Punishment in Dog Training; Inadvisable, Ineffective and Unethical
From a personal standpoint, I feel that punishing dogs is a needlessly distressing and futile training method. Devices reminiscent of medieval implements of torture, such as choke chains, prong collars, shock collars, (and I’m tempted to include crates here too! But more on that in my next post!), are still employed in dog training. When it comes down to it, many dog owners just want fast and effective solutions to their dog training problems, and these devices are often sold as just that. However, when you look at the scientific research, it becomes apparent that training with punishment may not be your best bet!
There are many issues with using punishment as a training tool. It causes the dog pain and/or creates fear, it stifles their curiosity and desire to learn, it undermines our relationship with them, and can cause dogs to completely shut down and live in a state of learned helplessness. However, one of the most alarming problems that can arise for owners using punishment in training is the increase in aggression that can stem from it, both as an immediate response, and as a more long-term effect.
In 2009, Dr Meghan Herron performed a study assessing the impact of various training methods, including physical corrections (e.g. hitting) on dogs’ behaviour. Her findings were that dogs who were subjected to confrontational training methods very often reac-ted aggressively. For example, 43% of dogs who were hit or kicked responded aggressively, 31% of dogs ‘alpha-rolled’ responded aggressively, 29% of dogs forced into a ‘dominance’ down and 26% of those grabbed by the scruff of the neck responded aggressively, as well as 15% of those who had ‘no’ shouted at them.
In contrast, it was found that reward-based training elicited aggression in very few dogs, regardless of the training issue.
Apart from the more immediate risk of an aggressive response to punitive training, there also appears to be a more long-term effect on the dog’s behaviour. Another study (Hsu & Sun, 2010), found that dogs subjected to physical punishment were more aggressive than those who weren’t. They did point out that this could be related to the fact that dog owners had a higher tendency to physically punish aggressive dogs than dogs who weren’t aggressive. However, the fact that they continued to behave aggressively implies that physical punishment is certainly not an effective means of reducing aggression.
Yet another study, by Blackwell et al. (2007), as well as reiterating the findings of the above studies, found a link between owner use of punishment and an increased level of reactivity towards dogs from outside the household.
This is just a sample of the studies out there implicating punitive training as a risk factor for aggressive behaviour. So much so, that Ó Súilleabháin, in a 2015 letter to the editor of Zoonoses and Public Health, went so far as to say that in light of the evidence linking punishment-based training and aggression, ‘punishment-based dog training would appear to be a clear public health risk factor.’
As well as having a greater chance of inciting aggressive behaviour in your dog, punishment-based training methods are less effective than their reward-based counterparts.
In their 2004 study on dog training methods, Hiby et al., found that:
- dogs who were trained using reward-based methods had the lowest occurrence of over-excitement;
- dogs who were trained using punishment-based methods (including verbal reprimands) had the highest occurrence of separation related problems
- In none of the tasks trained were punishment-based methods more effective than reward-based methods;
- dogs trained using exclusively reward-based methods were reported to be significantly more obedient than those trained using either punishment or a combination of reward and punishment.
They also found a correlation between the number of problematic behaviours reported and the number of times owners reported using punishment-based training methods. And furthermore, those who trained their dogs using exclusively punishment-based training or a combination of punishment and reward, ‘reported significantly more problems than those using only reward-based or miscellaneous methods.’
Additionally, a study by Haverbeke et al. (2008), showed that military dogs trained using aversive stimuli showed reduced learning performance.
In 2006, the UK passed the Animal Welfare Act. This provides for five welfare needs, one of which is ‘the need to be protected from pain, suffering, injury and disease’. It places a duty of care on people to meet these welfare needs in their animals.
Punishment is generally used to cause fear or pain to the object of punishment. Deliberately causing another creature ongoing fear or pain is a welfare issue. For a person responsible for the animal to do this is contrary to the duty of care we have to our animals.
Additionally, both fear and pain trigger the ‘fight or flight’ (stress) response in our bodies. Occasional stress is something we can cope with, and can be necessary to our survival. Chronic stress, on the other hand, is disruptive and harmful, and has physiological, mental, and behavioural affects. Subjecting our dogs to chronic stress is just another way of inflicting suffering.
Stress has a widespread effect on most of our organs, and when we are subjected to chronic stress (as a dog who is regularly punished would be), and the corresponding elevated levels of stress hormones, the immune system, gastrointestinal system, endocrine system, cardiovascular system and thyroid are all negatively impacted. Cortisol has a profound impact on the brain, affecting learning and memory, and increasing the risk of anxiety and depression, as well as compulsive behaviours.
So, if punishment-based training is less effective than reward-based methods, has an adverse effect on a dog’s health by causing them stress, and if it is a lot more likely to elicit an aggressive response, why do these methods remain in use?
Often, when we shout at a dog, or hit them, or jerk the leash, the behaviour is interrupted. For instance, the dog is barking at another dog and we shout and he stops because he is startled and perhaps frightened. But chances are, the next time a dog passes he will bark again. In fact, there’s a good chance the behaviour will escalate, as the dog is likely to associate the other dog’s presence with the owner’s unpleasant behaviour. However, the fact that the owner will probably feel better for having vented their anger, and the fact that the behaviour stops in the instant, make the act of punishing very rewarding for the punisher, if not an effective tool for long-term behaviour change.
I like to think that most people do not want to cause their dogs pain or fear, undermine the relationship they have with their dogs, or run the risk of their dog developing aggressive tendencies. Rather, I think punishment is something people often resort to when they don’t know what else to do. Now that research shows us a better, more compassionate way, surely it’s time to bin the choke chains and shock collars, stop jerking leads and shouting, and get out our treat bags instead!