Last updated Feb 9, 2004

PART TWO OF
"It's all just behavior to the dog"

Describe the steps required to move from a lure to a hand signal
when teaching a first time dog owner to teach the sit / down / stand.

1. Define - what is a lure?
2. Define - what is a cue?

Keep in mind as you do, that, as Leslie Nelson so wisely states, "humans are notoriously stingy"
While we are encouraging them to move from a lure to a cue, we are not talking about
"getting rid of the food" ... we must strive to keep them GENEROUS in their rewards while you
explain the process of moving from food LURE to food REWARD.

Great job, guys! [great stuff is in purple]

First to answer - Lisa !!!!!

Lure: Dog follows food to get into position and receives the reward (reinforcement) once he hits the correct position.

Cue: Stimulus for a behavior; something that causes the behavior to occur at a particular time. A stimulus effects (causes, or triggers) a behavior, reinforcement strengthens the behavior. Before you can put a behavior on cue, that is expect the dog to produce the behavior reliably in response to a specific word or signal, the behavior must be an operant -- that is, a behavior the dog emits voluntarily in order to trigger reinforcement. Only then can you introduce the further contingency that the behavior will be reinforced only when produced in the presence of a given stimulus (cue). (thank you, Morgan Spector)

Moving from a Lure to a Hand Signal:

(1) Demonstrate the lure and show them how at the same time they are building in hand signals.

(2) Lure the dog in to position many times and then slip in a "lure" without food in the luring hand. Quickly take a piece of food from another location and feed the dog once he is in position.

(3) Then explain that the dog MUST be proficient at luring..."we, humans tend to expect too much from our dogs too quickly. We think that if they do a sit a few times in the kitchen, or here in class, that they should automatically do it every where. Not so. You may be able to get the dog to sit using a hand signal here in class. But, it is unlikely you will be able to do it out on the sidewalk with other dogs walking by.

Be aware of the environmental context in which you have taught your dogs the commands. -- Environmental context means where you are and how much distraction is happening. You may be able to get good responses on the front lawn when the neighborhood is quiet, but your dog may go deaf in that same position with children riding by on bicycles. The children in the picture have changed the environmental context-- SO, practice in new situations using a lure and fading to the hand signal and give your dog the benefit of the doubt. It is very, very, very important that you continue to reinforce the sit, down, or stand generously. Fading the lure does not mean fading the reward. If you do not have food handy (demonstrate appropriate praise).

(4) Once your dog is responding to the hand signals and has received enough rewards from you so that he is responding quickly and enthusiastically, then and only then do you fade the reward. Give a reward every other time he responds. Gradually move to every second time. Do a few at every time. Then go to every third time. Every now and then throughout the dog's life reward what you want with something the dog thinks is valuable and it usually includes more than a pat on the head. The possibility of a big payoff is what keeps people going back to play the slot machines...."

And Nancy !!!

Steps required to move from a lure to a hand signal...
To "lure" an action is to use the dog's involuntary reaction to an object (usually food) in order to obtain a desired result.

So, a savory, aromatic, drool-worthy piece of treat stuck to the dog's nose will produce certain involuntary reactions (besides the drool) when that irresistible treat is moved in certain directions. The dog doesn't have much control over what happens... it just does... but he produces the desired end result and gets the treat. Repeat. Repeat. Blah, blah, blah, Ginger. Lure.

The savory, aromatic, drool-worthy piece of treat goes from being "on the dogs nose" during the lure movement to being concealed within the hand during the lure movement. Since the "If > Then" has already begun, the distance between the treat and the nose can increase. The dog's association, unbeknownst to the dog, begins to be less about the savory, aromatic, drool-worthy piece of treat and more related to the hand movement. This is subtle change is further reinforced when the reward is given at the end of the process... along with praise... when the desired result occurs.

Next, the savory, aromatic, drool-worthy piece of treat is no longer in the hand during the lure movement. That savory, aromatic, drool-worthy piece of treat is nearby... in a treat bag or pocket, but right there and readily available. We continue the lure movement without the treat in hand. The treat and praise are given when the desired result occurs. The dog's association between the hand movement and the behavior increases. The treat is beginning to become a separate issue.

Soon, the savory, aromatic, drool-worthy piece of treat which is nearby begins to come randomly. Every time the dog responds correctly to the lure movement, he is highly praised... but only every other or every third time will the dog get the savory, aromatic, drool-worthy piece of treat which is nearby. The lure movement is becoming a hand signal. The dog is developing the "If > Then" behavior of "If I see this (hand signal), Then I do this (action)". The treat is continuing to distance from the "If > Then" behavior.

Gradually, the savory, aromatic, drool-worthy piece of treat comes less frequently as the dog develops the "If > Then" behavior.

The lure movement is replaced by a hand signal. The hand signal becomes the "cue" to put into play the "If > Then" event.

Finally, a "cue" is an indicator... verbal or non-verbal which produces a desired action from the dog.

Now... do I get the savory, aromatic, drool-worthy piece of treat?"

TEN! and and extra jackpot for answering! *S*S*S*S*S*

And Kim !!

Dogs do not speak english, they don't even speak human. They speak dog. One of the ways they speak/learn is in pictures. So if we are to teach them to understand what sit is we must first "speak" their language, show them what we want, then translate that into human. So with the food on their nose, lure them into position. As soon as they hit the position say "YES!" to mark it as yes this is what you want, pop the treat and translate it, "good sit". They will soon learn that sit means tush on the ground. When they understand the command we start removing the food. You will not always need to carry treats with you or use treats as a reward. A simple smile and "good sit...thank you" works. Praise is the biggest reward. Telling that what they are doing is good and what you like is the most important thing.

That is what I say on week one. But when exactly do I remove the food from the students? Is that an individual thing? Do I do that as a class? I guess that is where I feel I am falling really short. Am I creating treataholics, "won't do it unless I have food" dogs? Or am I just stressing too much?? (I stress like this with the shelter dogs too....am I reading them correctly? Did I miss something? A dog I mark as a good dog, will it go into a family and then bite the child? Huge weight!! thats why I dont sleep well on Saturday nights. Wonder how well Lisa sleeps.) As far as the definitions-without looking them up- I would define them as: Lure- something desireable that guides the dog into position Cue- a physical or verbal signal to preform a specific task.

And Danaaaaaaa !!!

Moving from a Lure to a Cue:

Show the reward to the dog and the deliver the reward; Maybe even two or three times. This is to ‘prime the pump’ and get the dog in the game.

Lure the dog to perform the desired behaviour. As the dog achieves or approximates the desired behaviour, mark it with “yes!” then reward. Remember to move slowly enough that your puppy can follow the lure, yet not so slowly your puppy gets bored with following. Keep their attention.

Repeat the Lure/Reward until the dog easily performs the behaviour every time you lure and you notice some sort of recognition of “hey! I know this” in your puppy. Remember to mark every completed behaviour with a “yes” and be sure to reward. If it’s an exceptionally good response to your cue a ‘jackpot’ of multiple treats is always a welcomed reward. Don’t be stingy when teaching a new behaviour.

Changing a lure to a hand signal is easy. You’ll want to decide what signal works best for you and your dog. Be sure it’s clear to the dog. The more clear the signal, the easier to add the cue.

Have a hand-full of [treats] in one hand and place a really wonderful treat in your lure hand where your middle fingers meet your palm, placing your thumb over the treat to keep it from accidentally finding it’s way into your puppy’s mouth. Lure your puppy with the treat to get the behaviour and mark it with that ‘yes!’ and reward with a treat from the other hand. Don’t treat with the lure hand.

Repeat this a few times. Your puppy may appear to be confused when the food isn’t delivered from the lure hand. Explain it clearly by luring with one hand and rewarding with the other.

Now remove the treat from your lure hand. The lure hand [ cue ] should look the same to the puppy without the treat as it did with the treat. Offer the cue [ empty lure hand ] and mark the behaviour with a wonderful “yes” or “woo hoo!” and reward/jackpot.

Now it’s time to label what the puppy has already been doing. While using the treat to lure, add your verbal cue “sit/down/stand...etc.” at the moment the behaviour occurs. Mark it and reward. Repeat this step four or five times and remember to reward and add that ‘jackpot’ for that really good response.

Now you’ll start adding your cue “sit / down” before you lure the behaviour. The cue precedes the lure. You did remember to mark those sits with a good “yes!” and a reward, right? Good. After a few repetitions of this your puppy may start to recognize that the cue always comes before you lure them and they will begin to anticipate the lure. When they hear the cue “sit!” they will most likely offer the behaviour before you can lure them. You need to ‘jackpot’ this one and have a party - right here - right now! A: you’ve now moved from luring the behaviour to cuing the behaviour and B: your puppy is learning how to offer good behaviours for attention.

I think this is the first time in a LOOOOOOOOOOOONG time that there's no Cathy answer ... *sniff*

My reply:
(I always write my answer before I get yours.)

Why do our students continue to hold food in their hands for the entire class time? Having recently been an active student in a couple of classes, I can say from experience that it's because, unless you are sitting right next a table or ledge to put your food container, it's the only way to have the food really convenient without fishing for it, even in a treat tote. Open containers on chairs invariably end up with the dog-next-to-you's nose in it. This point is being kept in mind for our future classroom planning!

DEFINITIONS:

Lure: A piece of food or other object that the dog is interested in enough to follow with his nose. A target may also be used as a lure. The dog shifts its position in order to move toward and/or contact the lure. Using this physical orientation, the trainer is able to manipulate the dog into desired postures and positions, then marks it with a "yes" or click and delivers a reward. The goal: the dog learns "when I do this, I can make food happen."

Cue: A signal to the dog to perform a specific behavior. Stimulus control is when the behavior is totally "on cue"- i.e. the dog does not offer the behavior unless the cue is presented. The cue can be visual, auditory, environmental or even smell. The goal: the dog learns "when I see (or hear or smell) that, they want me to do this, and then food happens."

The cue is only added AFTER the dog is giving you reliable *offered* behavior, ie: when the dog is consciously CHOOSING to perform the desired behavior with some level of confidence in anticipation of the delivery of the reward. The dog has made the connection between his behavior and the consequence.

Reward: Anything the individual dog is willing to work to earn - not just food.
Reinforcer: Anything that strengthens a behavior.

Side note: I don't talk about "getting rid of the food" but about the importance of varying the types of rewards and the importance of real-life rewards (getting to go through the door, get to the car, get in the car). It is a transition from giving food rewards in the early stages of teaching a new behavior to using a fluent behavior in "real life" for "real life" rewards.

"Will I have to reward the dog forEVER?" Yes - not always with food and not for every behavior, but always with praise and sincere appreciation and, from time to time, you will surprise your dog with something he really values. Trainers never stop rewarding appropriate behavior. To stop rewarding is to cause behavior to deteriorate, or even become extinct. We don't continue to reward EVERY single sit with a cookie. We move to a different rate of reinforcement. We vary how much, how often, and for what level of performance with a variety of possible rewards. We make the work itself rewarding. We create work-a-holics by using a variable schedule like a slot machine. The goal: the dog learns "when I see (or hear or smell) that, they want me to do this, and then- if I do it good enough- sometimes food or something even BETTER happens - and THIS could be the time!."

How do you answer "I want him to do it because I said so - not for the food"?
Behavior is driven by consequences. "Because I said so" implies an "or else" or negative consequence. Negatively trained dogs continue to perform because they are sure if they don't, the negative consequence will occur. The positively trained dog continues to perform because they believe there's a really good chance a great consequence might happen.

We are building behavior on positive consequences - food is just one of the many choices of rewards. In class the easiest and most effective reward is food. It could be a privilege, game, or some other activity the dog wants - but for expedience in early stage training, food is the easiest to control and present in a timely manner.

ON to the assignment:
Helping the student move from lured behavior to cued behavior:

Your primary goal is get the food out of the lure hand and into the OTHER hand, making the hand signal a true CUE and not a lure. When we get beyond luring and on to cueing, real training begins.

Luring can get a dog to sit, but it doesn't TEACH him to sit.
The verbal marker "YES!" in beginning classes and "click" in Prep is what gives the dog the essential information. Stress to your students the importance of a clear behavior marker!

Learning happens when the dog makes the association between his behavior and predictable consequences. Simply following his nose is not enough. We don't fail our students by not "getting rid of the food" but by allowing them to continue to lure the dog for weeks without making it clear to them how to move to a clear cue followed by a predictable reward.

A, B, C = Antecedent (cue), Behavior (sit), Consequence (reward)

THE STEPS FROM LURED TO CUED BEHAVIOR:
(These are in the "Silence is Golden" handout mentioned in the last Q of the week answer and on the blue diagram "Teaching the Sit, Down & Stand" with illustrations, also in the folder.)

1. Lure the behavior: Quietly move the lure, letting the dog lick at the treats in your hand, guiding the dog into the position by slowly moving the food in the direction necessary to elicit the behavior you are after. Avoid chattering. Quiet encouragement is fine, but ignore all behavior that isn't what you are after and instead concentrate on telling your dog when it's on the right track.

2. Mark the behavior: At the precise instant the dog assumes the desired position, say "Yes!" (or "click") and then deliver the reward. You might give several rewards in a row while the dog remains in position to make it clear to the dog that this is the magic position that gets rewards and it will keep him there a few seconds longer. You may continue the training session by luring the dog into another position, or give a release to tell him that the training moment is concluded and he's free to choose his own position. Your release may be "all done!" or "ok" or "free dog"!

3. Morph the motion of the lure into a clear hand signal:

a. When the dog is confidently following the lure into the desired position, make your lure hand look like the specific hand position that you intend to become the cue. At this point you are still holding a lure in the signal hand. Hold your palm flat and place a piece of food in your palm and cover it with your thumb. (Like a boy scout oath.) The up, down or horizontal motion of the lure hand becomes the motion of the cue. Say "YES!" as the dog assumes the desired position and reward.

b. When the dog is confidently following the 'hand signal' into each position, stop feeding the dog the food lure that is in the signal hand. Instead deliver a treat from your other hand. The dog may momentarily switch its attention to the reward hand - just wait. He will go back to the hand that gives him the information. When he does, mark the behavior, "YES!" and deliver the food again from your reward hand. He may still be following the smell of the food in the signal hand, but now he never gets that piece of food so it becomes irrelevant.

c. Again, when the dog is CONFIDENT at this stage, (when you would be willing to bet $5 bucks that he will follow the signal) eliminate the food from your signal hand. In fact, show him it's empty. Now the signal becomes a clear CUE to perform the behavior and the the dog will focus on the work ("what does that mean - what does my handler want me to do next?" - not "where's the food, where's the food, where's the food?") At this point, the reward should be delivered AFTER the marker "yes" or click, and come from anywhere OTHER than your signal hand: your other hand, pocket, treat tote, counter, or dish clear across the room. We didn't "get rid of the food" we just separated it from the cue.

4. Change the picture: Practice all of the above kneeling, sitting, standing - next to, sideways, and facing your dog. Try it with your dog on a chair or at the top of the stairs. Do it in the living room, the kitchen, the patio, the deck, the front yard, the backyard, the sidewalk, and when he's really ready, the park. Lower your standards (temporarily) each time you change places/positions.Be generous with your rewards. The dog will still get a reward for following each response to the cue at this stage of his training.

All of the above steps are demonstrated week one and reiterated at the beginning of week two.
When they arrive with their dogs week two, 90% of students will still be luring and some will not have accomplished all behaviors on the lure. Work individually to troubleshoot luring problems so they are successful getting the behaviors (by luring) and then work as a group to show them step by step how to morph the lure into the hand signal. Many will be yapping "sit sit sit!" Remind them to work silently before you start so you don't have to single out the yappers individually. The only word the dog should hear is "yes" followed by quiet praise paired WITH the food delivery.

Remind the students: as soon as the dog understands the cue, there is no reason to have food in your signal hand. You may have to temporarily go back a step or two in highly distracting situations, but you should endeavor to get as quickly as you can to a clean cue - i.e. lure briefly to get the pup's attention and get him back in the game and then go straight back to cueing without food in your signal hand. Leave the food on a table or counter and don't pick it up until AFTER the dog has followed the cue.

5. Add the verbal LABEL after several days of working silently and focusing totally on the hand signals. [I do this to put the kabosh on "SIT SIT SIT SIT SIT" habits. If you can get them to be totally quiet through the initial luring/hand signal teaching process, the ones who get it will be quieter and more concise with their words later. Why is this so important? It's less confusing to the dog and it prevents the handler from assuming that the dog understands words he really doesn't even hear in the midst of a room full of verbal diarrhea.

Body language is a dog's first language, spoken language is totally foreign and harder to learn. Can the dog learn verbal commands the first week? Oh yes, of course. But jabbering "come on, sit, hey, listen up, sit sit SIT, spot, SPOT, sit" is so strong in our students that telling them not to talk at all at first is easier on the dog AND the instructor!]

Give the hand signal and then AS the dog assumes the desired position, label it using a pleasant tone of voice. Label it several times and then move on to the next progression. The dog must understand the hand signal before you move on to teaching a verbal cue as follows:

6. Teach the verbal CUE:

1. Say the cue word ONCE.
2. Follow it immediately with the hand signal.
3. Mark it "Yes!" (click) AS the dog assumes the position.
4. Reward the action.

The order is important. Soon you will notice the dog starting to respond to the verbal cue without waiting for the hand signal because he has made the association that one follows the other. When the dog begins to anticipate the hand signal (starts to sit or down following the word, before the hand signal) fade the visual prompt: gradually eliminate the hand signal -make it smaller and smaller til it is no longer needed and the dog responds to the verbal cue alone.

7. Fade the prompts: A prompt is an extraneous or unintended association. Like the nod of a head paired with a down command, the place you train, or your position relative to the dog (or picking up food right before you give a command.) When you fade the prompt (gradually eliminate - make it smaller and smaller til it is no longer needed) you clarify for the dog EXACTLY what part of your behavior is the actual cue and teach the dog to disregard all the rest.

8. Change the picture: (Just because he understands the hand signal anywhere and at any level of distraction, doesn't mean he will transfer that ability to the verbal cue.) Practice all of the above kneeling, sitting, standing - next to, sideways, and facing away from your dog. Try it with your dog on a chair or at the top of the stairs. Do it all around the house, inside, outside, the sidewalk, and when he's really ready, the park. Lower your standards (temporarily) each time you change places/positions. Be generous with your rewards. The dog should still get a reward every time for following each cue at this stage of his training while you generalize the context of the cue.

The next stages of acquiring a new behavior:
Fluency, generalization, and maintenance.
Fluency - the dog responds to the cue effortlessly, at a high level of criteria. He really "knows it."
Generalization: responds reliably and with confidence in any situation, environment or distraction level.
Maintenance - you don't just teach a dog a skill and expect it to stay at a high level of performance without ongoing practice and reinforcement.

Latency - is the time delay between the cue and the response; zero latency is instant response to a cue.

Duration: holding the position for a longer time period following the cue ("Sit" -two, three, four- "Yes!" reward) without getting distracted or offering another behavior in absence of another cue or release. In early training, if you build too long a duration too quickly, the dog may start guessing - running through his rolodex of behavior possibilities. If he does this, back up and lower your criteria.

You should not move from a continuous reinforcement schedule to a variable schedule (wean the dog off a reward for every response) until the dog completely understands the relationship between his behavior and the consequence - until he has reached some level of fluency. This varies with the individual dog and handler. An experienced handler may move to a clean signal and a variable schedule on the very first training session. A novice handler is going to take longer to get his part of it right, so it will take a bit longer. Don't set a time limit, make it contingent on the dog's understanding of what is being taught at each level.

The next step: 2-fers and 3-fers.
Moving from one on-cue (reliable) command to another with jackpots for excellence - only the best responses get a food reward (sloppy responses may get verbal praise for effort but not food, or a "no reward marker" (try again!) if they offer the wrong response to a cue.) Build strings of quick-response behavior with intermittent rewards and jackpots. Remind your students to build only one criteria at a time (cue response time/duration/position or ?) not work on improving multiple factors in the same session.

This is where the power of a jackpot will strengthen the behavior and add incentive to the dog's accuracy of performance - and it is the point where we begin to wean the dog off receiving a reward for every response. When you get to the point of raising criteria, rewarding excellence or adding additional steps between rewards when heeling or building time or distance on stays, the weaning process comes naturally.

It's hard to get students to build on tiny foundations of success, and even harder to keep them from expecting too much too soon and not rewarding generously enough. I don't want them focusing on "getting rid of the food" but on paying attention to what they are teaching and in shaping behavior accurately. It's hard enough to keep them generous without planting any idea that food rewards are "bad" and we need to get rid of them. When students put the food away and don't keep the dogs engaged, they fall into barking meaningless commands at their dogs and "managing chaos" ... as Kim so wisely put it.

Here might be where we get into the discussion of what ARE the different schedules of reinforcement and how, when and where do we apply the different types? No, that isn't the next Q of the week. But DO give it some thought and tell me what you think!