As someone who works with animals and with with people who work with animals, it is really important that I fully understand and utilize teaching techniques that will reduce frustration (in both my learners) and that will get the desired results quickly.

One of the ideas I have come across in the last 10 years which has really helped me and my students, both canine and human, is a deeper understanding of the implication of the learning cycle, or learning loop. There are a few graphical representations of the learning cycle, but here is the one I have found to be the most helpful.

Because it is a cycle, there really is no beginning or ending and each segment influences the next each time we go around, but most people start by talking about the ABCs. 

‘A’ stands for Antecedent...

This could be a cue, BUT it also includes all aspects of the environment that gives your learner information about what will earn reinforcement.

‘B’ stands for Behaviour... 

This can be a thought or an action that the learner does.

‘C’ stands for Consequence... 

This is the outcome of the behaviour from the learner’s perspective.

 When I first started training I focused a lot on this part of the Learning Loop. When you think about it, this is actually more of a line if you stop there. More recently, I’ve come to realize that when we are training we actually want to create a loop or cycle and not a line. Closing the loop so that the learner is ready to begin another repetition is an important aspect to consider because a quicker return to the ‘A’ each time means that within a 2-3 minute training session, we will give more treats overall (a higher rate of reinforcement), which makes for a more memorable session and more good feelings about training in general.

When I create a quick return to being ready to begin another loop also means that in a 2-3 minute session, you will actually get more reps of the “agenda behaviour” (the one you are trying to teach) which should speed up learning overall.

Let’s look at a specific example of how I have changed my approach to training a specific behaviour thinking more explicitly about how to create nice clean training loops while still progressing the behaviour forward.

Exploring the example of sending to a bed or mat

For years I taught sending to a bed or mat using the following general steps: Begin standing on the same side of the bed/mat as the dog, hide cookies on the bed/mat when the dog isn’t looking, approach the bed/mat together and encourage the dog to explore the bed/mat, mark and add new cookies to the mat, call the dog off the mat (release) back to the same side as you, repeat a few times, once the dog is trying to get to the mat to explore remove the planted lure cookies and just mark and reward for them getting to the mat, add an automatic down and distance to this mat behaviour.

Perhaps this progression sounds familiar, it has been around a long time and “it works”. But when I examined it closely with the new lens of the learning loop, there are many places where that progression fell short. There were many spots where I found I could make the learning as easy as possible for the dogs and the execution of the training as easy as possible to do well for the humans. And so in the past 3-4 years I have changed what I consider to be a more efficient and supportive progression for my canine learners. In fact, the new progression can be so easy and seamless, you may wonder if there is any learning going on there at all, and yet we see the behaviour changing.

The problem areas with the old procedure with regards to the learning loop

Antecedent Issues

Because the dog and the handler are on the same side of the bed/mat to start, I found that there was nothing in the general set up that encouraged the dog to explore the mat. Putting cookies on the mat to encourage the dog to explore the mat was used. This set-up can be effective but since we want the mat itself to help to cue the behaviour and not a mat with cookies on it we will have to fade the presence of the cookies at some point. We will have to change the antecedent. Additionally, I found that this progression tended to get a lot of movement towards the mat from the handler which becomes a part of the antecedent or cue that we later need to get rid of as well. Since many breeds of dogs have been bred for centuries to pay close attention to the movements of their handlers, I have found that this can be hard to do especially in dogs who are very concerned with where the food is coming from.

Behaviour Issues

The biggest behaviour issue with the old procedure is that because of the set up of the dog and handler on the same side of the mat, splitting the behaviour really finely tended to encourage the handler and the dog to start each loop very close to the mat, sometimes only requiring the dog to take a single step or two to get onto the mat. Since my end goal is for the dog to be able to travel 20 - 40 feet to their bed, this really doesn’t help me out as big movement toward the bed isn’t really a part of the behaviour from the beginning. This means building distance will be trickier down the road. When distance was added, we ended up with a lot of “failures” and handlers stepping in to “help” adding handler movement toward the bed to increase clarity of what was expected. This, of course muddies the antecedent as explained above.

Consequence Issues

One of the biggest challenges with the old procedure I believe, is that rewards were almost always delivered while the dog was on the bed/mat. If the dog is already on the bed/mat they cannot preform the behaviour again right away. They have to be released, or called off the mat in order to be able to practice another round of sending to the mat. Essentially, in the old procedure I used to lump together 2 different aspects, sending to the mat, and staying on the mat (duration) at the same time. I found that it usually resulted in dogs who didn’t send very well to the mat, but who would hold there fairly well once they had some help to get there. Again, the independent send is weakened or at the very least, more challenging to build.

When we reset for the next rep, we necessarily introduced a set of other behaviours. For example,  calling or releasing off the mat, which added challenge of correct execution for both my novice canine and my novice human learners.

How my new procedure creates seamless learning loops.

The procedure I use now has been adapted from a procedure from Hannah Brannigan, introduced to me by my friend and colleague Melissa Gillard.

1. 1. Pull your bed away from the wall about 2 “dog lengths”.
2. 2. Stand on the opposite side of the bed from the wall.
3. 3. Toss a cookie (in such a way as your dog can see the cookie being thrown) toward the wall so that once your dog eats the treat, the bed is between you and the dog.
4. 4. As your dog returns toward you, if they step on the bed “mark” with the word “yes” and throw the cookie back towards the wall.

As the dogs get more comfortable we shift criteria slightly for 4 feet on the bed, and in some classes like agility we might introduce the word “search” to allow better communication as to where the cookie is going to show up, but in a nutshell, the essential elements of the behaviour is trained in a matter of minutes.

Antecedent Improvements

Now when I teach the dog to go to a mat or bed, the bed is pulled away from the wall giving about 2 ‘dog lengths’ of space between the bed and the wall for the dog to maneuver. At the beginning stage of the training, the dog will most often be on the opposite side of the bed from the handler and the treat will usually be tossed to that side as well. This antecedent arrangement will allow the dog to take advantage of their natural inclination to return toward the handler and step onto the bed/mat on their way. 

Behaviour Improvements 

The behaviour of movement toward the bed and getting up onto the bed is so easy to train this way, we only encounter a slowing down of progress with dogs who are Leary of stepping up onto anything. When this happens we either use a bit of a lure a few times to help the dog explore the bed, or we change out the bed for another option that is easier for them and then progress forward from there. When we are ready to change the angle of approach to the bed toward the dog being on the same side of the bed as the handler, it is easy to do this gradually by changing where the food is delivered (more on that in the consequence section). It is so easy to progress in this way, it almost flows naturally from the initial behaviour which always included a fair bit of motion toward the bed.

Consequence Improvements

This is where the Learning Loop viewpoint really creates a different outcome. It helps us to take a look at what is the agenda behaviour REALLY and ask ourselves “how can we set the dog up to do THAT again as seamlessly as possible?” With this lens on, we mark the dog for getting up onto the bed, but we toss the treat AWAY from us to the opposite side of the mat to collect their treat. After they pick up their treat and eat it, they are right back where they need to be to restart the learning loop. They are off the mat, in a position to move toward the mat to earn their next reinforcement. 

Shifting the ABCs

As dogs become confident in this pattern, you can make small changes in handler position and the location of the treat toss (consequence) to extend their understanding of distance (antecedent/behaviour), angle of approach (antecedent/behaviour) and handler position relative to the bed (antecedent). It can be so gradual that is seamless, but also fairly quick taking only a few training sessions instead of weeks of work. When the dog doesn’t really have any questions and the training is very efficient, learning can advance quickly.

Creating confidence with tight learning loops

We have had such success with this procedure that even a completely novice handler and dog can look like pros in a matter of minutes. What does it mean for them to look like pros? It means they both look confident and comfortable with what they are doing. When our human learners feel competent and their dog is successful, the human animal connection is strengthened, which is why I do the work I do in the first place.

Do you have any procedures you use in teaching your dog(s) where you can create nice tight training loops? How does this make you feel as the human involved in the session? Let me know in the comments!

Until next time,

❤️ THAT 🐶