Imagine if you and all the rest of humanity were suddenly transported to an alien environment – I mean REALLY alien —where EVERYTHING is different and topsy-turvy, e.g. turning a door knob opens a trap door, turning on a light switch flushes the toilet, unknown objects have unknown effects with some of them involving pain.
This is the world our dogs enter and yet they try everyday to survive and make their human happy.
How would you, or any animal, figure out a set of behaviors to survive? You probably answered “trial and error” — Good answer because it refers to the law of effect, the crux of learning theory; but, before you can understand the complex, you have to start with the simple and regardless of how simplistic and mechanical TV personalities make dog training appear, it’s a little more complex than simply reward and punishment.
Teaching dogs revolves around the law of effect and operant conditioning. Omitting the scientific jargon , according to the law of effect, we, our dogs, anything alive, will move toward the pleasant and away from the painful. As any organism experiences the consequences of random behaviors, it develops a preferred repertoire of behaviors. The more pleasant the outcome, the greater the likelihood that the behavior will be repeated.
So, behavior is shaped by the consequence it produces, for good and “bad” behaviors alike, which is where many dog owners go wrong –because of poor timing or inconsistency, they unknowingly reinforce the “wrong” behavior.
There are four possible combinations of this relation, often illustrated as +/- quadrants, and some greatly misunderstood and misused when it comes to working with dogs.
The first relation is referred to as positive reinforcement: the behavior results in a reward, something pleasant. The words “positive” and “negative” in this context refer to adding or deleting, and not to whether it is good or bad; Your dog sits on cue and receives a treat. So his behavior of “sit” is reinforced by something pleasant, a snack, so the likelihood of the dog repeating the behavior increases.
The second relation, positive punishment, involves adding an unpleasant consequence in the hopes of eliminating the behavior –you run a red light, you get a fine or cause an accident –very unpleasant outcomes, a punishment. Spanking children or dogs falls into that category, so does getting drunk and throwing up. It doesn’t matter who inflicts the punishment – if it’s painful and results from the behavior, it’s punishment.
The third is referred to as negative reinforcement (“escape” is a better word and is actually used in learning theory) because it involves the removal of something unpleasant, and creates something pleasant so that the behavior is increased. Misguided trainers who don’t understand the difference between emotional responses and operant conditioning often use “flooding” to eliminate what they consider an “undesirable behavior” using negative reinforcement. A dog afraid of thunder is left in a room with a CD playing loud thunder and is not released until he has “learned” to be calm or until he’s catatonic.
The final relation is negative punishment, often referred to as omission –the behavior results in the removal of something pleasant in order to reduce the behavior. Children are fighting in the snow and the dad takes away their parkas so they freeze. The jackets were warm and comfortable, the behavior, fighting, resulted in the “omission” of warmth and comfort.
While the “positive” quadrants are pretty straightforward, the “negative” ones always seem to create confusion, especially for me until I realized you have to look at the entire outcome, desired or not, and whether it causes an increase or decrease in the behavior. If the action causes an increase in the behavior, it’s reinforcement. If it causes a decrease in the behavior, it’s punishment.
“Loose Leash Walking”
Let’s take an example of how each of these behavior-consequence relations applies to shaping behavior in your dog. Suppose we want to work with “loose leash” walking. How many times have you seen an owner being dragged by an exuberant Labrador Retriever or other large dog?
The most common and outdated method that trainers use to teach the “heel”, where the dog walks politely next to the owner without pulling on the leash, involves positive punishment via a choke chain. The dog pulls ahead, the owner jerks sharply on the leash which results in the choke chain snapping tightly around the dog’s neck. Most who use this method will quickly point out that if it’s done right, a quick snap, it doesn’t hurt the dog and is just a signal or reminder. I don’t know about you, but having a chain snap around my neck is definitely unpleasant and I would definitely not want to continue the behavior that caused this, IF I understood what that behavior was. It is therefore, by definition, a punishment, and it will decrease the undesirable behavior of pulling on the leash, but at a great cost to the dog’s emotional health, and sometimes damage to the trachea, because, frankly, I have seen very few owners (and trainers, too, for that matter), lay the groundwork for using this technique effectively and safely.
Unfortunately, punishment in this example deteriorates into negative reinforcement when the snap of the choke chain turns into an actual choke, and the owner doesn’t release until the dog stops pulling, probably because the dog is close to collapsing due to the choke chain restricting the flow of oxygen through his airways.
I have heard trainers use the term, “make like a tree,” i.e. as soon as the dog begins to pull ahead, the owner stops walking and stands still until the dog stops and/or returns to the owner’s side. The premise here is that the dog enjoys walking and is pulling ahead because he’s finding joy in discovery. By stopping and not moving forward until the dog stops pulling, the owner is removing the joy. . . an example of negative punishment. And again, this can rapidly deteriorate into positive punishment (he doesn’t stop pulling so you snap the choke chain to get his attention) or negative reinforcement (you keep choking him until he stops pulling). So, as this example demonstrates, the quadrants that involve punishment, seem to be intertwined.
Scientific research has proven that the most effective learning, both in terms of retention and emotional well-being is based on POSITIVE REINFORCEMENT, whether the subject is a child, adult or a pet.
So how do you use positive reinforcement to teach a dog to “heel” or simply walk by your side without pulling? First of all, you have to TEACH him instead of immediately “correcting” him. If you were learning to use a computer for the first time without being given any directions and every time you pressed the wrong key, your teacher smacked you on the back of the head, how eager would you be to learn? Yet, we do this to our dogs on a daily basis as if they were born with an inherent knowledge of what the words “heel”, “sit”, “down,” etc. mean.
Positive reinforcement means the owner is adding something very pleasant as long as the dog is not pulling, e.g. the dog is encouraged with a kind voice to stay by the owner’s side and fed tidbits of a favorite snack as long as he doesn’t pull ahead. In my classes, I do not permit the use of choke chains and encourage harnesses that have a clip in the front (“Sensation” harness). We also start all “loose leash” walking with a long line and/or “hands off the leash” if it’s a private session. This forces the owner/handler to “communicate’ with the dog instead of resorting to immediately pulling on the leash. It’s very difficult to convince owners to relinquish control and trust the relationship that they have built with their dog. The human psyche seems to thrive on dominance, to the point of damaging a dog.
I have had many well-intentioned clients who have mistrained their dogs because they didn’t understand the behavior-consequence relation. Every owner needs to understand how dogs learn, and even more importantly, make sure that they work with trainers who understand this as well.
Moving to a New Day
Teaching dogs should be fun for both owner and dog. I’ve worked with many clients whose dogs had become “problematic” because of faulty or aversive training. Whenever you move away from the positive reinforcement quadrant, and start adding or subtracting “punishments”, you run the risk of “poisoned cues” and confusion for the dog, or worse. When aversive techniques are used, dogs learn to suppress rather than change behavior. It’s called survival.
Operant conditioning in its most simple and fundamental form needs to be understood before moving to the next level of effective teaching which incorporates canine cognition and affective neuroscience, which will add a whole new dimension to our relationship with our dogs. We have to drop this whole idea of “me, master — you, beast” attitude with our dogs. I cringe every time I see “celebrity trainers” deliberately “push” a dog into a situation (despite the dog’s warning signals) just to dramatically demonstrate “dominance.”
It’s always easier to train/teach a dog correctly and wholesomely to begin with, rather than traumatize a pet and then have to “undo” everything to get back on track.
Mutual respect and cooperation . . . we as a society seem to have lost much of that in the way we deal with each other, as well as with our pets. When it comes to my own personal pets, I want them to think for themselves, to be by my side when we go on walks because they want to, to be happy and well-socialized. Positive reinforcement does not mean that you let them run amuck with no behavioral boundaries. Boundaries and limits are taught, and because I live with my dogs 24/7, they all learn a “decorum of behavior” by just being in my presence.
Dogs learn every minute of every day that they spend with you. Make sure that you are reinforcing the behaviors you consider “desirable.”
Please consider helping us with a MONTHLY GIVING program to support our dog food fund. All donations are tax-deductible since Extraordinary Dogs Inc. is a 501 (c)(3) non-profit public charity.
All of our dogs are “death row” survivors and are in training for service and/or therapy for their human companions.
Every donation is of tremendous help and is applied directly to the care of our dogs.
Thank you to all our supporters and friends who have joined us on this magnificent journey of love!