusive interventions are considered.

Response. LIMA requires you to document that less intrusive methods do not work for this specific case before you escalate. Routine use of aversive equipment without that documented hierarchy is not LIMA. It is the label of LIMA on a different practice.

Move 10. “All four quadrants”

Translation. Behavior science describes four operant procedures functionally. This is a descriptive taxonomy, not a normative endorsement.

Response. Pointing to the existence of four quadrants does not establish that all four should be used routinely on companion dogs. Descriptive science does not equal prescriptive endorsement. The welfare research has specifically established costs associated with positive punishment and negative reinforcement that are not associated with positive reinforcement.

Group D. Conceptual and Logical Deflections

Move 11. “Pack leader,” “alpha,” or “dominance” framing

Translation. The conceptual basis was Schenkel’s 1947 captive-wolf study, generalized into the alpha-pack-leader model that became part of popular dog training in the late twentieth century.

Response. The wildlife biologist L. David Mech, whose own earlier work helped popularize the alpha framing, has retracted it in the peer-reviewed literature (Mech, 1999) and in subsequent published commentary (Mech, 2008). Contemporary canine ethology does not support dominance-based confrontational handling on domestic dogs. The framing has been retracted by one of its own original popularizers, in print.

Move 12. “My dog tells me when to use it” / “the dog asks for the correction”

Translation. Behaviors interpreted as the dog asking for the correction are commonly displacement behaviors, conflict behaviors, or stress signals.

Response. Reading those signals as consent reverses the inferential direction. The behaviors are evidence of the welfare cost, not evidence of the dog’s endorsement of the procedure that produced them.

Move 13. “It worked for my client” / “I have hundreds of success stories”

Translation. Anecdotal individual cases. Visible behavioral suppression and welfare cost can coexist.

Response. A dog whose problem behavior has been suppressed by an aversive intervention is not, by that fact alone, a dog whose welfare has been preserved. The visible outcome and the unmeasured outcome (conditioned emotional response, threat-circuit engagement, affective state, relationship cost) are not the same. Population-level controlled and observational research is the appropriate evidence base for welfare claims.

Move 14. “Positive reinforcement does not work for serious cases”

Translation. Empirically not supported by the clinical practice of board-certified veterinary behaviorists.

Response. ACVB Diplomates treat the most severe canine aggression, anxiety, fear, and predatory cases without aversive equipment, using behavioral assessment, environmental management, reward-based behavior modification, and psychiatric medication when clinically indicated. The clinical specialty that handles the hardest cases does not use aversive equipment as standard of care.

Move 15. “Safety signal” or “the dog is positively reinforced by safety”

Translation. The most sophisticated current proponent argument. Cites contemporary neuroscience to argue that successful e-collar training transitions the dog to an anxiety state where the safety signal (the absence of shock) functions as positive reinforcement, and that this exempts the training from welfare scrutiny.

Response. The Sears et al. 2026 paper that this argument leans on says the opposite. Safety signals acquire their value entirely from their inverse relationship with the aversive contingency. Without the aversive event, no warning stimulus acquires threat value, no feedback cue acquires safety value, and the avoidance response is not reinforced. The aversive contingency is the precondition for the entire learning architecture. Calling it positive reinforcement by safety does not change that the aversive event must occur for the safety signal to mean anything. And the dorsolateral striatum circuit that mediates overtrained avoidance is the same circuit implicated in obsessive-compulsive disorder and in survivors of early life stress. The clinical population overlap is the welfare cost the argument tries to wave away.

4. The Twelve Objections: Counters and Kill Shots

These are the objections most likely to come up in live debate. For each: a one-paragraph response and a one-sentence kill shot for moments when time is short.

Objection 1. “You are cherry-picking studies”

Response. The argument does not rest on any single study. It rests on multiple independent lines of evidence using different methods, populations, and outcome measures: experimental studies, observational studies, guardian surveys, cognitive bias tasks, clinical referral data, peer-reviewed physical-effects research, pain neuroscience, threat-circuit imaging, and active avoidance neuroscience. A critic who wants to dismiss the convergence must explain why each independent methodological line is wrong, why the same welfare signal would appear across such different methods by coincidence, and why every major veterinary organization that has reviewed the evidence has reached the same conclusion.

Kill shot. Cherry-picking is selecting one study. Convergence is twenty-plus studies, different methods, same answer.

Objection 2. “Cortisol was not significant in Cooper 2014”

Response. Cortisol is one measure, and a blunt one. Behavioral stress indicators in Cooper et al. (2014) were significant in the electronic collar group, and these are validated welfare markers in canine research. Dismissing a study on the basis of one non-significant physiological measure while ignoring significant behavioral measures is not a methodological critique. It is selective reading.

Kill shot. Cortisol is one measure. The behavioral stress markers in Cooper hit significance. That is welfare harm, regardless of the hormone assay.

Objection 3. “The reward groups just had more reinforcement”

Response. This either concedes that the aversive equipment added no outcome benefit (in which case it carried welfare cost for no training advantage), or reduces to a hypothetical about a better-designed protocol that has not been tested. The trainers in the aversive equipment groups of Cooper and China were industry-nominated, and the high-reinforcement protocol was available to any trainer who wanted to deliver it. They did not.

Kill shot. If the reward group did better with more reinforcement, that is the answer. Use more reinforcement. The collar added nothing.

Objection 4. “Controllability and predictability make aversive use safe”

Response. Controllability and predictability reduce the magnitude of threat-related neural responding. They do not convert aversive stimulation into a neutral stimulus. Dr. Luiz Pessoa, senior author of Limbachia et al. (2021), and Dr. David Knight, senior author of Wood et al. (2014), have each confirmed in writing that their research cannot be used to support the proponent claim. Their statements are interpretive support: they confirm the reading of the published studies at the level of authorial intent.

Kill shot. Predictable shocks still hurt. The researchers being cited say so themselves, in writing.

Objection 5. “Low-level stim or gentle corrections are benign”

Response. If the stimulation is behaviorally meaningful enough to change behavior through escape, avoidance, or suppression, it is by functional definition aversive. Independently, the engineering measurement work by Lines, van Driel, and Cooper (2013) showed an eighty-seven-fold range across thirteen commercial collars at maximum settings, an eighty-one-fold median ratio within collars across user-dial settings, and manufacturing faults in two of thirteen new collars including one that delivered a maximum-strength impulse regardless of user setting. The user dial is not a reliable proxy for what the dog feels.

Kill shot. If it is mild enough to not bother the dog, it is mild enough to not work. And the dial does not tell you what the dog actually feels.

Objection 6. “It is not really aversive, the dog just finds it weird and wants to turn it off”

Response. “The dog wants to make it stop” is a plain-language description of negative reinforcement. The dog’s behavior is being maintained by termination of a stimulus. Any stimulus that functions as a negative reinforcer is, by definition, an aversive stimulus. There is no category of “weird but not aversive” in the experimental learning literature that supports avoidance learning. A merely novel stimulus produces an orienting response followed by habituation; the fact that e-collar stimulation continues to drive behavioral modification across hundreds of trials without habituating shows the dog’s nervous system is not classifying it as merely novel.

Kill shot. “The dog wants to turn it off” is the textbook definition of aversive. You just described negative reinforcement and called it neutral.

Objection 7. “An electronic collar is just like a TENS unit”

Response. TENS and electronic training collars have opposite therapeutic purposes and therefore opposite electrical parameters. TENS is calibrated to selectively activate large-diameter A-beta touch fibers without crossing the nociceptive threshold; the goal is reduction of pain perception. An electronic collar must deliver a stimulus the dog works to avoid, escape, or terminate, which by definition means crossing the nociceptive threshold. The two devices cannot share a biological mechanism because they have opposite biological goals. TENS is also FDA-regulated as a medical device with required disclosure of pulse parameters; electronic collars are not.

Kill shot. TENS is designed to reduce pain. An e-collar has to produce something the dog works to avoid. Opposite mechanisms. Not the same device.

Objection 8. “The prong collar just gets the dog’s attention”

Response. If the prong were only an attention-getting device without an aversive component, the dog could be trained to stop pulling with any other attention-getting device, including a tap, a verbal cue, or a sound. The specific efficacy of the prong collar is its mechanical delivery of an aversive event. The vocabulary of “attention” does not change the mechanism of action.

Kill shot. If it were attention, a hand clap would do it. The prong works because it hurts.

Objection 9. “The choke chain mimics how a mother dog corrects her pups”

Response. This claim is biologically inaccurate. Mother dogs do not perform sustained neck constrictions on their puppies. The alpha-wolf mythology underlying this claim originated with Schenkel’s 1947 captive-wolf study and was popularized in late twentieth-century dog training literature. L. David Mech, one of the original popularizers, has retracted the framing in the peer-reviewed literature (Mech, 1999) and in subsequent published commentary (Mech, 2008).

Kill shot. Mother dogs do not strangle their puppies. Mech retracted the mythology himself, in print.

Objection 10. “My dog looks happy in training”

Response. Visible engagement is not a physiological readout. Threat and stress circuitry do not announce themselves through tail posture or eye expression alone. Once avoidance learning is well established, a dog may perform fluently and quickly precisely because the behavior prevents the aversive event. Fluency in the instrumental response is compatible with ongoing threat prediction. This is the active avoidance picture set out in Cain (2019) and Sears et al. (2026): the calm-looking dog is in an anxiety state mediated by an effective avoidance response, not in the absence of threat representation.

Kill shot. What a dog looks like in fifteen seconds is not what is happening in its nervous system across time.

Objection 11. The Rescue Device Pattern (Unfalsifiability)

Response. When every study showing harm is set aside for a different methodological reason (Cooper had reinforcement differences, Vieira de Castro was not randomized, cortisol is complicated, neuroscience is in humans, Herron relied on guardian reports, Carter used a model neck), the methodology critique is being used as a rescue device rather than a genuine pursuit of better evidence. The diagnostic question is: what would count as evidence against the tools? If the answer is that no existing study, no convergent finding across disciplines, no professional consensus, and no regulatory precedent would be sufficient, then the position is not scientifically falsifiable.

Kill shot. What evidence would change your mind? If nothing would, this is not a scientific disagreement.

Objection 12. “Modern e-collar trained dogs are in a goal-directed anxiety state, not under threat”

Response. This is the most sophisticated current proponent argument. It invokes contemporary neuroscience (Cain, 2019; Sears et al., 2026) to claim that successfully trained dogs have transcended the aversive contingency through avoidance, and that the resulting calm anxiety state exempts the training from welfare scrutiny. The neuroscience does not support that reading. The shift from fear to anxiety during effective avoidance does not eliminate the underlying threat representation. The warning stimulus retains its conditioned threat value; what changes is that the dog has acquired a behavioral option that controls exposure to the aversive event. When the avoidance response is blocked or fails, the fear state returns along with inflexible defensive reactions. And the dorsolateral striatum circuit that mediates overtrained avoidance is the same circuit implicated in obsessive-compulsive disorder (Gillan et al., 2014) and in stronger avoidance habits in survivors of early life stress (Gordon et al., 2020). The clinical population overlap is the welfare cost the argument tries to wave away.

Kill shot. Anxiety state is not welfare neutral. The dog has built an avoidance habit using the same brain circuit OCD recruits.

5. The Citation Block

These are the studies and authorities to be able to name on demand. Author, year, and the one-sentence finding. The full citations are in the policy paper References section.

Controlled Experimental Studies

• Cooper, Cracknell, Hardiman, Wright, and Mills (2014). Industry-nominated e-collar trainers produced welfare-relevant behavioral stress in dogs and no training advantage over reward-based methods.
• China, Mills, and Cooper (2020). Re-analysis confirming that reward-based training produced equivalent or better outcomes more efficiently than e-collar training. No necessity advantage.

Observational and Survey Studies

• Schilder and van der Borg (2004). Shock-collar-trained working dogs showed elevated stress behaviors in training and in the broader training context, indicating conditioned emotional response.
• Vieira de Castro et al. (2020). Aversive-trained dogs showed elevated post-training cortisol, more stress behaviors, and more pessimistic cognitive bias than reward-based-trained dogs.
• Casey et al. (2021). Dogs trained with two or more aversive methods showed pessimistic cognitive bias compared with reward-based-trained dogs.
• Herron, Shofer, and Reisner (2009). Confrontational handling produced aggressive responses in a substantial percentage of dogs in clinical referral data, including 31 percent for alpha rolls and 43 percent for hitting or kicking.
• Masson, Nigron, and Gaultier (2018b). Among French e-collar users, 71.8 percent operated without professional advice and 7 percent of dogs showed physical wounds.
• Starinsky, Lord, and Herron (2017). Electronic fence escape rate 44 percent versus physical fence 23 percent; no clear protective effect of electronic containment.

Pain Neuroscience and Sensory Engagement

• Dubin and Patapoutian (2010). Nociceptors fire below the threshold of tissue injury; their function is to warn the organism away from potentially damaging stimuli before damage occurs.
• Raja et al. (2020). Revised IASP definition of pain explicitly includes potential tissue damage and applies to nonhuman animals.
• Affolter and Moore (1994). Canine haired-skin epidermis is approximately three to five cell layers thick, considerably thinner than human epidermis.
• Lines, van Driel, and Cooper (2013). 87-fold range across 13 commercial e-collars at maximum settings; 81x median ratio within collars; 2 of 13 with manufacturing faults; no point-of-sale disclosure of stimulus parameters.

Threat Circuitry and Active Avoidance

• LeDoux (2014). Amygdala-centered defensive circuitry processes aversive events; circuit activation is not welfare-neutral.
• Cain (2019). Active avoidance is goal-directed instrumental behavior under threat; the calm-looking dog is in an anxiety state, not in the absence of threat representation.
• Maier and Watkins (2005). Controllability modulates downstream consequences of aversive stressors but does not render the stressor benign or stress-free.
• Limbachia et al. (2021). Controllability over aversive stimulation attenuates threat-related neural responding without eliminating it.
• Wood, Ver Hoef, and Knight (2014). Amygdala mediates emotional modulation of threat-elicited responses.
• Sears et al. (2026). Safety signals acquire their value entirely from their inverse relationship with the aversive contingency. Overtrained avoidance becomes habitual via the dorsolateral striatum, the same circuit implicated in OCD and survivors of early life stress.
• Pessoa (personal communication, 2026). Senior author of Limbachia et al. confirmed in writing that his research does not support treating controllable aversive stimulation as neurologically neutral or welfare-benign. Interpretive support: confirms the reading of the published study at the level of authorial intent.
• Knight (personal communication, 2026). Senior author of Wood et al. (2014) confirmed in writing that his fear-conditioning research cannot be used to support the proposition that predictable aversive stimulation is neutral or benign. Interpretive support: confirms the reading of the published study at the level of authorial intent.

Physical Effects (Neck-Pressure Equipment)

• Pauli, Bentley, Diehl, and Miller (2006). Collar pressure during ordinary pulling significantly elevates intraocular pressure; harness pressure does not.
• Carter, McNally, and Roshier (2020). No collar tested produced neck pressures low enough to mitigate injury risk under realistic pull forces (83 to 832 kPa across 7 collar types).
• Hunter, Blake, and De Godoy (2019). Peak contact pressure of 44.61 N/cm² transmitted to the canine neck during ordinary on-leash walking.
• Grohmann et al. (2013). Peer-reviewed case report: fatal cerebral ischemia in a German Shepherd following punitive choke-chain hanging.
• Rozanski (2022). Repeated collar pressure recognized clinical concern for tracheal collapse; harness recommended over collar; standard of care.

Studies to Be Ready For Proponents to Invoke

• Johnson and Wynne (2024). Frequently cited by proponents as evidence of e-collar necessity for predatory chasing. The study establishes narrow efficacy under specific protocol conditions, NOT necessity, welfare neutrality, or broad real-world superiority. Methodology challenged in Bastos et al. (2024) peer-reviewed critique and Bangura (2025) SSRN methodological critique. Authors’ published response (Johnson and Wynne, 2025) did not resolve the substantive methodological concerns.
• Christiansen et al. (2001). Sometimes invoked to support controllable, predictable shock-collar use. The welfare measures relied largely on guardian report and temperament tests; methodological thinness, not a clean positive welfare conclusion, prevents the dataset from supporting a welfare-benign reading.
• Tortora (1983). Sometimes invoked for e-collar use in aggression. The protocol was a complex 9-stage process beginning with positive reinforcement, not simple aversive conditioning. Author’s own Experiment 3 undermines the proponent reading. 1983 design predates contemporary welfare-science methodology.
• Lindsay (2005), Volume 3. Textbook chapter expressing one author’s interpretation, not peer-reviewed welfare research. The characterization of low-level e-stim as a “tingle” conflicts with Dubin and Patapoutian (2010), Raja et al. (2020), and the engineering measurement findings of Lines et al. (2013).

Institutional Position Statements

• FVE, FECAVA, FEEVA, WSAVA (2024). Joint position paper unanimously calling for a complete ban on the sale and use of electric pulse training devices including electric shock collars.
• AVSAB (2021). Aversive methods including electronic collars, prong collars, choke chains, and leash corrections should not be used under any circumstances.
• ACVB (December 2025). Letter to AVMA: shock collars are not medically necessary, not evidence-based for preventing euthanasia, and not aligned with the standard of care for veterinary behavior medicine.
• AAHA (2015). Only acceptable training techniques are non-aversive, positive techniques.

6. Closers

Lines for ending an exchange when the opponent will not concede or move. These are not first moves. They are last moves, deployed after the substantive argument has been laid out.

• “The veterinary behavior specialty that handles the hardest cases in this country has formally rejected the equipment you are defending. That is the answer to the necessity claim.”
• “Wales banned shock collars in 2010. Switzerland in 2008. Germany under case law since 2006. Austria since 2005. More than fifteen years of regulatory experience across multiple jurisdictions, and no published study attributes measurable public-safety harm to the prohibition. The necessity-from-public-safety claim has been put before that record, and the published case for harm has not been made.”
• “The two senior authors of the neuroscience studies you are citing have both confirmed in writing that their research does not support your interpretation. I have those communications. The argument from neuroscience does not work for the position you are defending.”
• “If you are invoking safety signals as positive reinforcement, the safety signal acquires its value from the aversive event. Take away the aversive, the safety signal means nothing. The contemporary neuroscience does not exempt the training from welfare scrutiny. It explains why a successfully avoidance-trained dog can look calm while being in an anxiety state mediated by the same brain circuit OCD recruits.”
• “What evidence would change your mind? If nothing would, this is not a scientific disagreement, and we should call it what it is.”
• “Seventy percent of US guardians, in Petco’s own 2020 disclosure, believe shock collars harm their pet’s emotional or mental wellbeing. Sixty-nine percent consider them cruel. The market you are defending is not the market the consumer base actually wants.”
• “The convergent welfare evidence catalogued in the peer-reviewed literature, the international veterinary consensus formalized in 2024, and the regulatory record across more than a dozen jurisdictions all point in the same direction. The case is not novel and it is not radical. It is the case the evidence has been making for a long time.”

7. The Don’ts: Practitioner Mistakes That Derail Force-Free Debaters

These are the mistakes most often made by force-free practitioners in live debate. Each is named, with a one-line corrective.

Don’t 1. Pursuing the dog-as-anecdote rabbit hole

The proponent will offer a story about a specific dog whose life was saved by an e-collar. Engaging the anecdote is a trap; you cannot win an exchange about a dog you have never met. Stay at population level. Anecdotes are evidence about anecdotes, not about the welfare profile of the equipment in the population.

Don’t 2. Defending studies on methodology rather than convergence

Every study has methodological limits. If the proponent attacks Cooper alone, you lose if you defend Cooper alone. The case is not Cooper. The case is convergence across more than twenty studies. Move to the convergence frame and stay there.

Don’t 3. Operant quadrant pedantry without function

Getting drawn into a debate over which quadrant a procedure technically belongs to (positive punishment vs. negative reinforcement) is rarely productive in front of a public audience. The function of the procedure (does it depend on aversive stimulation?) is what matters. Stay functional.

Don’t 4. Conceding “professionals can use these tools safely”

This is the trap that splits the welfare argument into a tiered framework. The international veterinary consensus has explicitly rejected the tiered framing. The mechanism is aversive in any hands. Cooper and China studied industry-nominated trainers and the welfare cost was still there.

Don’t 5. Apologizing or hedging to seem reasonable

Hedging language (“maybe in some cases”, “it depends”, “sometimes”) reads as weakness in a debate context where the other side is making confident claims. The evidence is convergent. State it that way. Reasonableness is not the same as concession. (Note that softening categorical claims about absence of harm, as in Pillar 6 and the closers, is not hedging. It is choosing claims you can defend rather than claims that invite hostile fact-checking.)

Don’t 6. Letting the proponent pivot from one tool to another

When the e-collar argument is failing, the proponent will pivot to prong collars (“well, prongs are different”). When prongs are failing, they will pivot to choke chains. When all three are failing, they will pivot to confrontational handling. The unified policy frame closes that pivot route. The mechanism is aversive control, regardless of which tool delivers it. Treat them as one category in the debate, the way the international veterinary consensus does.

Don’t 7. Forgetting to name the consensus

Force-free practitioners sometimes argue the welfare science alone and forget to name the institutional consensus. The consensus argument is independent and complementary. Naming the AVSAB, ACVB, AAHA, FVE/WSAVA, and the major welfare and professional organizations forces the proponent to argue not only against the science but against the entire field. Most cannot.

Don’t 8. Engaging tone with tone

Combative balanced trainers often deploy mockery, insult, or condescension. Matching their tone is rarely useful for the audience. The audience is not the opponent. Speak to the audience. Calm, accurate, evidence-anchored language wins the long game in front of mixed rooms even when it does not feel like winning in the moment.

Don’t 9. Accepting the safety-signal framing without naming the inversion

When a sophisticated proponent invokes Sears 2026 and argues that safety signals positively reinforce the dog’s avoidance response, the practitioner mistake is to engage on operant-quadrant terms (“technically that is negative reinforcement, not positive reinforcement”). The right move is to name the inversion: the safety signal acquires its value entirely from the aversive contingency, and without the aversive event the safety signal means nothing. Do not get drawn into the operant taxonomy debate. The aversive contingency is the precondition for the entire learning architecture, and the dorsolateral striatum circuit that mediates overtrained avoidance is the same circuit implicated in OCD and survivors of early life stress.

End of Playbook

Companion to Bangura (2026), The Scientific Case Against Aversive Dog Training Equipment and Methods.

DEBATE RIP-CARD

Force-Free vs. Aversive-Based Trainers

THE FRAME

Refuse the wrong question. Suppression efficacy is not in dispute. The right question is welfare cost, necessity, and justified consumer access.

Opener: “I’m not arguing whether shock or prong collars work. They can change behavior. The question is how, what it costs the dog, and whether that cost is necessary when alternatives produce the same outcome without it.”

THE NINE PILLARS (THIRTY-SECOND VERSIONS)

1. If it changes behavior through avoidance, it bothers the dog. That is what aversive means.

2. Why does the collar have a dial? Because sometimes you have to turn it up. That is the answer.

3. If the dog barely feels it, it would not work. That it works tells you the dog feels it.

4. Pain does not require visible damage. The international pain definition (IASP) says so. Canine skin is much thinner than ours.

5. It is not one study. It is twenty-plus, in different countries, with different methods, all pointing the same direction. Sufficient for precautionary policy.

6. Tested with industry-nominated trainers, the tools added nothing. The veterinary specialists who treat the toughest cases do not use them.

7. Treats do not cancel out shocks. The dog’s brain registers both.

8. Most of the developed world has banned or restricted these tools. The United States is the outlier.

9. The safety signal is defined by the aversive event. Take away the aversive, the safety signal means nothing. The aversive is the precondition.

KILL SHOTS (PAIRED TO THE OBJECTION)

1. Cherry-picking? Cherry-picking is one study. Convergence is twenty-plus, different methods, same answer.

2. Cortisol? One measure. The behavioral stress markers in Cooper hit significance. Welfare harm, regardless of the hormone.

3. Reward groups had more reinforcement? Then use more reinforcement. The collar added nothing.

4. Predictable shocks? Still hurt. The researchers being cited (Pessoa, Knight) say so themselves, in writing.

5. Low-level is benign? If it is mild enough not to bother the dog, it is mild enough not to work. And the dial does not tell you what the dog feels (Lines 2013: 87-fold range across collars; 2 of 13 with manufacturing faults).

6. Just “weird”, not aversive? “The dog wants to turn it off” is the textbook definition of aversive.

7. Just like a TENS unit? TENS reduces pain. An e-collar produces something the dog works to avoid. Opposite mechanisms.

8. Prong gets attention? If it were attention, a hand clap would do it. The prong works because it hurts.

9. Choke mimics mother dogs? Mother dogs do not strangle puppies. Mech retracted that mythology himself, in print (1999, 2008).

10. My dog looks happy? What a dog looks like in fifteen seconds is not what is happening in its nervous system across time.

11. Studies dismissed one by one? What evidence would change your mind? If nothing would, this is not a scientific disagreement.

12. Goal-directed anxiety state? Anxiety state is not welfare neutral. The dog has built an avoidance habit using the same brain circuit OCD recruits.

CLOSERS

• The veterinary specialty that handles the hardest cases has formally rejected the equipment you are defending. That is the answer to necessity.

• Wales 2010. Switzerland 2008. Germany 2006 (case law). Austria 2005. Netherlands 2018/2021. Quebec 2024. France June 2025. Colombia 2025. Belgium-Flanders 2027. Fifteen-plus years of regulatory experience, and no published study attributes measurable public-safety harm to the prohibition.

• The two senior authors of the neuroscience studies you cite have confirmed in writing that their research does not support your interpretation. Interpretive support, on file.

• If you invoke safety signals as positive reinforcement: the safety signal acquires value from the aversive event. Take away the aversive, the safety signal means nothing.

• What evidence would change your mind? If nothing would, call it what it is.

CITATIONS TO NAME ON DEMAND

Cooper et al. 2014 • China et al. 2020 • Schilder & van der Borg 2004 • Vieira de Castro et al. 2020 • Casey et al. 2021 • Herron et al. 2009 • Masson et al. 2018b • Starinsky et al. 2017 • Pauli et al. 2006 • Carter et al. 2020 • Hunter et al. 2019 • Grohmann et al. 2013 • Rozanski 2022

Pain neuroscience: Dubin & Patapoutian 2010 • Raja et al. 2020 • Affolter & Moore 1994 • Lines, van Driel, & Cooper 2013

Threat circuitry: LeDoux 2014 • Cain 2019 • Maier & Watkins 2005 • Limbachia et al. 2021 • Wood et al. 2014 • Sears et al. 2026 • Pessoa pers. comm. 2026 (interpretive support) • Knight pers. comm. 2026 (interpretive support)

Consensus: FVE/WSAVA 2024 • AVSAB 2021 • ACVB Dec 2025 • AAHA 2015 • BVA • BSAVA • AVA • CVMA • NZVA

DON’TS

• Don’t engage anecdotes. Stay at population level.

• Don’t defend single studies. Defend the convergence.

• Don’t let “professionals can use it safely” split the welfare argument.

• Don’t let the proponent pivot from e-collars to prongs to chokes. Treat them as one category.

• Don’t forget the institutional consensus. Naming AVSAB, ACVB, FVE/WSAVA forces the opponent to argue against the field.

• Don’t match combative tone. Speak to the audience, not the opponent.

• Don’t accept the safety-signal framing without naming the inversion. The safety signal acquires value only from the aversive contingency.

Bangura (2026), Aversive Dog Training Policy Paper. Companion playbook available.

Jurisdiction Playbook

Legislative and Regulatory Restrictions on Aversive Dog Training Equipment

Companion to The Scientific Case Against Aversive Dog Training Equipment and Methods

Bangura, April 2026

Reference document. One profile per jurisdiction.

How to Use This Playbook

This document organizes the legislative and regulatory record on aversive dog training equipment by jurisdiction. The policy paper presents the same material as a comparative summary in Table 1; this playbook expands each entry into a profile suitable for legislative testimony, public comment letters, op-ed writing, professional advocacy correspondence, and conversations with state and provincial lawmakers.

When to use this document. When preparing for legislative testimony or public comment. When drafting an op-ed on dog training regulation. When writing to a state legislator or animal welfare advocate. When responding to the claim that bans on aversive equipment are unprecedented or radical. When supporting a colleague who is introducing legislation in their state. When teaching in a continuing education context that includes policy or advocacy material.

Structure of each profile. Jurisdiction name. Statutory authority with citation. Effective date. Scope of prohibition (what is banned, what is permitted, what carve-outs exist). Enforcement mechanism and penalty structure where known. Years of operation under the ban. Notes on any post-implementation review, comparative position, or supplementary regulatory context.

Organization. The playbook is organized by region: Europe, Latin America, North America at the subnational level, and Australia and New Zealand. The United States is treated separately, with a section on the existing state-level partial restrictions in the tethering context, and a section on pending United States legislation organized by legislative-design approach. A summary section at the front provides the comparative pattern across all jurisdictions covered.

Verification. Statutory citations and effective dates are drawn from primary sources where available and from the Welsh Government 2017 review and the FVE, FECAVA, FEEVA, and WSAVA 2024 joint position paper for jurisdictions where primary verification was conducted through those secondary references. Practitioners using these profiles for legislative or formal advocacy purposes should verify current statutory language, which may have been amended since the dates referenced here.

A note on framing. Where this playbook discusses the absence of harm from prohibition, the framing is deliberately literature-claim form rather than categorical. The defensible position is that no published peer-reviewed evidence of measurable public-safety harm from prohibition has been produced, not that no harm has occurred. The distinction matters for advocacy work. Categorical claims invite hostile fact-checking; literature-claim form does not.

Table A. Comparative Jurisdictional Summary

This table consolidates the legislative and regulatory record on aversive dog training equipment by jurisdiction. It is organized by region and provides the citation, effective date, scope, and years of operation for each jurisdiction profiled in this playbook. For deeper detail on any jurisdiction, see the corresponding profile in the regional sections that follow.

The pattern across these jurisdictions is consistent. Where evidence-based welfare considerations have been weighed by national, regional, or state legislative bodies, the consistent direction of policy has been toward restriction or prohibition, never toward expansion of access or normalization of use. Bolded entries indicate jurisdictions most useful for United States legislative testimony: long-standing comprehensive bans, recent enactments demonstrating active legislative trajectory, and the most active pending United States bills.

The Comparative Pattern at a Glance

Multiple jurisdictions on five continents have enacted legislative or regulatory restrictions on aversive dog training equipment. The pattern across these jurisdictions is consistent. Where evidence-based welfare considerations have been weighed by national, regional, or state legislative bodies, the consistent direction of policy has been toward restriction or prohibition, never toward expansion of access or normalization of use.

The earliest comprehensive restrictions date to the 2008 Swiss Animal Protection Ordinance and the 2010 Welsh Statutory Instrument banning electronic collars. More than fifteen years of regulatory experience in these jurisdictions has not produced peer-reviewed evidence of measurable public-safety harm attributable to the prohibitions. The defensible reading of that record is literature-claim form: no published study has attributed measurable public-safety harm to electronic collar prohibition in any jurisdiction that has banned the equipment. The necessity-from-public-safety claim has been put before the regulatory experience of more than a dozen jurisdictions, some operating under bans for over fifteen years, and the published case for harm from prohibition has not been made.

Jurisdictions Covered in This Playbook

Europe. Wales, Switzerland, Germany, Austria, France, Spain, Sweden, Finland, Norway, Denmark, the Netherlands, Slovenia, Belgium (Flanders, effective 2027).

Latin America. Colombia (2025).

North America (subnational). Quebec, Canada (2024).

Australia (federal). Commonwealth import prohibition on pronged collars.

Australia (state and territory). Australian Capital Territory, New South Wales, Queensland, South Australia, Tasmania, Victoria.

New Zealand. Long-standing veterinary and professional opposition under the Animal Welfare Act 1999 framework.

United States (subnational, partial restrictions in tethering context). Hawaii, Rhode Island, Connecticut, plus 23 states and the District of Columbia with tethering laws.

United States (pending legislation, 2024 to 2026 period). New York Senate Bill S 7723 / Assembly Bill A 6985; New Jersey Senate Bill S 3814, Assembly Bill A 4206, Assembly Bill A 4207; Massachusetts House Bill H 2342 and Senate Bill S 1459; Rhode Island House Bill H 7487.

Deployment line: “Wales has banned shock collars since 2010. Switzerland banned aversive collars years before that. Fifteen years of regulatory experience in those jurisdictions, and no published study attributes measurable public-safety harm to the prohibition. The necessity-from-public-safety claim has been put before the regulatory experience of more than a dozen jurisdictions, and the published case for harm from prohibition has not been made.”

Europe

Wales (United Kingdom)

Statutory authority. Animal Welfare (Electronic Collars) (Wales) Regulations 2010, Welsh Statutory Instrument 2010 No. 943 (W. 97), made under section 12 of the Animal Welfare Act 2006.

Effective date. March 24, 2010.

Scope of prohibition. Prohibits the use of electronic collars on dogs and cats. The prohibition covers remote-controlled, bark-activated, and containment electronic collars. Use is the prohibited act; possession is not directly criminalized but use is enforceable as a welfare offense.

Enforcement and penalty. Enforced under the Animal Welfare Act 2006 framework. Conviction can result in fines and, for serious or repeated welfare offenses, imprisonment.

Years of operation. Sixteen years as of 2026. The longest-running comprehensive electronic collar prohibition among English-speaking jurisdictions.

Notes. Wales was the first United Kingdom nation to enact a comprehensive electronic collar ban. Scotland and England have not enacted comparable legislation, though England has had repeated parliamentary discussion of similar bans, and a coalition of British Veterinary Association, Kennel Club, RSPCA, Battersea, Blue Cross, and Dogs Trust has called publicly for an England ban. The Welsh Government published a 2017 review of the welfare implications of electronic collars that supports the regulatory framework.

Switzerland

Statutory authority. Animal Protection Ordinance (Tierschutzverordnung, TSchV), 455.1, Article 76, of 23 April 2008, in force 1 September 2008. Underlying Animal Protection Act (Tierschutzgesetz, TSchG) of 16 December 2005.

Effective date. September 1, 2008.

Scope of prohibition. Prohibits the use of spike collars (collars with inward-facing prongs), pinch collars, and electronic collars. Article 76 also broadly prohibits equipment causing pain, fear, or major injury. Use is the prohibited act; the language reaches both training and routine handling contexts.

Enforcement and penalty. Enforced under the Federal Animal Protection Act framework. Cantonal veterinary authorities have enforcement responsibility.

Years of operation. Approximately eighteen years of operation as of 2026.

Notes. Switzerland’s Animal Protection Ordinance is one of the most comprehensive in Europe, covering pinch, prong, and electronic collars together with broader categorical language on equipment causing pain. The Swiss legislative model is frequently cited as the comprehensive standard against which other jurisdictions are compared.

Germany

Statutory authority. Animal Welfare Act (Tierschutzgesetz, TierSchG), §3 No. 11, originally 1972 and consolidated 18 May 2006; interpreted to cover electronic training devices by Federal Administrative Court (Bundesverwaltungsgericht) judgment of 23 February 2006, BVerwG 3 C 14.05.

Effective date. Statutory framework long-standing; case-law interpretation in force from 23 February 2006.

Scope of prohibition. Prohibits the use of electronic collars and equipment causing pain. The Federal Administrative Court decision interpreted the statute to cover electronic training devices on the basis of their design and function, regardless of how an individual user might attempt to apply them. Prong collars are restricted under the same framework based on judicial and regulatory interpretation. Some German states (Länder) have enacted more specific guidance under the federal framework.

Enforcement and penalty. Enforced under the Animal Welfare Act framework, with state-level veterinary authority responsibility.

Years of operation. Approximately twenty years of restriction as of 2026.

Notes. Germany’s framework is interpretive rather than explicitly enumerated; the federal Animal Welfare Act’s general prohibition on causing avoidable pain has been read to cover electronic and prong collars. This makes German legislation slightly less direct as a citable model than the Swiss ordinance, but the practical effect is comparable.

Austria

Statutory authority. Federal Animal Protection Act (Tierschutzgesetz), §5(2)(3)(a), BGBl. I 2004/118, of 28 September 2004.

Effective date. In force 1 January 2005.

Scope of prohibition. Prohibits the use on animals of spike collars, prong collars, electric and chemical training devices, and other equipment causing fear, pain, or harm.

Enforcement and penalty. Enforced under federal animal protection framework, with provincial-level (Bundesländer) veterinary authority responsibility.

Years of operation. Approximately twenty-one years of operation as of 2026.

Notes. Austrian law is part of the broader Central European pattern of comprehensive welfare-based restrictions.

Denmark

Statutory authority. Bekendtgørelse nr. 607 af 25. juni 2009 om forbud mod brug af visse aggregater, halsbånd mv. til dyr, under the Danish Animal Welfare Act (Dyreværnsloven), lov nr. 386 af 6. juni 1991.

Effective date. In force from 25 June 2009.

Scope of prohibition. Prohibits remote-controlled and automatically operating electric devices, sharp or pointed prong collars, and the advertising and sale of prohibited equipment.

Enforcement and penalty. Enforced under animal welfare framework.

Years of operation. Long-standing.

Notes. Denmark’s prohibition is part of a broader Nordic regulatory pattern restricting aversive training equipment.

Finland

Statutory authority. Animal Welfare Act 693/2023.

Effective date. In force 1 January 2024.

Scope of prohibition. Prohibits the use and sale of electric collars and spike collars on animals.

Enforcement and penalty. Enforced under Animal Welfare Act framework.

Years of operation. Two years as of 2026; the 2023 Act is the first Finnish statute to contain an explicit prohibition on electric and spike collars (the predecessor 1996 Act did not contain such a provision).

Notes. Part of the Nordic regulatory pattern.

France

Statutory authority. Arrêté du 19 juin 2025 fixant les règles sanitaires et de protection animale auxquelles doivent satisfaire les activités liées aux animaux de compagnie d’espèces domestiques, Article 14, under Code rural et de la pêche maritime, Articles L. 214-6-1 et seq.

Effective date. Published 19 June 2025; provisions in force on publication.

Scope of prohibition. Prohibits electric, prong, and strangling collars (without stopping buckle) in professional contexts: educators, breeders, kennels, refuges, presenters. Private use by individual guardians is not yet covered; the broader Assembly proposition de loi (passed 16 January 2023) covering private use remains pending in the Senate.

Enforcement and penalty. Enforced under domestic animal welfare framework with administrative and criminal penalties for violation.

Years of operation. In force from publication on 19 June 2025; one year as of 2026.

Notes. France’s 2025 arrêté is one of the most recent national-level prohibitions in Europe and demonstrates the continuing legislative trajectory toward restriction. The regulation was supported by the French Order of Veterinary Surgeons (Ordre national des vétérinaires) and major French animal welfare organizations.

Netherlands

Statutory authority. Besluit van 26 april 2018 amending Besluit houders van dieren, Article 1.3(h), under the Wet dieren, Article 2.1; further strengthened by Staatsblad 2021, 361.

Effective date. Pinch collar prohibition in force 1 July 2018; electric stimulation device prohibition initially with professional exception, later closed by the 2021 amendment.

Scope of prohibition. Prohibits the use of equipment delivering electric shocks to dogs and pinch collars.

Enforcement and penalty. Enforced under the Animals Act framework.

Years of operation. Approximately eight years for the pinch collar prohibition; five years under the 2021 closure of the e-collar professional exception.

Notes. The Netherlands enacted its prohibition following sustained advocacy by Dutch veterinary and welfare organizations.

Norway

Statutory authority. Animal Welfare Act (Lov om dyrevelferd), LOV-2009-06-19-97, of 19 June 2009, with implementing regulation. The predecessor 1974 Animal Welfare Act also restricted training collars.

Effective date. Act in force 1 January 2010.

Scope of prohibition. Prohibits electric training devices, anti-bark electric collars, invisible electric fences, and prong collars.

Enforcement and penalty. Enforced under Animal Welfare Act framework.

Years of operation. Long-standing.

Notes. Part of the Nordic regulatory pattern. The Norwegian Council on Animal Ethics has supported broader restriction.

Slovenia

Statutory authority. Zakon o zaščiti živali (ZZZiv), of 18 November 1999, published Uradni list RS št. 98/99 of 3 December 1999; most recent amendment ZZZiv-G in force 1 August 2025.

Effective date. Original Act 1999; current consolidation in force from 1 August 2025.

Scope of prohibition. Restricts electronic training collars under the general Animal Protection Act framework.

Enforcement and penalty. Enforced under Animal Protection Act framework.

Years of operation. Long-standing; twenty-seven years under the 1999 framework as of 2026.

Spain

Statutory authority. Ley 7/2023, de 28 de marzo, de protección de los derechos y el bienestar de los animales, Article 27(ñ).

Effective date. September 29, 2023.

Scope of prohibition. Prohibits the use of electric, impulse, punishment, and choke collars; hunting, herding, and guard dogs are exempt.

Enforcement and penalty. Administrative and criminal penalties for violation; serious-infraction penalties of €10,001 to €50,000 under Article 76.

Years of operation. Approximately three years of operation as of 2026.

Notes. Spain’s 2023 national animal welfare law is one of the most comprehensive in southern Europe and includes provisions on training methods alongside broader companion animal welfare standards.

Sweden

Statutory authority. Animal Welfare Act (Djurskyddslagen) 2018:1192, supplemented by the Animal Welfare Ordinance 2019 and Jordbruksverket regulations.

Effective date. In force 1 April 2019.

Scope of prohibition. Prohibits equipment delivering electric shocks and spike collars on animals.

Enforcement and penalty. Enforced under Animal Welfare Act framework.

Years of operation. Seven years under the 2018 Act as of 2026.

Notes. Part of the Nordic regulatory pattern.

Belgium (Flanders)

Statutory authority. Decree of 13 July 2018 establishing principle prohibition under the federal Animal Welfare Act of 14 August 1986; phase-out scenario set in 2021.

Effective date. Full prohibition in force 1 January 2027.

Scope of prohibition. Prohibits remote-controlled and bark-activated electric collars on dogs in the Flemish region, with no exception for military, police, or behaviour therapists; invisible-fence collars remain permitted.

Enforcement and penalty. Will be enforced under regional animal welfare framework once in force.

Years of operation. Enacted but not yet in force as of 2026.

Notes. Flanders is the first of the three Belgian regions to enact a comprehensive electronic collar prohibition, with delayed effective date of 2027 to allow transition. The Flemish framework is notable for not exempting military, police, or behaviour therapists, which closes a carve-out commonly found in other European frameworks.

Latin America

Colombia

Statutory authority. Ley 2480 de 2025 (Ley Kiara), Article 10, regulating professional pet care services; complemented by general anti-cruelty framework Ley 84 de 1989 as updated by Ley 2455 de 2025 (Ley Ángel) of 18 April 2025.

Effective date. In force in 2025 (Ley Kiara); Ley Ángel signed 18 April 2025.

Scope of prohibition. Ley 2480 (Ley Kiara) prohibits prong and electric collars in regulated pet care services, including kennels, training centres, transport, grooming, and spas. Private-use scenarios are addressed under the general anti-cruelty provisions.

Enforcement and penalty. Enforced under national animal welfare and consumer protection framework.

Years of operation. Approximately one year of operation as of 2026.

Notes. Colombia is the first Latin American country to enact a national-level prohibition on aversive training equipment, demonstrating that the regulatory trajectory extends beyond Europe and the English-speaking world. The Colombian law explicitly addresses both sale and use, providing a more comprehensive consumer-protection model than legislation that addresses use alone.

North America (Subnational)

Quebec, Canada

Statutory authority. Règlement sur le bien-être et la sécurité des animaux domestiques de compagnie et des équidés, chapter B-3.1, r. 0.1, made under the Loi sur le bien-être et la sécurité de l’animal (CQLR c. B-3.1), replacing the earlier P-42, r. 10.1 framework.

Effective date. February 10, 2024.

Scope of prohibition. Prohibits collars likely to cause pain, specifically identified by the Quebec Ministry of Agriculture, Fisheries and Food (MAPAQ) as étrangleur (choke), à pointes (prong), électrique (electric), and martingale collars. The regulation builds on a 2013 principle that an animal’s collar must not interfere with breathing or cause pain or injury.

Enforcement and penalty. First-offense fines range from $600 to $12,000, with tripled penalties for repeat offenses. Enforcement by MAPAQ.

Years of operation. Two years of operation as of 2026.

Notes. Quebec is currently the only Canadian province with comprehensive prohibition on prong, electric, choke, and martingale collars under animal welfare regulation. The MAPAQ identification of these specific collar types as falling within the breathing-or-pain-or-injury language is the operative regulatory standard for enforcement purposes. Quebec’s framework is the closest North American analog to the European regulatory model and is a useful citable precedent for United States state legislation.

Australia and New Zealand

Australia (Commonwealth)

Statutory authority. Federal customs law (Customs (Prohibited Imports) Regulations 1956, as amended).

Effective date. Long-standing import prohibition; specific regulatory listing of pronged collars dates from federal review.

Scope of prohibition. Prohibits the import of pronged collars into Australia. Sale and use within Australia are regulated at the state and territory level (see entries below).

Enforcement and penalty. Enforced by Australian Border Force at the federal level for import offenses.

Years of operation. Long-standing.

Notes. The federal import prohibition does not, on its own, prohibit possession or use of pronged collars already in the country. State and territory legislation handles the in-country prohibition. The combination of federal import prohibition and state-level use prohibition is a useful regulatory model for jurisdictions where federal authority over training equipment is constitutionally limited.

Australian Capital Territory

Statutory authority. Animal Welfare Act 1992 (ACT) §13, with prescribed permitted electric devices listed in Animal Welfare Regulation 2001; further strengthened by Animal Welfare Legislation Amendment Bill, passed 26 September 2019.

Effective date. Act 1992; regulation 2001; 2019 strengthening amendments.

Scope of prohibition. Prohibits the use of electric devices on companion animals; electronic training collars are not on the prescribed-permitted list. The 2019 amendments also recognised animal sentience.

Enforcement and penalty. Enforced under ACT animal welfare framework.

Years of operation. Long-standing.

New South Wales

Statutory authority. Prevention of Cruelty to Animals Act 1979 (NSW), section 16 (electric collar prohibition added by 2000 amendment).

Effective date. Act 1979; section 16 prohibition added 2000.

Scope of prohibition. Prohibits the use, sale, and possession of electric collars on dogs; containment systems permitted only inside a fence at least 1.5 metres high.

Enforcement and penalty. Enforced by RSPCA NSW and state authorities under the prevention of cruelty framework.

Years of operation. Long-standing.

Notes. The NSW framework includes a narrow exception for containment systems under specified conditions. The exception is narrower than the general permissive treatment that would apply absent regulation.

Queensland

Statutory authority. Animal Care and Protection Act 2001 (Qld), section 37A, as inserted by Animal Care and Protection Amendment Act 2022.

Effective date. Amendment Act passed 2 December 2022; section 37A in force 12 December 2022.

Scope of prohibition. Prohibits the possession, use, and supply of pronged dog collars. Electronic collars in Queensland are regulated rather than banned.

Enforcement and penalty. Enforced by RSPCA Queensland and state authorities.

Years of operation. Approximately three years of operation as of 2026 for the prong collar prohibition.

Notes. Queensland’s 2022 amendment is one of the more recent expansions of an Australian state framework and demonstrates the continuing legislative trajectory at the subnational level.

South Australia

Statutory authority. Prevention of Cruelty to Animals Regulations (No. 2) 2000 (SA), regulation 8(1)(a), under the Animal Welfare Act 1985 (SA).

Effective date. Regulations 2000.

Scope of prohibition. Prohibits placing on an animal a collar designed to impart an electric shock.

Enforcement and penalty. Enforced under Animal Welfare Act framework.

Years of operation. Long-standing.

Tasmania

Statutory authority. Animal Welfare Act 1993 (Tas), section 8(2)(ja), inserted by Act No. 36 of 2022.

Effective date. In force 30 November 2022.

Scope of prohibition. Prohibits the use of pronged collars and similar pinching collars on animals.

Enforcement and penalty. Enforced by RSPCA Tasmania and state authorities.

Years of operation. Approximately three years of operation as of 2026.

Victoria

Statutory authority. Prevention of Cruelty to Animals Regulations 2019 (Vic), under the Prevention of Cruelty to Animals Act 1986; technical specifications for any conditionally permitted electronic collar use set out in Ministerial Approval Notice S 56, Victorian Government Gazette, 6 February 2020.

Effective date. Regulations 2019; technical specifications notice 6 February 2020.

Scope of prohibition. Prohibits the use of pronged collars under regulation 11. Electronic collars are heavily regulated under regulations 23 to 29A, conditionally permitted only under the technical-specifications standards in Ministerial Notice S 56.

Enforcement and penalty. Enforced by RSPCA Victoria, state authorities, and local government.

Years of operation. Seven years under the 2019 framework as of 2026.

Notes. Victoria’s framework demonstrates a tiered approach: full prohibition on prong collars combined with conditional regulation of electronic collars. The conditional regulation includes specific protocol requirements for any permitted electronic collar use. This is a useful comparative reference for jurisdictions considering tiered legislative approaches.

New Zealand

Statutory authority. Animal Welfare Act 1999 framework, with position guidance from the New Zealand Veterinary Association and the Animal Welfare Advisory Committee.

Effective date. Long-standing framework; specific positions on electronic collars from professional bodies.

Scope of prohibition. Electronic collars are not subject to a formal national-level prohibition, but the New Zealand Veterinary Association does not support their use, and the Association of Pet Dog Trainers New Zealand has issued a position statement that the use of electronic training collars is not only unnecessary but a form of cruelty.

Enforcement and penalty. Animal Welfare Act 1999 framework provides general welfare-based enforcement authority.

Years of operation. Long-standing professional opposition; statutory framework allows enforcement on welfare grounds.

Notes. New Zealand is included here because the professional consensus position closely parallels formal regulatory prohibition in other jurisdictions, even though New Zealand has not enacted explicit equipment-specific legislation. The NZVA and APDTNZ positions are useful citations for jurisdictions weighing professional consensus alongside statutory frameworks.

United States (Subnational, Partial Restrictions in Tethering Context)

United States state-level legislation on aversive training equipment is, at the time of writing, limited to partial restrictions in the tethering context rather than comprehensive prohibitions on sale and use. These statutes are useful as evidence that United States legislatures have already recognized welfare concerns associated with aversive equipment and have begun legislating accordingly, though the existing statutes do not yet meet the comprehensive standard adopted in Europe and Australia.

Hawaii

Statutory authority. Hawaii Revised Statutes §711-1109(1)(j) (cruelty to animals in the second degree), as amended by Act 182, Session Laws 2021.

Effective date. 2021 amendment.

Scope of prohibition. Makes it a criminal offense of cruelty to animals in the second degree to tether or restrain a dog to a stationary object by means of a choke collar, pinch collar, or prong collar, unless the dog is engaged in an activity supervised by its owner or an agent of its owner. (The terms “owner” and “agent of its owner” are direct statutory language from HRS §711-1109.)

Enforcement and penalty. Criminal prosecution under the cruelty to animals statute.

Years of operation. Five years of operation as of 2026.

Notes. Hawaii’s 2021 amendment is one of the more direct United States statutory references to choke, pinch, and prong collars by name. The supervised-activity exception narrows the scope to the tethering context, but the legislative recognition of these specific tools as cruelty-relevant is a useful precedent.

Rhode Island

Statutory authority. Rhode Island General Laws §4-13-42, as substantially expanded by H 8095, Chapter 079 of the 2024 Public Laws of Rhode Island, in force 12 June 2024.

Effective date. Original tethering provisions long-standing under the General Laws framework; substantial expansion in force 12 June 2024.

Scope of prohibition. The 2024 expansion of §4-13-42 prohibits tethering a dog with a choke-type collar, head collar, or prong-type collar. It additionally restricts permanent tether area to no less than 113 square feet (or a six-foot trolley radius at ground level), prohibits tethering for more than ten hours in any twenty-four-hour period, prohibits tethering between 10:00 p.m. and 6:00 a.m. (with a fifteen-minute exception), and incorporates the Tufts Animal Care and Condition Weather Safety Scale to limit outdoor confinement under adverse weather conditions.

Enforcement and penalty. Civil and criminal penalties under the General Laws framework.

Years of operation. Two years under the 2024 substantial expansion as of 2026; the underlying tethering framework is long-standing.

Notes. The 2024 expansion is one of the more comprehensive United States state-level tethering statutes by virtue of incorporating the Tufts Weather Safety Scale and the time-of-day prohibition. Rhode Island H 7487 of 2026 is pending and would further escalate penalties for repeat violations and expand enforcement authority to municipal animal control officers (see the United States Pending Legislation section below).

Connecticut

Statutory authority. Connecticut General Statutes §22-350a, as amended by Public Act 10-100 (effective 1 October 2010) and subsequently amended; current version under Public Act 22-59 (effective 1 October 2022).

Effective date. October 1, 2010 (P.A. 10-100); current version effective October 1, 2022 (P.A. 22-59).

Scope of prohibition. Prohibits tethering a dog by means of a coat hanger, choke collar, prong-type collar, head halter, or any other collar, halter, or device that is not specifically designed or properly fitted for the restraint of the dog.

Enforcement and penalty. Civil and criminal penalties under the General Statutes framework.

Years of operation. Sixteen years under P.A. 10-100 as of 2026; current version under P.A. 22-59 in force four years.

Notes. Connecticut’s statute is notable for naming choke, prong, and head halter collars by category as inappropriate tethering equipment, alongside the catch-all language for any device not specifically designed or properly fitted for the restraint of the dog.

Other United States States with Tethering Statutes

Statutory authority. Various state statutes catalogued by the Animal Legal and Historical Center, Michigan State University College of Law (2022).

Effective date. Various.

Scope of prohibition. Twenty-three states and the District of Columbia have enacted laws regulating the tethering of dogs. Several states explicitly name choke, prong, or pinch collars as prohibited tethering equipment. Specific statutory language varies.

Enforcement and penalty. Varies by state.

Years of operation. Varies by state.

Notes. Practitioners considering legislative work in a specific state should consult the Animal Legal and Historical Center’s table of state dog tether laws for the current statutory language in their jurisdiction. The cumulative effect of these state statutes is that aversive equipment has already been recognized in United States law as welfare-relevant, even if the recognition is currently limited to the tethering context.

United States Pending Legislation (2024 to 2026 Period)

Pending United States legislation in the 2024 to 2026 period reflects three distinct legislative-design approaches to aversive equipment and aversive methods, beyond the enacted tethering statutes already discussed. The first approach is professional licensing of dog trainers tied to non-aversive standards. The second approach is restriction of aversive equipment within specified behavior-modification contexts. The third approach is enhancement of existing tethering and care statutes. None of the proposals catalogued in this section had been enacted as of the date of this paper. Collectively, however, they establish that United States state-level legislative interest in regulating aversive training equipment, aversive methods, and the dog training profession itself is active across multiple states and across multiple legislative-design models.

Note on bill numbering format. All bill identifiers in this section use the canonical letter-prefix-and-space form (“S 7723,” “A 6985,” “S 3814”) that matches each state’s legislative reference style.

Approach 1. Trainer Licensure Under Non-Aversive, Evidence-Based Standards

The trainer-licensure approach addresses the dog training regulatory vacuum at the state level by adopting force-free methodology as the standard of professional practice and using state licensure to enforce that standard. This is the same regulatory model used for veterinary medicine, mental health counseling, and social work in the United States. New York and New Jersey have introduced parallel proposals under this approach.

New York Senate Bill S 7723 / Assembly Bill A 6985

Statutory authority. New York State Senate Bill S 7723 / Assembly Bill A 6985, 2025-2026 Regular Session. An act to amend the agriculture and markets law, in relation to requiring the commissioner of agriculture and markets to establish licensing and educational standards for individuals providing canine training for non-service and non-police dogs.

Effective date. Pending.

Scope. Would add Agriculture and Markets Law section 113-a, requiring licensing and educational standards for individuals providing canine training to non-service and non-police dogs, with the statutory language explicitly mandating non-aversive, evidence-based, positive reinforcement techniques as the basis of those standards.

Procedural status. Assembly Bill A 6985 had its enacting clause stricken on 20 February 2026. Senate Bill S 7723 remains in the Senate Agriculture Committee.

Enforcement and penalty. Would create a state regulatory framework for trainer licensure with associated disciplinary mechanisms.

Notes. The licensure approach is structurally significant: rather than legislating against specific tools, it adopts force-free methodology as the standard of professional practice and uses licensure to enforce that standard. This is the model recommended in Section 10 of the policy paper. The procedural setback to A 6985 in February 2026 does not foreclose the approach; the parallel S 7723 remains active in the Senate Agriculture Committee, and the New Jersey proposals (below) carry the same structural model forward.

New Jersey Senate Bill S 3814 (2024-2025 Session)

Statutory authority. New Jersey Senate Bill S 3814, 2024-2025 Session. Establishes Dog Training Licensing Board to license and regulate dog trainers.

Effective date. Held by sponsor.

Scope. Would have established a Dog Training Licensing Board with an evidence-based humane training code precluding aversive methods.

Procedural status. Introduced 24 October 2024. Held by sponsor following committee testimony in early 2025 and not advanced from committee.

Notes. The withdrawal of S 3814 in early 2025 was followed approximately one year later by the introduction of two parallel Assembly bills (A 4206 and A 4207, below) on 19 February 2026, indicating that the New Jersey legislative interest in the licensure approach did not end with the withdrawal of S 3814 but is being carried forward through the Assembly.

New Jersey Assembly Bill A 4206 (Dog Trainer Licensing Act)

Statutory authority. New Jersey Assembly Bill A 4206, 2026-2027 Session. Dog Trainer Licensing Act, introduced by Assemblywoman Carol A. Murphy.

Effective date. Pending.

Scope. Would establish a Board of Examiners of Dog Trainers under the Dog Trainer Licensing Act.

Procedural status. Introduced 19 February 2026; referred to the Assembly Regulated Professions Committee.

Notes. A 4206 is one of two parallel Assembly bills introduced on the same day (the other being A 4207, below). The two bills represent slightly different licensure-board structures under closely related statutory frameworks.

New Jersey Assembly Bill A 4207 (Dog Training Licensure Act)

Statutory authority. New Jersey Assembly Bill A 4207, 2026-2027 Session. Dog Training Licensure Act creating the New Jersey Dog Trainer Licensure Board.

Effective date. Pending.

Scope. Would establish a New Jersey Dog Trainer Licensure Board under the Dog Training Licensure Act and would expressly tie licensure standards to professional codes of ethics that incorporate the Least Intrusive, Minimally Aversive Effective Behavior Intervention Policy adopted jointly by the Association of Professional Dog Trainers (APDT), the Certification Council for Professional Dog Trainers (CCPDT), and the International Association of Animal Behavior Consultants (IAABC).

Procedural status. Introduced 19 February 2026; referred to the Assembly Regulated Professions Committee.

Notes. A 4207 is structurally significant because it ties the licensure standard to a specific named professional standards document, the LIMA Effective Behavior Intervention Policy, rather than leaving the standard to be determined by regulation. This is a useful citable precedent for any state legislature considering trainer licensure with substantive standards-of-practice requirements built into the enabling statute.

Approach 2. Restriction of Aversive Equipment in Behavior-Modification Plans

The equipment-restriction-in-behavior-modification-plans approach addresses aversive equipment use in a specific high-stakes context: court-ordered or regulatory behavior modification plans for dogs that have been deemed dangerous. Massachusetts has introduced parallel House and Senate bills under this approach in the 194th General Court.

Massachusetts House Bill H 2342

Statutory authority. Massachusetts House Bill H 2342, 194th General Court, 2025-2026 Session. An Act Relative to Dangerous Dogs.

Effective date. Pending.

Scope. Would require that any dangerous-dog behavior modification plan ordered under the proposed dangerous-dog statute use exclusively evidence-based training techniques that do not result in pain, discomfort, fear, or anxiety, and would explicitly exclude electric, prong, and choke collars from such plans, with required adherence to the principles of the American Veterinary Society of Animal Behavior (AVSAB) and the American College of Veterinary Behaviorists (ACVB).

Procedural status. Reported favorably by the Joint Committee on Municipalities and Regional Government and referred to the House Committee on Ways and Means on 14 August 2025.

Notes. The Massachusetts approach is narrower than full equipment prohibition, but it operates in a high-stakes context (dangerous-dog behavior modification) where the choice of method directly affects public safety as well as canine welfare. The explicit incorporation of AVSAB and ACVB principles is structurally similar to the LIMA reference in NJ A 4207 and is a useful citable precedent for states considering equipment-restriction language tied to professional standards.

Massachusetts Senate Bill S 1459

Statutory authority. Massachusetts Senate Bill S 1459, 194th General Court, 2025-2026 Session. An Act Relative to Dangerous Dogs.

Effective date. Pending.

Scope. Senate companion to H 2342, with the same equipment exclusions and standards-of-practice requirement.

Procedural status. Reported favorably by the Joint Committee on Municipalities and Regional Government and referred to the Senate Committee on Rules on 19 November 2025.

Notes. The favorable Joint Committee report on both H 2342 and S 1459 indicates substantive legislative momentum in the 194th General Court, even if neither bill has yet reached a floor vote.

Approach 3. Enhancement of Existing Tethering and Care Statutes

The third pending-legislation approach is enhancement of existing tethering and care statutes, increasing penalties or expanding enforcement authority without changing the categorical scope of the underlying prohibition. Rhode Island has introduced legislation under this approach.

Rhode Island House Bill H 7487

Statutory authority. Rhode Island House Bill H 7487, 2026 January Session. Penalty escalation and expanded enforcement authority for the existing dog care and tethering statute.

Effective date. Pending.

Scope. Would increase penalties for repeat violations of the existing dog care and tethering statute (Rhode Island General Laws §4-13-42, as substantially expanded by H 8095 in 2024) and would expand enforcement authority to include city and town animal control officers.

Procedural status. Introduced 4 February 2026; referred to the House Judiciary Committee.

Notes. H 7487 is the second pending Rhode Island bill in two years (following the 2024 substantial expansion of §4-13-42 by H 8095, Chapter 079 of 2024). The 2024 expansion added the substantive scope; H 7487 of 2026 would add the penalty escalation and the enforcement-authority expansion. Together, the two bills represent sustained legislative attention to the tethering-statute framework rather than a single one-time enactment.

Cross-state pattern. Pending legislation in the 2024 to 2026 period spans four states (New York, New Jersey, Massachusetts, Rhode Island) and three legislative-design approaches (trainer licensure, equipment restriction in behavior-modification plans, tethering-statute enhancement). The pattern indicates active United States state-level legislative interest in this regulatory space, distributed across multiple design models, even though no single comprehensive prohibition has yet been enacted at the United States state level. For a legislator or staffer assessing whether action in their state would be premature or out of step, the pending-legislation record across these four states establishes that action is neither premature nor unprecedented.

Deployment Notes

For Legislative Testimony

The most powerful argument from this jurisdictional record is the absence of published evidence of measurable public-safety harm from the existing prohibitions. Wales has banned shock collars since 2010. Switzerland since 2008. Multiple Australian states for over a decade. Quebec since 2024. France since June 2025. The Netherlands since 2018 (pinch) and 2021 (electronic). Colombia since 2025. No published study attributes measurable public-safety harm in any of these jurisdictions to the prohibition. The defensible framing is literature-claim form, not categorical: the published case for harm from prohibition has not been made. Categorical claims invite hostile fact-checking; literature-claim form does not.

Citing two or three specific jurisdictions by name, with effective dates and years of operation, anchors the testimony in concrete regulatory precedent rather than in abstract policy argument. The Welsh, Swiss, and Quebec examples together cover Anglo-American, Continental European, and North American legislative traditions, which makes them particularly useful for United States audiences.

For Op-Eds and Public Comment

The strongest framing for general audiences is comparative. Most of the developed world has restricted or banned this equipment, and the United States is increasingly the outlier. Use the Wales, Switzerland, Germany, France, Netherlands, Quebec, and Australia examples to establish that the regulatory direction is global and consistent, not novel or radical.

For Conversations with State Legislators

State legislators considering action will want a model to follow. Three operative legislative-design models are now in active circulation in the 2024 to 2026 period: (1) trainer licensure under non-aversive standards (NY S 7723 / A 6985, NJ S 3814, NJ A 4206, NJ A 4207); (2) equipment restriction in court-ordered behavior modification plans (MA H 2342, MA S 1459); and (3) penalty escalation and enforcement expansion for tethering statutes (RI H 7487, building on the 2024 substantial expansion of RI §4-13-42). The three models are complementary, not alternative. The policy paper recommends an integrated approach combining equipment prohibition with trainer licensure.

State legislators may also want to know that comparable regulatory frameworks already exist for adjacent professions: veterinary medicine, mental health counseling, social work, and others. State licensure of dog trainers is not a novel regulatory category; it is an extension of existing welfare-affecting professional regulation to a profession that currently lacks it. NJ A 4207 in particular ties the licensure standard to a specific named professional standards document (the LIMA Effective Behavior Intervention Policy adopted jointly by APDT, CCPDT, and IAABC), and is a useful citable precedent for any state legislature wanting substantive standards-of-practice requirements built into the enabling statute.

For Conversations with Animal Welfare Advocates in Other States

Practitioners working with state-level animal welfare advocates who are considering legislative work in their state can use this playbook to identify the closest existing precedent for the legislative model under consideration. The advocacy community sometimes works in isolation; pointing to the comparative jurisdictional record helps locate any state-level effort within a larger international pattern. The 2024 to 2026 pending-legislation record across four states (NY, NJ, MA, RI) is particularly useful because it establishes that legislative interest in this space is active and distributed across multiple design models, not concentrated in a single state’s effort.

For Responding to the “Radical” or “Unprecedented” Framing

Aversive equipment proponents sometimes frame any United States legislation as radical or unprecedented. The jurisdictional record refutes that framing directly. The policy is not novel and it is not radical. It is the policy that has been adopted, in some cases for more than fifteen years, by jurisdictions across Europe, the United Kingdom, Australia, North America, and Latin America. The United States legislative trajectory described in the policy paper is conservative in the technical sense: it follows established international regulatory practice.

Closing line for legislative use: “The United States is not being asked to do something novel. It is being asked to do what most of the developed world has already done. The convergent welfare science, the international veterinary consensus, and the regulatory experience of more than a dozen jurisdictions all point in the same direction. The case is conservative, not radical.”

Table B. Quick-Reference Deployment Summary

This table is built for legislative testimony, op-ed writing, and conversations with policymakers. It is denser and shorter than Table A, designed to fit the comparative jurisdictional record into a format that can be scanned in seconds. Use this table to pick which two or three jurisdictions to cite for a specific argument.

Bolded entries indicate the most useful citations for legislative testimony in the United States: long-standing comprehensive bans (Wales, Switzerland), recent enactments demonstrating active legislative trajectory (France, Netherlands, Spain, Quebec, Colombia, Hawaii, Rhode Island), and the most active and structurally significant pending United States bills (NY S 7723 / A 6985 for the licensure model, NJ A 4207 for licensure tied to LIMA, MA H 2342 for equipment exclusion in dangerous-dog plans).

End of Jurisdiction Playbook

Companion to Bangura (2026), The Scientific Case Against Aversive Dog Training Equipment and Methods.

Studies Playbook

Peer-Reviewed Literature Behind the Force-Free Position, Organized by Study

Companion to The Scientific Case Against Aversive Dog Training Equipment and Methods

Bangura, April 2026

Reference document. One profile per study.

How to Use This Playbook

This document organizes the peer-reviewed literature behind the force-free position by study, with one profile per study. The policy paper organizes the same material by argument; this playbook organizes it by source. Both are useful in different contexts.

When to use this document. When an opponent says “show me a study,” when an opponent challenges a specific paper (“Cooper had problems,” “the cortisol was not significant,” “Johnson and Wynne proves you wrong”), when preparing for a podcast appearance or formal debate, when writing a response that requires citing a specific finding accurately, when teaching the literature in a continuing education context, or when mentoring a junior practitioner who is learning the evidence base.

Structure of each profile. Title and full citation with hyperlinked DOI. Design and sample in one to two sentences. Key findings. Where applicable, the proponent reading and the response. Deployment, meaning what argument the study most strongly supports. Limits and honest acknowledgments where applicable, included so the practitioner is not surprised by methodological critiques.

The honest treatment of limits matters. The practitioner who already knows a study’s weaknesses cannot be surprised by a critic who points them out. The convergent welfare evidence does not require any individual study to be perfect. It requires the agreement across studies to be robust, which it is. The pattern across the literature is sufficiently convergent to support precautionary welfare policy.

Profile groups in this document. Group A: Controlled experimental studies of aversive equipment. Group B: Direct observational and clinical studies. Group C: Population-level survey studies. Group D: Affective-state, cognitive bias, and attachment studies. Group E: Mechanical and physical effects (neck-pressure equipment). Group F: Pain neuroscience and sensory engagement. Group G: Threat circuitry, controllability, and avoidance learning, including the Pessoa and Knight personal communications. Group H: The contested study (Johnson and Wynne 2024). Group I: Proponent-cited studies and how to address them.

A note on the personal communications. The Pessoa and Knight personal communications cited in Group G are interpretive support, not primary empirical evidence. They confirm the reading of the published studies (Limbachia et al. 2021 and Wood et al. 2014) as the present author represents that reading in the policy paper. They do not introduce new empirical findings. Their value is that the senior authors of the studies most often cited by proponents have explicitly disclaimed the proponent reading of their own work.

Table A. Comprehensive Studies Summary

This table consolidates the studies covered in the playbook by group, with the design and the key finding in compact form. The Group column corresponds to the group letters used in the playbook body (A through I), allowing fast navigation from the table to the corresponding profile. Studies in Groups A through G support the force-free position; Group H is contested; Group I consists of studies cited by proponents that, on examination, do not support the proponent reading.

Bolded entries are the strongest single citations for each group: anchor studies that any practitioner should be able to name on demand. The convergence across groups (experimental, observational, survey, affective-state, mechanical, pain neuroscience, threat circuitry) is the structural feature of the evidence base; no single study carries the case alone, and the agreement across methodologies is what makes it convergent.

Group A. Controlled Experimental Studies of Aversive Equipment

Cooper, Cracknell, Hardiman, Wright, and Mills (2014)

Citation. The welfare consequences and efficacy of training pet dogs with remote electronic training collars in comparison to reward based training. PLOS ONE, 9(9), e102722. 10.1371/journal.pone.0102722

Design and sample. Randomized controlled study of 63 pet dogs presenting with recall and chasing problems. Three groups: (1) industry-nominated trainers using e-collars, (2) the same trainers without e-collars, and (3) APDT-affiliated reward-based trainers. The industry-nominated trainers were nominated by the Electronic Collar Manufacturers Association as representing best practice.

Key findings. Dogs trained without e-collars achieved equivalent or better training outcomes. The e-collar group showed significantly elevated stress-related behaviors during training, including yawning, panting, and tense body posture. The cortisol comparison did not reach significance in the larger main study (the preliminary nine-dog phase did show elevation post-stimulation, but the larger phase did not replicate this clearly).

Proponent reading. The cortisol non-significance is sometimes cited to dismiss the welfare findings, and the comparison group differences in reinforcement rate are sometimes used to argue the comparison was unfair.

Response. Cortisol is one measure and a blunt one. Behavioral stress markers are validated welfare indicators on their own, and they reached significance. The reinforcement rate critique either concedes that the e-collar added no benefit (in which case it carried welfare cost for nothing) or reduces to a hypothetical about a better protocol that the industry-nominated trainers themselves did not deliver.

Deployment. This is the anchor study for the necessity claim against electronic collars. The argument is decisive: industry-nominated, manufacturer-endorsed best-practice trainers, working on the problems most favorable to the tool (recall and chasing), did not produce better outcomes than reward-based trainers, and produced welfare harm in the process.

Limits and honest acknowledgments. The cortisol finding is real and should be acknowledged. The honest framing is that cortisol can be suppressed, lagged, or buffered by context, and a single blunt physiological measure does not erase the validated behavioral stress findings. Dissociation between behavioral and physiological stress markers is a known feature of canine welfare research, not a flaw in this study.

Cross-reference: Policy paper Section 3.1, Section 5.1, and Section 9.2.

China, Mills, and Cooper (2020)

Citation. Efficacy of dog training with and without remote electronic collars vs. a focus on positive reinforcement. Frontiers in Veterinary Science, 7, 508. 10.3389/fvets.2020.00508

Design and sample. Re-analysis and continued analysis of the Cooper 2014 dataset, focusing specifically on training efficacy across the three groups (industry e-collar trainers, same trainers without e-collars, reward-based trainers).

Key findings. Reward-based training achieved superior outcomes on multiple efficacy measures, including faster latency to sit, fewer hand and lead signals required, and faster general obedience progress. The e-collar group did not show better learning outcomes than the matched no-collar group operated by the same trainers.

Proponent reading. Sometimes ignored or dismissed as repackaging the Cooper data.

Response. It is intentional re-analysis of the same dataset focused on a different question (efficacy) than the welfare focus of the 2014 paper. The finding is decisive on the necessity claim: whatever benefit the industry-nominated trainers brought to the training was not coming from the collar, because the same trainers did not produce better outcomes when they had it.

Deployment. Use this study together with Cooper 2014 to close the necessity claim for electronic collars at the professional level. The combination tests the argument “these tools work in expert hands” directly, using the experts the industry put forward, and the argument fails.

Cross-reference: Policy paper Section 3.1 and Section 5.1.

Group B. Direct Observational and Clinical Studies

Schilder and van der Borg (2004)

Citation. Training dogs with help of the shock collar: Short and long term behavioural effects. Applied Animal Behaviour Science, 85(3-4), 319-334. 10.1016/j.applanim.2003.10.004

Design and sample. Behavioral observation study of protection-trained working dogs in guard-dog training programs (the published abstract describes the sample as German Shepherd dogs) during and after shock collar training, compared to matched dogs trained without shock.

Key findings. Shock-collar-trained dogs showed lower body postures, high-pitched yelps, avoidance behaviors, and other behavioral signs of fear and stress on shock application. Dogs that had received shocks showed more behavioral signs of fear and stress in the broader training context, even when no shock was being delivered, indicating a conditioned emotional response associated with the training context, the trainer, and the commands.

Deployment. This is the foundational study for conditioned emotional response to aversive training. Use it to counter the “the dog is fine when the collar is off” argument and to explain why aversive equipment produces welfare costs that persist beyond the moment of application.

Limits and honest acknowledgments. Working dog population, not pet dogs. Proponents will sometimes argue this limits generalizability to companion dog training. The honest acknowledgment is that the population is specific, but the underlying mechanism (conditioned emotional response to aversive stimuli) is general across mammalian learning, and the finding is corroborated by Vieira de Castro 2020 in pet populations.

Cross-reference: Policy paper Section 3.1 and Section 3.4.

Deldalle and Gaunet (2014)

Citation. Effects of 2 training methods on stress-related behaviors of the dog (Canis familiaris) and on the dog-owner relationship. Journal of Veterinary Behavior, 9(2), 58-65. 10.1016/j.jveb.2013.11.004

Design and sample. Direct observational study at two French training schools, one using negative reinforcement and one using positive reinforcement. Behavioral coding of dogs during training sessions.

Key findings. Dogs in the negative-reinforcement school showed significantly more stress-related behaviors and significantly less gaze toward the guardian during training compared to dogs in the positive-reinforcement school.

Deployment. Use this study to counter the “the dog looks happy and engaged” argument. The direct observational design rules out reliance on guardian self-report. The finding on guardian-directed gaze is particularly useful because it speaks to relationship quality, not just stress. Reduced guardian-directed gaze is one of the canonical behavioral indicators of insecure attachment and uncomfortable engagement with the handler.

Cross-reference: Policy paper Section 3.1, Table 2.

Rooney and Cowan (2011)

Citation. Training methods and owner-dog interactions: Links with dog behaviour and learning ability. Applied Animal Behaviour Science, 132(3-4), 169-177. 10.1016/j.applanim.2011.03.007

Design and sample. Home-based observational study of 53 dog-guardian pairs, assessing training methods, dog learning ability on a novel task, and behavior problems.

Key findings. Dogs of guardians who reported using more physical punishment were less playful with their guardian and interacted less with the experimenter. Dogs of guardians who reported using more rewards performed better on a novel training task.

Deployment. Use this study to counter the “punishment teaches the dog faster” argument. The novel-task design rules out the possibility that reward-based dogs are simply better-trained on practiced behaviors; they perform better on a task they have not seen before, which is the actual measure of learning ability.

Cross-reference: Policy paper Section 3.2, Table 2.

Herron, Shofer, and Reisner (2009)

Citation. Survey of the use and outcome of confrontational and non-confrontational training methods in client-owned dogs showing undesired behaviors. Applied Animal Behaviour Science, 117(1-2), 47-54. 10.1016/j.applanim.2008.12.011

Design and sample. Clinical referral survey of 140 client-owned dogs presenting to the University of Pennsylvania Veterinary Behavior Clinic, assessing training techniques previously used by guardians.

Key findings. Confrontational handling techniques produced aggressive responses at the following rates: hitting or kicking 43 percent, growling at the dog 41 percent, forcing the release of an item 39 percent, alpha roll 31 percent, staring the dog down 30 percent, dominance down 29 percent, grabbing jowls or scruff and shaking 26 percent, choke or pinch collar use 11 percent, and shock collar use 10 percent. Dogs presenting for aggression to familiar people were significantly more likely to respond aggressively to the alpha roll and to yelling “no.”

Deployment. This is the anchor study for the confrontational handling argument. Use it to establish that techniques justified by dominance theory are clinically identified independent risk factors for guardian-directed aggression, and that they are particularly contraindicated for dogs already presenting with aggression.

Limits and honest acknowledgments. The study relies on guardian report of past technique use, which has known limitations (recall bias, social desirability bias). The clinical population is also self-selected for behavior problems. The honest framing is that these limits affect prevalence estimates, not the direction of the finding. Subsequent population-level work (Casey et al. 2014) using multivariable analysis confirmed the increased odds of family-member aggression in dogs trained with aversive methods.

Cross-reference: Policy paper Section 3.2, Section 6.2, Table 2.

Group C. Population-Level Survey Studies

Hiby, Rooney, and Bradshaw (2004)

Citation. Dog training methods: Their use, effectiveness and interaction with behaviour and welfare. Animal Welfare, 13(1), 63-69. 10.1017/S0962728600026683

Design and sample. Survey of 364 UK dog guardians assessing training methods, obedience ratings, and behavior problems.

Key findings. Reward-based methods correlated with higher reported obedience scores. Punishment-based methods correlated with significantly more behavior problems. The number of behaviors taught using rewards correlated with obedience; the number taught using punishment correlated with problem behaviors.

Deployment. Foundational early population-level study. Use it as the historical anchor: the survey-level association between aversive methods and worse behavioral outcomes has been documented in the literature for more than two decades.

Cross-reference: Policy paper Section 3.2, Table 2.

Blackwell, Twells, Seawright, and Casey (2008)

Citation. The relationship between training methods and the occurrence of behavior problems, as reported by owners, in a population of domestic dogs. Journal of Veterinary Behavior, 3(5), 207-217. 10.1016/j.jveb.2007.10.008

Design and sample. Survey of 192 UK dog guardians assessing training methods and behavior problem occurrence.

Key findings. Guardians using punishment-based training reported significantly higher rates of behavior problems including aggression. The highest aggression scores were reported in dogs whose guardians combined positive reinforcement with positive punishment (the methodology often labeled “balanced”).

Deployment. Use this study to counter the balanced training claim that mixing rewards with punishment produces moderate, considered training. The empirical finding is that the combination is associated with the highest aggression scores in this sample. Adding rewards to a punishment-based protocol does not neutralize the welfare or behavioral cost.

Cross-reference: Policy paper Section 3.2, Table 2.

Arhant, Bubna-Littitz, Bartels, Futschik, and Troxler (2010)

Citation. Behaviour of smaller and larger dogs: Effects of training methods, inconsistency of owner behaviour and level of engagement in activities with the dog. Applied Animal Behaviour Science, 123(3-4), 131-142. 10.1016/j.applanim.2010.01.003

Design and sample. Survey of 1,276 Austrian dog guardians, with smaller and larger dogs analyzed separately.

Key findings. High-frequency aversive training correlated with increased aggression, excitability, and anxiety. Reward-based training correlated with higher obedience without those side effects. The pattern held across both smaller and larger dogs.

Deployment. Use this study to counter “this only applies to small dogs” or “this only applies to large dogs.” The size-stratified analysis rules out the proponent escape route that breed or size moderates the welfare cost.

Cross-reference: Policy paper Section 3.2, Table 2.

Blackwell, Bolster, Richards, Loftus, and Casey (2012)

Citation. The use of electronic collars for training domestic dogs: Estimated prevalence, reasons and risk factors for use, and owner perceived success as compared to other training methods. BMC Veterinary Research, 8, 93. 10.1186/1746-6148-8-93

Design and sample. Survey of 3,897 UK dog guardians focused specifically on electronic collar use.

Key findings. Electronic collar users reported lower training success than reward-based trainers for comparable problems. Guardian attendance at training classes and guardian gender were the strongest predictors of e-collar use, not dog characteristics.

Deployment. Use this study to counter the claim that e-collars are reserved for dogs that genuinely need them. Guardian characteristics (specifically, source of training advice or absence of advice) drive equipment selection more than dog characteristics. The tool is selected by the human, not chosen by the dog’s problem.

Cross-reference: Policy paper Section 8.1, Table 2.

Casey, Loftus, Bolster, Richards, and Blackwell (2014)

Citation. Human directed aggression in domestic dogs (Canis familiaris): Occurrence in different contexts and risk factors. Applied Animal Behaviour Science, 152, 52-63. 10.1016/j.applanim.2013.12.003

Design and sample. Population-level survey with multivariable analysis of risk factors for human-directed aggression in domestic dogs.

Key findings. Adjusted increased odds of family-member aggression in dogs whose guardians used aversive methods, after controlling for confounding variables.

Deployment. Use this study to corroborate Herron et al. (2009) at the population level. The Herron clinical data are sometimes critiqued for sample selection; the Casey 2014 multivariable population analysis answers that critique by reaching the same direction of finding with appropriate statistical controls.

Cross-reference: Policy paper Section 3.2.

Masson, Nigron, and Gaultier (2018b)

Citation. Questionnaire survey on the use of different e-collar types in France in everyday life with a view to providing recommendations for possible future regulations. Journal of Veterinary Behavior, 26, 48-60. 10.1016/j.jveb.2018.05.004

Design and sample. Survey of 1,251 French respondents about electronic collar use, including acquisition patterns, application contexts, and reported outcomes.

Key findings. Among e-collar users, 71.8 percent operated the equipment without professional advice. 75 percent had tried two or fewer alternative methods before reaching for the collar. 7 percent of dogs on which collars had been used presented with physical wounds.

Deployment. This is the anchor study for the real-world use argument. Use it to establish that the actual exposure profile in the population (lay use, minimal alternative-method exposure, documented physical wounds) is materially different from the experimentally idealized version of e-collar use that proponents reference. Policy that addresses only the idealized version ignores the version that exists in the marketplace.

Cross-reference: Policy paper Section 8.2, Table 2.

Starinsky, Lord, and Herron (2017)

Citation. Escape rates and biting histories of dogs confined to their owner’s property through the use of various containment methods. Journal of the American Veterinary Medical Association, 250(3), 297-302. 10.2460/javma.250.3.297

Design and sample. Survey of 974 US dog guardians comparing containment methods (electronic fence, physical fence, and tethering) and outcomes.

Key findings. Escape rates by containment method: electronic fence 44 percent, physical fence 23 percent, tethered 27 percent. Electronic fences did not produce a clear protective effect on bite or escape outcomes compared to physical fencing.

Deployment. Use this study to counter the “electronic containment improves safety” argument. Physical fencing outperforms electronic on the available data. This is also the best peer-reviewed reference for guardians weighing physical fencing against electronic boundary systems.

Limits and honest acknowledgments. The study is survey-based and relies on guardian self-report of escape and bite incidents. The honest framing is that the magnitude of the difference (44 percent versus 23 percent) is large enough to be informative even granting the limits of guardian recall, and the direction of the finding is the opposite of what the proponent reading would predict.

Cross-reference: Policy paper Section 3, Table 2.

Group D. Affective-State, Cognitive Bias, and Attachment Studies

Vieira de Castro, Barrett, de Sousa, and Olsson (2019)

Citation. Carrots versus sticks: The relationship between training methods and dog-owner attachment. Applied Animal Behaviour Science, 219, 104831. 10.1016/j.applanim.2019.104831

Design and sample. Survey and observational study using a Strange Situation Procedure adapted for dogs to assess attachment patterns toward the guardian.

Key findings. Aversive-based training was associated with significantly weaker dog-guardian attachment compared to reward-based training. Secure attachment patterns were more consistently observed in reward-trained dogs.

Deployment. Use this study to counter the “aversive training builds respect” or “stronger relationship” claim. The Strange Situation Procedure is methodologically more rigorous than guardian self-report and is the standard tool in attachment research. Aversive methods produce weaker, not stronger, attachment.

Cross-reference: Policy paper Section 3.2, Table 2.

Vieira de Castro, Fuchs, Morello, Pastur, de Sousa, and Olsson (2020)

Citation. Does training method matter? Evidence for the negative impact of aversive-based methods on companion dog welfare. PLOS ONE, 15(12), e0225023. 10.1371/journal.pone.0225023

Design and sample. Multi-measure welfare study of 92 pet dogs from 7 Portuguese training schools (3 reward-based, 4 aversive-based of which 2 were mixed and 2 were high-aversive). Behavioral observation, salivary cortisol, and cognitive bias testing.

Key findings. Dogs in aversive-based schools showed significantly more stress behaviors during training, significantly higher post-training cortisol, and significantly more pessimistic cognitive bias. Dogs in mixed-method schools also showed significantly more stress behaviors and panted more than reward-only schools, although the cortisol difference reached significance only for the high-aversive group.

Deployment. This is the strongest single welfare study because it combines behavioral, physiological, and affective-state measures and finds convergence across all three. Use it as the anchor for the argument that aversive training produces welfare cost across multiple independent indicators, not just one.

Limits and honest acknowledgments. The cognitive-bias finding was specific to the high-aversive group, with the mixed group showing behavioral and physiological cost but not the cognitive-bias effect at significance. Honest framing: the welfare cost is dose-dependent, with higher aversive exposure producing the broader pattern. This does not exonerate the mixed protocol; it locates the cognitive-bias effect at the highest exposure level.

Cross-reference: Policy paper Section 3.2, Table 2.

Casey, Naj-Oleari, Campbell, Mendl, and Blackwell (2021)

Citation. Dogs are more pessimistic if their owners use two or more aversive training methods. Scientific Reports, 11, 19023. 10.1038/s41598-021-97743-0

Design and sample. Cognitive bias test (judgment bias paradigm) of 104 dogs across training method categories.

Key findings. Dogs trained with two or more aversive methods showed significantly more pessimistic cognitive bias than dogs trained with reward-based methods. Cognitive bias is a validated indicator of persistent affective state.

Deployment. Use this study to counter the “the welfare effect is short-term” or “the dog gets over it” argument. Cognitive bias measures persistent affective state outside the training context, meaning the welfare cost extends beyond the training session itself. This is also the cleanest demonstration that the cumulative-exposure argument is empirically supported.

Cross-reference: Policy paper Section 3.1, Section 3.5, Table 2.

Group E. Mechanical and Physical Effects (Neck-Pressure Equipment)

Important scope note. The studies in this group apply specifically to neck-pressure equipment (prong and choke collars). They do not apply to electronic collars, which operate by a different mechanism. The welfare case against electronic collars rests on nociception and threat-circuit engagement (Groups F and G), not on mechanical injury. Conflating the two arguments is a common practitioner mistake.

Pauli, Bentley, Diehl, and Miller (2006)

Citation. Effects of the application of neck pressure by a collar or harness on intraocular pressure in dogs. Journal of the American Animal Hospital Association, 42(3), 207-211. 10.5326/0420207

Design and sample. Measurement of intraocular pressure in 51 eyes of 26 dogs while the dogs pulled against a collar or a harness.

Key findings. Intraocular pressure rose significantly from baseline when pressure was applied via a collar, but not when equivalent pressure was applied via a harness. The proposed mechanism is ventral neck pressure compressing the jugular veins and obstructing ocular aqueous outflow.

Deployment. Use this study to establish that ordinary collar use under pull produces measurable physiological consequences in healthy dogs, not only in dogs with pre-existing ocular disease. The study explicitly recommends harnesses over collars for dogs with weak or thin corneas, glaucoma, or any condition where elevated intraocular pressure could be harmful.

Limits and honest acknowledgments. The study tested flat collars, not prong or choke collars specifically. The honest framing is that prong and choke collars concentrate or constrict force in ways that flat collars do not, which on the underlying mechanism would be expected to produce equal or greater elevation. The peer-reviewed literature has not yet directly tested prong or choke collars against the same protocol, and the burden of demonstrating mechanical safety lies with the manufacturers, not with the welfare science community.

Cross-reference: Policy paper Section 4.2.

Carter, McNally, and Roshier (2020)

Citation. Canine collars: An investigation of collar type and the forces applied to a simulated neck model. Veterinary Record, 187(7). 10.1136/vr.105681

Design and sample. Tested seven collar types and a slip lead on a simulated canine neck model with a pressure sensor beneath the collar. Force levels: firm pull (40 N), strong pull (70 N), and jerk (141 N average).

Key findings. Collars produced pressures between 83 and 832 kilopascals on the model neck. Collar type and applied force each had significant effects on the pressure delivered. The authors concluded that no collar tested produced a pressure low enough to mitigate the risk of injury when the dog pulls on the lead.

Deployment. Use this study to establish that the pressures produced by ordinary on-leash equipment, under pull forces representative of real-world handling, fall in injury-relevant ranges. The conclusion that no collar tested produces injury-mitigating pressures is decisive on the equipment-design question.

Limits and honest acknowledgments. Simulated neck model rather than live dogs. Some proponents will use this to dismiss the finding. Honest framing: the model standardizes the mechanical question (what pressure is delivered to the neck under specified force) in a way that live-dog measurement cannot. The mechanical finding is robust; what live-dog studies could add is biological response, not mechanical force calibration.

Cross-reference: Policy paper Section 4.2.

Hunter, Blake, and De Godoy (2019)

Citation. Pressure and force on the canine neck when exercised using a collar and leash. Veterinary and Animal Science, 8, 100082. 10.1016/j.vas.2019.100082

Design and sample. Measurement of force and pressure on the canine neck during ordinary on-leash walking using different collar constructions.

Key findings. Peak contact pressure reached 44.61 newtons per square centimeter, with significant differences in how different collar constructions transmit force to the neck.

Deployment. Use this study together with Carter 2020 to establish that ordinary collar use during ordinary walking transmits substantial pressure to the canine neck, not only during training corrections.

Cross-reference: Policy paper Section 4.2.

Grohmann, Dickomeit, Schmidt, and Kramer (2013)

Citation. Severe brain damage after punitive training technique with a choke chain collar in a German shepherd dog. Journal of Veterinary Behavior, 8(3), 180-184. 10.1016/j.jveb.2013.01.002

Design and sample. Peer-reviewed case report of a 1-year-old German Shepherd subjected to a punitive training technique in which the guardian lifted the dog off the ground by the choke chain.

Key findings. The dog initially appeared normal, then became progressively ataxic, began circling to the left, and showed reduced consciousness. Magnetic resonance imaging showed multifocal T2 and diffusion-weighted changes consistent with severe cerebral edema from ischemia. The injury mechanism was carotid artery compression producing cerebral hypoxia. Because of the severity of the neurological findings, the dog was euthanized.

Deployment. This is the documented case-report endpoint of the mechanical-injury argument. Use it to establish that punitive choke-chain techniques can produce fatal outcomes, not as a hypothetical risk but as a peer-reviewed clinical fact.

Limits and honest acknowledgments. Single case report. Proponents will sometimes argue this is unrepresentative. Honest framing: case reports do not establish prevalence; they establish that the injury mechanism is real and has been documented to produce the outcome of euthanasia. The peer-reviewed literature does not need additional cases to establish the mechanism.

Cross-reference: Policy paper Section 4.2.

Rozanski (2022)

Citation. Tracheal collapse. Today’s Veterinary Practice, February 2022. https://todaysveterinarypractice.com/respiratory-medicine/tracheal-collapse/

Design and sample. Clinical veterinary review of tracheal collapse in dogs.

Key findings. Repeated collar pressure is recognized in the clinical veterinary literature as a concern for tracheal collapse, and harnesses are commonly recommended in place of collars for dogs diagnosed with tracheal collapse. Cough induced by collar pressure is a recognized diagnostic feature of the condition in veterinary medicine.

Deployment. Use this clinical reference together with Pauli 2006, Carter 2020, and Hunter 2019 to establish that the recommendation to use a harness rather than a collar, particularly for dogs with diagnosed or suspected tracheal collapse, is mainstream veterinary clinical practice. This is not advocacy; it is standard of care.

Cross-reference: Policy paper Section 4.2.

Group F. Pain Neuroscience and Sensory Engagement

Important context note. The studies in this group support the welfare case at the level of biology. They establish that aversive training equipment crosses the nociceptive threshold by design, that the human and canine sensory anatomies are not equivalent, and that the relationship between user setting and delivered stimulus on a typical e-collar is not standardized in ways that would justify the proponent appeal to a low intensity setting. This group answers the proponent argument that low-level stimulation is mild, predictable, or comparable to a TENS unit at the level the argument is actually made: sensation severity, not tissue damage.

Dubin and Patapoutian (2010)

Citation. Nociceptors: The sensors of the pain pathway. Journal of Clinical Investigation, 120(11), 3760-3772. 10.1172/JCI42843

Design and sample. Peer-reviewed review of nociceptor neurobiology in the Journal of Clinical Investigation.

Key findings. Nociceptors are specialized peripheral sensory neurons that detect potentially damaging stimuli at the skin, including extremes of temperature, pressure, chemical, and electrical signals, and transduce these stimuli into neural signals carried to higher brain centers. Critically, the system fires below the threshold of actual tissue injury. Its biological function is to warn the organism away from potentially harmful events before damage occurs.

Proponent reading. The proponent argument that low-level stimulation does not cause tissue damage and is therefore welfare-neutral assumes that nociception requires injury to engage. The neuroscience does not support that assumption.

Response. Nociceptors do not require tissue damage to fire. C-fiber and A-delta fiber nociceptors respond to electrical, mechanical, thermal, and chemical stimulation at intensities well below any injury threshold. The system exists precisely to warn the organism before damage occurs. A stimulus that crosses the nociceptive threshold engages the welfare-relevant neural machinery whether or not tissue damage results.

Deployment. Use this paper as the anchor reference for the nociception argument. The welfare case against aversive equipment does not require a showing of tissue damage. It rests on the biological reality that the equipment is functionally calibrated to deliver a stimulus the dog experiences as unpleasant enough to change its behavior, and unpleasant enough to change behavior through avoidance, escape, or suppression is, by definition, noxious.

Cross-reference: Policy paper Section 4.1.

Raja et al. (2020)

Citation. The revised International Association for the Study of Pain definition of pain: Concepts, challenges, and compromises. Pain, 161(9), 1976-1982. 10.1097/j.pain.0000000000001939

Design and sample. Revised International Association for the Study of Pain definition; published in Pain (the leading journal in the field), authored by the multidisciplinary IASP Task Force.

Key findings. Pain is defined as an unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage. The phrase “potential tissue damage” is intentional and core to the definition. The IASP also affirms that the definition applies to both human and nonhuman animals.

Proponent reading. Some proponent arguments equate “pain” with “tissue injury,” treating the absence of injury as the absence of welfare cost.

Response. The international scientific definition of pain explicitly includes potential tissue damage. The phrase was added precisely because pain is a warning signal, not an injury report. A stimulus delivered at an intensity that the canine peripheral nervous system encodes as noxious meets the IASP definition of pain whether or not tissue damage follows.

Deployment. Pair this study with Dubin and Patapoutian (2010) when an opponent reduces the welfare argument to a tissue-damage argument. The international scientific definition is on the welfare side of the debate. The proponent reading conflicts with the consensus definition of the field that studies pain.

Cross-reference: Policy paper Section 4.1.

Affolter and Moore (1994)

Citation. Histologic features of normal canine and feline skin. Clinics in Dermatology, 12(4), 491-497. 10.1016/0738-081X(94)90215-1

Design and sample. Peer-reviewed review of histologic features of canine and feline skin in Clinics in Dermatology, organized by anatomic region.

Key findings. Canine haired-skin epidermis is approximately three to five cell layers thick, considerably thinner than human epidermis. The cutaneous structures that mediate nociception in dogs are anatomically distinct from human cutaneous anatomy.

Proponent reading. Self-test demonstrations on the human forearm, palm, or wrist are commonly cited by proponents to argue that low-level e-collar stimulation is a mild sensation and therefore welfare-neutral for dogs.

Response. Human and canine cutaneous anatomy are not equivalent. The same delivered electrical or mechanical energy will reach deeper structures in canine skin than it does in human skin, simply because the canine epidermis is a fraction of the thickness. A self-test that crosses the human nociceptive threshold at one intensity will reach deeper canine tissue at the same delivered energy. Self-testing on human skin systematically underestimates what the canine nervous system receives.

Deployment. Use this study whenever a proponent invokes a self-test demonstration on their own forearm or wrist as evidence that the stimulus is mild. The anatomical comparison is not in the proponent’s favor.

Cross-reference: Policy paper Section 4.1, Section 2.4.

Lines, van Driel, and Cooper (2013)

Citation. Characteristics of electronic training collars for dogs. Veterinary Record, 172(11), 288. 10.1136/vr.101144

Design and sample. Engineering measurement study of electrical characteristics of thirteen commercially available electronic training collar models in the United Kingdom; impedance measurements on twenty-seven dogs; published in the Veterinary Record.

Key findings. Stimulus energy at maximum settings ranged from 3.3 millijoules to 287 millijoules at a 50 kilohm load representative of canine neck impedance, an eighty-seven-fold range across products. Within a single collar, the median maximum-to-minimum energy ratio across the available strength settings was 81, with individual collars ranging from 8 to 1,114. Two of thirteen new collars examined contained manufacturing faults; in one case the collar could deliver a maximum-strength impulse regardless of the user’s setting. The authors reported that user-disclosed comparison data such as voltage, pulse parameters, and waveform are not available at the point of sale. The authors concluded that a given strength setting cannot be assumed to deliver a similar stimulus across collar models or brands.

Proponent reading. Modern e-collars are commonly defended as operating at low stimulation levels, with skilled trainers using the minimum effective setting and the dog feeling something no worse than a tap on the shoulder.

Response. From the canine nervous system perspective, what determines whether the nociceptive threshold is crossed is the actual electrical signal at the skin, not the user’s intensity setting. With an eighty-seven-fold range across products, an eighty-one-fold median ratio within products, manufacturing faults documented in two of thirteen new collars, and no point-of-sale disclosure of stimulus parameters, the proponent appeal to a low intensity setting, even granting good faith user technique, is not informative about the welfare-relevant question of whether the stimulus is noxious to the dog.

Deployment. This is the strongest single citation for the consumer-protection argument and for the “it’s just a tingle on a low setting” rebuttal. The fact pattern, particularly the manufacturing-fault rate and the absence of point-of-sale disclosure, is also legislatively useful: it speaks directly to the regulatory vacuum and is the kind of evidence that lands with legislators evaluating consumer access.

Cross-reference: Policy paper Section 4.1, Section 8.3.

Group G. Threat Circuitry, Controllability, and Avoidance Learning

Important context note. The studies in this group support the welfare case at the level of fear and threat neuroscience. They establish that controllability and predictability attenuate, but do not eliminate, threat-related neural responding; that active avoidance is a goal-directed instrumental behavior under threat rather than a welfare-neutral compliance state; and that overtrained avoidance becomes habitual via the same dorsolateral striatum circuit implicated in obsessive-compulsive disorder and in the stronger avoidance habits observed in survivors of early life stress. This group answers the proponent argument that controllable, predictable aversive stimulation is welfare-neutral. The Pessoa and Knight personal communications in this group are interpretive support, not primary empirical evidence; they confirm the reading of the published studies as represented in this paper. Both senior authors of the studies most often cited by proponents have explicitly disclaimed the proponent reading of their own work.

LeDoux (2014)

Citation. Coming to terms with fear. Proceedings of the National Academy of Sciences, 111(8), 2871-2878. 10.1073/pnas.1400335111

Design and sample. Theoretical and review paper on the neuroscience of threat processing, fear conditioning, and the distinction between defensive-circuit activation and the conscious experience of fear.

Key findings. The amygdala and connected circuits respond to predicted aversive events, encoding threat associations and driving avoidance learning. LeDoux distinguishes defensive-circuit activation from the conscious experience of fear, but does not characterize circuit activation as welfare-neutral.

Deployment. Use this paper as the anchor reference for the threat-circuit argument. The neural circuitry that responds to aversive training events is the same circuitry that processes threat in mammalian models generally. This is not extrapolation; it is the core of the field.

Cross-reference: Policy paper Section 4.3.

Cain (2019)

Citation. Avoidance problems reconsidered. Current Opinion in Behavioral Sciences, 26, 9-17. 10.1016/j.cobeha.2018.09.002

Design and sample. Review of contemporary active avoidance research.

Key findings. Active avoidance is goal-directed instrumental behavior under threat. The shift from a fear state to an anxiety state during effective avoidance does not eliminate the underlying threat representation. The warning stimulus retains its conditioned threat value; what changes is that the animal has acquired a behavioral option that controls exposure to the aversive event. When the avoidance response is blocked or fails, the fear state returns along with the inflexible defensive reactions characteristic of fear.

Deployment. Use this paper to counter the “the dog looks calm and engaged during e-collar work, so the welfare cost is gone” argument. The contemporary fear-and-avoidance literature has moved beyond the older framing of avoidance as a reflexive, fear-driven response. Avoidance is goal-directed; the calm-looking dog is in an anxiety state mediated by an effective avoidance response, not in the absence of threat representation. The welfare cost is in the underlying threat representation that the warning stimulus retains throughout.

Cross-reference: Policy paper Section 4.3.

Maier and Watkins (2005)

Citation. Stressor controllability and learned helplessness: The roles of the dorsal raphe nucleus, serotonin, and corticotropin-releasing factor. Neuroscience and Biobehavioral Reviews, 29(4-5), 829-841. 10.1016/j.neubiorev.2005.03.021

Design and sample. Review of stressor controllability research integrating dorsal raphe, serotonergic, and corticotropin-releasing factor system findings across decades of work.

Key findings. Controllability modulates downstream consequences of aversive stressors but does not render the stressor benign or stress-free. The animal recruits stress-system machinery in response to controllable aversive events. What controllability attenuates is particular sequelae, including the spread of activation into prefrontal regions that produces the broader behavioral signature of learned helplessness.

Deployment. Use this paper to counter the “if the dog can control the aversive by behavioral compliance, the aversive is no longer welfare-relevant” argument. The neurobiological literature establishes a robust modulation finding: controllable aversive events produce a different downstream profile than uncontrollable aversive events. The work does not establish that controllable aversive events are stress-free or welfare-neutral. The aversive remains aversive.

Cross-reference: Policy paper Section 4.3.

Limbachia, Morrow, Khibovska, Meyer, Padmala, and Pessoa (2021)

Citation. Controllability over stressor decreases responses in key threat-related brain areas. Communications Biology, 4, 42. 10.1038/s42003-020-01537-5

Design and sample. Functional MRI study of human participants in conditions varying in controllability over aversive stimulation.

Key findings. When participants had control over aversive stimulation, the magnitude of threat-related neural responding was attenuated compared to uncontrollable aversive conditions, but not eliminated. The threat circuitry continues to respond; it responds less strongly.

Deployment. Use this study to address the proponent argument that controllability or predictability renders aversive stimulation welfare-neutral. The peer-reviewed neuroscience says attenuation, not elimination. Reduced response is not the absence of fear or stress.

Limits and honest acknowledgments. Human participants, not dogs. Proponents will sometimes argue this limits generalizability. Honest framing: the threat-circuit machinery (amygdala, ACC, periaqueductal gray, locus coeruleus) is conserved across mammals, and the principle of attenuation-without-elimination is the established finding in the field. The senior author has confirmed in writing that his research cannot be used to support the proponent reading (see Pessoa entry below).

Cross-reference: Policy paper Section 4.3.

Pessoa (Personal Communication, 2026)

Citation. Personal communication with Will Bangura regarding the interpretation of Limbachia et al. (2021) and the aversive character of controllable stressors. On file with author.

Nature of the citation. Direct written correspondence with the senior author of Limbachia et al. (2021). This is interpretive support, not primary empirical evidence. It does not introduce new empirical findings; it confirms the reading of the published study as represented in this paper.

Substance of the communication. Dr. Pessoa confirmed that the attenuation finding in his published work should not be interpreted as rendering controllable aversive stimulation neurologically neutral or welfare-benign. Controllability reduces the magnitude of threat-related neural responding; it does not eliminate it.

Deployment. Use the Pessoa communication to settle interpretive disputes about Limbachia et al. (2021) when the study is invoked in support of the controllability defense. The senior author has explicitly disclaimed the proponent interpretation. Note that this is interpretive support: the empirical finding lives in the published study, and the proponent reading conflicts with the published study and the senior author’s own statement of how the study should be read.

Cross-reference: Policy paper Section 4.3 and Section 9.4.

Wood, Ver Hoef, and Knight (2014)

Citation. The amygdala mediates the emotional modulation of threat-elicited skin conductance response. Emotion, 14(4), 693-700. 10.1037/a0036636

Design and sample. Functional MRI and skin conductance study of threat-elicited responses and their emotional modulation in human participants.

Key findings. The amygdala mediates the emotional modulation of threat-elicited responses, situating the amygdala as the key node through which aversive stimulation produces emotional response.

Deployment. Use this paper together with Limbachia et al. (2021) to establish the neuroscientific foundation for the threat-circuit argument. The senior author has confirmed in writing that his research cannot be used to support the proposition that predictable aversive stimulation is neutral or benign (see Knight entry below).

Cross-reference: Policy paper Section 4.3.

Knight (Personal Communication, 2026)

Citation. Personal communication with Will Bangura regarding the interpretation of Wood et al. (2014) and the aversive character of predictable threat. On file with author.

Nature of the citation. Direct written correspondence with the senior author of Wood et al. (2014). This is interpretive support, not primary empirical evidence. It does not introduce new empirical findings; it confirms the reading of the published study as represented in this paper.

Substance of the communication. Dr. Knight confirmed that his research, including the broader research program on conditioned diminution of unconditioned responses, cannot be used to support the proposition that predictable aversive stimulation is neurologically neutral or benign.

Deployment. Same use as the Pessoa communication. Together, the Pessoa and Knight communications close the interpretive disputes around the neuroscience-based defense of controllable or predictable aversive use. Both senior authors of the studies most often cited by proponents have explicitly disclaimed the proponent reading of their own work. Note that the underlying empirical findings live in the published studies; the personal communications are interpretive support, useful to settle a reading dispute, not to introduce a new empirical claim.

Cross-reference: Policy paper Section 4.3 and Section 9.4.

Sears, Andrade, Samels, Laughlin, Moloney, Wilson, Alwood, Moscarello, and Cain (2026)

Citation. Devaluation of response-produced safety signals reveals circuits for goal-directed versus habitual avoidance in dorsal striatum. Nature Communications, 17, 2542. 10.1038/s41467-026-69119-3

Design and sample. Shuttlebox active avoidance with rats; novel safety-signal devaluation procedure; chemogenetic suppression of dorsomedial and dorsolateral striatum; both sexes tested, with the devaluation effect reported in males.

Key findings. Active avoidance is positively reinforced by response-produced feedback cues that the brain transforms into safety signals through their inverse relationship with the aversive event. Moderately trained avoidance is goal-directed and depends on posterior dorsomedial striatum. Overtrained avoidance becomes habitual, depends on dorsolateral striatum, and is insensitive to safety-signal devaluation. The same overtrained-habit circuit is implicated in obsessive-compulsive disorder (Gillan et al., 2014) and in survivors of early life stress (Gordon et al., 2020).

Proponent reading. A sophisticated proponent argument runs that the dog’s avoidance behavior in shock-collar training is positively reinforced by the safety signal (the absence of shock), and therefore the training is not actually aversive; the dog is working for safety, not avoiding pain.

Response. The argument inverts the structure of the learning architecture. Safety signals acquire their value entirely from their inverse relationship with the aversive contingency. Without the aversive event, no warning stimulus acquires threat value, no feedback cue acquires safety value, and the avoidance response is not reinforced. The aversive contingency is the precondition for the entire learning architecture. To describe avoidance as positively reinforced by safety, in the technical sense Sears and colleagues use, is not to claim that the underlying training regime was not aversive. Additionally, the same dorsolateral striatum-mediated habitual circuit is the substrate of clinically maladaptive avoidance in obsessive-compulsive disorder and in survivors of early life stress. The argument is not that aversive training causes obsessive-compulsive disorder in dogs. The argument is that the learning architecture aversive-based training depends on, particularly under prolonged or overtrained conditions, is the same architecture implicated in clinical populations as the substrate for persistent maladaptive avoidance.

Deployment. This is the strongest available counter to the most sophisticated proponent argument in current circulation. Use it together with Cain (2019) and the Limbachia/Pessoa and Wood/Knight pairs to close the controllability and predictability defenses at the level of contemporary neuroscience.

Cross-reference: Policy paper Section 4.3 and Section 9.6.

Christiansen, Bakken, and Braastad (2001)

Citation. Behavioural changes and aversive conditioning in hunting dogs by the second-year confrontation with domestic sheep. Applied Animal Behaviour Science, 72(2), 131-143. 10.1016/S0168-1591(00)00203-3

Design and sample. Field study of remote shock collar use in 114 hunting dogs (Norwegian elkhounds, English setters, and hare hunting dogs) across two consecutive years of pasture confrontation testing with sheep, under conditions designed to maximize controllability and predictability of the aversive stimulus.

Key findings. The authors’ own welfare measures were limited, relying largely on guardian report and temperament tests, and the study did not detect a significant fear or anxiety effect using those measures.

Proponent reading. Sometimes invoked by proponents to argue that controllable, predictable shock-collar use in field conditions is welfare-benign.

Response. The methodological thinness of the welfare assessment, rather than a clean positive welfare conclusion, is what prevents the data set from supporting a welfare-benign reading. The study did not include physiological stress measures, did not include direct behavioral coding of stress markers, and did not include cognitive bias or other validated affective-state measures. Absence of evidence under thin measures is not evidence of welfare neutrality. The authors themselves recommend the device be used only for the specific purpose of preventing livestock attack, in connection with positive reinforcement, which is not how the equipment is sold or used in consumer markets.

Deployment. When this study is invoked, point to the methodological limits of the welfare assessment in the original paper, and to the authors’ own restricted recommendation. Even granting the Christiansen finding for its specific specialized context, the finding does not transfer to consumer marketplace conditions.

Cross-reference: Policy paper Section 4.3 and Section 9.13.

Group H. The Contested Study

Johnson and Wynne (2024)

Citation. Comparison of the efficacy and welfare of different training methods in stopping chasing behavior in dogs. Animals, 14(18), 2632. 10.3390/ani14182632

Design and sample. Comparison of training methods for stopping predatory chasing behavior in dogs. The study examined a narrow problem profile under specific experimental conditions.

Key findings. The authors reported efficacy under specific protocol conditions for the e-collar training method. The study reaches a narrow efficacy conclusion under the specific protocol conditions tested. It does not establish necessity, welfare neutrality, or broad real-world superiority of e-collar training.

Proponent reading. Frequently cited as decisive evidence that electronic collars are necessary or uniquely effective for predatory chasing intervention, and sometimes extended into broader claims about e-collar superiority across training contexts.

Response. The study establishes narrow efficacy under specific protocol conditions, not necessity, welfare neutrality, or broad real-world superiority. Its protocol design has been challenged in the peer-reviewed literature by Bastos, Warren, and Krupenye (2024), with specific methodological concerns about the reward-based comparison condition, the duration of the training trial, and the baseline comparability of groups. The present author’s separately published methodological critique (Bangura, 2025, SSRN) raises additional concerns about internal validity and generalizability to general pet dog populations. The authors’ published response to Bastos (Johnson and Wynne, 2025) did not resolve the substantive methodological concerns about the original protocol. A single contested efficacy finding under narrow experimental conditions is not a foundation for policy that grants broad consumer access to a device whose mechanism engages nociception and threat circuitry.

Deployment. When this study is invoked, do not concede the proponent reading. Reframe to what the study actually establishes (narrow efficacy under specific conditions), reference the Bastos peer-reviewed critique, reference the SSRN methodological critique, and note that the authors’ 2025 response did not resolve the substantive methodological concerns. The narrow efficacy finding does not extend to necessity, welfare neutrality, or broad real-world superiority, and the policy question is not whether the protocol can produce a behavioral outcome under tightly controlled conditions but whether the equipment should be available in the consumer marketplace where the protocol conditions are not present.

Cross-reference: Policy paper Section 5.3 and Section 9.13.

Bastos, Warren, and Krupenye (2024)

Citation. What evidence can validate a dog training method? Learning and Behavior, 53(2), 227-228. (Advance online publication, November 20, 2024.) 10.3758/s13420-024-00658-9

Design and sample. Peer-reviewed methodological critique of Johnson and Wynne (2024).

Key findings. Identifies specific methodological concerns about the reward-based comparison condition, the duration of the training trial, and the baseline comparability of groups.

Deployment. Use this paper together with the Bangura (2025) SSRN methodological critique whenever Johnson and Wynne 2024 is invoked. The peer-reviewed critique establishes that the methodological concerns about Johnson and Wynne are not idiosyncratic; they are recognized in the literature.

Cross-reference: Policy paper Section 5.3 and Section 9.13.

Johnson and Wynne (2025) Authors’ Response

Citation. Chasing solutions: A response to Bastos et al. (2024). Learning and Behavior. Advance online publication. 10.3758/s13420-025-00672-5

Design and sample. Authors’ published response to Bastos, Warren, and Krupenye (2024).

Key findings. The response did not resolve the substantive methodological concerns raised by Bastos et al. about the original protocol.

Deployment. Mention the existence of the response when Johnson and Wynne 2024 is invoked, so that an opponent cannot claim the methodological concerns were addressed in print. They were not. Note that the published exchange between Bastos and Johnson/Wynne is fully available in the peer-reviewed literature; the methodological concerns survive the exchange.

Cross-reference: Policy paper Section 5.3 and Section 9.13.

Bangura (2025) SSRN Methodological Critique

Citation. A critical evaluation of Johnson and Wynne’s (2024) methodology in comparison of the efficacy and welfare of different training methods in stopping chasing behavior in dogs. SSRN, posted February 24, 2025. 10.2139/ssrn.5154127

Design and sample. Separately published methodological critique by the present author.

Key findings. Raises additional concerns about internal validity and generalizability of Johnson and Wynne (2024) to general pet dog populations. Note that SSRN posting is preprint-style hosting; the critique is publicly available and citable but is not itself peer-reviewed.

Deployment. Use this critique alongside the Bastos peer-reviewed critique when Johnson and Wynne 2024 is invoked. The two critiques cover different methodological concerns and together provide a more complete account of why the study does not support the proponent reading.

Cross-reference: Policy paper Section 5.3 and Section 9.13.

Group I. Proponent-Cited Studies and How to Address Them

These are the studies most commonly cited by balanced trainers in defense of aversive equipment. The pattern across them is consistent: the proponent reading typically depends on selective extraction of a single finding from a larger study whose overall conclusions point in a different direction, extension of a narrow methodological finding into broad claims the study does not support, or citation of older work that predates contemporary canine welfare science methodology.

Salgirli, Schalke, Boehm, and Hackbarth (2012)

Citation. Comparison of learning effects and stress between 3 different training methods (electronic training collar, pinch collar and quitting signal) in Belgian Malinois police dogs. Revue de Médecine Vétérinaire, 163(11), 530-535.

Design and sample. Compared three training methods for distraction-based training in 42 Belgian Malinois police dogs in Germany, with cortisol and behavioral measures recorded across three training sessions.

Key findings. The comparison was among three aversive interventions in working dog populations under trained-handler conditions.

Proponent reading. Cited to argue that electronic collar use produced less stress than pinch collar use, positioning electronic or pinch collars as welfare-acceptable.

Response. The comparison was among three aversive interventions, not against reward-based training. The study cannot establish welfare neutrality of any of the three. The relevant comparison for companion dog policy is aversive versus non-aversive, not e-collar versus pinch collar. The senior author Esther Schalke has been a public critic of electronic collar use in companion dog contexts; her broader research program supports the welfare case against aversive equipment.

Deployment. When this study is invoked, reframe the comparison. The study does not address whether aversive equipment is welfare-acceptable; it addresses whether one aversive method is less stressful than another. That is not the policy question.

Cross-reference: Policy paper Section 9.13.

Steiss, Schaffer, Ahmad, and Voith (2007)

Citation. Evaluation of plasma cortisol levels and behavior in dogs wearing bark control collars. Applied Animal Behaviour Science, 106(1-3), 96-106. 10.1016/j.applanim.2006.06.018

Design and sample. Evaluated plasma cortisol levels and behavior in 24 kennel dogs (8 per group) wearing electronic bark collars, citronella spray bark collars, or inactivated control collars across a structured exposure protocol.

Key findings. Under the specific test conditions (kennel dogs, brief structured exposure, plasma cortisol at scheduled intervals), cortisol elevation was modest.

Proponent reading. Cited to argue that bark collars do not significantly elevate cortisol above baseline and are therefore welfare-neutral.

Response. Cortisol is a single physiological measure of stress, not a comprehensive welfare assessment. Behavioral stress markers and cortisol do not always converge. Absence of cortisol evidence is not evidence of absence of welfare cost. The full set of behavioral, affective, and physiological measures is what welfare assessment requires.

Deployment. When this study is invoked, reframe to the broader welfare assessment standard. The Cooper 2014 study illustrates the same point in the e-collar context: a non-significant cortisol finding does not erase validated behavioral stress findings.

Cross-reference: Policy paper Section 9.13.

Schalke, Stichnoth, Ott, and Jones-Baade (2007)

Citation. Clinical signs caused by the use of electric training collars on dogs in everyday life situations. Applied Animal Behaviour Science, 105(4), 369-380. 10.1016/j.applanim.2006.11.002

Design and sample. Examined dogs trained with electronic collars under three conditions varying in predictability and contingency of stimulus delivery.

Key findings. All three groups experienced welfare cost, with the lower-contingency conditions showing the largest effects. The study’s overall conclusion was that the variability of welfare risk across use conditions, combined with the difficulty of guaranteeing optimal contingency conditions in real-world use, argues against approving the equipment for general use.

Proponent reading. Cited selectively for the finding that dogs receiving stimulation under contingency-clear conditions showed less stress than dogs in lower-contingency conditions, and read as supporting controllable e-collar use as welfare-acceptable.

Response. Selective citation. The study’s overall conclusion supports the welfare case, not the proponent case. The senior author Esther Schalke has been a public critic of electronic collar use. Controllability and predictability attenuate but do not eliminate threat-circuit engagement, as confirmed by the Pessoa personal communication.

Deployment. When this study is invoked, point to the authors’ own conclusions and to the senior author’s broader public position. Selective extraction of one sub-finding misrepresents what the paper actually argues.

Cross-reference: Policy paper Section 9.13.

Tortora (1983)

Citation. Safety training: The elimination of avoidance-motivated aggression in dogs. Journal of Experimental Psychology: General, 112(2), 176-214. 10.1037/0096-3445.112.2.176

Design and sample. Multi-stage avoidance-learning protocol involving electronic collar negative reinforcement combined with extensive positive reinforcement in 36 dogs the author defined as having avoidance-motivated aggression.

Key findings. Complex nine-stage process beginning with positive reinforcement training, using both play and choke collars in early stages, introducing a conditioned safety signal as a negative reinforcer, and reserving electronic stimulation for later stages. Tortora’s own Experiment 3 within the same paper showed that simple aversive use of the electronic collar produced only slight decreases in aggression.

Proponent reading. Cited to argue that electronic collar training resolves aggression and improves welfare.

Response. The protocol was not simple aversive conditioning; it was a complex multi-stage process that began with positive reinforcement. The study’s own Experiment 3 undermines the proponent reading that simple aversive conditioning of aggressive behavior is effective. The study is from 1983, predating contemporary canine welfare science methodology by decades, with no behavioral welfare measures, no physiological measures, no follow-up assessment of conditioned emotional responses, and methodological standards that would not meet contemporary peer-review expectations.

Deployment. The convergent welfare evidence base from 2004 forward, using contemporary methodology, supersedes a methodologically thin 1983 design. Citing Tortora as evidence against current force-free aggression behavior modification asks practitioners to weigh a single 1983 multi-stage protocol study without modern welfare measures against forty years of subsequent peer-reviewed convergent welfare research.

Cross-reference: Policy paper Section 9.13.

Lindsay (2005), Handbook of Applied Dog Behavior and Training, Volume 3

Citation. Lindsay, S. R. (2005). Handbook of Applied Dog Behavior and Training, Volume 3: Procedures and Protocols. Blackwell Publishing. ISBN 978-0813807386.

Design and sample. Three-volume textbook on applied dog behavior and training. Volume 3 is on procedures and protocols.

Key findings. Lindsay defends electronic collar use and characterizes low-level electronic stimulation as a “pulsing tingling or tickling sensation” rather than a noxious event.

Proponent reading. Cited as authoritative support for the position that low-level electronic stimulation is not aversive.

Response. A textbook chapter expressing the author’s interpretation of training methodology, not a peer-reviewed welfare research finding. Lindsay’s characterization of low-level electronic stimulation has not been substantiated by peer-reviewed nociception research. Electrical stimulation strong enough to drive avoidance learning is, by functional definition, crossing the nociceptive threshold; the characterization of the same stimulation as a non-aversive sensation cannot be reconciled with the operant requirement that the stimulation function as an aversive event. See also Dubin and Patapoutian (2010), Raja et al. (2020), and Lines, van Driel, and Cooper (2013) in Group F.

Deployment. A textbook chapter expressing one author’s interpretation is not a peer-reviewed welfare finding. The convergent welfare research catalogued in this playbook supersedes textbook interpretation when the two are in conflict.

Cross-reference: Policy paper Section 9.13.

Closing Notes on Deployment

The pattern across the studies in this playbook supports a single conclusion: convergent welfare evidence across multiple methodologies, populations, countries, and outcome measures favors the force-free position. The case does not depend on any individual study being unimpeachable. It depends on the agreement across studies being robust, which it is. The pattern is sufficiently convergent to support precautionary welfare policy.

When a critic attacks any single study in this playbook, the response is the same: the case is not built on this study alone. Use the convergence frame, name the other studies that point in the same direction, and let the critic explain why the same welfare signal would appear across so many independent methods by coincidence.

When a critic invokes a proponent-cited study from Group I, do not concede the proponent reading. Each of those studies, on careful examination, supports a narrower conclusion than the proponent reading attributes to it, and several actually support the welfare case against aversive equipment when read in their entirety.

The Pessoa and Knight personal communications in Group G are interpretive support, not primary empirical evidence. Their value is that the senior authors of the studies most often cited by proponents have explicitly disclaimed the proponent reading of their own work. Use them to settle interpretive disputes about Limbachia (2021) and Wood (2014) when those studies are invoked, while remembering that the underlying empirical findings live in the published studies.

The pain neuroscience studies in Group F (Dubin and Patapoutian 2010, Raja et al. 2020, Affolter and Moore 1994, Lines van Driel and Cooper 2013) answer the proponent argument at the level the argument is actually made: sensation severity rather than tissue damage. Use them when an opponent reduces the welfare argument to an injury argument, or when an opponent invokes a self-test on the human forearm, or when an opponent appeals to the user’s low intensity setting on the dial.

The expanded threat-circuitry studies in Group G (Cain 2019, Maier and Watkins 2005, Sears et al. 2026) answer the most sophisticated proponent argument in current circulation: the claim that controllable, predictable aversive stimulation is welfare-neutral or that successful avoidance-trained dogs have transcended the aversive contingency. They have not. The aversive contingency is the precondition for the entire learning architecture, and overtrained avoidance recruits the same dorsolateral striatum circuit implicated in obsessive-compulsive disorder and in survivors of early life stress.

Table B. Quick-Reference Deployment Summary

This table is built for live deployment. Use it during preparation for podcasts, written exchanges, or formal debate to identify which studies to cite for a specific argument. The Strongest Argument column names the single most powerful argument each study supports.

Bolded entries are the strongest single citations for live deployment. The Pessoa and Knight personal communications are interpretive support, not primary empirical evidence; their value is that the senior authors of the published studies have explicitly disclaimed the proponent reading.

Table C. Argument-to-Studies Index

This table reverses the lookup direction. Instead of asking what does this study show, it answers what study do I need for this argument. Use it when an opponent makes a specific claim and you want to find the citations that counter it most directly.

Use this table in real time during exchanges. The argument the opponent makes is in the left column; the studies that counter it are in the right column. Cross-reference back to the corresponding profile in the playbook body for the full citation, deployment language, and honest acknowledgment of limits.

End of Studies Playbook

Companion to Bangura (2026), The Scientific Case Against Aversive Dog Training Equipment and Methods.