equine clicker training

using precision and positive reinforcement to teach horses and people

B is for Baselines

If you’re like me, new training projects are exciting and you can’t wait to get started. I love the early stages of learning when things can change so quickly and my brain is busy with possibilities and ideas for the future.

But …

Whether I am training a new horse or training a new behavior, I’ve learned that it’s important to take the time to accurately assess my starting point. Gathering baseline data not only tells me where I need to start, it can also help me get ideas for how, where, and when to train. It’s easy to think I “know” what the horse can do, but unless I carefully observe the horse for a few days, I probably don’t have completely accurate information.

Why am I writing about this? I bought a new horse, of course!

In the fall of 2020, I bought a 4 year old Lusitano gelding named Madagascar. When I met him, he already knew how to walk, trot, and canter under saddle and he had been to a small show. He was the first “already trained’ horse I have bought in a long time (20 years) and I felt it was important to accurately assess what he already knew before I introduced clicker training and started teaching new behaviors.

This was an interesting opportunity for me because my more recently acquired personal horses have been ones that I bought when they were very young – weanlings or yearlings. Some of them had been handled a little bit, but mostly I treated them as if they knew nothing and I trained everything. That’s a lot of work, but it was easy in that I didn’t have to sort through past training and decide what needed my attention and what was good enough.

Having a crossover horse seemed like a good opportunity to document the way I gathered and sorted information so I could make decisions about how to proceed with his training. It would also mean I could answer the common questions, “Do you have to clicker train everything?” and “How do you decide what to train or re-train?” with more recent personal experience.

With that in mind, I carefully documented everything I did with Madagascar for the first year. My intention had been to share it as a training journal, but life got complicated and I put it aside. This fall I finally have time to start going through my notes and I hope to be able to write some blogs on the training I did with him. With that in mind, it seemed like a good idea to start at the beginning, and that was collecting data about what he could do and who he was.

Here are some pictures of him that I took in the fall of 2020. The first is the day he arrived. The second is an informal session with a ball.

Where to start?

In my case, I had a new horse, and I didn’t know very much about him. I had some ideas about what he had probably been taught and how. Once he was here, I needed to discover the strength and quality of those behaviors. This may not be your situation. Perhaps you have had your horse for a while and have definite ideas about what behaviors you want to train.

But … I still think it’s important to take a look at your horse’s entire repertoire in an objective way. Taking the time to collect some baseline data will help you identify common issues that affect multiple behaviors and make it easier to see any unexpected changes in behaviors that you are not addressing directly in your training. It’s surprising how often we start training based on what we think our horse can do, without actually taking the time to systematically evaluate how the horse does each behavior.

One way to start gathering information is to make a list of all the behaviors you do on a daily basis and keep track of what the horse can do. I recommend you do this for a few days as every behavior will show slight variations. Paying a little more attention to the daily routine helps me identify areas that need improvement and can also provide some nice surprises about what my horse is doing well.

Before I continue, I want to mention that it’s not a bad idea to occasionally conduct this kind of review for existing as well as in-progress behaviors. You may realize that a behavior had been slowly deteriorating (a good chance to figure out why) or improving (a good chance to celebrate) without you really noticing. I also find it’s helpful to do this kind of “where are we now?” review for behaviors where I am making progress but feel like I’ve lost direction a bit. Often I teach a behavior with a lot of precision and direction in the early stages, but once it’s “good enough,” I lose momentum. Doing a review can get me back on track by forcing me to think more about my end goals and consider the future of the behavior.

Collecting data

Since Madagascar was new to me, I had a lot to learn about what he could do and how he did it. At the same time, I knew that most horses need time to settle into new environments and that some “issues” that come up when a horse first arrives may resolve on their own after the horse settles in. With that in mind, I wrote up a list of behaviors to evaluate and started collecting data.

I will mention that when I made my list, I took into account the fact that I bought Madagascar in October and winter was coming. I didn’t have an indoor space for training so I wanted to be sure I used what time I had (before the bad weather) wisely. This is one reason I included some groundwork or riding related behaviors. It’s also why I included blanketing. The list includes some behaviors that could be broken down into smaller behaviors, so under each category I had specific points I wanted to check. I’ll give an example of a sub-list as well as the main list.

My main list:

  • Approaching
  • Touching
  • Haltering
  • Leading to and from field
  • Walking in the ring, both directions and with me on both sides
  • Tying and cross-tying
  • Picking up feet
  • Grooming
  • Saddling
  • Lunging
  • Blanketing
  • Mounting block
  • Dinnertime
  • Hand feeding/behavior around food

For each behavior, I had specific things I wanted to observe. These included things like the horse’s willingness to participate (or obvious avoidance), signs of relaxation or tension, duration for which he could maintain the behavior (if relevant), and ease with which he did the behavior.

Here’s an example of a sub-list for haltering:

  • How does he react when I approach with the halter? (move toward me, away from me, wait)
  • What does he do when I present the halter as if to put it on? (lower his head, lift his head, back up, wait, etc.)
  • As I am putting the halter on, does he assist, avoid or seem non-reactive to haltering? I use halters that have to slide over the ears so I wanted to see how comfortable he was with that.
  • Are there any indications that haltering has pleasant or unpleasant associations?

The last item on the list is hand feeding/behavior around food. Before I started formal clicker training sessions, I wanted to get a sense of how Madagascar was around food. In particular, I wanted to know:

  • How does he take treats from my hand?
  • What treats will he eat?
  • If I offer a treat, does he start to mug me for more?
  • Are there any other behavioral changes (desirable or undesirable) if I feed a treat or two?

My list changed a bit as I got to know him. I started with the basic behaviors I thought he needed for life on my farm and then I updated it as I started to identify areas that might need attention. I also did some experimenting to see if he was better at doing certain behaviors under one condition vs. another.

Organizing Your Data

I spent a week or two collecting data and establishing baselines. Some things were easy. Others took more time, or I waited until he had been here longer before asking for them. While I had no reason to believe that Madagascar had not been well handled or wasn’t capable of doing basic behaviors, he still showed gaps in his education or comfort level. I now had a list of possible training projects, and the next step was to prioritize them.

While it’s possible to be working on many behaviors during a session, I wanted to be careful that I didn’t confuse him by trying to teach too many things at once. I also thought carefully when setting priorities. I find people sometimes want to address the biggest issue with a horse right away, but often it makes more sense to wait a bit until the horse is settled in and has had some positive training experiences.

Note: this is not a complete list of everything I did with him in the first month. I’ve simplified it a bit for the purposes of this article. I did a lot of other exploratory things with him, including assessing how comfortable he was walking around our property, lunging, bridling, showing him novel objects, etc.

I divided the behaviors up into three different categories. They were:

First priority: requires immediate attention:

  • Behaviors he needed for daily life on the farm
  • I set aside time to have training sessions working specifically on these behaviors

In his case, I chose to start working on:

  • standing on the cross-ties
  • improving his foot care skills
  • leading in the ring – he tended to crowd me and was pretty unbalanced

Second priority: needs work, but can be improved slowly:

  • These were behaviors that were of lower quality than I would like, but not urgent
  • They were also behaviors that I wanted to train but didn’t need right away
  • For these behaviors, I came up with ways to slowly improve the behavior during our normal sessions – a few repetitions a day would change the behavior slowly over time, which was fine.

I chose to work on:

  • haltering – he tolerated it, but wasn’t very cooperative
  • leading from the right hand side
  • standing square
  • manners at feeding time (he tended to crowd me)
  • blanketing – I didn’t think he had worn one before
  • grooming – he seemed uncomfortable with grooming on his neck and around his ears

Third priority: behaviors that I’d like to teach him but could wait:

  • These were behaviors that he either didn’t know or that I wanted to retrain
  • They were typically not urgent, but useful

On my list, I had:

  • targeting
  • mat work
  • head lowering
  • backing

Once I had my list and had divided the behaviors up into categories, I was ready to start making training plans. The good thing about taking the time to observe and collect baseline data was that it gave me something constructive to do while Madagascar and I were getting to know each other. Once we had been together for a few weeks, I had a better sense of how to approach his training.

I realize I haven’t answered the question about whether or not I used clicker training for everything. The answer is no. If he knew a behavior well and seemed comfortable doing it, then I maintained it with “normal” handling. If he seemed to know a behavior but was a little sticky or needed some clarity, then I would mark and reinforce with scratches (he was very itchy), praise, or a break. Madagascar turned out to be pretty polite around food so I did slowly start using some food reinforcers after about two weeks. I was careful to choose behaviors where mugging was less likely to occur and I carefully observed him for signs of overexcitement or tension.

As for Madagascar, he settled in well. He made some new friends, got a new wardrobe, and proved to be a quick learner.

Want to learn more about clicker training? check out my books.

Categories: Uncategorized

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s