The British Small Animal Veterinary Association (BSAVA) recommends the use of positive reinforcement training methods. Reinforcement or reward-based methods of training use positive stimuli such as physical and verbal praise, treats, toys and play to increase the frequency of desirable behaviours. Research indicates that reward-based methods of training are both more effective and less likely to have undesired consequences than aversive methods of training. Reward-based methods allow owners to develop a strong bond with their pet, based on building confidence and trust, rather than using methods likely to create anxiety or fear.

The BSAVA recommends against the use of aversive methods for training animals. Aversive methods of training animals are based on the principle of applying an unpleasant anxiety/fear-provoking stimulus to stop or prevent unwanted behaviour. Examples of aversive training methods include verbal reprimands (shouting), physical corrections such as hitting, shaking, smacking or lead jerking, and the application of aversive devices such as electric shock collars, prong collars, spray collars, choke chains and electric containment fences. To be effective in stopping unwanted behaviour, shocks and other aversive stimuli received during training are likely to be acutely stressful, painful and frightening for the animals. This may cause injuries and frequently result in long term adverse effects on behavioural and emotional responses such as fear and aggression (Vieira de Castro et al. 2020; Casey et al. 2021). This has implications for both public safety and animal welfare.

In order to safeguard animal welfare, the BSAVA supports the need for legislation which bans and sale and use of devices which enable aversive training, such as electronic shock collars.

Background Information

Operant conditioning is a type of associative learning; an animal learns to associate its own behaviour with a particular outcome. Reinforced behaviours are more likely to recur, and punished behaviours less likely to recur. The four quadrants of operant conditioning are:

● Positive reinforcement: the application of a pleasant stimulus to increase the likelihood of a desired behaviour recurring, e.g. rewarding an animal for a specific behaviour with praise, play, food

● Negative reinforcement: the removal of an unpleasant stimulus to increase the likelihood of a desired behaviour recurring, e.g. the release of pressure from a headcollar when a dog stops pulling

● Positive punishment: the application of an unpleasant stimulus to decrease the likelihood of an undesirable behaviour recurring, e.g. a shock or vibration when a dog barks

● Negative punishment: the removal of a pleasant stimulus to decrease the likelihood of an undesirable behaviour from recurring, e.g. stopping a game if an animal play-bites.

These quadrants are effectively two combined pairs. The pairs are positive reinforcement-negative punishment (reward-based training), and positive punishment-negative reinforcement (aversive training). Reward-based methods by their nature adhere to the welfare ethos of being ‘Least Invasive, Minimally Aversive (LIMA)’. There is substantial research evidence of the superiority of reward-based training methods as compared with the positive punishment-negative reinforcement combination. This includes an association between the use of aversive training methods with a more ‘pessimistic’ cognitive state, indicative of compromised welfare, as well as behavioural and physiological indicators of stress.



Legislation and Codes of Practise

Section 9 of the Animal Welfare Act (2006) places a duty of care on people to ensure that they take reasonable steps to meet the welfare needs of animals for which they are responsible. This includes protecting animals from pain, suffering, injury and disease. Aversive training methods and devices have the potential to cause physiological and psychological suffering. Furthermore, Defra’s Code of practice for the welfare of dogs and the Code of practice for the welfare of cats advocate reward-based methods of training.

The use of electronic collars in training and behavioural modification has been identified in specific research as particularly aversive, resulting in cognitive, behavioural and physiological indicators of negative emotional states. The rules on the use of electric shock collars are different across the UK:

● In England, the government announced in 2023 its intention to introduce legislation which prohibits the use of electronic collars for dogs and cats. Subject to government and parliamentary processes the ban was due to come into force on 1st February 2024 but has been delayed. This follows government announcements made in 2018 outlining intentions and in May 2021, the government’s Action Plan for Animal Welfare which included a statement regarding banning remote controlled e-collars given their scope to harm cats and dogs.

● The Welsh Assembly introduced legislation to prohibit the use of electronic collars in dogs and cats with Animal Welfare (Electronic Collars) (Wales) Regulations 2010 made under section 12 of the Animal Welfare Act, 2006. An offence under these regulations carries a maximum period of imprisonment of 51 weeks, a fine, or both.

● The Scottish Government issued guidance on training methods and training aids for dogs, with particular focus on the welfare issues that may arise from the use of aversive methods including e-collars, in 2018.

● In Northern Ireland, there is no specific legislation relating to electric shock collars or other aversive training and containment aids.


Reviewed by members of BSAVA Scientific Committee (Rachel Casey, Gillian Diesel, Ben Garland, Melanie Hezzell, Jeremy Kirk, Caroline Kisielewicz, Michael Rampersad, Sarah Underhill, James Warland) 2024.

BSAVA members Member Only

If you are a BSAVA member please log in to access additional content. If you are not currently a BSAVA member and would like to join us please click here.

Quick links