From a general level, positive reinforcement carries the least risk for undesirable outcomes such as aggression, escape behavior, fear of the punisher, apathy, etc. Punishment (the addition of something unpleasant, or the removal of something pleasant) might work at deterring behavior, but it will almost always be accompanied by additional, unintended consequences for the dog. Important to note here is that negative reinforcement in dogs can also carry negative consequences. Negative reinforcement can be considered something as simple as removing a distracting item, in hopes of increasing calm behaviors, and in theory, is not always harmful. However, in training or practical contexts, in order to remove something unpleasant, it often must first be added. Due to the necessity of adding something aversive in order to use this method, it can also lead to the negative effects listed above. For this reason – many experts suggest using only positive reinforcement due to its non-aversive, inherently ethical nature.
Gal Ziv from Wingate Institute in Netanya, Israel put together a 2017 review outlining the effects of aversive dog training from a number of different studies. Consistent findings across multiple survey studies indicate that punishment and negative reinforcement lead to more instances of aggression and behavioral problems when compared to reward-based methods. Important to note, one study did indicate that reward-based methods such as calming or distracting your dog increased aggression, but overwhelmingly, aversive methods are associated with more aggression and increased behavioral problems. Interestingly, one study also found that inconsistency of training methods (switching between positive reinforcement and punishment methods) leads to higher levels of aggressive behavior. This is thought to be due to the dog’s inability to predict what kind of consequence will occur, possibly causing them to react in hostility.
Field studies, where researchers observe dogs and their trainers in real-time, also indicate negative outcomes for these more aversive training methods. In one study, dogs who were punished by their owners interacted less with the experimenter and strangers, and even played less with their owners. In the same study, dogs who were positively reinforced learned new tasks more quickly. In a separate study observing two dog training schools (1 using negative reinforcement, 1 using positive reinforcement), dogs showed more stress-related behaviors and low body posture when negatively reinforced. The relationship between owner & dog may also be compromised by negative reinforcement – the same study showed that fewer dogs gazed and made eye contact with their owners at the negative reinforcement school. Eye contact and gaze behavior increases oxytocin (bonding hormone) in both pups and people, so the lack of this gazing behavior likely compromises the owner-dog bond. Neither of these studies showed aversive training to be any more effective than reward-based training, but they did effectively show potential welfare and behavioral implications for dogs trained aversively.
So – how should we train our dogs? In the interest of ethical treatment, promoting strong bonds, and reducing aggression and behavioral problems, we should consistently use positive reinforcement and reward-based methods. While there may be no difference in the effectiveness of training methods, aversive training methods also come with unwanted, “side effect” behaviors and reduce the bond you have with your dog.
Source: Ziv, G. (2017). The effects of using aversive training methods in dogs – A review. Journal of Veterinary Behavior, 19, 50-60. http://dx.doi.org/10.1016/j.jveb.2017.02.0 04
Thanks to @tucker_whv for the awesome photo, and to @weazelbedelia for taking it! Find them on Instagram.