What are Reward-Based Training Methods for Dogs (and Cats)?
It's best for your pet if you only use reward-based methods to train them. Here’s what that means.
 |
| Photo: OlgaOvcharenko/Shutterstock |
By Zazie Todd PhD
Get Companion Animal Psychology in your inbox.
Dog training methods can basically be divided into those that use fear and pain and those that don’t. Or aversive methods and reward-based ones.
What do we mean when we talk about reward-based training methods? It's an important definition because there’s an ever-increasing body of research that tells us that we should use reward-based methods because aversive methods have risks to dogs’ welfare. Those risks include fear, anxiety, stress, aggression, stress-related illnesses, pessimism, and a worse relationship with the owner.
There are also some studies that suggest similar risks for cats.
Many animal behaviour and welfare organizations recommend the use of reward-based training methods. The AVSAB position statement on humane dog training says,
“Based on current scientific evidence, AVSAB recommends that only reward-based training methods are used for all dog training, including the treatment of behavior problems.”
They aren’t the only ones. The BC SPCA, Dogs Trust, RSPCA, RSPCA Australia and the CVMA are among the many organizations that recommend training with rewards.
But sometimes I see people debate whether particular approaches count as reward-based or not. And because dog training is not regulated, sometimes people try to argue that certain methods are ‘kind’ or ‘humane’ when they really aren’t. So it’s important to know what reward-based training methods are.
How are reward-based dog training methods defined?
The simple definition is that reward-based methods are positive reinforcement and negative punishment, while aversive methods involve positive punishment and negative reinforcement.
Of course, that definition relies on an understanding of operant conditioning. Another way of describing reward-based methods is to say that they involve giving rewards for behaviours you want to see more of and removing the rewards for behaviours you want to see less of.
Examples of aversive and reward-based methods
A study from Dr. Rachel Casey et al (2021) compared two groups of dogs: those who were trained with reward-based methods only, and those who were trained using two or more aversive methods (from a list of physical punishment, electronic or citronella collars (bark-activated or remote), a pet corrector, water pistol, choke/checkchain, and rattle can or similar ‘distraction’ noise).
When people used two or more of these methods dogs were more pessimistic.
The scientists have a clear explanation of why the different methods are grouped together:
“Positive punishment and negative reinforcement inevitably occur together depending on the focal behaviour described by these definitions: as one behaviour within a context increases, another will be decreased. For example, the action of spraying a dog with water may both reduce (positively punish) jumping up, but also increase (negatively reinforce) standing on all four paws when the spraying is stopped. Similarly, a dog could be positively reinforced with attention for sitting to greet people (behaviour increasing with the intervention), and negatively punished for jumping up by withdrawal of attention if the dog does not sit.”
When I wrote about the barriers to the adoption of humane training methods (Todd, 2018) and why more people don't use only positive reinforcement, I included management techniques in my definition of humane methods:
“Humane training refers to the use of positive reinforcement and negative punishment in training, along with management strategies that are not aversive. This is also known as reward-based training and is the same approach taken by a number of professional bodies... Humane management strategies include but are not limited to the use of nopull harnesses, putting lids on garbage cans, use of pet gates to keep dogs separate from children or other animals, and the use of a muzzle with appropriate prior conditioning using only nonpunitive, nonscary techniques.”
Another example comes from the research of Vieira de Castro et al (2020), which looked at dogs who attended different training schools and their relationship with their guardian. The scientists assessed the training methods based on videos of the classes. The examples of aversive methods used in this study are leash jerks, yelling at the dog, hitting the dog (all positive punishment), and pulling the leash (leash jerks/choke collar) until the dog sat down (negative reinforcement). Schools that did not use any of these aversive methods were classed as reward-based.
Training dogs and cats in everyday life
Of course, explaining the ins and outs of these methods to ordinary dog guardians is sometimes tricky, especially given the lack of regulation in the field.
This is why I always tell people that my tip for finding a dog trainer is to pick one who will use food to train your dog. Of course, there are times when play is an appropriate reward too. But at least it’s easy for people to see if food is being used or not.
Cats should also only be trained with reward-based methods. The use of loud noises like rattle cans, squirts of water, or shouting at your cat are also bad for your cat’s welfare and may damage your relationship with your cat. And in case you’re wondering, yes of course you can train cats… Teaching them to like their cat carrier is a great place to start (there’s a plan in the back of my book, Purr).
Since dog training isn’t regulated we can’t control how trainers describe their methods. This makes it all the more important for those of us who use kind, humane methods to be clear about what they are—and for pet guardians to insist on these methods too.
If you liked this post, check out my books Wag: The Science of Making Your Dog Happy and Purr: The Science of Making Your Cat Happy, which include plenty of tips for training and enrichment—and the science of why they matter.
Further reading
If you'd like to learn more, check out these posts:
- Dog training science resources and cat training science resources (for links to the scientific literature on training methods and places where you can read about it)
- Seven reasons to use reward-based training methods
- The ultimate dog training tip (on why it’s a good idea to use food in training).
- The best dog training treats
- The best cat training treats
- 3 easy ways to get started with training your cat
References
Casey, R.A., Naj-Oleari, M., Campbell, S. et al. Dogs are more pessimistic if their owners use two or more aversive training methods. Sci Rep 11, 19023 (2021).
Todd, Z. (2018). Barriers to the adoption of humane dog training methods. Journal of Veterinary Behavior, 25, 28-34.
Vieira de Castro, A. C., Barrett, J., de Sousa, L., & Olsson, I. A. S. (2019). Carrots versus sticks: The relationship between training methods and dog-owner attachment. Applied Animal Behaviour Science, 219, 104831.
