How do dogs learn? When it comes to learning, dogs can be taught in several ways. Dogs do not come with an operating instruction manual nor with a troubleshooting guide. It’s up to us to understand how to teach a dog so that the dog can learn. It is only by understanding the underlying machine that motivates dogs that we can really understand the engineering behind the art of dog training. Understanding how dogs learn is not difficult, nor should it be intimidating for dog owners. Today, we’ll be taking a peak at four ways dogs may learn that a particular behavior has a consequence. By fully understanding dog learning theory, basically, the four consequences that can take place when dogs interact with their owners and their environment, you can make the learning process easier. Also, you can identify which dog friendly training methods to use so that you can better bond with your companion.
Life is About Consequences
If we take a look at life, we will notice how many things we do are driven by consequences. If we go to work, we get paid, if we put on mosquito spray we make mosquitoes go away, if we make many mistakes, we fail a test, if we are rude to customers, they may leave and perhaps even never come back again.
In other words, things happen because we do things, and dogs as well learn the relationship between their actions and their consequences. In the world of learning theory, this is known as “operant conditioning” that is, learning (conditioning) based on the ways we “operate.”
Thorndike once said: “responses that produce a satisfying effect in a particular situation become more likely to occur again in that situation, and responses that produce a discomforting effect become less likely to occur again in that situation.”
A Word About Reinforcement
Reinforcement is a behavior change process that takes place when the likelihood of a behavior increases in rate.
Reinforcement encourages behaviors and therefore it increases the likeliness of the behavior occurring.
Reinforcement can occur as the addition or subtraction of stimuli.
In order to be effective, reinforcement must occur in a timely manner and it must be contingent upon the behavior occurring.
For reinforcement to occur, the reinforcing stimulus should be added or subtracted during or immediately after a behavior.
When it comes to how dogs learn, there are two scenarios where dogs are more likely to increase and strengthen behaviors: positive reinforcement and negative reinforcement.
Did you know? Reinforcement can sometimes be generated directly by the behavior itself within the dog (self-reinforcement), rather than being socially mediated by the interactions between dog and trainer.
A Word About Punishment
Punishment is a behavior change process that takes place when the rate of a behavior decreases in rate.
Punishment suppresses behaviors, and therefore, it decreases the likeliness of the behavior occurring.
Punishment can occur as the addition or subtraction of stimuli.In order to be effective, punishment must occur in a timely manner and it must be contingent upon the behavior occurring.
For punishment to occur, the punisher stimulus must be added or subtracted during or immediately after a behavior. When it comes to how dogs learn, there are two scenarios where dogs are more likely to decrease and weaken behaviors: positive punishment and negative punishment.
Confused by all this? Let’as take a closer look at how dogs learn with some practical dog operant conditioning examples.
1) Positive Reinforcement
Also known as added reinforcement, in this case, we are talking about the addition (thus, the term positive) of a stimulus, during or immediately following a response.
In order to be considered positive reinforcement, there must be the addition of a stimulus that increases the probability of the behavior repeating (reinforcement).
For sake of an example, a cookie (the stimulus) is given to the dog (added) the moment the dog sits for the purpose of increasing and strengthening the sitting behavior (reinforcement). As the dog learns the association between sitting and getting the cookie, the dog will soon be sitting more and more.
*Non socially mediated example in a dog’s environment: a dog has found some tasty rabbit poop by the fence line. The behavior of visiting the fence line should therefore become more frequent when the dog is hungry.
Remember: it’s up to your dog to determine what he considers to be reinforcing. If you use your dog ‘s kibble and your dog is a finicky eater to start with, he might find the fact of being offered kibble not reinforcing enough to make sitting repeatedly for it worth it. In such a case, you may have to experiment with high-value treats or other rewards, to get the sitting behavior to increase. Rather than speculating on whether your dog likes something or not, sometimes it’s far more productive to evaluate whether a behavior is increasing or not.
2) Negative Reinforcement
Also known as subtracted reinforcement, in this case, we’re talking about the the removal (negative) of a stimulus, during or immediately following a response, that increases the probability of behavior (reinforcement).
In order to be considered negative reinforcement, there must be the removal of a stimulus that results in an increase in the probability of the behavior repeating (reinforcement)
For sake of an example, pressure on the dog’s back (the stimulus) is removed (subtracted) the moment the dog sits for the purpose of increasing (reinforcing) and straightening the sitting behavior. As the dog learns the association between sitting and removal of pressure from his back, the dog will soon be sitting more and more.
*Non socially mediated example in a dog’s environment: a dog retreats to his dog house when he feel the heat of the sun burning on his back. The behavior of retreating to dog house should therefore become more and more frequent when it’s hot.
Remember, it’s up to your dog to determine what he considers reinforcing. If your dog is really, really eager to be touched, even if that includes, pushing on his back, his sitting behavior will likely not increase. The behavior will increase only if the dog perceives the pressure as unpleasant and wants to escape it. Same with the example of the sun, in order to retreat to the dog house the dog must find the heat unpleasant. A Nordic dog with a heavy coat may therefore retreat more frequently or with minimum heat compared to a Chihuahua.
3) Negative Punishment
Also known as subtracted punishment, in this case, we’re talking about the removal (negative) of a stimulus during or immediately following a response, that decreases (punishment) the probability of a behavior.
In order to be considered negative punishment, there must be the removal of a stimulus that results in the decrease in the probability of the behavior repeating (punishment).
For sake of an example, every time a dog owner comes home from work, his dog greets her by jumping up on her. The owner therefore decides to implement a new strategy. He asks his dog to sit, but if his dog fails to sit and jumps on him, she says “too bad!” makes a quick about face and exits the room contingent upon the jumping behavior.
*Non socially mediated example in a dog’s environment: every time a dog pounces nearby a wild animal, the wild animal retreats in a hole. With time, the behavior of pouncing may reduce and stop as the dog plans a more effective hunting strategy.
Remember: it’s up to the dog to determine whether he considers the removal of a stimulus punishment. If the dog is home alone all day and socially motivated, there should be a significant decrease in the rate of jumping as the dog doesn’t like to be left alone. If the owner is home all day though, and the dog is often corrected for doing things which creates stress, the dog may perceive the owner’s absence as reinforcing, as he gets a feeling of relief.
4) Positive Punishment
Also known as added punishment, in this case, we’re talking about the addition (positive) of a stimulus during or immediately following a response, that decreases the probability of behavior (punishment).
In order to be considered positive punishment, there must be the addition of a stimulus that results in the decrease in the probability of the behavior repeating.
For sake of an example, when Rover is asked to sit and jumps on this owner instead, the owner scolds the dog by pointing his finger to him and saying in a firm, tone of voice “bad dog!” right the moment the dog jumps.
*Non socially mediated example in a dog’s environment: a dog approaches a skunk and the skunk emits a powerful mixture of sulfur-containing chemicals which causes the dog’s eyes to burn and drooling. Afterward, the dog’s behavior of approaching the skunk reduces and the dog stays at a safe distance.
Remember, it’s up to the dog to determine whether the added stimulus is punishing enough to make the behavior of jumping stop. If the dog in question is a soft dog who is also frightened by his owner when he uses a firm voice, the behavior of jumping may reduce and eventually stop. If the dog has been alone all day and is eager to greet the owner, even being scolded may be perceived as music to ears, so the scolding may be actually reinforcing if it gives him a slice of attention. In this case, the behavior will like not stop.
As seen, dogs can learn from us and their environment in several different ways. One important question one must ask is: is the frequency of the dog’s behavior increasing or decreasing? This can reveal a whole lot about the whether the training is working or not. Among the four quadrants of dog training though, only two have a reputation for being “dog friendly”: positive reinforcement and negative punishment. Negative reinforcement and positive punishment involve the use of aversives. Here are 13 negative effects of aversive dog training, a worthy read to understand why many behavior professionals frown on these. Well, this is not all folks! Dogs learn in other ways, but we will see them in another article or two. So stay tuned, for part 2 on how do dogs learn.
“The American Veterinary Society of Animal Behavior recommends that veterinarians identify and refer clients only to trainers and behavior consultants who understand the principles of learning theory and who focus on reinforcing desirable behaviors and removing the reinforcement for undesirable behaviors. ” AVSAB Position Statement
So to recap:
- Excel-Erated Learning: Explaining in Plain English How Dogs Learn and How Best to Teach by Pamela J. Reid, James Kenneth Publishers (July 25, 2011)
- How Dogs Learn (Howell reference books)by Mary R. Burch, Howell Book House; 1 edition (April 21, 2008)
- Flickr Creative Commons, TheRebelRobin Graduation! Blaze graduated from Dog Training 101, CCBY2.0