At a dog training seminar that I attended recently I found myself engaged in a familiar conversation, namely the effect of various types of dog training methods on the behavior of pet dogs. Nowadays the line seems to be drawn between two camps, one advocating "positive dog training" (which uses rewards such as food and play) versus "discipline-based training" (which incorporates slip collars and leash tugs and other physical interventions to enforce compliance).
To give you a bit of a technical background, in reality there are only four methods that are used to teach dogs the kinds of tasks and behaviors that we humans are interested in having them learn. The first is called positive reinforcement, where the term "reinforcement" refers to anything that increases the likelihood that the dog will repeat a behavior. The "positive" refers to the fact that we give the dog something that he wants, like a treat. The second method is called negative reinforcement, where the "negative" refers to taking something unwanted or annoying away, so for example, if you pull up on the leash causing a choke chain to tighten and push down on the dog's hindquarters while you tell him to sit, the negative reinforcement comes when the dog goes into the sitting position you take away the pressure around his neck and on his lower back. The other two methods involve punishment. A punisher is anything that reduces the likelihood that a dog will repeat a behavior. Positive punishment refers to the fact that when a dog does something which we don't want him to do we apply something the dog doesn't like, which could be a slap or a loud reprimand. Negative punishment involves taking something that the individual wants away from him. A human example would be if the child acts out the dinner table he doesn't get dessert when everybody else does. Positive dog training is almost always based on positive reinforcement, while discipline-based training uses a combination of negative reinforcement and positive punishment.
Prior to the mid-1940s, most dog training was done using discipline-based training, since most of the early training models came from military dog trainers who had the idea that a dog should be trained using the same kind of discipline-based procedures that were used for human recruits. The change toward more positive training came about because of a series of books written by Blanche Saunders. Although by today's positive training standards she was still a bit harsh, she clearly recognized the value of rewards and was much softer on her canine students than most trainers before her. Over time, positive dog training has come to dominate the canine training scene, following much along the model of the techniques used by Ian Dunbar and others. However over the past few years, due to the influence of certain high profile dog trainers who have popular television series like Cesar Millan, discipline-based training has begun to gain in popularity.
One of the people in the group that I was speaking with insisted that discipline-based training procedures should not have been abandoned, and that no real proof exists showing negative effects on dogs. She complained that there was a bias among canine researchers, who she referred to as "foodies" since they usually reward the dogs with a stream of food treats. "Just because we live in a kinder and gentler world doesn't mean that we can't teach a dog that when he does something we don't like it has negative consequences," she said. "Properly applied, by people who know what they're doing, there's nothing wrong with negative reinforcement or a little bit of punishment. The problem is that most researchers and the people they get to train the dogs they test probably don't really believe in discipline-based training, and so they either overdo it, or don't work as hard at using it properly."
As luck would have it I had just finished going over an article that was recently accepted for publication in the Journal of Veterinary Behavior* that directly dealt with this issue. It was done by two French researchers, Stéphanie Deldalle and Florence Gaunet who wanted to test the effect of the two training styles. They wanted to use the most naturalistic setting that they could find, and to collect data based on observation rather than upon changes in blood chemistry or heart rate or other invasive procedures. What they did first was to attend a number of dog training classes in their vicinity and observe the training procedures that were used. In the end they selected one class which used positive dog training methods almost exclusively and contrasted it to another dog training class which used discipline-based methods (mostly negative reinforcement). They reasoned that the instructors in both of these classes were teaching their students using the method that they favored because, as trainers, they believed it was the most efficient system and would produce the best results.
Once the classes were selected, dog and owner pairs were brought in and tested on familiar exercises such as heeling and sitting on command. There were 26 dogs trained using discipline-based procedures and 24 using positive training. The researchers were not looking at actual learning performance, but rather were measuring the amount of stress that dogs appeared to show when performing their learned exercises. They used easily observable behaviors associated with stress, such as mouth licking, yawning, scratching, sniffing, shivering, whining, low posture, attempts to run away, and whether or not the dog avoided making eye contact with their handler.
The results were rather straightforward. If we simply look at whether a dog showed any of the stress related behaviors we find that 65% of the discipline-based trained dogs showed at least one such sign, as compared to only 8% of the positively train dogs. For some selected behaviors the differences were quite striking, such as in mouth licking (38% discipline; 8% positive), yawning (23% discipline; 0% positive), and low posture (46% discipline; 8% positive). One of the results which I found most interesting had to do with whether or not the dog looked at the owner's face. Both humans and dogs have a tendency to avoid looking at things that raise their stress levels or make them uncomfortable, so the fact that only 38% of the discipline trained dogs looked at their owners faces as compared to 88% of the positively trained dogs seems telling.
This is a small study, but because it is done using actual class trained pet dogs and instructors who believe in their particular training method, it is quite interesting. It seems to be just one more study that suggests that using punishment and negative reinforcement can produce potentially harmful and unwanted emotional changes in dogs. For additional findings on similar topics click here or click here.
Stanley Coren is the author of many books including: The Wisdom of Dogs; Do Dogs Dream? Born to Bark; The Modern Dog; Why Do Dogs Have Wet Noses? The Pawprints of History; How Dogs Think; How To Speak Dog; Why We Love the Dogs We Do; What Do Dogs Know? The Intelligence of Dogs; Why Does My Dog Act That Way? Understanding Dogs for Dummies; Sleep Thieves; The Left-hander Syndrome
Copyright SC Psychological Enterprises Ltd. May not be reprinted or reposted without permission
* Data from: Stéphanie Deldalle & Florence Gaunet (2014). Effects of 2 training methods on stress-related behaviors of the dog (Canis Familiaris) and on the dog-owner relationship. Journal of Veterinary Behavior, 9, 58 – 65.