Getting Started With Markers!
The intelligent use of markers is one of the easiest things anyone can do to increase their effectiveness as a trainer or handler. Markers give us the ability to give the dog instant feedback about their choices allowing them to adjust quickly to changing circumstances. The more I learn about them the more amazed I am that any training gets done without them. Yet, I spent most of my career using no deliberate markers. I say “deliberate” because markers are all around us. Trainers use them, pet owners use them, vets use them, groomers use them. Anyone who handles animals uses markers from time to time. However, in most cases they are used in a scattered and disorganized way. So the average person never really sees their full potential.
A marker is nothing more than a signal that something has just changed or is about to change. Your dog already probably knows a bunch of markers you didn’t even mean to teach. Most dogs know what it means their owner grabs their keys or their purse. They know the sound of cabinet in which food is kept being opened. They know the sound of their owner’s car in the driveway. They know the doorbell. Each of these events communicates to the dog information about changes in their environment. In many dogs, each of these markers produces a predictable response as well. Most people can count on their dog behaving a certain way in response to each of these events. So markers often become cues to perform habitual behavior. When this happens, the behaviors are usually extremely reliable even in the face of stern punishment. This is a clue about building reliable behaviors. There is power in the marker!
The markers I listed in the last paragraph are quite powerful but as I said before, aren’t typically deliberate, and therefore, tend to produce reliable but often undesirable results. Trainers are found of reminding their clients that they are always teaching the dog whether they realize it or not. The predictable responses most dogs have to these signals are great evidence of that fact.
But how do we use markers to build the correct behavior? Let’s roll up our sleeves and get to it!
The first thing I want to make clear is that if you follow the instructions that follow, there is a good chance they will begin to act as cues as well. While this is likely and desirable, we need to make sure it isn’t actually a goal. Expecting it, or looking for it puts undue pressure on you and the dog. Teach the dog what the marker means, instead of trying for a specific response.
The No Reward Marker
The first marker we are going to talk about is the “No Reward Marker” or the NRM. I generally start with this marker for two reasons. The first is that people are just naturally programmed for saying “no.” Getting someone to mark unwanted behavior is much easier than getting them to mark wanted behavior. It’s just something about humans as a species. The other reason is that the NRM can offer very quick relief for many behavior problems. This relief is often necessary if we want the dog to succeed. Before people can invest in their dog in a positive way, it is often necessary to get them over the hump of unwanted behaviors. If the owner can’t see living with the dog as a good thing in general, they are less likely to put work in. So starting with something that can give them relief from problem behaviors is a great start. Furthermore, as you will see, by teaching them the NRM the way we advise, we are also necessarily teaching them how and when to reward the dog as well.
So the NRM serves 3 purposes, it teaches the owners how to use markers in a non-emotional way, it provides (in most cases) nearly immediate relief from many problem behaviors, and it teaches allows us to work on their timing for rewards and reward events.
The NRM marker is something that doesn’t really exist in human behavior so it sometimes takes time for people to understand it. Many people want to use “No” as the NRM. I generally discourage this, “No” is an entirely different marker that we will discuss late. The NRM is unique because there is no implied “or else”. When humans tell each other to stop doing something there is an inherently implied “or else.” The implied consequence may only be “I’ll be unhappy with you,” but there is almost always the implication that you won’t like the results if you keep doing that thing. The NRM offers no such condition. This may be the toughest thing for many handlers to master. “Or else” is so ingrained in our way of thinking, most people aren’t even aware of it. But it comes out when I start having them use the NRM. I use the word “uh-uh” and I instruct them to say it in an upbeat way. But most start with a very harsh tone. They want to “load” the or else in the tone. Sometimes, I require people to say “oopsie” instead because it’s next to impossible to sound menacing while saying “oopsie”!
Learning how to tell the dog he’s making a mistake without issuing a threat is a valuable skill that every handler should learn. Sadly, it took me many years as a working professional to even understand that we should be doing that.
This may be at the center of some of the controversy surrounding the NRM. Many trainers say the NRM discourages the dog and causes a flat performance. My experience is the exact opposite of theirs though. I find that dogs respond very well to the NRM in most cases. In fact, it often creates a renewed vigor and joy in the opportunity to earn reward by getting back on track. I think there are two reasons my experience is different than that of so many other trainers. The first is that I we deliver it without the typical human implied or else. The second is that in the beginning EVERY “uh-uh” is followed by a “Yes” and a reward at the appropriate time. So the NRM doesn’t merely mean “You’re wrong” it also means, “but there’s an opportunity to gain something good here.” So there’s no threat and actually a hope for a reward. This protects the dog from discouragement a great deal.
The NRM means “The thing you are currently doing will not produce a reward.” It’s not a threat in any way. If I tell you “that Coke machine is broken, you can put your money in it, but no drink will come out,” you won’t see that as a threat or even see me as the bad guy. You may experience some disappointment, but that will be directed at the circumstance not at me for passing the information along. This is the purpose of the NRM. It tells the dog he’s off the race course and can’t win until he gets back in bounds. We want him to win. In fact, it’s our job to ensure he wins as much as possible!
So with the dog on the leash, walk him towards a distraction. When the dog starts to consider leaving your side to go for the distraction, plant you feet firmly, prepare to stop the dog’s forward motion and quietly say “Uh-uh”. If the dog pulls, keep your feet planted and apply just enough back pressure to keep the dog from getting to the distraction.
Now … we wait …
The dog may pull for a long time, or short time. He may be super committed or he may be a bit less committed due to early intervention and our unusual response. Let him work it out. It’s very important that whenever safety allows, the dog DECIDE on his own to give up on the distraction.
The moment he looks back at us, we say “Yesss!” and offer something of value in return. This may mean a food treat or a game of fetch or just a lot of love. When using affection as a reward, it’s very important to make sure it’s meaningful. I see many people who absently give a dog a scratch on the head and an half-hearted “good boy” who think that’s a valuable reward for their dog. In most cases they are vastly overrating their own importance to the dog. Give the dog a real good scratch with lots of joy and enthusiasm. We want to foster a joy of working in the dog, a sedate style of handling isn’t likely to encourage that. This should be no problem if you find yourself genuinely on their side. Their victories, no matter how small are your victories. Embrace them and rejoice in them!
In any case, the idea here is to make sure the dog sees the upside to ignoring distractions marked with “uh-uh”. He should be aware that he’s choosing between a reward that’s not actually attainable and one that is easily attainable. To continue the Coke machine metaphor; I tell you the machine is broken, but being a human being you decide I might not know what I’m talking about. So you go put your money in and push the button. Nothing happens. Now few people are going to be satisfied and stop there. Most are going jiggle the change lever, push the button a few more times, try other buttons, some people will try to kick the machine or slap it “just right.” I stand there and wait patiently while you try to coax the machine into doing your bidding. But at some point you give up. You shrug your shoulders and turn your attention back to me, and I say suddenly “But you know what! I just remembered I have a Pepsi here, do you want it?” Maybe you really wanted a Coke, but the Pepsi is right there. So you thank me and take it.
If this pattern were to repeat over a period of time, two things would become pretty clear to you rapidly. The first is that I’m never wrong about the operational status of Coke machines. The second is that I’m happy to give you a Pepsi when you’re thirsty, all you have to do is stop messing with the Coke machines. There may always be a bit of disappointment about the Coke machine being broken, but the fact that a cold drink is still available, does a lot to mitigate that disappointment. And of course in many cases you’ll still be able to get a Coke when you want it.
This practice of always offering a reward for ignoring the distraction will do a lot to build a consistent habit in your dog and it doesn’t usually take that long. Once the NRM is learned, it can be applied in any number of circumstances with great results.
A few words of caution though. If I tell you the Coke machine is broken and put your money in and get a drink, you aren’t likely to believe me next time when I tell you it’s broken. So it is vital to your success that in the beginning you only use the NRM when you are in a position to ensure the dog doesn’t get the distraction he’s trying to get to. Also remember that sniffing is fun for dogs, so if you’re using the NRM to stop sniffing, you will need to use enough leash pressure to get his nose off the ground. Allowing him to sniff while you say “Uh-uh” is the same thing as feeding him a treat while saying it. It will make the “uh-uh” pretty meaningless at least as far as sniffing goes.
The Success Marker
The second half the NRM is the Success marker. I use the word “Yesss!” This marker indicates the dog has succeeded in something and has earned a reward. In the beginning we want to offer a lot of rewards for even small effort. As we move along, rewards become harder to earn and further apart. But in the beginning we want the dog excited about the prospect of earning many rewards easily. We are building something known as a “Tolerance for Non-reinforcement” (TFNR). The TNFR is a complex thing and I won’t go into a lot of detail about it here. Just be aware that the dog’s confidence in his ability to earn the reward is a huge part of his ability to work without immediate rewards later. This confidence comes from lots of rewards early on. It’s almost impossible to over-reward the dog in the beginning.
So in first few sessions, we are going to want to offer a lot of rewards. This will help build the dog’s motivation.
The timing of markers is always important, but I think it’s most important with the “yesss” marker. People often want to wait until they can actually give the dog the reward to mark. So if the dog is at the end of the leash and looks at them, they want to wait until the dog is right next to them so they can offer the reward. This is certainly understandable, but unnecessary and in fact, it can be harmful to overall success. The success marker should be presented as soon as the dog earns the reward even if he’s too far away to actually receive it. This is the reason markers are so valuable in the first place. They are more precise than just offering a reward. With good timing you can tell the dog exactly when he does the right thing. If he’s at the end of the leash and turns to look at you, mark that with “Yesss” immediately. In most cases he will rush back to you to get the reward he’s already earned. He will know exactly what he needs to do to turn that reward on in the future. So be very quick with your success markers!
With a little practice, these two markers can help you solve a lot of behavior problems and in the process will improve your timing and your communication with the dog.
Here is a video of how this works in real time.
*** The Social Nature Of Dogs ***
Want to start a fight online?
Go to a dog training forum and state an opinion (any opinion) about “pack dynamics”.
Hell, mention the word “pack” and you’ll see the fireworks start pretty quickly.
It seems like everyone has a strong opinion about the inherent structure of canine social groups, and they are ever ready to spring into action to defend those beliefs. The word “pack” seems to be a key element in this divide.
For the uninitiated, I’ll break down the broad strokes of the conflict. There are many people who believe that because dogs are directly descended from wolves (there is actually room for debate about this but that’s another matter) that their innate social structure is reminiscent of that of wolves. They will go on to suggest that many behavior problems we see in the domestic dog are the result of our failure to recognize their need to have a stronger social hierarchy similar to their wild ancestors. While the details of this model vary in degree, the idea of a linear hierarchy is central to the concept. Trainers who adopt this idea are likely to talk a lot about “leadership” or “dominance” or perhaps they will admonish clients to be “alpha”. At the core the idea is responsible, compassionate dog owners have an obligation to provide a clear social hierarchy for the dog because it’s absence will necessarily create stress and behavior problems. Because compliance is related to status, this philosophy is more likely to produce labels such as “attitude problem”, “stubborn”, or “disrespectful.”
On the other side of the coin there are those who insist that the domestic dog is not a “pack” animal, and retains few of the wolf’s social characteristics. They will argue that 1000’s of years of domestication has removed such traits. The idea is that living within human society requires a completely different skill set from living with a bunch of wolves. We live in closer quarters, so dogs have to be more accepting of social groups outside their primary “pack” or “family”.
In other words, wolf packs don’t share territory or intermingle with wolves from other packs, but dogs are expected to share plenty of public spaces with other dogs. Most of them socialize or want to socialize with random dogs they meet on the street.
The debate becomes trickier still when you try to compare experiences. People who raise their dogs according to a strict hierarchy concept will often report that whenever they get a new dog, the old dogs teach the new dog the rules. They will observe that strict hierarchies are practiced in their household.
On the other hand, people who don’t believe in the linear hierarchy model, see a more loosely ordered social structure. While they might observe occasional contests over certain high value items, for the most part they will observe their dogs seem to have a much more free flowing sort of relationship.
Appealing to the study of feral dogs doesn’t seem to help much either. There doesn’t seem to be much consensus on the matter from those reporting on the subject. Conclusions range from suggestions that feral dogs have clear hierarchies to an idea that feral dogs tend to congregate into loose social units and change groups frequently or even go solo for awhile.
One issue we face is that the behavioral and ethological studies of domestic dogs are rarely undertaken from a purely informational point of view. The design of the study, the data collection methods and interpretation of data always seems to lend itself to a particular predetermined outcome. On the whole those who deny the pack model always seem to find exactly what they are looking for. Likewise those who preach it, happen to collect data that supports that theory.
While the problem with stilted studies and cherry picked data are a barrier to answering the question about feral dog social behavior, I think there is a bigger issue. I think maybe the entire debate is based on an erroneous assumption.
Let me explain.
Dr. Robert Sapolsky is a Neuroscientist and Primatologist and Stanford University. Sapolsky was studying a particular baboon troop when something interesting occurred. The social nature of baboons has been a pretty settled fact for some time. The typical baboon troop resembles a more violent version of the structure often ascribes to wolf packs. In a baboon troop, status is everything. If the leader gets mad, he takes it out on an underling, who goes and knocks around one of his subordinates, and so it goes to the bottom of the order. A juvenile baboon who is minding his own business can get attacked out of nowhere just because someone else was having a bad day.
The alpha males in the troop Sapolsky was studying discovered that a neighboring troop was eating out of a garbage dump from a tourist resort and wanted to get in on the easy food. They would organize raids on the other troop’s territory in order to steal food. This is highly aggressive and dangerous behavior. Only the most assertive males participated.
It turned out the resort threw away a bunch of meat that had been infected with tuberculosis. As a result all of the baboons who raided the garbage dump got tuberculosis and died very quickly. Within a matter of days, Sapolsky’s troop was left with no alpha males. What remained were juveniles, females and mid-level males who preferred grooming over fighting. Overnight, the culture of that troop changed. Suddenly, there was less conflict and fewer fights. The troop was far more cooperative than any that had documented before. When a new baboon would join the troop, if he tried to start conflict, it would be shut down and within 6 months new baboons would be playing by the new rules. Life expectancies increased throughout the troop. Eventually, this new culture was being carried on even though no one who was part of the original troop when the shift occurred remained. This was the first documented multi-generational cultural shift in any species other than humans.
This story raises and obvious question that eventually we must turn upon our own thoughts towards dogs: What if there never was an inherent social structure for the baboons? What if they all acted that way because that’s what happens when you have a bunch of assertive dominant types and no one to enforce fairness. What if it’s not a structure embedded in their DNA but the simple effect of group dynamics where there are plenty of incentives to bullying and none to being altruistic? The answer seems to be an obvious “yes”.
Turning the same question towards dogs, it would seem even more likely. Leaving aside the structure of the wolf pack for a moment, let’s consider the selective pressure that had to have been placed on the domestic dog from the outset. Humans would have been more likely to keep and care for dogs who were more biddable and tractable. There is no real dispute on this point, but most seem to look at the smaller behaviors instead of the larger ones. For most it means being easily handled, easily trained and willing to work with and for humans. Of course these traits would be important. But those traits would also require the dog to adapt to many aspects of human societies that were foreign and alien to the social world of their ancestors. In fact the acceptance of handling, biddability and trainability by another species would require a great deal of social adaptiveness as a precursor of those traits, not the other way around. What if the dog’s greatest asset is that of adaptation. It hardly seems a stretch to suggest that the fundamental social feature of the domestic dog is to adapt and adjust to fit the culture he is adopted into.
In fact, of all the theories about dog social nature, it is the one that seems to best explain the observable data.
* Dogs who’s owners enforce a strict social hierarchy have been observed helping teach new dogs the same rules.
* Dogs who’s owners don’t enforce a strict hierarchy are observed to have more fluid interactions that appear more contextual than status driven.
* There seems to be no universal data about the behavior of feral dogs despite plenty of feral dogs in world to observe.
* The very traits listed as the ones most likely to result in successful domestication all hinge upon a truly flexible temperament.
I think it’s time we stop trying to figure out what the “native culture” of the dog is and realize that just like with the baboons, such a thing doesn’t exist. If it ever did 10’s of 1000’s of years of selective pressure have replaced it with something more domesticated.
So the question should no longer be about what they would do without human intervention, but rather, “what social structure best benefits the dog?”
I believe the bulk of the dog training community has been looking at this question all wrong, after all, the “natural habitat” of the domestic dog is the same as ours … and it always has been.