Will I Always Have to Use Treats?

If you like the T Shirts or Products
shown please visit www.muttzrus.com

This is a common question that rarely gets a complete answer, and to give you the best answer, this post will be a little heavy in terms and psycho-babble…I’ll apologize up front, but I have a BA in Psych from KU (Rock Chalk!) So, training in a positive reinforcement manner requires two things; a behavior that you want and a reward to maintain or reinforce that behavior. The problem is, many people think that rewarding a dog is synonymous with bribing them and once started, it will have to go on forever! This is an untrue statement (bordering on a lie) that is commonly used by those who want to punish or correct a behavior in order to gain control. It is my job in this post, and as a trainer, to convince anyone reading that using treats or rewards is the best and correct path to take; so let’s get right to it!

There are three things you can do to a behavior; reward it, punish it or ignore it. Let’s take the example of showing up to work on time (or not.) If I punish someone who does not show up on time, I become the bad guy and my employees learn to avoid (or hide from) me, or they only show up late when I am not around. If I ignore it, my employees see that nothing happens when they are late and learn they can continue with the behavior with no effect, good or bad. Finally, I can reward showing up on time with praise or a reward (a raise) and they will continue to replicate the behavior because it feels good. The reason for sharing this example is to show the futility in ignoring or punishing behavior…those responses do not necessarily reduce or eliminate a behavior over the long term. You may get some immediate change in the behavior, but no lasting effect. Unfortunately it breeds avoidance, indifference or lack of motivation. I think each of us have had a boss that fits this bill at one time or another.

So let’s get back to dog training. When I first begin teaching a behavior, I reward for every success. If I am teaching sit and the dog’s rump hits the ground, I follow up the behavior with a tasty treat. This is where most opponents to treat training start screaming “see, I told you they bribed the dog” or “my dog does what I tell them because I am in control” (sounds like that boss we were talking about, doesn’t it?) What those folks don’t see is that all good positive reinforcement trainers will begin to fade the reward as the dog becomes more reliable with the requested behavior; showing the reward doesn’t have to go on forever. Humans figured out long ago that if we continue to reward every time a behavior is given, the reward eventually loses its effectiveness. Think about Vegas. If you walked into a casino and saw a slot machine that paid out a dime for every nickel you put in, you would probably sit down and start playing! But if you found out that you could only put in one nickel per play, after a period of time the dime pay out would get old and boring. If, while feeding one nickel into your machine at a time, you notice the person next to you playing a quarter slot and all of a sudden their machine went nuts and started making noise and paid out 46 quarters, which one of these machines would you want to play? This is an example, simply, of what psychology refers to as continual vs. intermittent reward schedules.

How does this relate to dogs and training? Simple, when we first teach a behavior, we reward every success (continual). We continue this until a predetermined level is reached, then we switch to rewarding only randomly (intermittent). For me, this predetermined level is an 80% success rate. This means that when a dog is successfully performing a behavior 80% of the time I ask for it, I believe they know the material and are ready to move to “Vegas Style Training”. In other words, their rewards are ready to come randomly. This improves behavior because now you choose which version of behavior to reward. In other words, a slow “sit” would get nothing while a prompt “sit” would be rewarded!

So why 80% and how often will I have to reward…read on! 80% is a level that I picked based on my own personal experience (that you might all be able to relate to!) When I was a kid, if I got straight A’s (90% or above) my parents made a huge deal out of it, it was a PARTY!!! If I got all C’s or lower (>70%) it was not a pleasant night in the Deathe household. If I got all B’s (80%) nothing much was said because I knew the material, not well enough for a Party, but good enough to NOT get in the dog house. At 80% the dog is ready to move on to an intermittent reward schedule. So how much should you treat your dog once you move them to the intermittent reward schedule? I use a treat 20-30% of the time (2-3 treats per 10 correct responses). The magic is that once you have this level mastered at 80%, you replace treats with affection and life rewards for the dog. By this, I mean love and pets become the reward used in most cases for positive reinforcement. Even though my primary reward system for my dogs has moved to affection, I continue to use food treats periodically to keep interest and motivation going strong, and especially when I’m trying to reinforce a behavior or when I’m teaching a new trick! So I ask you…are using treats in dog training the right or wrong way to relate to your four legged friends? Only you can decide!

Sit Part 3…Signal & Cue!

If you like any of the shirts or products
shown please visit us at www.muttzrus.com


Now you have to decide which verbal cue or hand signal you will be using for “sit”. As with all of my training, I focus on hand signals first! Remember, DOGS DON’T SPEAK ENGLISH! I simply modified the luring movement of my hand with the treat, to become my hand signal for “sit”. Imagine taking a treat between your thumb and index finger, palm facing down, and then flip your hand palm up. My hand moves about 12 to 18 inches upward from waist level to nearly shoulder height. If you have been consistent with using the lure and clicker, the addition of the hand signal should be very quick. So, is my hand signal the only one you can use? No, I am simply explaining the hand signal that I use. There are many out there and none are any better than the other. But you need to make sure that each signal you use is distinctive and not at all similar. If they are too close in similarity, it could cause real confusion for the dog. Imagine if your hand signal for sit is pointing down with your index finger and your hand signal for down is pointing down with an open hand. Not very different are they? Once you determine the hand signal you are using, you should spend at least a day working with the lure (treat) and clicker and several days thereafter solidifying the hand signal. Within the week, the hand signal should be working at least 70-75% of the time if you are practicing in a non-distracting environment. It would be foolish to expect this to work at, let’s say, a dog park this early in the training. The topic of distraction will be covered in a future post as well as “The 3 D’s” of dog training.

Notice that at this point there is still no verbal command…no “sit”! This is intentional. I don’t want to add a second cue until Fido is proficient with the hand signal. I always start out with a hand signal first because body language and body position are the easiest forms of communication for dogs to learn. Remember, DOGS DON’T SPEAK ENGLISH! It is important here to understand that until the first cue (hand signal) is successful at least 80% of the time, trying to combine or add another cue is not advisable. Let’s assume though, that Fido is now at or above the magic 80% level and we are ready to name this behavior! Now it doesn’t matter what we call this behavior because DOGS DON’T SPEAK ENGLISH! I could call it “sit” or I could pay homage to one of my favorite musicians and call it “Margaritaville”. The word itself matters little to the dog; it is just easier on us to call it “sit”. So how do you do it? The key is knowing which cue to lead with, and trust me it is important. To combine cues, make sure to use the unknown cue first. Just say the word sit (or Margaritaville) right before you give the hand signal for the command and in no time the dog will “combine” the cues and learn that each means the same thing! Saying sit before the hand signal is the key. If you were to give the hand signal first and then say “sit”, the verbal cue would be ignored because it would become background noise, since the hand signal is the only thing that has meaning at this point. Think of a school crossing guard with the hand signal to stop and the stop sign (verbal command). Granted, they do not use a verbal signal to the cars, but the hand signal is the universal command for stop and the stop sign is a written translation of a verbal command. So, people first learn to stop with a hand signal, as dogs learn to sit with a hand signal. Later, people learn to read a sign indicating they need to stop (equivalent to a verbal command) just as Fido learns to recognize the word “sit”. In both cases over time, the cues are combined and end up meaning the same thing.

Continue to work with your dog, using treats, practicing both the hand signal and verbal command together until the dog reaches the 80% success mark. At that point, I recommend spending several weeks where you continue using treats, but practice the hand signal and verbal command separate from the other to make sure each is understood independently. When we can say Fido has each version of sit down pat, we will begin to do what is known as fading the reward. This idea will be covered in the next post; “Will I always have to use treats?”

Don’t forget to take the time to vote on the poll question….What topic do you want covered next?

Sit Part 2....How it's done and using a clicker!

If you like any of the shirts or products
shown please visit www.muttzrus.com

OK, at this point you know that a polite “sit” is the way your dog asks for something or says please. Now for the big question, how do we teach sit? Well, we know that “sit” is when a dog puts his/her rump on the ground, right? So, let’s use a little bit of physics to help us get this behavior. Put simply, where the nose goes, the butt follows! I start with a basic luring technique. I hold a treat near the dog’s nose until Fido is interested in it, and then move my hand with the treat in it, back over the dog’s head towards the tail. You will notice that as the dog tries to follow the treat with their nose they fold into a sit so they can get to the treat! This sounds really simple and it is; but you have to consider the value of the reward and the speed in which you lure the dog with the treat! With a couple of tries and some patience you should be able to consistently lure the dog into the sit position! The mechanics are easy; the process is a little more complicated! When do I click? Do I say “sit”? When do I treat? Do I use a hand signal? The answer to all of these is yes, but the order is specific and important!

As you are luring the dog to this position you need to also be using the clicker! I know that some people are not fans of this device and debate its usefulness, but if you have been reading my blog at all, you know I am a big believer in these simple noise makers as a way to tell your dog they have done something correctly. To those who are new, see my post on terminology for explanation of positive markers. The idea of a clicker is to bridge the time period between performing the right behavior and getting the reward. The click will be used as a way to let the dog know that what they have done is correct, and that a treat or reward is on its way. So, as you lure the dog into a “sit”, make sure to click the moment Fido’s rump hits the floor, and follow with a tasty treat or toy, to let them know they got it right!

I need to mention at this point that a release word must also be worked into the picture. If not, you will find the dog releasing their “sit” at the sound of the clicker. Remember, the click is the positive mark (letting the dog know they have done something right); but how do we let the dog know it is OK to come out of the sit and get the treat? Enter the idea of a release word. The release word is used in conjunction with the positive mark (clicker). The idea here is the clicker tells the dog something correct has happened, and a treat is on the way, while the release word tells the dog the behavior is complete and now you can come get the treat or reward. So, after the rump hits the ground and you click, you will need to say the release word (all done is mine) then allow the dog to have the treat. As the dog becomes better at the command sit you will begin to add time between the click and the release word, which will start building time (duration of the command) or what we call stay (this concept will be covered in a later post.) As you go forward you will find that many dogs will begin to work just for the noise of the clicker! At this point, we not only have a dog that can be lured into a sit but one that is working for the positive mark of the click.

So now you’ve got “sit” mastered with the click, release word and treat. What is the next step? Check out the next post to learn the importance of timing as well as the difference between the hand signal and the verbal command. You’ll then determine if it is more important to speak human or dog!