JOY FULL Horses: Part 3: Going Micro: Unit 2: Base Behaviors

Base Behaviors
In the previous post I introduced you to Kay Laurence’s definition of microshaping.  Often when people are freeshaping behavior, it is very hit or miss training.  They are too macro in their shaping plan, resulting in long dry spells between clicks.

Microshaping takes a very different approach.  In microshaping you are using very small steps, clear criterion, and well thought out training plans to create a success rate of 98% or better.  Base behaviors help to create this high degree of success.

A base behavior is similar to the tap root behavior I described earlier.  (Refer to: Cues Evolve: Part 2 published Sept. 1, 2016 https://theclickercenterblog.com/2016/09/01/) Returning to these behaviors over and over again creates a deep history of reinforcement and helps builds the learner’s confidence.

The base behavior is just what the name implies.  It is a behavior you keep returning to that serves as the end of one movement cycle and the starting point for the next.  Kay will often click as the dog returns to the base behavior, but she will toss the treat out away from the dog.  The dog runs to get the treat and then returns promptly back to the base behavior for another click.  If you’re clever, – and Kay is – you can build a huge repertoire of reaction patterns using the concept of base behaviors.

A base behavior might be a dog coming to a balanced stop with all four feet on a platform.  The dog returns to the platform because he knows he will get clicked.  As he lands on the platform, Kay clicks and tosses the treat so the dog must leave the platform to get it.  As soon as he gets the treat, the dog returns to the platform to get clicked again.

Kay tosses the treat out very strategically.  The path the dog takes to get back to the platform produces the behavior Kay wants.  That might be trotting around a cone to get back to the platform.  Kay gradually moves the cone out to the side.  The dog could by-pass the cone and go straight to the mat, but he’s learning that going out in a wide arc around the cone is part of the behavior.

Another function for base behaviors is they let you bring your learner into stillness so you can draw his attention to a small muscle movement. Kay offers a great example of how this works.  Suppose you want to teach your dog to wag his tail on cue.  When he’s with you, tail wagging is almost always present, so it may be hard for him to realize that’s what he’s being clicked for.  The solution: have your dog lie flat on his side.  This is your base position.  The rest of your dog is still.  Only his tail is moving.  As you click each time his tail lifts up off the floor, he’ll begin to realize that’s the desired behavior.

Grown-ups and standing on a mat are two great base behaviors we use with the horses.  Both keep the feet still so you can focus on ears forward, the pilates pose, or any other muscle pattern you may be working on.
robin-on-mat-1a
Tossing Treats
With horses we don’t generally toss treats the way you can with dogs.  It isn’t that horses aren’t capable of tracking a thrown treat.  It’s more that we don’t usually work on the kinds of surfaces that we want horses eating off of.  Tossing a treat out into a sand footing raises concerns about sand colics.  And the trouble with tossing a treat out into grass doesn’t need explaining.

Kay uses tossed treats to tremendous advantage when she’s working with her dogs.  Suppose she wants to teach her dog to back up.  She’ll teach this using a mat.  Once again, coming to a balanced stop with all four feet on the mat is the base behavior.  She clicks, and then tosses the treat out in front of the mat so he has to leave the mat to get to it.

Once he’s got his treat, the dog is going to return to the base behavior.  Initially the food is tossed just a step or two away from the mat so taking a step back is the easiest way to return to the mat.  Click.  The food is tossed out a step or two from the mat and the cycle repeats.  The behavior Kay wants is backing.  She get this through the strategic use of treats and the return to the base behavior of stopping with all four feet on the mat.

Once the dog is consistently taking a step or two back to get the mat, Kay will toss the treat a little further away.  Note, the change she makes is in the distance she tosses the treat.  She doesn’t move the mat.  The dog knows where it is.  He’s learning how to return to it from increasingly greater distances and varied directions.  This creates a very confident backer.

Precision
A 98% success rate depends upon precise criteria.

So picture this.  Kay has just tossed a treat out in front of her dog.  It’s a short toss.  He only has to go a step or two forward to reach the food.  Returning to the mat means backing up a similar number of steps.

Many of us would watch the front feet, and as soon as they were both on the mat, we would click.  And we would be wrong.

We would be clicking an outcome not a behavior.  The outcome is the dog standing with all four feet on the mat.

The behavior Kay looks for is the right front steps back onto the mat, followed by the left front stepping back onto the mat – click.  That’s precision training.

What is the difference?  This level of precision means the same behavior is consistently being marked.  In the outcome driven view of the world the dog might back right front, left front this time; followed by left front, right front the next; followed by a half step forward, then right front, left front.  Those are three different approaches to the mat.  Precision training creates clean, precise results.

Dynamic Food Delivery
We may not be able to toss treats to our horses in the same way that Kay uses them with her dogs, but we can certainly put the concept of strategic treat delivery to work and use it to our advantage.  When I’m teaching backing, I turn and walk with my horse as he backs up.  After I click, I’ll turn forward and walk as I get the treat from my pocket.  I move my feet WHILE I am getting the treat out of my pocket.

I’ll end up exactly where we started when I initiated the backing.  This resets the whole behavior so I can repeat that unit over and over again until it is clean and fluid.  Then I’ll follow the mantra of loopy training: when a loop is clean, you get to move on.  Not only do you get to move on, you should move on.

I’m not asking the horse to back up, click and treat, and then back up again without going forward.  That may seem as though it’s the same behavior – backing. To the horse each additional step back is a step into unknown territory.  Is he about to step into a hole or onto unstable footing?  Is he getting too close to the electric fence or the other horses?  Each step presents a new question which means each step is a completely new behavior.

Resetting via the food delivery means I can ask for the same step over and over again, creating a consistent response for this simple reaction pattern.  I’m not being outcome driven – backing through a specific pattern.  Instead I’m focused on the reaction pattern that will create for me the ability to ask for backing at any time, for any number of desired steps, in any direction and in good balance.

Training via reaction patterns means I can be precise in my criteria.  It isn’t any step back I am looking for.   It is this foot stepping back.  And more than that, I will be watching how the horse lifts his foot.  I’ll time my click so I’m marking the upward lift of the foot, not the part of each step where his foot is returning to the ground.  I will do this until my loop is clean.  At that point it will be time to change criteria and move on to an expanded version of this movement cycle.

The video illustrates how to use strategic food delivery in combination with a request to back.  It also illustrates the difference between training for reaction patterns versus outcome.  The overall function of this set-up is to teach my horse to back in a square.  If I were training with that narrow goal in my sights, I would ask my horse to continue backing around the perimeter of the square.  Instead I build the underlying reaction pattern that will make this an easy outcome to achieve.

Precision and Play Go Together
Precision training creates precision results.

This sounds hard.  It sounds as though you have to really be focused and thinking every second.  Clicker training is supposed to be FUN! How can you play if you have to be thinking every second about which foot is moving and what part of the arc to click in?

Remember that convenience store where all the lottery tickets are printed with winning numbers?  You look at your numbers and you’ve won!  Hurray!  You don’t have to think about what your lucky number is.  You don’t have to wonder if your non random numbers look random enough.  You can choose the same set of numbers over and over again.  Each time you’ve got a winner!  Easy!

Your horse steps back onto his mat – left front, right front.  You’ve won!

Hand him the food a step or two forward.

He steps back – left front, right front.
You’ve won again!

This is easy.  This is child’s play!

Learn to set up behavior, and it truly does become playful because laughter thrives on success.  Laughter dies away when there’s confusion.  When the clicks are few and far between, when your learner decides he’d rather go sniff the nearest manure pile than interact with you, that’s when play disappears.

Play evolves out of success.  Play evolves when both learner and teacher are relaxed and confident in the process.  Good technique, attention to detail, attention to your learner’s emotional needs are the breeding ground for play.

Coming Next: Part 3: Going Micro: Patterns

P.S. To learn more about Kay Laurence’s training visit: learningaboutdogs.com

Remember, if you are new to the JOY Full Horse blog, click on the JOY Full Horses tab at the top of this page to find the full table of contents and links to each of the articles I have published so far.

I hope you will want to share these articles by sending links to this blog to your friends.  But please remember this is copyrighted material.  All rights are reserved. Please do not copy any of the “JOY Full Horses” articles without first getting written permission from Alexandra  Kurland, via theclickercenter.com

Also note: these articles are not intended as an instruction guide for introducing your horse to clicker training.  If you are new to clicker training and you are looking for how-to instructions, you will find what you need at my web sites:

theclickercenter.com                    theclickercentercourse.com

JOY Full Horses Part 3: Going Micro: Unit 1 – The Many Forms of Micro

This begins Part 3 of JOY FULL Horses.  I ended Part 1 by posing the question: what are ten things you would want someone who is new to clicker training to understand about cues?  I asked you to make your own list.   In Part 2, unit by unit, I shared my answer to this question.  How did my list match up with yours?

The exploration of cues took us on quite a journey.  In any journey, once you’ve taken in the overall look of a landscape, you can begin to focus in on the finer details.  That’s what we’ll be doing in Part 3.  We’re going Micro.

The Many Forms of Micro

monty-with-miniWhy Micro
Going Micro doesn’t mean we’re all going to take up training miniature horses.  Going Micro refers to the level of focus we bring to the training.

At first blush becoming more detail-oriented would seem to be anti-play.  It’s easy to visualize a horse cantering through an agility-style obstacle course and think of that as play.

We tend to be very goal oriented.  We see someone “playing” tennis, and we think that individual is having fun.  But when that individual is a professional tennis player who spent his childhood being forced to practice for hours at a time every day, there is nothing playful in the behavior.  It is work, work, and more work.  Read Andre Agassi’s autobiography,  Open, to discover how far from Play this form of forced practice can take you.

HOW something has been taught is much more important than WHAT the behavior is.  A horse trotting through a series of obstacles may be doing so because he loves the game.  Or he may be trying hard to avoid the sharp lash of his trainer’s whip.  In this section we’re going to look at what it means to go MICRO and how that can lead directly to a creative Play state for both the handler and the horse.

MicroShaping
Kay Laurence coined the term MicroShaping.  For her it refers to a very specific form of training.  The conversation I was having with Poco, the ear-shy horse you met at the end of Part 2, depended upon my being able to ask small questions.  In clicker training we often call this thin slicing.

Kay Laurence uses this term in her training, but when Kay talks about microshaping, she doesn’t just mean breaking the training down into super small steps.  She means setting the training up so the success rate is 98% or higher.

Yes, you read that correctly.  98%.

If you want a 98% success rate, you need a well thought out training plan.  You aren’t waiting, waiting, waiting, hoping that your horse will give you something you want.  You are setting up the training environment in such a way that, of course, he presents clickable moment after clickable moment.

It also means you are looking for underlying reaction patterns – not the final outcome behavior.

Outcome Versus Reaction Pattern
When you focus on reaction patterns versus outcome, what difference does it make to your training?

If you want your horse to back up in a square, you won’t simply ask your horse to back up, and then back up some more until you have somehow gotten him around the pattern. (Or more likely trapped in a corner.)  That’s being goal driven.

If you are training reaction patterns, you’ll view backing as a movement cycle. Movement cycles are just that.  They don’t have a linear beginning, middle and end.  Rather the behaviors within the cycle loop back to the beginning point so the whole pattern can occur again.  A cycle is not complete until the individual is in position to repeat the entire pattern.  Sitting in a chair from a standing position isn’t the complete movement cycle.  You’re only half way there.  The person needs to stand up again, returning him to a position from which he can sit down.  Only then can he begin the next loop in the cycle.

So with backing you’ll begin with a very small unit.  You’ll ask your horse to shift his weight slightly back.  Click.  You’ll feed so that he’s brought forward again.  In this way you can keep repeating this small unit.  You will know you can expand to the next criterion when it is already occurring on a consistent basis.  You’ll end up with a clean, finished behavior because you trained clean throughout.

It’s easy to read fast through this last paragraph and miss completely the significance of what I am saying.  Suppose you are reinforcing your horse for standing beside you in the stationary behavior of the “grown-ups are talking”.  You want him to keep his head balanced evenly between his shoulders.  Check, he’s doing that consistently.

Behavior varies.  Doing the same exact thing over and over again just doesn’t happen.  Even Olympic athletes, as consistent as they are, show some slight variability performance to performance.  So, while your horse is standing beside you, your eye may be caught by the movement of his ear flicking forward.  He may have been listening to the sounds of horses moving in a paddock behind him, and that’s why his ears were back.  Now something has caught his attention in front of him, and his ear flicks forward to listen.  You’ve been focused on his head position.  Now that that overall position is becoming consistent, your attention can broaden out to include the flick of his ears.

If his ears were always back, it would be hard to make that the next criterion.   If a behavior isn’t happening at all, waiting for some piece of it to pop out can take your learner straight into the frustration of an extinction process.  Your learner will begin throwing behaviors at you trying to get you to click.  One of those behaviors just might be a flick of an ear, but it’s a messy process.  You’re more likely to get pinned ears as he becomes increasingly frustrated trying to figure out what you want.  Frustration can lead to a whole lot of behaviors you don’t want. They’ll become unwanted guests attaching themselves at odd times to the desired behavior.

Instead of this form of hit or miss shaping, you’ll wait until the behavior begins to pop out as you consistently reinforce your current criterion.  If the new behavior begins to happen often enough for you to notice it, then it is likely that it will happen again fairly soon.  So even if you go a tiny bit down the extinction road by withholding your click as the previous criterion occurs, your learner will very quickly land on the right answer.  Click then treat.  He won’t feel the frustration of an unsolvable puzzle.  Instead he’ll experience the pleasure of another successful answer.

Waiting until the next shift in criterion is already occurring removes a lot of training frustration for everyone – horse and handler alike.  Once you get the hang of this approach, you’ll begin to understand what it means to train via reaction patterns.  With that shift you’ll also discover that you are reaching your overall training goals faster.

Kay Laurence has a great video example illustrating the difference between training that is goal driven versus training for reaction patterns.  The final outcome that she is working towards is having her dog put his foot up on a small box.  In the goal-driven example, Kay waits to click until her dog lifts his paw onto the box.  Her dog doesn’t understand that the box is important or that she wants him to put his paw on it, so even though he has received a click for approximating that behavior, he’s not sure what he should do next.

In the video Kay trains in sixty second units.   During his first session, he presents many behaviors.  He looks around, he noses the box, he lifts his other foot, he paws at the box, he walks sideways, he backs up, etc..  At the end of the sixty seconds the number of behaviors he’s offered versus the number that were clicked puts him down around a 20% success rate.  But you don’t need to do the math to know that he’s a confused dog who isn’t having much fun.  You just have to watch the clip to see that he’s teetering on the brink of shutting down altogether and going off to have a nap.

In the second video the box is still there, but now Kay is looking for something very different.  In order for the dog to put his front foot up on the box, he first has to lift his leg up.  So she is clicking for any shift of balance through his shoulder that brings his foot up off the ground.

This transforms the training.  Suddenly the dog is getting click after click.  It’s as though you’ve walked into a convenience store to buy a scratch-off lottery ticket only to discover that every card is printed with the winning numbers!  I would eagerly buy that kind of lottery ticket.  I like knowing I’m going to win, and so do our animals.

In this second video clip instead of looking lost, the dog is fully engaged.  Very quickly he’s consistently lifting his paw.  So now it’s an easy step to move the box so it’s in the path of the paw lift.  Click, treat – you have your outcome behavior.

The dog didn’t begin by putting his paw on the box.  Kay wasn’t waiting, hoping he would do something, anything she could click.  He was set up to succeed at a 98% or higher success rate.

An important element in creating this kind of training success is the use of base behaviors.  I’ll define what that means and give some training examples in the next section.

Coming Next: Part 3: Going Micro: Base Behaviors

Remember, if you are new to the JOY Full Horse blog, click on the JOY Full Horses tab at the top of this page to find the full table of contents and links to each of the articles I have published so far.

I hope you will want to share these articles by sending links to this blog to your friends.  But please remember this is copyrighted material.  All rights are reserved. Please do not copy any of the “JOY Full Horses” articles without first getting written permission from Alexandra  Kurland, via theclickercenter.com

Also note: these articles are not intended as an instruction guide for introducing your horse to clicker training.  If you are new to clicker training and you are looking for how-to instructions, you will find what you need at my web sites:

theclickercenter.com                    theclickercentercourse.com