In part one of this topic, I described non fear-based aggressive dogs, such as those dogs who have been selected for guarding and protection work, like terriers, and mixed breed dogs who may include the genetics of some of these breeds. When we understand that aggressive behavior has been genetically selected for as a desired behavior, we can see that some of the time problem aggression may originate from somewhere other than from fear. Sometimes, dogs exhibiting aggression may be enjoying the activity! Let’s take a look at how we can make a shift to change non fear-based aggression in dogs.
Dealing with a Puppy Who Enjoys Biting
I often think about the two police dog prospects I raised many years ago, long before my current training knowledge. After the first week, we referred to them as the “pinchy puppies”. We had been instructed to never suppress any jumping up, biting or aggressive behavior. When faced with any conflict whatsoever, these pups would leap up and grab your legs, your arms, your feet, or your sleeves! When they were excited, they would leap gleefully at you and grab whatever was closest; hands, sweaters, scarves; everything was fair game for their little, pinchy mouths. They were obnoxious, confident and aggressive puppies. If you withheld a food bowl until the puppy sat, these pups would bite. If you didn’t open the crate door fast enough, they would grab the bars of the door, latch on and growl. If they wanted something and could not get to it, they would bark, scream and carry on, and if you reached over the pen to restrain them, they would target your hands and arms. I carried so many bruises and punctures on my arms that my doctor commented on the situation when I saw her for a routine blood pressure check. These dogs had been genetically selected to bite first and ask questions later, and they did not exhibit any signs of fear, anxiety or discomfort in doing so. When offered a tug toy, at 12 weeks, they would happily latch on and bite and tug and growl with abandon. Later on, during police training, they were taught operantly what they should and should not bite and when this was a desired behavior. They both went on to become solid police dogs; you might not want to cuddle up to them, but they never bit anyone out of context.
The Aggression Modification Method You May Not Have Tried
In the case of dogs like this, the genetic predisposition to bite is incredibly similar to the genetic predisposition of a retriever to fetch. When we talk about young dogs who are inclined to pick up everything they can get access to and carry it around, no one would suggest that counter conditioning is the best alternative for dealing with a puppy who is carrying shoes or the vacuum cleaner attachments. We reward the pup for picking up toys, and we implement a “Response Cost” by removing the item from the dog any time they take something they ought not take. Response Cost is the procedure of decreasing a target behavior by removing a reinforcer, such as when we remove an “illegal” item when the dog chooses to pick it up[1]. If done calmly, as to not accidentally reinforce the behavior with undue attention, eventually, the dog learns not to pick things up unless those things are “legal” toys.
When working with dogs who have been bred as guarding dogs, Response Cost works very well for dealing with unwanted bites, when the dog latches on he will lose access to the thing he is biting. Usually this means access to me! I have raised many, many German Shepherds who if encouraged would have been exactly like the pinchy puppies. Instead, every time my young dogs put their teeth on me, they get timed out for thirty seconds to three minutes. This means that I will go behind a gate or they might get put into a crate. The cost of putting even one accidental tooth on me is losing access to me. They immediately get a turn to try again with access to me, and learn very quickly that when frustrated, excited, angry or confused, teeth would not solve the problem. When counseling clients, I often have them go to the bathroom and close the door for thirty seconds to three minutes, depending on the age and reaction of the dog. This penalty is rarely upsetting to a puppy, but they do learn quickly that their people disappear when teeth touch a person. Sometimes, aggression is simply not about fear or anxiety; it is about the hard wired, genetically selected behavior that we use in protection work.
Response Cost is easy to teach to clients. Pups readily learn a Delta Signal, or warning, and then if they don’t change their behavior, you give a verbal marker that indicates that a Response Cost will next be implemented. In practice it works like this: suppose you are playing with the puppy and they grab hold of your foot. You gently and kindly say “that’s enough”. Being a new puppy, they may not know what this means and continue to bite your foot. You then say “oh, too bad” and gently but systematically implement the Response Cost, either by going into another room and closing the door, or by putting the puppy in his crate. I never recommend leaving the crate; I simply want the puppy to learn that he will lose a turn and be put into the penalty box. If you sit with the puppy, he should not panic, and if he does panic, you are right there to rescue him. This is not the time to teach the puppy that making noise in his crate will not open the door. This is simply about teaching the puppy that he will lose access to me if he puts a tooth on me, even if it is an accident. It may even be preferred to use an empty bathroom nearby or a different crate than their usual one as to not create a negative association.
Keeping in mind that many of these puppies are extremely bright, it should be noted that there are a few puppies who will figure out that biting will get them put in their crates, and if the door is closed, they will bite in order to get access to a good napping spot. Or they will nip to get attention. This kind of operant use of an unwanted behavior really highlights how it is an operant, not emotional behavior.
What About Helping Aggressive Adult Dogs?
Understanding the dog’s heritage, be it purebred or a cross of a guarding and non-guarding breed, can help us to better understand why a dog might be behaving in an aggressive manner, but implementing a Response Cost intervention with a 150lb mastiff who is intent on hurting you may not be practical. When talking to trainers with a background in protection work, they will talk openly about the use of aversives to control this behavior. If your understanding of aggressive behavior is limited to breeds who bite in the same manner that retrievers fetch, and your training is limited to reinforcing biting and punishing unwanted bites, considering counter conditioning can be a cognitive and value-based leap that not all are ready to make. Separately, if your current framework is to address aggressive behaviors using desensitization and counter conditioning, consider the leap you may have to make in order to entertain that aggression might be resolved operantly! This can be a huge mind-shift for some. Luckily, there are safe and effective operant procedures you can use to address the problem.
When working with a large, difficult, guarding breed who has figured out that biting will cause you to increase distance, or that biting will result in resources such as food, toys or space coming into his control, there can be significant safety risks to the person working with the dog. If you are not experienced in working with dogs like this, it is better to refer them to someone who is, than trying to prove your worth by putting your own safety or others at risk.
Behavior Management for the Win
This is where management becomes the single most important tool in your tool box. The first question I always have when working with such a dog is “who is able to work safely with this dog?” If the answer is no one, then the only option is to work with the dog inside a barrier and me on the outside. Although I have occasionally made progress with dogs like this, generally, these dogs have a very poor prognosis. It may just not be practical or safe to work with such dogs.
Extinction and Differential Reinforcement of Alternate Behaviors
When I have had to work with these dogs, I have found extinction and a Differential Reinforcement of Alternate behavior (DRA) works most effectively. The alternate behavior I like to start with is turning away from me. With the dog safely confined behind a fence that the dog is unable to get through, I will set up a chair near the confinement area and wait for the dog to stop engaging in aggressive behavior towards me. This is not what I would do with a dog who is exhibiting any distress or anxiety. This is a protocol for a dog who is enjoying barking, lunging and carrying on, so you need to be really sure that the dog is engaging in the behavior because he likes to do so. As you might imagine, with any extinction protocol, the behavior is going to go through several stages; you will see an initial outburst, and then you will see the behavior spike, then when the behavior ceases, and the dog turns away, I use a marker such as a clicker and toss something of very high value into the confinement area. Usually, the dog will grab the reward and then the cycle will reinitiate. One reinforcement does not a behavior change make! You and the guardians will go through this cycle multiple times, over several training sessions before you see consistent decrease of the behavior. During this time, I never try to interact with or engage the dog. Eventually, the dog will most likely stop barking and lunging at me altogether, and at that point I begin capturing and shaping control behaviors such as go to mat, lie down and stay, and sit and stay. I look for prosocial behaviors such as choosing to seek out my attention for training and allowing me to touch the dog when he is leaning against the barrier. I need to be really certain that the dog is comfortable with me before I will try anything else. I have had very good success using a French Linen Tug with a rope tied to the handle, slipped under the barrier entrance to engage these dogs. This allows me to teach the dog what they MAY bite, which later can help me when I move from protected contact to semi-protected contact where the dog is back tied (tethered to a sturdy object) while I work with them.
Once I am “in the club” and I am comfortable working with the dog while he is back tied, my next strategy is to use a relaxation protocol to teach the dogs to anticipate that I am not always going to be exciting and offer tugging or training. Sometimes, I am going to show up with a book or my knitting and hang out with them. I find that teaching dogs context for relaxation is critical for long term success. I want to give the dog as many good examples of what to expect when working with and around humans before I try to work more directly with the dog. Ideally I want to give the dog as many different people to learn this with as possible, too!
Success Stories in Aggressive Dog Behavior Modification
One of the really powerful things about an extinction and DRA protocol is that it often transfers very quickly to other people, and if you are working in full protected contact as described, you can transfer what the dog learns with you to a wide variety of people. Once a confidently aggressive dog learns “this is not the target, don’t bark and lunge at them” then the dog will often learn that the rule applies to many people. When people become a resource for training activities (and most of these dogs LOVE to train!), then they begin offering prosocial behaviors to people in general.
It is important to recognize that not all confidently aggressive dogs will read the training manual when you go to implement an extinction and DRA protocol and there can be nuances. Luckily, there is now a growing movement of educated trainers (Denise Fenzi is one who is incredibly generous with her time and expertise when it comes to working with protection dogs using R+ strategies for instance) in sports such as Internationale Gebrauchshund Pruefung (IGP) who are using reinforcement based strategies to teach their dogs when they should and should not bite. Many of these trainers are willing to share their knowledge of the breeds and of their sports with other respectful pro trainers. The key here is to approach our colleagues with respect and be prepared to learn. You may not agree with everything they do, but you may also learn a lot about how to manage these very unique types of dogs. If you are new to the world of working with aggressive dogs, and you come across a dog who seems to be amping up instead of slowing down with a counter conditioning protocol, get a second set of eyes on the dog before you throw yourself into the world of operantly trying to change this behavior.
Reinforcement of Training is Key
It should be noted that if you are seeing a dog like this in a shelter or rescue scenario, the dog may not be a good candidate for rehoming, even in the event that you overcome the aggression. Because the training is not conducted in a consistent, calm home environment, when frustrated they may return to a previously successful behavioral strategy. Resurgence, a.k.a. Extinction Burst, the return of a behavior that was extinguished, when a more recently reinforced alternative response is not reinforced, is common[2]. When aggression worked for the dog, and particularly if it worked for a long period of time, this is a strategy the dog will try again if the alternate behaviors taught by the trainer are not being maintained through reinforcement. We need to take great care with dogs like this so that we do not rely too much on the amount of progress that the dog makes in a structured environment and in the hands of an experienced trainer. The training must be reinforced.
4 Reasons to Use an Extinction and DRA Protocol
My education as a dog trainer and behavior consultant started in the mid 1980s, in the era when we would simply use a collar such as a choke chain or a prong collar to enforce the alternate behavior, and many trainers today still do this. They simply cannot see why anyone would take a long period of time to carry out an extinction and DRA protocol as outlined above. There are several reasons to do so.
First is safety. If you have a giant aggressive mastiff, you are unlikely to be able to use a positive punisher quickly enough and with sufficient force to decrease aggression safely. If you cannot do it, almost certainly, your clients cannot. Many trainers who still train this way use their battle scars for bragging rights as though no one might take them seriously if they haven’t been injured on the job, but they don’t tell you that some of the time, their clients are also carrying scars they did not sign up for as a part of their training sessions. The second reason is simply that dogs who are genetically predetermined to be aggressive when faced with a conflict are going to meet force with more force. Dogs are significantly faster and ounce for ounce, stronger than we are. The third reason not to use punishment to decrease aggression, even if it might work, is simply that like all training, punishment strategies have to be practiced or the behavior will revert to what it originally was. This is why many board and train situations don’t solve the problem over the long term even if the dog stopped exhibiting the behavior during and for a short period of time after the training. Finally, causing pain and suffering in the name of rehabilitating aggressive dogs is ethically questionable.
There are lots of reasons not to confront aggression with force, and there are operant strategies we can use to change behavior without force. Be open and connect with trainers who have successfully used operant methods. Ask them to share their journey with you. Be open as they share the satisfaction achieved in helping aggressive dogs in this way.
Ref. 1
First Course in Applied Behavior Analysis, First Edition, Paul Chance, 1998, Waveland Press, University of Illinois
Ref. 2
Handbook of Applied Behaviour Analysis, Second Edition, Fisher, Piazza and Roane, 2021, The Guilford Press
Sue Alexander, CPDT-KSA, CBCC-KA
Sue has been teaching and consulting about dog training and behaviour since 1993. Sue has an extensive speaking portfolio including speaking at the Canadian Association of Professional Pet Dog Trainers National Conference in 2009, presenting at the Dogs in the Park Service Dog Conference for 6 years, and being interviewed on countless podcasts, television shows and radio programs. Sue is devoting her energies for the past two years to consulting to the dog training industry, helping other trainers to learn, grow and optimize their dog training professionalism.
