The importance of using reward-based methods to train dogs is widely known, yet some people still use aversive methods. By definition, both reward-based and aversive methods work to change behavior, but the problem is that aversives such as leash “corrections”, shock collars, and prong collars have risks of harm to the dog’s welfare (Ziv, 2017). So why are some people still using them? There’s some evidence that people’s beliefs about dog training methods haven’t caught up to what the research tells us.
Dog training methods and behavior issues
Research shows that people who use aversive methods such as shock and prong collars to train their dog report being less satisfied with their dog’s obedience (Hiby et al, 2004). People who use aversive methods at least some of the time say that their dog shows more fear, aggression, and attention-seeking than those who use only reward-based methods (Blackwell et al, 2008).
Research has also found that the effects of positive reinforcement training are dose-dependent; in other words, the more of it, the better! When people use positive reinforcement more frequently, their dogs are less likely to be anxious and/or aggressive and are more likely to be obedient, compared to people who use reward-based training less often (Arhant et al, 2010).
Are reward-based methods more effective for training dogs?
These days there is a wide range of evidence, such that many organizations, including the American Veterinary Society of Animal Behavior, warn against the use of aversives. It is easy to understand that stress might be associated with poorer learning, and that dogs might associate unpleasant stimuli (such as a tug on the leash or a shock from a collar) with the person who is present when this happens, thus affecting the dog-owner relationship. As well, anyone who uses treats and/or toys to train dogs can see just how motivating these can be as reinforcement for behaviors we’d like to see more of. This raises the idea that perhaps, in addition to avoiding the risks of aversives, reward-based methods might just work better.
There are some experimental or quasi-experimental studies that suggest that reward-based dog training methods are more effective. For example, one study found that dogs trained with positive reinforcement were quicker to come when called and quicker to get their bottom on the floor when asked to sit than those trained with shock collars (China et al, 2020). Another study found that dogs who had a history of being trained with positive reinforcement were quicker to learn a new task than those who didn’t have such a history (Rooney & Cowan, 2011).
Are people who use aversive methods satisfied with the results?
In recent studies, almost everyone uses positive reinforcement to train their dog; it’s just that some still also use aversives. One clue as to why some people still use aversive methods comes from a study of shock collar usage in France. This found that a majority (58%) of electronic collar users (and a surprising minority of non-users, 37%) said they thought a shock collar was the best way to resolve behavior issues (Masson et al, 2018). But the results of the people who used an electronic collar did not match up to this belief. The findings also suggest that unfortunately, some dogs were receiving a large number of shocks over a long period of time.
The authors concluded that,
“Based on this survey, it appears that in a real-life setting, electronic collars' ability to modify behaviors is limited.”
One concern in this study is that 72% of the dog guardians who used a shock collar did not seek professional advice before using it. Most bought the collar online or at a pet store or garden centre and had tried few other methods to help their dog. In contrast, people who used the internet more and read more dog training books were less likely to use an electronic collar. This highlights the need for more education on dog training and dog behavior, and that dog guardians have a responsibility to seek it out.
A range of factors influences people’s choices of dog training methods, including societal factors such as regulation and individual factors such as knowledge and experience (Todd, 2018). Dog guardians’ personality does not seem to be linked to training methods, but men are more likely to use aversives, especially if they didn’t go to puppy class (Woodward et al, 2021) or if they have depression (Dodman et a,l 2018). It seems that traditional gender roles may influence people’s choice of training methods. The use of aversive methods is also higher amongst older age groups, who perhaps still believe in outdated “dominance” ideas, whereas households with children are more likely to stick to reward-based methods (Woodward et al, 2021).
Knowledge and dog training methods
It seems that people who use reward-based methods are happier with the results than those who use aversive methods—but this isn’t widely known. False beliefs about effectiveness may be one reason why people use aversive methods.
Dog training is not regulated, and people often tell me that the reason they tried aversive methods is because their dog trainer recommended them. This highlights the importance of choosing a dog trainer who will only use modern, humane approaches.
Animal Behavior Essential Reads
It seems that more knowledge of dog behavior and of training methods is needed in order to ensure that more people use only reward-based methods to train their dog. This is important for good animal welfare and for a better human-animal relationship, and although more research is needed, it seems it might be more effective too.
You can learn more about the scientific research on dog training methods (and how to put it into practice) in my book Bark! The Science of Helping Your Anxious, Fearful, or Reactive Dog.