The Winkler Aggression Model
The term “Aggression”
In our society, the word “aggression” is being used very casually. A person who puts a great effort into his work is said to “aggressively attack” his job. Someone raises their voice, and immediately that someone is labeled as being “aggressive”. The word has lost meaning. Shift now to the world of dog training and the term has become a catch phrase. A puppy refuses to come, or he rushes out he front door when it is opened and a self appointed “expert” proclaims that said puppy is “aggressively challenging the owners authority. Of course for the “expert” such a big behavioral problem translates into a healthy fee. So many things, from simple playfulness to stubbornness all the way to botched training will have some form of “aggression” to blame.
Then in another training genre, the protection and police dog world, an expert in “aggression” is of course a sought after commodity and again the term “aggression” is used to colorfully talk about what is happening in a dog’s mind. I read one book a number of years ago; the author identified 14 different types of trainable “aggression”. Making it sound like these are inborn qualities in every dog. Terms like “civil aggression” are used to depict a dog’s natural instinct to show aggression towards a human without visible protection training equipment on. Anyone who takes an objective look at such claims and asks himself “what is that even supposed to mean?” will realize that there is no validity to such talk.
The truth of the matter is, that many of us really do not have a clear understanding what “aggression” really is, what it is meant to do and how it shows itself. I have searched many books and talked to many trainers and instructors, PhDs and behaviorists. Nobody offers a clear definition. Almost everyone has a different take on it and identifies it differently.
The term “aggression” has become nothing more than a colorful word that is used for effect during lectures and discussions.
Why are we talking about “Aggression”?
Why does it matter how the word is used? Why should any of us care? Well, in the law enforcement and military service world, aggression is a very necessary element. But we need to be clear what aggression has to mean in this world. We have to draw a line. We have to gain an understanding that so far has been elusive or at a minimum has gotten lost in translation.
In the service dog world aggression is absolutely necessary and essential for a dog to engage a human being effectively. Without it failures will occur and failures in this world usually seriously compromise the safety and welfare of the dog’s human partners.
Therefore, it is very important that we identify what aggression truly is and what it is not. We have to find a useful definition of what we mean when we refer to the behavior. And once we have defined it, we have to identify how it comes about. Find the source or sources of aggression that we can utilize in training to better prepare dogs for the difficult tasks they face.
I searched dictionaries and books to try and find a useful definition that we could apply and utilize in discussions about training and in the design of training programs that require aggression as an element. The American Heritage Dictionary defined aggression as “the initiation of unprovoked hostilities”. A fairly kind and gentle definition. In my opinion somewhat too human and too politically correct to be useful for our purposes. Another text offered the following: “aggression is defined as behavior aimed at causing harm or pain, psychological harm, or personal injury.” This actually comes close to the point and gives the user of the term an idea of what is meant. However, for our purposes, for the training of police and military service dogs we need a very clear line of distinction, we need clarity what is meant when we use the word. It is important and even crucial that all personnel involved in assessing, purchasing, training and handling dogs in this service industry have a clear understanding what the term should mean so that little guesswork is left.
So, with that in mind here is the “Winkler” definition of aggression:
“Aggression is a state of emotional excitement that will lead an individual to perpetrate violence upon another”
Sources of aggression
In order to clearly identify what actually motivates an animal, in particular a dog to reach an emotional state which leads to the above behavior, namely aggression we have to understand what makes dogs do what they do. Understand their drives. And in doing so differentiate between real science and pop psycho “babble”.
What are drives? Unfortunately, people have forgotten what the term is meant to describe. It shouldn’t be used to give a name to every little thing a dog does. Drives are the internal impulses and urges that motivate animals-- in this case dogs-- to take certain actions. In order for something to be classified as a drive there has to be a drive specific stimulus, drive specific action, and a drive goal. We can manipulate the drives in our dogs during training to suit our purposes and to get them to perform tasks that are the results of these manipulated drives. However, we should never lose sight of the fact that a drive has biological significance for the animal and its species. It is this biological significance that is specific to every drive that gives us a better idea of what we can and cannot achieve by manipulating the drive. Drives can be split into two main categories. The criteria that create the division are the drive goals. One category contains the drives that lead to the gain of something positive or pleasurable; for example: sex drive, prey drive, pack drive (in this case the desire to be with members of the same species). In this category there is a tone of excitement and lust (or passion) during the drive action and deep satisfaction when the goal is reached. The other category contains the drives that lead to the prevention of something negative or harmful; for example: defense drive, flight drive, the desire to remain unscathed. In this category there is a tone of stress and tension during the drive action and relief when the goal is reached.
Let me say one thing right off the top that I have stated in previous articles. I do not believe that there is a natural drive to “fight”.
I believe that what we refer to as fighting “drive” is a package made up of a number of components which are in turn individual drives, drive-products, and behaviors. How good or strong a dog’s fighting “drive” is depends on how many of the components are usable in the dog’s training, how strongly the individual components are present in the dog, and how well promoted all usable components are when the dog is being assessed.
I will now discuss the major components that I have been able to notice separately. I will wrap up my thoughts on fighting “drive” at the end.
A lot has been written about prey drive, so I won’t re-hash all of that again.
Prey drive is part of the food acquisition behaviors of dogs.
The stimuli triggering prey drive are erratic, fast evasive movements. The prey drive actions are chasing, pouncing, biting, pulling down, shaking to death, re-biting, and carrying. As I mentioned above, prey drive is a lust oriented drive. This means that all drive specific actions will be performed in a lustful (passionate) manner. During training this should be kept in mind to ensure that we are in fact working in prey drive. The end goal of the drive is possession of the prey with the intent to eat it (at least from a biological standpoint before human manipulation).
I’d like to make one point here. Prey drive in itself will generally not motivate a dog to bite a human being (I said generally, there are some exceptions). Since prey drive has consumption (or eating) as the end goal of the drive, and canines are not cannibalistic by nature, a properly socialized dog who views humans as his own (even though only adopted) kind will experience very strong inhibitions when it comes to carrying out the follow through drive actions. The fast evasive movement of a human can stimulate prey drive and it can even lead to pursuit and chase. But when it comes to biting, there is a mental block. I will get to why I brought this point up a bit later in the article.
Now, this article is about assessing the traits in dogs. So let’s look at the things to consider where prey drive is concerned. Naturally we have to look at the stimulation threshold of the drive. By this I mean how easy or how difficult it is to trigger the drive. Then we should look at the intensity with which the dog carries out the drive actions. How fast the dog pursues and how hard the dog bites, relative to his physical capabilities needs to be looked at. We can assess how strong the dog’s prey drive is. In other words, how much difficulty is the dog willing to overcome in order to engage in drive specific activities and to satisfy the drive. And one final assessment category for prey drive is the drive endurance. How quickly does the dog “have enough” of doing prey drive actions? Or how soon does the drive exhaust? Drive intensity, strength, and endurance even though related can appear in different levels. So a dog
with low intensity can have strong and enduring drive, etc..
A lot has also been written about this topic, however, I feel it is an aspect of protection training that is often misinterpreted. Therefore I will spend some time discussing the defense drive.
Defense drive definitely falls within the category of aggressive behaviors in dogs. But I think the biological significance of this drive needs to be examined closer in order to get a proper perspective. Defense drive can appear in conjunction with other behaviors and drives, or as self defense. Defense of prey, defense of territory, and defense of a weaker pack member (such as a puppy) are common overlaps during which defensive behaviors will appear. I will address these overlaps a bit more later.
For now I want to discuss self defense further.
Self defense behavior does not only belong in the realm of aggressive behaviors. It also falls into the realm of self preservation mechanisms.
The trigger stimulus for defense drive is threat or the perception of threat. I’m sure you are familiar with a variety of techniques that are used to threaten dogs for the purposes of protection training. So I don’t need to go into too much detail. One thing I want to point out though is that once the dog experiences a threat, he feels a worry or concern that harm may come to him. So the true trigger of defense behavior is the feeling of worry.
The goal of defense drive is always the same, namely making the worry go away. This is achieved when a safe distance is reached between the dog and threatener or when fear is caused in the threatener.
I use the drive specific actions as a way to split dogs’ defense behaviors into three divisions. The first major division is between the active defense reaction and the passive defense reaction.
Active defense reaction
The active defense reaction is a very aggressive form of defense behavior. This type of aggression falls in the category of re-active aggression. My description of the active defense reaction is that once the dog gets the trigger stimulus for the defense drive, he uses physical violence as a means to achieve his drive goal. I am deliberately using the term violence here to make a point. Dogs who show this reaction will resort to biting as the first or one of the first responses that their programmed behavioral pattern dictates for this drive. This is my personal line of distinction that I use when I assess dogs. The reaction is strong and powerful and stems from confidence in the dog. Dogs exhibiting this form of defensive reaction will go towards the threat and attack the threat physically. They show a clear “offence is the best defense” mentality.
Passive defense reaction
The passive defense reaction is split into two separate forms to give us our three divisions.
First there is the strong passive defense reaction. This reaction is one that we see in confident and strong dogs. The dog uses threatening displays such as barking, growling and gesturing while confidently standing his ground. The big distinction is that behaviors other than biting appear as the first responses to the trigger stimulus. And because the initial response is not physical, I classify this reaction as passive. Why a passive or non-physical response appears before the biting response can have different causes. One major one is simply the predetermined behavioral response pattern that the dog was born with. Another cause in highly social dogs is that they realize the threatener is human, and the biting response is inhibited, so other forms of defense behaviors are used first. I do not believe that dogs who show this type of reaction are any less tough or strong than dogs showing the active reaction. Generally dogs who fall into this category of passive defense can be taught to bite in defense quite readily. They will bite when a threat cannot be driven back by other means and continues to advance.
Second there is the weak passive defense reaction. During this reaction we can see the dog using threatening displays such as barking and growling, etc., but he is retreating to maintain a safe distance. How quickly a dog will retreat will vary. These types of dogs are definitely weaker and have less confidence than the two types discussed previously. They will only bite as a last resort when retreat is blocked and the threat continues to advance. This falls into the category of fear biting and is everything but an active response.
There are a few more points I’d like to mention regarding defense drive. I strongly believe that these three defense drive categories are pre- determined and that this predetermination sets limits to how much we can change through training. Comments like “we need to put more defense into this dog” make me cringe and feel sorry for the dog. All we can do is work with what the dog brings with him. Less mature dogs generally cannot show either of the two strong reactions, so the only defensive
reaction that can be elicited from them is the weak passive reaction. The strong reactions rarely appear in dogs before 18 months of age. An important point to remember when training young dogs.
As with all drives, the stimulation threshold is a factor during the assessment process. In this case the threshold is at what point the dog feels concern or worry. This threshold can be raised through deliberate confidence building exercises.
The next component of the package we call fighting “drive” is frustration aggression. Frustration aggression is also a form of reactive aggression that is created by depriving the dog in one of the lust oriented drives or at least by preventing their satisfaction. This form of aggression in my opinion serves the purpose of relieving built- up drive energy that has no proper outlet. The most useful drive to create this type of aggression in training is naturally prey drive. But other positive gain motivated drives can also be used. Such as hunger, sex drive, social reunification (or pack-)drive, or simply a high desire to expend physical energy through movement. When the satisfaction of these motivations is blocked, the dog experiences a sense of frustration. This frustration will reach a point where the dog reacts aggressively. Once this stage is reached, the aggression appears in the same form as all forms of aggression. Barking, growling and biting are the actions that are visible.
I made a short point during my discussion of prey drive, that a properly socialized dog will generally not bite a human being in prey drive. However, if the prey drive builds up to a certain level, and no outlet is presented, frustration aggression will set in. And once a dog is in a state of aggression, he will bite a human being.
Good examples of this model are high drive Malinois, who reach an aggressive state very quickly, because of their high drive and the fact that they seem to frustrate easily.
This is a very useful form of aggression, as it presents a much less risky training methodology than, for example, defense drive, yet still adds intensity and seriousness to many dogs.
Social aggression is the only type of aggression that can be categorized as active aggression. Even though the term active aggression is used frequently, it really only applies here. The reason social aggression is called active aggression is because it really does not require any specific action as a trigger stimulus. Social aggression serves two purposes of biological significance. One is ensuring the even distribution of a species across a given territory by repelling equally strong individuals. And the other is to establish and maintain order in social units such as a pack. Social aggression is always directed at the individual’s own kind. In the breeds that were created for police and military service, selection took
place that expanded the direction of social aggression to also included the dog’s adopted kind, humans. As an example of contrast, in the dog fighting breeds, selection took place to ensure that the social aggression would not include humans.
Let me give you a couple of other reasons why I hold this view. In virtually all older texts describing the police service dog breeds a few points were always made. They were that the dogs show mistrust and aggression against strangers and that they are very devoted and loyal with the family and very loving with children. To me this combination of qualities stem from a very strong closed pack oriented social behavior. That means loyalty and devotion to members in the pack and aggression against all outsiders, even those belonging to the same species.
Unfortunately, this form of aggression is not very common in our dogs anymore, because many people find it to be “socially” unacceptable. Dogs today are supposed to be social and to a certain degree friendly. And while I see nothing wrong with a social dog, I personally also see nothing wrong with a socially aggressive dog. These dogs are not unpredictable menaces to society or vicious animals. They simply have inborn motivations that include this form of aggression. Social aggression is a trainable trait, meaning it can be directed and controlled. Naturally that requires the right handler, so that accidents are prevented.
Socially aggressive dogs have an urge to be aggressive towards strangers. This can be controlled and the dog can be taught to tolerate strangers. However, the dog will not become a social or friendly dog with strangers, no matter what type of training is done. The only way this urge to confront a stranger aggressively when not under control would go away is if the stranger meets the confrontation and social order is established. This happens either if the person can subdue the dog and subordinate him or if the person unequivocally submits to the dog. (At that point the person is no longer a stranger though but an integrated pack member).
The trend in breeding has been to breed dogs who do not have social aggression. And that may be what many people want. The point I would like to make is that social aggression is nothing that should be made out to be something evil. It is a valuable trait in dogs that are in the right hands. Such dogs do demand a high degree of responsibility and vigilance on the part of the handler. Socially aggressive dogs who are also dominant are difficult to handle and to train and should be in the hands of experts.
Dominance behavior falls in the category of social interaction behaviors. It can appear together with social aggression, but it does not have to. It can appear on its own as well. In many ways dominance behavior resembles aggression, but it really is not a form of aggression in itself. Dominance behavior stems from an internal urge to prove superiority and status. In discussion I use the phrase “this dog likes to throw his weight around”. The reason I am making dominance behavior a component of fighting
“drive” is that it has an impact on how a dog physically interacts with other individuals and therefore it becomes part of the picture we see.
Dominance behavior includes climbing up on the helper, eye contact, puffing up to impress, and physical dominance through power. Satisfaction seems to occur when the dog gets a sense of power over the helper. This trait is almost always more strongly developed in males than in females.
Dominance behavior can appear on its own or it can overlap with other components. For example, in a dog with a sense of dominance and good prey drive and a personality that frustrates easily we can see that the dog becomes aggressive only if he cannot get a sense of power. This is not the same as social aggression. This type of dog likes to assert his strength while working in prey, the frustration occurs when he cannot express his power over his adversary for the prey. We can also see a dog with a sense of dominance but only capable of the weak passive defense reaction. This type of dog must naturally have a lower threshold for defensive reactions. For him not being able to feel physical power over the helper triggers the sense of worry which in turn triggers the defensive reaction, which in this example would include retreat.
Naturally there are countless examples of different combinations. There is no need to list them all, the point I am trying to make is that dominance is not automatically aggression. It is not an isolated trait, and always occurs in conjunction with another motivation. But it warrants examination on its own. I feel this is important particularly because the re-active forms of aggression can occur without any expression of dominance (Socially aggressive dogs always have some sense of dominance).
This is the last major component that I want to include in my discussion of fighting “drive”. What is rage? Webster’s defines rage as “a furious, uncontrolled anger” or “a brief spell of raving fury”. I think that definition gives us a pretty good point to start examining what I am talking about. Rage is a re- active form of aggression. Even though most dogs have some form of rage, only a few dogs have it as a useful trait for training. Rage can be directed, but it is very difficult to control. It is therefore not a trait
that is selected for in breeding for performance dogs. Biologically this trait is if anything a contradiction. It does not seem to have a biological goal. Rage is triggered by mistreatment, pain, and opposition.
It can appear as a result of frustration overload. What is unique about rage is that it is extremely forceful and violent and also that it seems to be able to override self preservation instincts. No other form of aggression has that quality. As soon as the negatives become too much and self preservation is threatened dogs will chose other options if they exist in defense, social aggression or frustration aggression. But rage seems to be able to shut out “good sense” even if only for a short period of time. Another unique quality to this form of aggression is that it has a certain vengeful or retaliatory tone to it. Vengeance may be the only goal we can give it. But that is very hard to quantify, and I may be anthropomorphizing a bit here.
I am sure that rage differs from the active defense reaction, because I have seen dogs who clearly showed the weak passive defense reaction to all threatening stimuli. But when pain was caused without any further reduction of safety distance, the dogs suddenly became enraged and came forward aggressively without any regard for their welfare. I have seen the term pain aggression used in a similar way. However I feel that is an inadequate description of rage. The Germans use the word “Wut”, which means rage or anger as part of their protection terminology.
Fighting “drive”, conclusion
This brings me to the end of the discussion of fighting “drive”. The major contributing components I have been able to isolate are the six I just described: prey drive, defense drive**, frustration aggression**, social aggression**, dominance behavior, and rage**. All dogs will have these components in some form or another. But we have to draw the line at the point where the components stop being useful. I hope you will agree with me when I say that a dog that everyone describes as having great fighting “drive” will display most if not all of these components in a useful form. I feel that it is important to look at this concept as a package made up of components, and not one drive. The individual components need to be promoted, solidified and manipulated to where a dog can freely switch between all the components he has in order to deal with virtually every situation he may
encounter. Only then does a dog have fighting “drive”. We also have to accept that some dogs will have fewer dimensions to their individual fighting “drive” package than others. Knowing which components are workable is very important in choosing the right training approach.
I would say that in the good modern day sport dog the package consists of prey drive, defense drive (with the strong passive reaction being the most common), and frustration aggression. Social aggression, dominance behavior and rage are more rare. But when we see a dog that has all six
components, we will not forget him soon, because the fighting “drive” that is displayed when those dogs work leaves a lasting impression.
(italics with asterisk indicate s potential source of aggression)
Every dog functions for his own reasons. All have different drives, different potentials and different expressions. No handler or trainer can wish a trait into his dog. A dog is born with his potential and his drives. All science allows us to do is activate and promote and manipulate what is naturally in a dog through his genetics. A worse approach is sometimes when trainers or handlers try to force something into a dog. Any time I hear the words “let’s put more of this (the reader can fill the blank here) into the
dog”, I cringe and realize that the person talking really does not understand what is happening inside the dogs mind and psyche.
Handlers and trainers also need to maintain perspective that other drives, even the ones that are not aggression sources are also important to a dog’s skill development, his training and many aspects of the tasks a service dog has to perform.
It is a trainer’s job to identify what drive and behavioral components a dog has from his genetics that will serve as aggression sources. He also has to understand what meaning these traits have to a dog biologically before their manipulation. The importance of this aspect cannot be overstated, where can aggression come from in a dog and how can we plan to trigger such aggression in our training to make it command responsive.
Thank you Dr. Pavlov
In our training we try to ensure reliable performance during real life deployment. However, in real life deployment many of the factors and circumstances which can contribute to whether or not aggression appears may not be present. So how can we ensure that a dog will be able to go into a state of aggression on command. This is where the great scientist Dr. Ivan Pavlov taught us a lesson more than half a century ago. His experiment which coined the phrase “classical conditioning” was ground breaking. Many people of course have heard of the experiment where a dog salivated after hearing a dinner bell. Many people have heard about it and to most people it appears pretty logical that a dog would figure out that when the diner bell rings dinner is coming. So the dog anticipates his dinner and therefore salivates. But that is NOT what happened.
What happened instead was that a reflex triggered by the parasympathetic nervous system (that’s the one that happens automatically without thought, like the heart slowing or accelerating upon demand) was replaced by a replacement stimulus. A process Dr. Pavlov labeled a “conditioned reflex”.
The real stimulus that triggers salivation is the smell and taste of food. The dog’s body cannot fight the reflex to salivate, he does not even know it is happening. The process of “classical conditioning” replaces the real and meaningful stimulus with an arbitrary and prior to conditioning meaningless stimulus. Until the replacement stimulus triggers the same reflexive response.
So how is this important for training aggression in a police or military service dog? Well, the drives and drive products that are triggered by real stimuli in training can be replaced by arbitrary replacement stimuli. So that in the end, we can train a dog to feel aggressive when the replacement stimulus is given (usually a voice command) even in the absence of all the real stimuli typically present only in training.
But, to ensure that, certain scientific principles have to be followed. The dog has to in fact be stimulated into aggression during training. Trainers have to be aware of how that is accomplished with any given dogs and the training assistants (decoys) must be able to assess the dog’s emotional state well enough to determine whether the dog had really reached the aggressive state.
Should the dog fail to reach the aggressive state, then the replacement stimulus will only trigger whatever motivation is active when the replacement stimulus was given.
Assessment and feedback are therefore crucial elements in this principle.
There is no room for folklore and training for a picture. We have to understand the science of aggression and the science of classical conditioning to be able to reproduce this behavioral state reliably on command in real life deployment.
Trainers and handler need to develop a keen awareness of what their K9’s job function is. Understand the skill and function the dog has to perform. Don’t train a picture that looks similar to the function
itself. A dog running after a man and biting an object on the man’s arm is far different than a dog biting the man. Yet the picture looks so very similar. Don’t be fooled and do not fall into the trap others have. As modern day trainers we all need to become behavioral scientists, we need to understand how our world looks to the dog, and what makes them do the things that can be useful to us in our world.
Aggression is a naturally occurring behavior in dogs which has multiple sources. It is not evil, and it is not abnormal. Scientifically training dogs to perform aggressive duties in the service of the men and women of law enforcement and military does not alter a dog’s psyche, personality or temperament.