—NAPOLEON BONAPARTE, WHO WAS SAVED BY HIS DOG AFTER FALLING OVERBOARD HIS SHIP (HE COULD NOT SWIM)
Every dog owner needs to have an understanding of some basic dog training techniques and their application. The most challenging task for a dog is to connect words to actions, so we begin with markers or marking words.
And you thought dogs got to do all the marking? Since dogs do not have an instinctual association with verbal praise or punishment, it must be taught. Markers allow dogs to connect to and define words when they are paired to a particular action. The marker word is the word we choose to say the moment before we give a reward.
After a dog executes a command I consistently say, “All right!” and give the dog a treat. It will know what “All right!” means. The mark of saying “All right!” builds a bridge between the act of executing upon the command and the receipt of a treat. There is always a little downtime between the dog’s positive action and the reward; the marker creates a continuum between the positive action and the delivery of the reward. The fact that it comes before the treat makes it the most important link in this little daisy chain of requests (“Doggy, sit”), action (sitting), marker (“All right!”), and outcome (treat goes into waiting mouth).
Delivering the treat right after saying the marker word(s) is essential. Using markers in conjunction with rewards will cultivate a dog’s vocabulary and act as a homing device to the right behavior.
Should I say “all right” to my dog when the dog is outside the context of marking, the word’s meaning and impact will be diluted. Needless to say, phrases like “good dog” are easy to overuse, and can completely nullify the intended use of a marking word or phrase. The way I coo the rather generic “goooood giiirrrllll” (or “boy”) is reserved solely for training. Remember, to a dog, a straight “good dog” sounds completely different than “gooood doooggggg.”
When it’s well timed, a marker acts as the click of a camera shutter for the dog. Dogs are visual learners and will snap a picture in their mind of what they were doing the moment the marker hit home. With repetition, a bitmap of images forms and crystallizes around the proper behaviors.
Warnings are a type of marker that can let a dog know it’s missing the mark, so to speak. When a dog veers off course, a quick “ehh-ehh” signifies to the dog that the reward does not come along this path. The warning builds a bridge from “the land of the lost” to the “promised land,” taking the dog from being off course to on.
After saying the marking word, always give the treat. Should you serve up a warning and the dog catches on and comes correct, make sure you state the marker before dispensing the treat. Do not forget that for a marker to be productive, the word, phrase, or sound must contain a unique, dedicated meaning. My friend Kokomo uses “boo-yaa” as a marker word. When he says, “Doggy, sit,” and the dog sits, Kokomo says, “Boo-yaa!” to mark the behavior and create a bridge to the reward. The dog receives the treat and the circle is closed. The marker also signifies the end of the repetition. Feel free to be creative in making your own marker.
When things don’t go smoothly: “Doggy, come.” The dog veers off course and I warn, “ehh-ehh,” to build a bridge to the right behavior. When the dog gets back on course and starts heading my way, I can use another bridge word, like “yesss,” to offer a positive hint. With practice, the dog will understand “yesss” to mean that he’s en route to a reward and “ehh-ehh” to mean that he’s moving away from the reward.
Common mistake: delivering the marker, warning, or treat too late. To quote B. F. Skinner, “To be effective a reinforcement must be given almost simultaneously with the desired behavior; a delay of even one second destroys much of the effect. This means that offering food in the usual way is likely to be ineffective; it is not fast enough.” That is a little dramatic, but it is the reason we use markers in the first place or we’d really have to rush to deliver the treat. The marker buys you some needed time. Delays of even a few seconds may render the point moot, so be ready to repeat in order to get it right.
The clicker has gained popularity in recent years, and for good reason. A clicker is a small handheld device that makes a—you guessed it—clicking sound. It’s a pretty sharp sound that reminds me of a metronome and it acts as a marker. For some, having the clicker in hand helps with timing: In one hand, I’m holding the treat, and in the other, a clicker. The moment the dog executes the command, I click and offer the reward. Is it better than your voice? If it improves timing, then it may well be.
Given how I’ve harped on timing and rapidly closing windows of learning, using a clicker to close the gap between command, mark, and reward can be very useful. For those of us who get caught up in the excitement of the moment (“Such a good doggy, you did so good, who’s my doggy?”) or those who can’t hide their frustration, the discreet click removes the element of emotion and provides an unmistakable marker for the dog.
If clickers can improve our timing in dispensing markers and remove the fallibilities in voice, what could be wrong? Nothing terribly wrong, but it’s one more thing that you need to hold. You’ve got a treat in one hand, maybe a leash in the other, possibly a pouch on your hip, you’re administering instruction while watching the dog, and now you’ve got to remember to click. It may be inconvenient for some. I get satisfaction from personally communicating with the dog, and I believe our voices can deepen the connection, but that’s just my belief. I’ve used a clicker before and consider it an individual choice.
One potential drawback to the clicker is the need to offer a reward every time you click. That is nonnegotiable or the clicker will lose its value. Naturally, my voice is something I use all the time with my dogs, so weaning them off rewards is easier because the dog will not automatically associate my voice with a reward.
Whether you use a clicker or a voice to mark, the learning process is the same. At first the sound has no value or meaning to the dog. When you begin to click and dispense the reward, it comes to symbolize a sign that reads “Treats just ahead.” This is established with repetition, and when the dog is conditioned to it, the clicker is what’s known as “charged.” Just like we don’t overuse our marking words, make sure not to dilute the value of the clicker by using it in lieu of snapping to your favorite song. One person told me that she used her clicker compulsively when she worked from home, often snapping it off thirty to forty times a minute as she sat at her computer. She then wondered why it wasn’t effective with the dog.
Dogs are nature’s homage to the success of trial and error. As dogs get a handle on markers, they will build a compendium of favorable behaviors and constantly rifle through their archives in search of behaviors that produce reward. Once a strong base is laid and a sound vocabulary is developed, their demeanor and overall behavior become not only more predictable but generally better. It takes some work to get there, but once a dog’s learning engine is on, it will aim to please and look for ways to do so. Until this engine light goes on, we have to do our best to draw them in, and this means luring.
Luring is the introduction of a reward in order to draw in (or lure) the dog into training. Placing a piece of food by a dog’s nose is usually good enough to garner interest, and this is exactly what luring is intended to do. Certainly, it has a purpose, but I caution owners against overusing the simple technique. Dogs that have been “overlured” can come to balk at anything unless they see a treat. To use luring effectively, a dog’s interest must be piqued.
When training, it is a good idea to hold the treat out at your side. Initially, the dog will eye the treat before looking to your eyes for instructions on how to earn it. When I teach a puppy to come by holding a piece of food and repeating the word “come,” he will surely do as told, but it’s difficult to cull exactly what he has learned. Is the dog associating anything with the word “come,” or is he following his nose? Typically, a bit of both. With repetition, he will get the right message. Where luring truly works wonders is with the “sit” and “down” commands. If you hold a small piece of food by a dog’s nose and simply raise the food, the dog’s eyes will follow it. Quadrupeds, when looking directly overhead, will naturally go into a sitting position. In this case, even if the dog is following its nose and not your instructions, it will catch on just the same. The food goes overhead, and when the dog looks, it will surely sit. After the dog sits, please be sure to mark the behavior with “Boo-yaa!” or similar and dispense the treat.
To help our dogs learn, we can augment their never-ending treasure hunts for rewards by “capturing” the behavior we seek. For example, if you’re trying to teach your dog to sit and you catch the dog in the act of sitting, offer a reward. It is an example of continuous reinforcement: We reward the act even when we did not request it, in order to speed learning. Should you catch your dog lying on his back with his legs straight out you can reward the behavior, and the dog may learn a neat trick. From there, give the trick a name like “dead,” and soon enough, the dog will perform a pretty sound imitation of a stiff.
As mentioned in schedules of reinforcement, shaping a behavior is to reward the dog’s actions that come close but don’t perfectly fulfill the requirements. Your dog paid attention to you and went into lying position instead of sitting as requested. Shaping says you would reward the behavior and give the dog credit for trying. As the dog learns, these “close but no cigar” efforts are rewarded. Shaping says we give full credit for the partial execution of a command, but I often take it a step further by rewarding initial eye contact.
Remember, “The eyes have it.” When I begin to teach, I am often standing and holding a treat in my hand with my arm extended away from my body. When a dog is willing to look away from a treat in favor of looking at me, I know we’re on the right track. Before I attempt much of anything, I will wait until we are eye to eye, because I want to guide the dog to the treat. I may have to wait for a dog to exhaust all avenues first: Staring at the treat, jumping, whimpering, running in circles, and a whole host of inaccurate reward-seeking behaviors. Eventually the dog will finally look to me and say with his eyes, “What can I do to get that?”
To an outside observer, it looks like I’m trying to cast some sort of spell on the dog by looking him down. I’m not. The dog wants the reward, and he’s trying to perform for it. He’s on the case like an old gumshoe and will look to me for clues. Once he does this, we start building a relationship. It is my job to guide him to and through the command so eye contact is the first step and I’m happy to reward it. How long eye contact can be maintained varies, but if we can lock eyes for just a few seconds, I deliver the reward. This is the first piece of shaping.
Using the “sit” command as an example, let’s go through it. I say the dog’s name as I hold a treat out at my side. The dog looks to the food and I wait until his eyes are on mine. Ideally, we get a little stare-down going. I reward eye contact and ready another treat. This time, I raise the food up past the dog’s nose and toward my eyes as I say, “Sit.” The dog gets to look at the food and my eyes, a no-lose situation. Should the dog only approximate sitting, I mark that action with “Gooood boooyy,” and I give the reward. Eventually, these attempts will produce a successful sit.
Allowable margins of error are part of being a good parent to your child, and it’s no different with dogs. The beauty of teaching with shaping is that it creates a paradigm where “close” can win you the cigar.
With a target behavior in mind, shaping allows the dog to be led in the right direction and encouraged. Our own miscues are expressed through the dog, so when we can’t find a good reason to reward the dog, we’re doing something wrong. It’s easy to figuratively shake our heads until the working partner (the dog) in the relationship does something right. However, when we take responsibility for playing the lead role in this game of charades, then it becomes a team sport.
I want my dog to come to me when he’s called, now what’s the plan? Will I require the dog to come and sit on my feet, or will it suffice if he gets within arm’s length? How long is the dog required to stay if he comes to me? I must know this and plan with the individual dog in mind. For example, my observation of high-energy dogs has taught me that getting eye contact is one thing and keeping it another. Many high-energy dogs concentrate intensely in short bursts so I won’t sweat the small stuff. Dogs on the shy side may be intimidated by your physical presence, so turning sideways and squatting down makes you more approachable. When a dog is nervous or preoccupied, I will practice some patience and tolerance and give it some time to work through its discomfort.
In the real world, dogs jump, run in circles, and do all sorts of things, especially when you’re trying to teach them. Do not give up. Ignore the dog until he does something right. Rethink what good behavior means: The dog stops being nuts for a moment—reward it. There is no need to get into a bribing scenario where we try to distract the dog from its state of distraction. Unless you’ve determined that you are attempting to teach your dog in a competitive environment, paying the dog no mind and waiting to reward is still training. By “competitive environment,” I mean an area where there are loads of stimuli that would have genuine value to the dog. In the case of Maya, the second trainer attempted to train her outside a dog run, and had no chance. Picking a suitable environment or figuring out the best motivator is guesswork until it becomes intuitive.
Once I have a clear picture of what I am teaching and a basic plan to get there, establishing a basis for success is next. When things are not coming along, I will backtrack. Each basic step of a command should be performed soundly and consistently before moving on. “Soundly and consistently” means an eighty percent success rate. Eighty percent success rates are the goal. In order for your dog to be an A student, it needs to earn a B-minus.
Owners are often tempted to speed the process along or quit when they see the finish line. They speed the process by nearly performing the command for the dog and are ready to move on the moment their dog shows any sign of willing participation. Should I physically pull my dog over and stuff a treat in his mouth, he will not know the “come” command but will come to expect treats for being pulled. We have to teach until the dog is truly eighty percent of the way there or this omission will come back to bite us you know where. The dogs may think, “All I have to do is let this ding-dong pull me around while he shouts ‘come,’ and I get free food. I like this game.” Listen with your eyes, assess your dog’s level of interest, and wait until something right happens. Reward the good and move incrementally down the path. Once things are under way, raise the stakes by having the dog work a little harder for each subsequent reward.
To determine and understand your dog’s window of learning, conduct a little experiment: How long can you lock eyes with your dog? This is roughly how long you have to get your point across. Since dogs are domesticated to take cues from our eyes, they have a great facility to improve upon this. In my experience, even the most anxious dogs will calm down when their powers of concentration are directed onto our eyes.
With shaping, we form a nexus between steps of the process while steps themselves are the links. The comprehensive “come” command is linked by a few steps: stepping away from the dog a proposed distance, turning to the dog, getting its attention, calling the dog with “come,” the dog coming to you at a desired pace, having it stop and then sit at a set point. These steps are all linked, and dogs are visual, so they will associate our cues to link the actions into a sequence. Teaching the steps is a matter of creating cues. How much refining each step calls for is personal. For me, the “come” command requires the dog to come to me but not necessarily sit. Others want the dog to walk, not run, to them, come to a full sit, and be willing to stay put. Getting a dog to walk, not sprint, can be accomplished by marking the word “slow” when the dog slows down; even an “ehh-ehh” will naturally slow a dog. The dog forms relationships between actions and sequences, then connects the dots that produce a reward. The most complex routines of agility dogs are testimony to the power of sound linking.
Breaking down a command into individual components allows one to reward the right responses and correct the wrong. When a dog performs four of six steps correctly, how can it learn where it went wrong? By defining the individual steps of the complete sequence. Teach the steps in order, as each step is a cue to the next and a confirmation or mark of the prior action. It is fine to teach either forward or backward, but never out of sequence, or the links will break.
Once a dog learns a command, it will be punctuated with a word like “come.” There can be more than one word: “Come, Fido” or “Slow, sit, stay.” Repeating words like “come” is useless until the dog knows what the words mean. Only after the dog has demonstrated eighty percent efficacy will the word stamp the sequence for the dog. In teaching the individual steps, your physical cues, followed by the dog’s physical actions, are the real indicators of learning. This is what makes hand signals so effective. A definitive wave over will create an association with the word “come” a lot faster than the sole vocalization of a command. The first connection is usually made to the hand signal, not the word, so we link a hand wave to the word “come.”
The successful execution of a command does not necessarily mean a dog has it down pat. How you play in practice and how you play in the game can be very different, and this is where I see the most flummoxed dog owners. A dog learns to sit in the kitchen with a certain person, and it knows how to do just that: sit in the kitchen for that person. It does not automatically make the connection to “sit” as a concept. Dogs act locally and aren’t so great about thinking globally. What a dog learns in one place may not hold up in another. “Yeah, I understand what you’re saying, but you should see how good she is with so-and-so in the den” is the cri de coeur of many dissatisfied dog owners. How do you get a dog to take its learning on the road? Practice, practice, practice.
Remember how a dog can take a snapshot and create a bitmap in its head? When a dog is instructed to perform a command in the park that is learned in the living room, it might notice there are no sofas in the park and ignore you. It might even be laughing at you on the inside, wondering how you confused a park with your living room. It will recollect body language and tone of voice down to the minutiae and incorporate those details in constructing visual snapshots. It is in this picture that the dog understands the command, and outside this context, it may be at a loss.
To proof (or prove) the behavior, it is necessary to teach the same commands in a number of places, ideally with different people. Locales that are loaded with distractions are not ideal at first. Reward the dog regularly and increase the degree of difficulty as it gains mastery over the new environments. Buy some new treats and give them with vim and verve.
“Aversion” and “corrections” are two more words that push the buttons of controversy in the world of dog training. There are outspoken pundits who decry the application of what has been dubbed “aversion-based training.” I won’t engage this controversy but will tell you that an “aversive” is an action that intends to remove a behavior. An abusive aversive would be yanking a dog’s leash and making a mean face while saying, “Cut that out,” as is pushing a dog’s nose into its own feces. Conversely, if a dog is getting wound up and is “pinning”—or staring down another dog with visibly bad intentions—a sharp “hey” and the offer of an alternative also provide an aversive. The alternative could be a reward, a command, or any activity that will redirect the dog away from subpar behaviors. Aversives can be effective in dealing with dogs that begin to fixate or obsess on things. Physical aversives can be abusive, and they also tend to ramp up a dog’s stress levels when the idea is to bring them down. Aversives have become synonymous with abuse and punishment, and that is not what they mean.
A correction is exactly what we all know it to mean. In the vernacular of dog training, it has come to mean something abusive or negative, although it did not start out that way. If someone is walking a dog and intends to go in a different direction from the dog, the pull on the leash that the dog feels is a correction. It may be course correction, but it is a correction nonetheless. When the same dog beelines to get nose-to-nose with a pile of garbage and the owner spots shards of broken glass, and pulls the dog away, she has just corrected the dog.
When corrections turn stupid is when the dog is being corrected for something it did not know in the first place. We can’t correct something that was never learned; we can only teach it. From an evolutionary perspective, a dog’s very survival is dependent upon being of value to us. Too many trainers forget this simple fact and proceed to take out their incompetence on a dog that will lose interest and begin to avoid learning altogether. I once read that corrections are not for mistakes, and I could not agree more. Once a dog becomes well versed in a command or activity, then corrections take on the tenor of refinement as you lead the dog back to a correct behavior. These corrections lead the dog to a reward.
Dogs absolutely need to comprehend the correction, or, before you know it, you’ll be correcting the dog for not knowing the correction. The dog needs to understand both the action and the correction. Cold. When I am attempting to maintain eye contact while holding a treat, a dog’s hungry eyes may veer away from mine and to the food. When I say “ehh-ehh,” that constitutes a correction that tells the dog “eyes on me.” I am telling the dog that the trail to the reward is getting cold and I want to lead him back to warm, warmer, hot, and “Boo-yaa,” the treat. Corrections are best when paired with praise. My “ehh-ehh” to a dog straying from a reward should be followed by a positive marker once he’s back on course.
Teaching and learning from a dog that is truly engaged is one of life’s joys, and when this is realized, the activity becomes the reward. For both of you.