Do Confrontational Training Methods Work?

In my dog training classes, people would often come to me and tell me that they tried hitting their dog or yelling at him to get him to stop a particular behavior, but nothing seemed to work. My response has always been that rewards are better than punishment. However, this was based on my own experience and on learning theory. All I had was anecdotal data, and a conviction that dogs should not be treated this way.

Now an article in the February 2009 issue of the journal Applied Animal Behaviour Science addresses this matter in a scientific survey (Herron, M.E., F.S. Shofer and I.R. Reisner. 2009. Survey of the use and outcome of confrontational and non-confrontational training methods in client-owned dogs showing undesirable behaviors. Applied Animal Behaviour Science 117: 47-54).

In this article, the authors surveyed 140 dog owners whose dogs had a variety of behavioral issues, such as aggression to familiar people, aggression to unfamiliar people, aggression to dogs, separation anxiety, or a variety of fears such as to thunderstorms and loud noises. The authors asked the owners to list the methods that they used to deal with these behavioral issues. The owners were also asked if they were bitten, snapped-at, or growled-at as a result of using these methods.

The methods that were used could be grouped into two categories: confrontational and non-confrontational.

Confrontational methods were subdivided into direct confrontation and indirect confrontation.

Direct confrontation included the alpha roll, pushing the dog down into a lying position, hitting or kicking the dog, jabbing the neck, using a prong or choke collar, using leash corrections, using a muzzle, forcibly removing something from a dog’s mouth, forcibly pulling the dog down with a leash, grabbing the dog by the scruff of the neck, using a shock collar, and rubbing the dog’s nose in soiled areas.

Indirect confrontation included yelling “no”, spraying the dog with a water pistol or bottle, growling at the dog, or staring down the dog.

Non-confrontational methods included using the “Look” or “Watch Me” commands, food rewards, food-stuffed toys, use of food as a trade for an item that the dog is holding, using the “Sit” command, and clicker training.

Among the direct confrontational methods, many provoked an aggressive response (bite, snap, growl, bare the teeth, or lunge at person) from some dogs. Each of the following direct confrontational methods is listed with the percent of dogs who responded with an aggressive response: Hitting or kicking the dog, 43%; Forcibly removing something from the dog’s mouth, 38%; Muzzle use, 36%; Alpha roll, 31%; Dominance down, 29%; Grabbing the dog by the scruff of the neck, 26%; Forcing the dog down with a leash, 17%; Choke or pinch collar, 11%; Shock collar, 10%.

Indirect confrontational methods also provoked aggressive responses from some dogs. Each of the following indirect confrontational methods is listed with the percent of dogs who responded with an aggressive response: Growl at dog, 41%; Stare down the dog, 30%; Water pistol or bottle, 20%; Yelling “no”, 15%.

Among the non-confrontational methods, the percent of dogs responding with aggression ranged from 0-6%. The highest aggressive response was when food was used as a trade for an object that the dog was holding (6%). Using the “Look” or “Watch Me” commands and using clicker training showed a 0% aggressive response, while using food rewards, the “Sit” command, and food-stuffed toys each had a 2% aggressive response.

The results of this survey show that using aggressive methods on a dog can result in an aggressive response. Some of this might be aggression as a result of fear on the part of the dog. An earlier study published in 2007 reported that using punishment in training dogs can give rise to fear-related responses (Blackwell et al. 2007. The relationship between training methods and the occurrence of behaviour problems in a population of domestic dogs. In: Proceedings of the 6^th International Veterinary Behaviour Meeting, Fondazione Iniziative Zooprofilattiche e Zootecniche, Brescia, Italy, pp. 51-52).

Interestingly enough, the dog owners were asked in the survey where they learned about the methods that they used.

Many owners reported that they learned about the direct confrontation methods from dog trainers. The percent of owners who learned about these methods from trainers is as follows: Force dog down with a leash, 70%; Choke or pinch collar, 66%; Leash correction, 50%; Knee dog in the chest for jumping, 31%; Bark-activated shock collar, 40%; Remote-activated shock collar, 29%.

Use of many of the non-confrontational methods fell into the category that the authors of the survey called “Self,” in other words, the owners thought of using these methods themselves, without any recommendations from anyone. The percent of owners who fell into the “Self” category for the different non-confrontational methods are as follows: Using food to trade for an item held by the dog, 63%; Food rewards, 56%; Using the “Sit” command, 54%; Food-stuffed toys, 54%; “Look” or “Watch Me” commands, 34%; Clicker training, 23%. However, between 31 and 60 percent of the owners reported that these methods were also recommended by dog trainers.

The value of this study is that it presents quantitative data showing that rewards work better than punishment in training dogs.

Let’s say that you are constantly exposed to someone who yells at you, can hit or kick you at any moment, hurts you, and generally scares you much of the time. Would you be willing and happy to do things for that person? If you answered “No,” then why should we expect dogs to be any different?

--Con Slobodchikoff

Comments

[dog training collars]

This blog really captures my attention. Confrontational training I think really works. It's just that in the first place that the difficulty of training pets is always there. Good to know they are learning from training lately.

[keep dogs]

this one interests me. confrontational leads to its causality..

this helps too.

keep dogs

[Miranda]

I found this article interesting. But I wonder about the direction of the causality. It may be that owners use confrontational techniques with dogs who are aggressive or in instances where the dog is already "primed" for aggression. The owners who use the non-confrontational techniques, on the other hand, may use them because they are effective with their dogs, and these dogs may have never been aggressive.

Speaking from my own experience, I have had four dogs in my life. The first three were never aggressive and I never used any confrontational techniques to discipline, train, or protect them. My fourth and current dog gets into fights with other dogs and I use confrontational techniques to break up the fights and calm her immediately after. The "alpha role," for example, gets her to focus on me, not on the other dogs in the area, and she calms down quickly.

All said, while this study may have been "scientific," in that it appeared in a journal, it could quite possibly have been confounded by the dog's prior behavior, leading to bias in the results.

[Mel F]

So agree with your observations Con!

I first learned about this study last fall after attending a Dog Aggression class given by a local, but well-known, vet in MN who also happens to have a degree in Animal Behavior.

I have always believed aggressive training methods were wrong, but now there is data to back it up!

[Sarah]

I recently came across your blog and have been reading along. I thought I would leave my first comment. I don't know what to say except that I have enjoyed reading. Nice blog. I will keep visiting this blog very often.

Sarah

http://adoptpet.info