The Royal Society for the Prevention of Cruelty to Animals (RSPCA) has recently launched a joint initiative described as an Animal Welfare Evidence Centre, accompanied by guidance material, imagery, and public-facing advice relating to dog training tools — including electronic training collars and prong collars. ( https://www.rspca.org.uk/getinvolved/campaign/shockcollars) The images and statements addressed in this article are taken directly from that newly released material.
The Association of Responsible Dog Owners (ARDO) considers it important to respond clearly and publicly to this initiative, not because evidence-led welfare discussion is unwelcome, but because the authority claimed by this “evidence centre” warrants careful scrutiny. While presented as an objective, science-based resource, the initiative brings together organisations whose core remits lie in animal sheltering, livestock advocacy, wildlife campaigning, opposition to trophy hunting, and farm animal transport — not in applied dog training, behaviour modification, or real-world canine management.
Among the organisations contributing guidance is Four Paws, which offers advice on dog training despite not being a dog training organisation, behaviour body, or professional association in that field. Its recommendation — visible in the campaign material — that owners consult a veterinary behaviourist or a vet for behavioural problems illustrates a recurring appeal to authority: that the inclusion of the word “vet” is treated as a proxy for practical expertise. Veterinarians play a vital role in animal health and welfare, but their primary training is medical. The assessment and resolution of complex, real-world dog training and behaviour problems — particularly those involving learning, risk management, and environmental context — is not inherently a veterinary function.
This same issue arises with the involvement of organisations such as Compassion in World Farming and Humane World for Animals. Whatever their stated aims, these groups are not dog trainers, do not represent applied canine behaviour professionals, and do not operate in the practical domain of training dogs safely in everyday environments. Yet through their inclusion in this evidence centre, they are collectively positioned as arbiters of appropriate training standards and acceptable tools — a role for which they have no demonstrable applied authority.
It is against this backdrop that ARDO offers the following response. This article does not dispute the importance of animal welfare, nor does it seek to defend misuse or abuse of any training tool. Rather, it examines whether the claims made through the RSPCA-led Animal Welfare Evidence Centre — particularly in relation to electronic training collars — are supported by sound evidence, coherent logic, and an honest engagement with real-world welfare outcomes.
At the centre of the RSPCA’s case is language. “Shock collar” is not a neutral descriptor. It conjures an image of sudden, painful, uncontrolled electrical injury. Modern electronic training collars do not function in this way. They deliver a static electrical pulse, adjustable in intensity and duration, often imperceptible at low levels, and designed — when used correctly — to act as a predictable, momentary signal to which the animal has been comprehensively trained to perform a specific behaviour that immediately terminates that signal, rather than functioning as a delayed or indiscriminate negative consequence imposed after the event.
This is not to deny that electronic training collars can also be used as a direct consequence for behaviour that presents a genuine welfare, safety, or life-threatening risk. There is nothing inherently unethical about that position. We live in a society in which behaviour — human behaviour — is routinely shaped and constrained through consequences, ranging from educational sanctions to the criminal justice system. To pretend otherwise is not compassionate; it is disingenuous. What is striking is not that animals may learn through consequence, but that we appear increasingly uncomfortable acknowledging that the same learning mechanisms which govern our own behaviour also apply to them [1][2].
The RSPCA goes on to assert that electronic collars “cause pain and fear”. This is stated as a universal truth, rather than a conditional possibility. Yet pain and fear are not inherent properties of an object. They are subjective experiences shaped by intensity, predictability, controllability, prior learning history, and context [1][3]. This is not controversial; it is foundational behavioural science. Even the research most commonly cited in this debate does not demonstrate sustained welfare harm when collars are applied in controlled, predictable ways [4].
At this point, however, readers should tread carefully. The RSPCA routinely relies on the DEFRA-funded Lincoln research led by Cooper and Mills (2014) as cornerstone evidence in its opposition to electronic training collars. That research has been comprehensively and critically challenged by independent academic peers. Notably, Sargisson and McLean published a detailed critique of the later China et al. paper, which itself was an MSc-linked reanalysis of the same dataset and methodology overseen by the same Lincoln researchers [5]. These critiques raise substantive concerns about study design, interpretation, and the validity of the conclusions drawn.
Further, Professor Douglas Elliffe of the University of Auckland has stated unequivocally that the Lincoln research should not be relied upon for political or policy decision-making, citing methodological limitations and over-reach in its conclusions [6]. This criticism is not peripheral; it strikes at the core of how that research has been used in lobbying and legislative contexts. Crucially, even DEFRA — the government body that commissioned and funded the work — concluded in written correspondence up until early 2018, that the evidence was insufficient and inconclusive to justify a ban on electronic training collars [7]. In other words, the RSPCA continues to cite research that its own funder determined did not support the policy outcome now being advocated.
This pattern matters, because it reveals a selective approach to science: research that aligns with a preferred narrative is foregrounded, while peer-reviewed critiques and conflicting evidence are ignored. That is not how evidence-based welfare assessment is meant to function.
One of the most striking features of the RSPCA’s material is its reliance on technical maximums presented as if they represent everyday use. We are told that collars can operate over long distances, that stimulation can last many seconds, and that activation may occur out of sight of the handler. These statements are technically true — but profoundly misleading.
A motor vehicle is capable of travelling at 130 or 140 miles per hour through a residential street, ploughing through school fencing and anything else in its path. That capacity exists. Yet no serious person would argue that this potential defines the everyday use of cars, nor that the appropriate response is to ban vehicles and force everyone to walk at a regulated pace because running might result in collisions and injury. Society understands — instinctively — that maximum capability does not describe normal application. We regulate behaviour, we set standards of competence, and we prosecute misuse.
The logic applied by the RSPCA collapses when followed to its conclusion. If a tool must be banned because it could be misused at an extreme fringe, then no tool survives scrutiny. Leads can choke, harnesses can injure shoulders, crates can confine for excessive periods, and even reward-based training can generate significant frustration when expected outcomes fail to materialise [2]. There is no principled stopping point to this argument — and the absence of one should concern anyone serious about welfare policy.
The claim that electronic collars are “completely unnecessary” because reward-based training can be used instead rests on a similarly fragile assumption: that all dogs, in all contexts, under all levels of arousal and risk, can be reliably trained using food or play alone. The real world does not comply. High-distraction environments — livestock, wildlife, long-distance recall — are precisely the contexts in which failure carries the greatest welfare cost. When recall fails in these circumstances, the consequences are not theoretical: livestock are injured or killed, dogs are seized or destroyed, and road traffic collisions occur, placing both animals and people at serious risk. This is documented reality, not sensationalist exaggeration — and it is something glaringly absent from the RSPCA’s public campaign material [8].
The RSPCA frequently suggests that electronic collars worsen behaviour, cause anxiety, or increase aggression. What is missing from this claim is twofold. First, there is no scientific evidence cited to support it. There is not a single peer-reviewed study demonstrating that the proportionate, responsible use of electronic training collars in everyday settings leads to increased aggression or anxiety as a common — or even uncommon — outcome. Second, there is no acknowledgement that poorly applied interventions of any type can produce behavioural fallout. Inconsistent signals, unclear contingencies, and chronic management without resolution are well-documented contributors to behavioural deterioration [2][3]. Correlation is not causation, and selective attribution does not advance welfare understanding.
It is also notable that the RSPCA appears unwilling to confront the limitations of its own preferred training framework. Reward-based training is valuable, but it is not infallible. Behavioural problems remain one of the principal reasons healthy dogs — particularly those under three years of age — are euthanised or die prematurely in the UK. This is not conjecture. A large-scale VetCompass study of primary-care veterinary records in England found that undesirable behaviours accounted for a substantial proportion of deaths in dogs under three years of age, with aggression the most prevalent category recorded [9]. These are animals for whom training has failed, not because people did not care, but because existing approaches proved insufficient to manage risk safely.
Against this backdrop, the assertion that “the only people who benefit from e-collars are those who profit from them” is difficult to sustain. Farmers benefit when livestock are protected. Wildlife benefits when predation is prevented. Dogs benefit when effective intervention prevents escalation to legal destruction. Owners benefit when they can meet their responsibilities safely and humanely. Crucially, the RSPCA itself promotes the Five Freedoms framework, one of which is the freedom to express normal behaviour. UK welfare law likewise places a duty of care on keepers to meet an animal’s welfare needs, including the need to be able to exhibit normal behaviour patterns [10]. Responsible use of electronic training collars can, in many real-world cases, enable exactly that: allowing dogs to explore their environment, exercise agency, and engage in species-typical behaviour without permanent restriction, confinement, or social limitation. Somewhat ironically, the RSPCA’s blanket stance risks undermining one of the welfare principles it publicly relies upon.
None of this is an argument for indiscriminate use. Electronic training collars are not benign, and they are not appropriate in all hands or all situations. But neither are they the caricature presented in campaigning material. The RSPCA’s case relies on emotive language, exaggerated outliers, selective science, and a reluctance to engage with legitimate academic critique. That may be effective campaigning. It is not rigorous welfare analysis.
Animal welfare deserves better than slogans. It deserves honesty about trade-offs, respect for individual variation, and a willingness to ask uncomfortable questions about what happens when effective tools are removed. A conversation that begins and ends with the word “shock” is not a conversation at all.
If you agree with what you have read here — or if you recognise the flaws in the arguments put forward by organisations such as the RSPCA regarding how dog owners should train, manage, and care for their dogs — we ask that you support the Association of Responsible Dog Owners (ARDO). You can do so simply by emailing info@joinardo.com with the word “Support”. We will add you to our supporter database and contact you when appropriate to help ensure that your voice is heard alongside ours during parliamentary proceedings and lobbying activity aimed at restricting the lawful, humane ways in which you protect and care for your dog — often by those who neither understand these tools nor live with the consequences of denying their responsible use.
Seligman, M.E.P. (1975). Helplessness: On Depression, Development, and Death. San Francisco: W.H. Freeman.
Baron, A. & Galizio, M. (2006). The distinction between positive and negative reinforcement. The Behavior Analyst, 29(2), pp. 141–164.
Moberg, G.P. & Mench, J.A. (2000). The Biology of Animal Stress: Basic Principles and Implications for Animal Welfare. Wallingford: CABI Publishing.
Cooper, J.J., Cracknell, N., Hardiman, J., Wright, H. & Mills, D.S. (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.
Sargisson, R.J. & McLean, I.G. (2020). Commentary on the interpretation of e-collar research data. Journal of Veterinary Behavior, 36, pp. 72–75.
Elliffe, D. (Unpublished expert review). University of Auckland. Commentary on the limitations of DEFRA-funded e-collar research and its suitability for policy decision-making.
DEFRA (2014–2018). Written ministerial correspondence regarding electronic training collars and evidential thresholds for prohibition.
Department for Transport (2022). Reported Road Casualties Great Britain. London: DfT.
Boyd, C., Jarvis, S., McGreevy, P.D., Heath, S., Church, D.B., Brodbelt, D.C. & O’Neill, D.G. (2018). Mortality resulting from undesirable behaviours in dogs aged under three years attending primary-care veterinary practices in England. Animal Welfare, 27(3), pp. 251–262. doi.org/10.7120/09627286.27.3.251
UK Government (2006). Animal Welfare Act 2006. London: HMSO.