Dog Behavior: Can you Reinforce Aggression?
Can you Reinforce Aggression in Dogs?
A great read for owners of reactive dogs!
What is Dog Aggression?
In my latest hub, I discussed the issue of "Can your reinforce fear in dogs" and today, I wish to address the topic of aggression. What is aggression in dogs? The definition of aggression, from a broad standpoint, is a disposition or behavior that tends to be hostile. Generally, there is an intent to harm. While from a human standpoint, aggression may be triggered from disruption of moral values, injustice, hate, jealousy and the need for retaliation, in dogs, aggressive displays are more down to earth, less sophisticated, and therefore, more primal and linked to survival. When dogs aggress, this is often triggered from a fight or flight response.
Fight or flight is a biological response to stress first coined by Walter Bradford Cannon. The response is triggered by a perceived threat that stimulates the sympathetic nervous system to release the hormones adrenaline and noradrenline. These hormones are responsible for preparing the dog for intense muscular action. The effects of these hormones when the dog is in fight mode often translates into:
- Dilated pupils
- Tense body
- Piloerection (makes dog look bigger and thus more threatening)
- Acceleration of heart rate
- The digestive system shuts down (ever wondered why an aggressive/fearful dog will not take treats?)
The decision to aggress or move away depends on a variety of factors. In some cases, it depends on temperament or the dog's history. Some dogs may be predisposed to choose fight as their first course of action when faced with a threat. This could be due to a genetic predisposition (weak nerves or a history of being selectively bred to aggress towards other dogs) or it could be a learned response due to the environment, poor socialization, fear, territorial tendencies. For instance, a dog that has rehearsed "distance increasing" behaviors over and over, may have learned it is effective in keeping threats at bay. An example would be a defensive dog that has learned to keep people or other dogs away by lunging and barking. Because this behavior has successfully worked at increasing distance (most people move away from a dog acting as such) the dog uses it more and more.
Several dogs prefer to move away rather than aggress and confront a perceived threat, but when cornered, will aggress. For example, a dog that is not too keen around children, my run away from them, but when the children find the dog under the bed and try to pull him out, he may bite because he no longer has an escape route. I have seen many dogs that tend to react aggressively in the same way in homes when there are guests, but then are fine with the same guests when they are outdoors. Same goes with being on leash or off leash. Some dogs change dramatically when they are off leash around other dogs as they have more space to work with. Of course, none of these approaches should be taken without the guidance of a dog professional well versed in dog behavior.
And then we have dogs that resort to aggression to protect their territory, protect their offspring, protect thier food. These are innate behaviors, that are considered normal from a dog' standpoint, but that are often frowned upon by humans. Yet, as humans we are not the best examples as we tend to protect our possessions in even a more prominent way than dogs! For instance, we are extremely territorial; indeed we lock our doors, erect fences to keep intruders away, install alarm systems. We rightfully protect our children from harm; we tell them not to talk to strangers, we keep them away from bad friends, we enroll them in the best schools. And yes, we guards our possessions avidly; we keep our valuables in banks, use passwords to protect our bank accounts, we protect our identities from theft.
Of course, there are many other forms of aggression in dogs. Re-directed aggression, frustration-elicited aggression, pain-elicited aggression, and in some cases, aggression may be triggered by an underlying medical condition.
*Note: It is wrong to label any expression of dog aggression as due to "dominance". Thousands of websites still promote this belief and as a result many dog owners are being misguided.
Sohia Yin's Example of How Classical Conditioning Wins!
Can you Reinforce Aggression?
Yes, an aggressive display can be reinforced, but not in the way we imagine. For instance, if a dog is growling because he wants you to leave the property, moving away will reinforce the growling behavior. If a dog is showing his teeth at you to protect his bone, his teeth-baring behavior will be reinforced. If a mother dog attempts to bite when you pick up a puppy, putting the puppy down will reinforce the attempting to bite behavior. When a behavior is reinforced, it will repeat over time. However, we are not looking at the big picture; just the behavior. Growling, barking, lunging, indeed are all outwards behaviors of an underlying emotion. If we work on changing the underlying emotional response, the outward signs will extinguish over time.
My First Hand Experience of This
I have experienced this first hand. The first time occurred when years ago at the beginning of my dog training career I was helping rehabilitate a 3 year old Labrador that was reported to be dog aggressive. This poor dog was not walked for over a year and had seen a dog trainer that immediately suggested an electronic collar. The neighborhood he lived in had fenced dogs everywhere and because walking him past these dogs caused him to lunge, bark, jump and act totally out of control, they had given up and made him live a whole year in the yard.
Our first walk was a disaster, he obviously did not take treats (as mentioned earlier, being way over the threshold in a fight or flight response, the digestive system shuts down) and he acted like some sort of wild bronco horse. Yet, the way this neighborhood was distributed made it impossible to avoid the dogs behind fences without him going over threshold. Even putting him in a car to work in a better location entailed passing a bunch of dogs! After much troubleshooting and assessing the situation, I finally, found a compromise: the combination of an empty stomach that morning, high-value treats that made him drool (liver anyone?), and a distance from the dogs that made him a little bit more responsive to the treats. From this distance (just slightly out of the property he was kept in) he was not lunging, jumping, but he vocalized with barks and growls.
*Note: I would normally work a dog better under the threshold, but the unique situation of the place this dog lived in forced me to work up close.
What happened was I ended up initially giving him cooked liver when he was barking/growling. I knew that some of the trainers I knew were going to call me "a bad trainer" because I was "reinforcing the aggressive display". But if what these trainers claimed truthfully held true, you would expect to see an "increase in the growling behavior" because reinforcement causes a behavior to increase. Instead, to my dismay, the behavior diminished over time! What was happening?
What was happening was the fact that I was not really reinforcing the growling behavior but was instead creating positive associations between the dogs and the treats. Instead of thinking "every time I bark I get a treat and so I will bark more and more" the dog was thinking " every time I see the dog I get a treat, awesome!" Classical conditioning was therefore stronger than operant conditioning!
Further down the years, I found it comforting to learn that some reputable dog behavior experts were seeing things the same way. For instance, Dr. Sophia Yin an Internationally-acclaimed veterinarian and animal behaviorist claimed in her article "The Dominance Theory" quoted:
''Doesn't giving food to aggressive dogs make them worse?"
"Many people think if you give treats when a dog is barking or growling out of fear or other aggressive reasons, you will reward aggressive behavior. Actually, if you change the underlying emotional state you'll change the outward behavior.'' She also offers on her website an interesting video (featured to the right) of how she classically conditioned a dog to stop growling at her from blowing in its face using treats delivered after blowing in his face and he was growling. The dog starts ultimately liking when she blows in its face, and therefore, she successfully shifted the emotional state to her advantage."
Ian Dunbar, another reputable professional which is a veterinarian, animal behaviorist, and dog trainer, further explained clearly what had happened in the training done with the Lab. In his own words, in an article for Dog Star Daily: Quoted:
''Some people are afraid that offering kibble during classical conditioning might unintentionally reinforce bad behaviors. However, by using food when classically conditioning, you can only reinforce good behavior because a dog cannot bark and lunge or eyeball another dog at the same time as turning to face you to take food."
Ian Dunbar then adds an example: "let’s say we are trying to classically condition a dog that is barking and lunging at another dog. We offer food, but the dog ignores our offerings and continues barking and lunging. Eventually though, the dog barks himself out and sniffs the food, whereupon he turns away from the other dog to take the food. Taking the food does not reinforce the dog’s barking and lunging. On the contrary, the food reinforces the dog for stopping barking and lunging, for turning away from the other dog and for turning towards his owner. After a couple of dozen repetitions, the dog will begin to form positive associations with the sight of other dogs. “I love it when other dogs approach because then my owner feeds me dinner.” And as a bonus, the dog’s trained response to seeing another dog is to turn away from the dog and to sit quietly and expectantly facing his owner."Operant Conditioning Rocks! But…Classical Conditioning Rules!''
This was true. The Lab I was training had to stop barking in order to eat the liver, so the behavior of barking/growling was not really being reinforced; rather, the silence and turning his head towards me was. With time, he had learned a "default behavior" to the barking/growling, which was to look at the dog and then look at me in anticipation of the treat. After almost a month of rehearsing this, we were able to bring operant conditioning to the picture, and we gradually got closer to the dogs, when he finally was able to sit and heel politely without acting as a wild bronco!
Now would I normally, work over the threshold allowing the dog to display part of his outward manifestations? No, I would not. Why? Just because I find that it unnecessarily stresses the dog initially. I prefer not to flood the dog even tough I know that this is only temporary. To learn more about flooding in dog behavior read" Flooding the dog, does it work?"
If circumstances, allow it, I would ideally work under the threshold with the dog comfortable. I like to see the dog learn at a distance he is more comfortable, rather than struggle. The video to below shows my work with Asso, the Labrador I was talking about. This video was of several years ago when I was just beginning my career as a dog trainer. There are many things I would change as of today (videos can do wonders in making you a better trainer), and one of them would be keeping the leash looser. I was a bit tense back then, because an Anatolian Shepherd dog behind one of those fences had actually bitten a dog in the past week and the dog lost its ear. There should be little space for tension when training reactive dogs! Any how the dog recovered nicely (he is walked again! and even taken to the dog beach!) and the owners still gives me updates when they can!
Working on Labrador to change emotional response
Why Punishment Does not Work When Dealing With Aggression
So if classical conditioning wins, then that would mean that if every time your dog lunges towards a dog you correct your dog with an alpha roll, leash correction,or shock, your dog is basically associating the presence of dogs with something unpleasant. This is why using aversives when dealing with dog aggression is not recommended and rather, counterproductive! I am so happy the Lab owners said no to the trainer recommending shock collars! If so, the situation would have been much more critical to solve! To learn more about this read: the Risks of Electronic Shock Collars
Yes, some dog owners claim "my dog has stopped lunging, reacting aggressively towards other dogs with the use of corrections" but what about the dog's emotional response, has that changed? Very likely yes, and for the worse! Now, not only your dog is tense upon seeing the dog, but now on top of that, he is expecting that correction that further exacerbates his emotional state. Yes, all the "outwards" signs are gone but you are only dealing with the tip of the iceberg!
For instance, if you correct growling, do you think your dog will automatically be peaceful and safe? No, because you have suppressed an outwards manifestation of an underlying emotion. Depriving a dog from a growl is indeed, a big mistake and will only teach your dog to bite without growling first! To learn more about the dynamics behind this read " Why Dog Growling Should Never be Suppressed" Dog trainer and owner of Peaceable Paws, Pat Miller, calls the growl "a gift" in her article ''The Gift of Growl" a must-read for all dog owners.
The same goes with all other outward manifestation of aggression. Some dog owners have suppressed these signs so well, correction after correction, that they assume their dog can now stay around other dogs, only to find out later that aggression has raised its ugly head again, because the emotional response has not changed. You cannot suppress an outwards manifestation without taking care of the inward problem! Its like putting a bandage over an infected wound! Just as in people, suppressing behavior may actually lead to an "explosion" afterwards due to cumulative stress. This is why some dog owners that often expose their dogs to a lot of stimuli and fail to recognize stress in their dogs, are surprised when at the end of the night the dog ends up biting somebody; dogs can also accumulate stress over time!
Disclaimer: this article is not to be used as a substitute for professional hands-on advice; if your dog is displaying any type of aggression always consult with a reputable trainer well-versed in positive dog behavior modification programs or better, a Certified Applied Animal Behaviorist (CAAB) or Veterinary Behaviorist. By reading this article, you automatically accept this disclaimer.
© Alexadry All Rights Reserved
Author's own experience as a certified dog trainer
Ian Dunbar, Dog Star Daily; "Classical Conditioning"
Sophia Yin, The Art and Science of Anmal Behaivor, "The Dominance Theory"
Pat Miller, Peaceable Paws, 'The Gift of Growl"
For further reading
- Understanding Threshold Levels in Dogs | Suite101.com
What are threshold levels and why are they important in dog behavior modification? Learn how to identify threshold levels and how to stay below them.
- Dog Behavior: Can Dog Behavior Problems be Cured Onc...
Can aggressive dogs be truthfully fixed once and for all? Learn why you should stay away from trainers making promising statements and guarantees.
- Dog Behavior: Can you Reinforce Fear?
We are always told not to pet, cuddle or comfort a fearful dog because this may reinforce fear. But can fear really be reinforced? Learn what the experts say about this.
- Important Dog Behavior Modification Program Follow U...
Dog to dog aggression is a common problem, and is often due to poor socialization. A behavior modification program is a must.
- Dog Behavior: A Guide to Behavior Modification Techn...
Say NO to coercion and say YES to rewards. Learn effective modern based dog behavior modification techniques. As the words imply, behavior modification entails modifying a dog's behavior for the purposes of increasing or decreasing wanted and...