equine clicker training

using precision and positive reinforcement to teach horses and people

Taking ClickerExpo Back to the Horses

Some ideas for integrating the Microshaping Strategy, Backchaining and Conditioned Secondary Reinforcers into my training

In March of 2008, I attended Clicker Expo in Kentucky. For more information on that, please see my Clicker Expo 2008 Report which describes the sessions I attended.   I love Clicker Expo and it always inspires me to come home and do more freeshaping and experimenting with new ways to get behavior, but I often find that it is easier to start applying my new knowledge by working with my dogs.  So for one month, my dogs (2 border collies) get lots of attention. This year, I was determined to take the Clicker Expo material and integrate it into my horse training more directly. I find that the material I get at Clicker Expo deepens my understanding of clicker training and changes the way I think about approaching training, but it is not always easy to figure out how to incorporate some of it into the work with the horses, especially the ridden work.

But this year, things came together a bit more for me.  I think it helped that I had spent part of the winter watching clicker training DVD’s and thinking about how clicker trainers working with other animals structure their sessions and build behaviors. In particular, I spent a lot of time watching Kathy Sdao and some of her discrimination, cue and early target training.  I also watched Alexandra Kurland’s new DVD’s on cues and microshaping.  This winter research combined well with the material I got at Clicker Expo and gave me some clear ideas for new things to try. 

For me there has always been a bit of a disconnect between how I train horses and how I think people train dogs with clicker training.  By that I mean that horse people just seem to do things differently and it is hard to figure out if it is because we start with a different traditional background, if horses really do require different training strategies, or if riding is just so different from dog sports that there is less direct carryover between working with dogs and riding horses.  I read and watch a lot of clicker training material geared towards dog training and I do find that a lot of this information is directly applicable to horses, especially husbandry behaviors and work on the ground.  But when I get to riding, there seem to be more differences.  I think part of the issue for me is that while I know a lot about training horses and have clicker trained my own dogs, I have not clicker trained my dogs to do any specific activity such as obedience, agility, flyball or other dog sports. 

So I feel like I am missing a piece in my understanding of how dog trainers go from initially teaching new behaviors to getting the kind of reliability and enthusiasm they need to get performance level behavior with dogs.  This is why the first sentence of the previous paragraph refers to how “I think people clicker train dogs” and not “how people clicker train dogs,” because in all honesty, I don’t know how the top competition and working dogs are trained as I have never observed or done any dog training at that level.  I think this makes it hard to relate some aspects of clicker training dogs to clicker training horses.   On the clicker lists, we get into discussions with new to horse, but experienced dog clicker trainers on the differences between dogs and horses and into even more complicated discussions about what riding is all about, because for most advanced riders, riding becomes more than a specific set of discrete cues.  Instead riding ends up being a steady flow of information passed back and forth between horse and rider and while there are certainly specific cues for specific behaviors, there is also a very subtle conversation going on that might have started out as discrete cues but evolves into something else.  This could be a topic for an article in itself so I am going to leave that thought out there, but I think horse people could benefit from looking a bit more carefully at what clicker trainers of other species are doing. 

Since I do spend most of my time working with horses (plus a bit of dog, cat, guinea pig, and kid clicking), going to Clicker Expo is my opportunity to get a better understanding of how other clicker trainers work.  Each year I see more areas of overlap or ways to use their tools and strategies in my own training.  Coming home from this Clicker Expo, I had a better understanding of how getting animals to work for longer periods of time without food rewards is not just about fading the click or using variable reinforcement schedules.  It has more to do with carefully constructing chains that contain behaviors that have several roles. These behaviors can be a step toward a final goal, but they can also be reinforcing in themselves, provide feedback for the animal about previously performed behaviors and indicate that more reinforcement is coming at the end of the chain.  This understanding about how sequences of behavior or chains can work together did not come out of any specific presentation, but came from seeing several people present related information and suddenly understanding how all the little puzzle pieces fit together to create a bigger picture.

This new view of the bigger picture gave me some definite ideas for things I wanted to try and I am going to share some of the things I am exploring this summer with my horses. This article is not a “how to” and my goal is not present a set procedure for how to do things.  My goal is just to share some of my thoughts and ideas about how to connect some pieces with the hope that it will generate ideas for how other clicker trainers can structure their own training sessions or do some problem solving.

The three main areas that I wanted to explore when I got home were using the “microshaping strategy”, secondary reinforcers, and backchaining.  Interestingly enough, as I have played around with them, I have found that they are easily interwoven together in a training session so I can be using more than one strategy at once.  I also want to add that I wanted to become more aware of poisoned cues and come up with better ways to train to prevent poisoned cues and also to see if I could “unpoison” some of the cues that were already poisoned in my horses. 

The “microshaping strategy” was developed by Alexandra Kurland who explains it in her new DVD on microshaping. It is based on microshaping, a term first used by Kay Laurence, that refers to clicking for individual muscle movements and building behavior from tiny little pieces.  All shaping is about building behaviors from successive approximations but there is a tendency to lump, either because the trainer is not skilled enough at recognizing small details, or because the animal jumps ahead to the end behavior without getting all the little intermediate steps down to create a solid foundation.  Initially having the animal jump to the goal behavior seems like a real bonus, but it can lead to problems later on if the animal doesn’t have all the building blocks to reproduce the behavior in other settings and those basic building blocks can also be useful for shaping other behaviors.  Therefore, it is worthwhile to take the time to build behaviors slowly and carefully.

Microshaping’s emphasis on building behavior out of very small and precise building blocks (often just a muscle twitch) can eliminate frustration on the animal’s part and create cleaner and stronger behaviors.   But when animals are first being microshaped, they sometimes go through a period where the reinforcement rate is low because they are not used to being clicked for tiny changes (and therefore don’t repeat a clicked behavior). Or the reinforcement rate could be low because the trainer is new to microshaping and doesn’t recognize the little pieces that will evolve into the final behavior. To counteract this, Alexandra took the idea of microshaping and paired it with another easy behavior to help keep the rate of reinforcement high in the early stages when the animal might be struggling with learning the new behavior. In her DVD, she trains horses to back using microshaping (clicking muscle movement) and she alternates between short sets where she reinforces backing with sets where the horse does target touches.  The horses are familiar with the targeting behavior and it provides a break from working on backing as well as a way to highlight better efforts at backing by marking them with a chance to target.  She refers to this as the “microshaping strategy” and one important component of the “microshaping strategy” is that the targeting or high rate of reinforcement behavior becomes a conditioned secondary reinforcer.

If this is not clear to you, there is a lot of information on the “microshaping strategy “on Alex’s yahoo list “the_click_that _teaches” and on her DVD.  For the purpose of this article, all you really need to know is that mixed in with teaching new behaviors, Alex inserts periods of target touching. I had played around with using the “microshaping strategy” with my horses last summer and found that not only did they appreciate the targeting breaks, but their targeting skills improved.   Even though I found it interesting, I got sidetracked on other projects and had not followed up on the use of targeting or using another behavior to reinforce the main behavior I was shaping.  

However, Kathy Sdao’s DVD’s showed how small intense work sessions can be very productive and made me start to think about how to structure my training to incorporate some periods of more rapid reinforcement and this brought me back to the “microshaping strategy.”  Usually my ride is one long training session where I will concentrate on one exercise or behavior for a while and then move to something else.    Of course the horses get breaks but even during the breaks, I might be working on something like walking quietly in head down.  There is some variability in how often I click, but unless I am training a brand new behavior, I don’t have periods of very high rate of reinforcement where I am clicking 5-10 times in rapid succession. My tendency is to ask for something a few times, get some small improvement and then move on to something else. What I wanted to try doing was to pick a few behaviors and train them in more intensive time blocks to see what happened to those behaviors and the rest of the training session.

This line of thinking led directly into some ideas for backchaining which is another area that I found interesting at Clicker Expo.  The session that Morten and Cecilie presented on backchaining was very enlightening because it was clear that the backchaining did build both enthusiasm and reliability.  Prior to this, I had not played around with backchaining much, other than to teach fetch or a few other simple tricks but after watching their session, I started to see how backchaining might be used in horses. One of the issues I have encountered with my horses is that there is some reluctance or initial decrease in enthusiasm when I start to reinforce less frequently, either because I am building chains or training sequences or because I am being more variable about how often I treat. I sometimes get the feeling that the horses are saying  “what? I have to do more to get my treat?” as opposed to “wow, I get to do another thing that might earn me a treat.” 

In the past, I have accepted that this is a phase the horses go through when they first learn about variable reinforcement and I usually introduce variable reinforcement gradually.  If the horse gets frustrated, I might go back to a higher rate of reinforcement for a bit and then try again, with a change in reward (food quantity or quality) or criteria to show the horse that it is worth his while to keep trying.  My goal is to prepare the horse to understand that sometimes you have to do more than one thing to get a click and treat.  In some cases, I have felt that the horse is not actually enjoying the work and I need to find a way to motivate him more or make the work more interesting.  With this in mind, I had been reading and thinking about ways to set up my training for the horses so that they would work for longer periods of time but maintain their level of interest and enthusiasm.  In his talk on conditioned secondary reinforcers, Ken Ramirez talked about seeing this problem when moving animals to variable reinforcement schedules and explains how he uses reinforcement substitutes (a.k.a. conditioned secondary reinforcers) to help animals make this transition and to make it so the animal is not really working for less reinforcement after all.  Ken’s talk was one of the important pieces for me in starting to look at the components of a chain and seeing how they help maintain the chain.

A better understanding of backchaining was the next piece because Morten and Cecilie showed how to put those components together to create a stronger chain. The idea of backchaining was intriguing and their presentation cleared up two of my concerns about using chains.  My previous experience with chains, and especially backchains, had been that the whole chain ends up with one cue delivered by the trainer and then completion of each behavior cues the next behavior. But that is just one kind of backchain. They do not train them that way.  They explained how they keep the cues in the backchain so that the animal does not start to do each behavior on its own, because the trainer still controls the timing of when the animal goes to the next part of the chain.  They also explained about the testing phase which is when the animal tries to drop out pieces of the chain to get to the end more quickly. If you understand that testing is part of the process and handle it appropriately, the chain is strengthened by allowing the animal to go through the testing phase.

With this in mind, I wanted particularly to see if I could apply backchaining to riding.  Now, in general, riding is a forward progressing chain. I ride from A->B->C and I can’t practice going from B->C until I get to B, so training B->C before training A->B is hard.    This applies to riding both geographically and gymnastically.  There are lots of exercises in riding where the final behavior requires other behaviors to set it up, so there is a limit to what you can backchain, but I still thought there were some applications for backchaining.   Morten and Cecilie’s examples were using backchaining to train set patterns for competition obedience with dogs.   Since riding is not about riding set patterns, except in some types of competition and I am not currently competing, I am not interested in backchaining a set pattern so I didn’t want to copy exactly what they did, but I did come away with the idea that I wanted to access the controlled anticipation and enthusiasm that comes from using backchains.  My interest is in exploring how to use backchaining to change a horse’s attitude about working longer sequences and as a problem solving strategy.   With this in mind, my focus is not on backchaining finished behaviors, but on using backchaining to help a horse motivate a horse to finish a sequence of behaviors. My idea was to come up with some solid behaviors to use as the end of chains and then use the horse’s desire to get to those behaviors to reinforce the previous behaviors in the chain. This would allow me to have the horse do multiple behaviors, all which were being reinforced, but without stopping for clicks and treats all the time.

Thinking about what behaviors to use in my backchains, especially as the final behavior, brought me back to the idea of conditioned secondary reinforcers which is something I had already been exploring.   I attended Ken’s lecture on advanced training tools at Clicker Expo 2007 and he introduced the idea of conditioned secondary reinforcers in that session.  I thought they would be useful with horses as there are times when we cannot use food, or my horses will not take food. I spent some time last summer conditioning Rosie to accept “good girl” and a pat on the neck by associating it with her food reward. I am not sure I could shape a new behavior with these secondary reinforcers but I do think she understands they mean she has been correct.  After attending Ken’s talk this year, I decided to see if I could start using this conditioned secondary reinforcer in my riding in a more systematic way.

Putting this all together, I came up with a few ideas for things to try.  I started out with the idea of working on each one separately, but it turned out that each one led into the next and I came up with some exercises that incorporated several of the ideas I got from Clicker Expo.  I think this came about pretty naturally because I started by doing some intense training of several behaviors that could become conditioned secondary reinforcers and then realized I could use these to strengthen chains and that there were some situations where backchaining seemed to work better than forward chaining.

When I got back from Clicker Expo, I started by looking at the “microshaping strategy.”   I wanted to insert some “blocks” of rapid reinforcement into Rosie’s rides. In order to get a high rate of reinforcement, I had to pick something that was easy to do several times in a row and I wanted to pick something that was going to be easy and reinforcing for Rosie.  I thought about using targeting and I could have started with that by having Rosie target my hand while halted, or carrying a small target. But if I was going to spend a lot of time on this behavior, I wanted it to be something very useful under saddle.  This behavior was going to become even stronger and could possibly become a conditioned secondary reinforcer if I did it correctly. 

The behavior I chose to use was the horse organizing herself in response to the single rein pickup.  I wanted the response to the rein pickup to become a very highly reinforced behavior.  For many clicker trained horses who have been trained using Alex’s program, the rein pickup has been highly reinforced and already has strong positive associations. I have done a lot of single rein riding with Rosie, with 100’s or 1000’s of rein pickups, but somewhere along the way, this behavior became problematic and unreliable.  On some rides, she might respond well to the rein pickup for an entire session, but at other times or on different rides, I would slide down the rein and she would ignore me, usually because she seemed to be fixated on some item of interest that she was watching. 

Because Rosie likes to look at things and is easily distracted, I had been treating this problem as an attention issue.  But after watching Jesus Rosales-Ruiz’s presentation on poisoned cues and putting together a few other details. it seemed to me that there was something a bit too avoidance-like about her attitude. And I knew it was possible that the rein pickup had gotten poisoned as she has had some saddle fit and physical issues that might have made her reluctant to organize herself and walk off.  I was aware that the rein pickup was not an ideal behavior to choose for the high reinforcement behavior because of the possibility it was poisoned.  I knew I would be better off starting with a behavior that was already reinforcing and easy for her, but by lowering the criteria for the rein pickup, I hoped to be able to get around this potential problem.

In addition, getting her more responsive to the pickup of the reins was something I wanted to improve and this seemed like a good opportunity to experiment.  The first thing I did was I used rein pickups in the same way that Alex used targeting in the microshaping strategy. One behavior I have been working on improving with Rosie is a good forward marching walk. I often start our under saddle work by clicking her for energy at the walk. I have seen some improvement but she doesn’t offer more energy unless I cue her and sometimes I feel like she is slow in responding. So I started adding in a short session where we worked on the rein pickup if she walked forward with energy. 

The set up was like this.  I would get on and ask her to walk and click for walking with more energy. If she gave me a particularly good effort, I would click, stop her and do 5-10 rein pickups, clicking each one where she responded.  My initial criteria was very low. If I picked up my buckle hand and slid down the inside rein and she responded in any way, I clicked and treated. If she set her jaw and ignored me, I dropped the rein and asked again.  This was a departure from previous work in this area where I would have stabilized my inside hand and waited for a response. I decided that since this was a known behavior, I was not going to use the rein pickup as negative reinforcement, but just as a cue. If she ignored the cue, I would ask again because I did not want to end up rewarding slow responses. If she continued to ignore me, I would ask her to walk forward and ask again.  It took me a while to figure out what the best solution was to no response at all to the rein pickup and I would not necessarily have done it this way with my other horses, but this worked for her.

The idea here was to do several things at once. First, I wanted to see if the walk would improve significantly by adding in the reinforcement of a rein pickup session as a reward for moving with energy. The blocks where I did multiple rein pickups were chances to earn a lot of reinforcement in a short amount of time and I hoped it would make her more motivated to walk with energy.  I also wanted to get in lots of repetitions of rein pickups to try and overcome any prior poisoning, improve the quality of her rein pickups and show her that picking up the rein did not mean that she was going to have to go back to work right away.  In my initial sessions, I would give her a long rein after the rein pickups and just let her walk off forward before asking her to do anything else. This has changed over time and I have slowly been changing the criteria for what I want from the rein pickup and adding in some collected starts and backing, but I started this very simply.

I also had a long term goal which was to make her response positively to a rein pickup so that I could use it as the anchor behavior in backchains. A rein pickup is very similar to asking a horse to collect at the halt and I had been working on getting Rosie to stop in an organized fashion instead of landing in a heap when I clicked.  I wanted to be able to end chains with a balanced halt and teaching the horse to organize at the halt with a rein pickup is similar to halting and asking the horse to come back into balance if they have lost it in the down transition.  Even though I started with the rein pickups as an isolated exercise, I wanted to get to the point where I could halt and ask for one until she got into the habit of halting and then organizing herself.

So I put all these things together and I have had some good results. First, Rosie’s walk did improve.  I did get some testing in that the first few times I inserted the blocks of 5-10 rein pickups, she offered a lot of halting and slow walking, anticipating the rein pickups. But I was able to just keep asking and clicking for more energy at the walk and wait her out.  I was careful to make sure that I didn’t go to the rein pickups until she was thinking about walking forward to earn her click.  Second, her rein pickups improved a huge amount. Not only did she stop ignoring me a good percentage of the time, but I have been able to shape the rein pickups and they have evolved into something more like flexions. Depending upon what she needs, I can use the rein pickups to improve her elevation, flexion, or softness. I have been able to change the mechanics of the rein pickup itself so that I can ask for these changes on both one rein and two reins. 

As an interesting aside here, I want to point out that while Alex’s “microshaping strategy” was the basis for this work, I was not sure if I should be calling it the “microshaping strategy” since the two behaviors I was training were not being shaped with the fine detail and attention to muscle movement that I associate with microshaping.  In the walking forward, I was just clicking for energy and in the rein pickup, I was clicking for any response.   Because of how I wanted to set up the backchain, and because it was easier to do the rapid reinforcement for the rein pickup, I started with forward walking as the behavior being shaped and the rein pickups as the equivalent of how Alex uses targeting. But somewhere along the way, as Rosie got better at the rein pickups, I found that in addition to using them to generate a high rate of reinforcement, I was doing a lot of microshaping of her response to the rein pickup and that behavior was becoming the microshaped behavior and the walking forward was now the easy behavior that gave her a break.

I progressed from clicking any response to a rein pickup to being more selective about what I clicked and I ended up microshaping her elevation, bend and whether or not she just gave at the poll or carried the connection through her whole body.  Once we were at this point, I started experimenting with inserting the rein pickup into other behaviors. I played around with turn on the forehand to see if I could keep her softer as she gave her hip and to see if adding the rein pickup made her better about giving her hip. I started at the halt by asking her to come up into a balance halt by sliding down the rein. If she did so, click and treat. If she didn’t, I just asked again. Once she was doing that, I alternated asking for a rein pickup with one step over with her hips. I clicked each little piece.  Then I started omitting some clicks. I would ask her to organize herself and step over once with her hind foot, click and treat.  Pairing the two together actually worked well as she would pick herself  and could then easily give her hip and get her click and treat.  She did test this and try to give her hip while bracing in her front end and I just stopped her and asked her to soften again.

This led to a few other backchaining experiments with Rosie.  Two that come to mind both involve her reluctance to work in the back of my ring.  My riding ring is oriented so that the long sides point toward the back hedgerow, which is where interesting things always seem to happen.  In Rosie’s mind, it is very important to keep a close eye on any activity back there.  And to be honest, I can’t entirely blame her. As a young horse, she would be nervous about working back there and just when I convinced her it was ok, something new would happen. We have had hot air balloons appear over the trees, as well as someone using a tractor, chain saw, bobcat, or dirt bike in the field behind it.  One day a herd of turkeys came by and another day a fox sat and watched us. This is not counting the numerous occasions when cats have been climbing around, children have been making noise and there have been other less scary but worthy of interest activities. 

She has improved a lot but there still are days when she comes out and I have to deal with her inattentiveness down there.  In the past, I have reinforced her for working down there or trotting in that direction and this has helped.  But, I have always done this by asking her to do an easy behavior in the near end and then asking her to do the same behavior a bit farther down.  I would start with a behavior she knew and then ask for a bit more. So as an experiment, I decided to see if I could find a behavior that she already COULD do down there and then build a chain so that she had to go down there to complete the chain and get her click and treat.    I have done a lot of ground and walk work with her and she is usually pretty good about listening at the walk, but I seem to have more trouble at the trot.   With this in mind, I decided to start by walking down there and trotting out.  This was easy for her to do and very reinforcing.   I walked her down to that end and through the corner, picking up a trot going toward the barn and clicking and treating that. I did this a few times until she was reliably picking up the trot and nicely trotting back toward the barn. Then I added another segment of trot a bit sooner. So instead of picking up the trot in the second short corner on the long side, I asked for it as we crossed the center line, praised her when we got to the corner and clicked her as we went down the long side.

I continued this pattern, adding additional pieces of trotting until I could trot down one long side, through the corner, short side, corner, and down the next long side.  If she did not perform a behavior as requested, by coming off the rail, stopping or slowing down, or spooking to the side, I just stopped her and we went back to the beginning again.  I did sometimes click an individual piece that was difficult, but I tried to stick to the idea of only rewarding the final behavior. This was important to me because part of the challenge with Rosie is that if she anticipates the click, I sometimes get slowing down both from her looking at things and because she is thinking about stopping when I click. I wanted to use the backchaining to get more flow and keep her thinking about working toward the end behavior. 

I do want to point out that this particular example worked for me and Rosie because she understands about chains and I was able to set it up so that she could not skip and go directly to trotting toward the barn. I can see where a person could get in a lot of trouble clicking the horse for accelerating toward the barn and I am not recommending that.   I am sharing this training scenario because I think it is a good example of using something the horse wants, to get the behavior the rider wants. If I had concerns about clicking for trotting toward the barn, I could have set the chain up so the end behavior was walking down the long side toward the barn, or even halting facing the barn.  On a horse that wants to spook by going quicker past something,  I would choose the end behavior in the backchain to be something that would ask the horse to slow down.  Rosie tends to suck back and want to stop when she sees something, so for her, I prefer to click for forward movement.

I used the same idea one day when I was schooling serpentines. I started by doing the last and hardest loop first and then added the previous loops one at a time.  Even though I started with the hardest loop, I broke that down into as many little pieces as I needed so that I started with a behavior she could do.   In her case, an easy behavior was completing the circle at the end of the last loop.   Then I worked up to doing the whole final loop of the serpentine. I don’t think it is important to backchain every little piece of this. Some parts are going to be easier to put together as forward chains and others are going to be more suited for backchains.  My main interest was in putting together some separate pieces into a backchain to see if I got better results than when I put them together as a forward chain.

Once she could do that end loop, I added the second loop and continued on to the end of the serpentine.  Then I did all three loops. Rosie has done a lot of serpentines so I didn’t actually have to put a lot of training into the first and second loops. It was more just a matter of adding them into the chain. I think the challenge with using backchaining as a strategy is figuring out where to start.  I am using backchaining in both these examples to help her keep going forward through a difficult part of an exercise because I want her to be more forward thinking, but I have to be careful how I set up the backchain.  I have to start with something she can do well or finds reinforcing in itself.   Then I want to put the difficult exercise as close to that behavior as I can so that there is only a short part of the chain between the difficult behavior and the click.  I could also put more reinforcing behaviors between the difficult part and the click,  which would be added motivation to get past the sticky part of the chain. It does require some thought and planning but I have found that having to figure out how to set up a backchain has made me more aware of what behaviors I can use to anchor backchains and how to set them up to get the most out of them.

Even though I have not talked specifically about using conditioned secondary reinforcers in my backchain experiments, I did use them to mark completing parts of the chain correctly. One thing I find difficult about using verbal secondary reinforcers under saddle is that once I give myself permission to speak (instead of just clicking), I get a bit sloppy about my language. I started out with a carefully trained “keep going signal” which meant “keep doing what you are doing.”  And I had a conditioned secondary reinforcer which meant she had finished one behavior correctly and we were moving on, as in “great, you are done with that but keep going.”  But once I am riding and thinking about many things at once, they both seem to end up as general praise and encouragement, which seems to work, but perhaps is not as clear to her as I might like.   I would like to clean that up and am working on it, but in the meantime, having some form of conditioned secondary reinforcers does seem to help when putting together these chains.  On the trotting exercise and the serpentine, I did use “good” when she finished each segment of the chain.

So far I have been very pleased with how easy it has been to integrate some of the new ideas from Clicker Expo into my riding sessions. I think adding in periods where she is on a high rate of reinforcement has been a nice addition as it seems to motivate Rosie to do more and it has also helped her focus. As she gets more experienced under saddle, I am working on extending duration and she doesn’t need the same kind of focus she had when she was learning new behaviors, but at the same time, I do want to stay mentally connected with her.  Adding in the sessions of high rate of reinforcement and making them contingent on other behaviors seems to help keep this connection.

I also think there is a huge potential in using backchains for horses when working through specific issues where you don’t want to keep clicking each individual behaviors. I think backchaining is an advanced training tool when used in this context, but even novice clicker horses could get introduced to the idea of simple backchains.  Backchaining is easier if you have some highly reinforced behaviors or conditioned secondary reinforcers, and using the combination together has made some significant changes in my training strategies.  I had been looking for ways to improve the structure of my training sessions to make better use of my clicking and treating and I think this is a step in the right direction.  I am finding that I am at a point in my training where I have to think very carefully about how to get the most from each click both in conveying information to my horse and in motivating her to keep working.  I hope this article has given you some ideas for changes you might want to make in your training or at least made you think about clicker training from a slightly different angle.

    If you have any questions, comments, or feedback, you can email me at kabart315@gmail.com.  I would love to hear about if anyone else is playing with these ideas and how it is going.

Katie Bartlett, 2008 – please do not copy or distribute without my permission