q
Welcome to moose theme.
Enter any text or widget here.
RPlus | Match to sample
Marlitt Wendt, Conny Ranz, Pferdsein, RPlus, R+, Clickertraining, clicker training, Clicker, Clickern, Positives Pferdetraining, Positive Verstärkung, Pferdeverhalten, Pferde-Ethologie, Pferdeethologie, Equine ethology, Native horses, Shaping, Target, Pferdetraining, empowered equestrians, Wildpferde, positive reinforcement, positive reinforcement training, Zirkuslektionen, Bodenarbeit, Freiheitsdressur, Freiarbeit, Wenn Pferde lächeln, Belohnung, Belohnungslernen
353089
post-template-default,single,single-post,postid-353089,single-format-standard,eltd-cpt-1.0,ajax_leftright,page_not_loaded,,moose-ver-2.1, vertical_menu_with_scroll,smooth_scroll,side_menu_slide_with_content,width_370,blog_installed,wpb-js-composer js-comp-ver-5.5.2,vc_responsive

Object discrimination

 

Our horses’ cognitive performance is grossly underestimated until today. Yet, their intellectual capacities carry a beauty within that can best be described with the following image: It is like taking a stroll at the beach when every now and then a horsy flash of insight or a small empathic jewel washes up on the shore so we can pick it up. However, only the vast expanse of the ocean reflects their unexplored mental depths, and we can hardly guess its true extent. Today we want to venture out into this exciting world by exploring our horses’ abstractive ability and their capacity for object discrimination.

 

Today we at R+ are devoting ourselves to exciting brain challenges.
Your horse is going to learn to discriminate colors.
An experimental setup.

So, what does belong together?

 

Experimental animal psychology has used standardized experimental setups to investigate the thinking ability and memory performance of animals for years; these setups are called “matching to sample” or “match to sample”. Horses learn to independently form categories, and to discriminate between objects by identifying matching pairs.

Brainy horses

 

In this process they are able to learn, for example, that a certain object always belongs to an identical counterpart, or to a visually different counterpart; and they are able to develop criteria for the classification of objects according to form or color. After going through this challenging learning process they will be able to make the connection between a sample stimulus and a target stimulus so to perform the corresponding action we have selected.

Match to sample for horses

 

Horses are really clever, and it is a lot of fun to watch them think; it is also good fun to observe them explore, step by step, how to solve a certain memory-task, or to watch them as they learn to discriminate colors upon specific signals. For example, they are able to indicate the yellow object among a variety of differently colored objects. The horse learns “What belongs together?” as a basic principle; it begins to form categories, and informs us about the categories by the means of this little exercise.

Discriminate colors

 

We aim to turn dry theory into luscious praxis; and we want to present you with a good example you can follow: This is the discrimination of colors. Harry, my gelding, has learned in training to indicate a same-colored foam cube after he has been presented with an identical example to stimulate a response. Step by step, we can build upon the following exercise plan, include more colors and let the horse indicate them; or we could experiment with different objects that share the same color and investigate in what way our horse has understood the criterion “match the duplicate”. Next, we will show you how you can structure your training in order to learn these exercises.

Origin art design cube: Designed by starline / Freepik

What do we need?

 

First, we need three identical pairs in three different colors, for example foam cubes, buckets, or stuffed animals. Blue and yellow are good colors to start because your horse is able to discriminate very well between these two colors. Later, we can include more.

PHASE 1

 

 

Train to indicate

 

First, the horse needs to learn how to indicate a solution, and only afterwards we can start to think about matching pairs. So, the horse should learn to touch the object with its nose without changing the object’s position, and without picking the object up with the mouth, and tossing it around.

How we start

 

To this end, we present the object as a classical target, in our case the first object is a cube with the first color, yellow. We hold the cube near the horse’s nose, and we click and reward whenever the curious horse-nose touches the cube. Little by little, we change the position of the cube but keep rewarding whenever the object is touched with the nose. We never reward other activities with the cube.

Increase the challenge

 

We can put the cube down on the floor in front of us as soon as the horse follows every movement of the cube in all directions with its head. We continue to reward touching the cube with the nose, and slowly change the cube’s position on the floor. Next, we start throwing the cube sideward so the horse needs to move in order to reach the cube. We can move on to the next phase in training as soon as the horse repeatedly moves after the cube in a determined manner and without being distracted even when it has to cover larger distances.

PHASE 2

 

 

Introduce the sample stimulus as signal

 

Next, the horse needs to learn that it should not simply run after the cube when it is thrown but to wait for the corresponding signal. So, the horse moves to the cube only after we present the corresponding identical example cube.

The target stimulus

 

The sample stimulus, so its scientific name, indicates for the horse a target stimulus, as it is scientifically called. So, Harry needs to understand that he should touch the yellow cube only when I present the second yellow cube in my hand as a signal for action.

A new signal

 

We only introduce our new signal with the sample stimulus when the horse constantly and reliably performs the behavior we have been training before. So, the horse has accomplished phase one. Now, we can hold up the sample cube to introduce the new signal whenever the horse starts moving towards the cube that lies somewhere on the floor. Bit by bit, we start rewarding only when the horse stops and waits for the signal. Occasionally, we can also reward the horse for standing and waiting patiently next to us. So, the horse learns that only our presentation of the sample cube is an invitation to indicate the target cube.

 

PHASE 3

 

Discriminate colors

 

It is only now that we start training the exercise “match to sample” where the horse needs to reliably locate and indicate a specific object among a variety of differing objects. For example, Harry gets the signal “sample cube yellow”, and should touch the yellow target cube even if other cubes in different colors are present.

Hot Target

 

In this step, we present the yellow target cube as “hot target”. This means it is the target object that should have the greatest appeal for the horse. So, we deliberately reinforce moments when the horse touches this target. The other cubes in different colors serve as distracting stimuli; so to say, they are stimuli we simultaneously present to distract the horse.

Understand complex relations

 

They are here to lead our horse on the wrong track. It is an easy and straightforward task to touch a yellow cube if there is only a yellow cube. However, the exercise is a lot more complex and includes new challenges if more than one possibility is available. They are here to lead our horse on the wrong track. It is an easy and straightforward task to touch a yellow cube if there is only a yellow cube. However, the exercise is a lot more complex and includes new challenges if more than one possibility is available.

Create moments of success

 

It is likely that only in this phase of training many horses will understand what aspect we are really looking for, in our example it is color. Maybe you remember playing “I spy with my little eye …”? Similarly, the child needs to locate and name a specifically colored object among other visible objects. In order to teach our horse to discriminate between colors it might be best to first leave the distracting stimuli in the background, and to place the yellow target cube in the next possible position.

Vary distracting stimuli

 

This reduces the probability that our horse just walks over to the next cube it can spot, and thereby reduces the probability for making frustrating mistakes. So, in the first rounds of training we change the position of the yellow cube but we always put the yellow cube nearest to the horse while simultaneously also switching all the other cubes with other colors. In short, the main focus is not on creating an unnecessarily demanding experimental setup but on creating as many successful moments as possible.

Mistakes can happen

 

What is important, we do not click and reward if the horse makes a mistake, yet, we quickly start a new attempt. This is to prevent our horse from simply turning to the next object, which might be the right one, if it does not receive a click for its first mistaken try. The second touch, which we assume would be on the right object, would then lead to a click and a reward, and the horse would learn to try and touch objects without thinking until it hits the target by chance.

Slowly introduce criteria

 

We can see that we have entered into this learning phase too quickly if our horse apparently struggles to understand the task, and aimlessly runs around touching different objects by chance. Particularly, starting an exercise again is stressful for horses. So, we should introduce new challenges only when we are quite certain that the next level of training can easily be mastered, and when the moments of success will outweigh the mistakes. For instance, only after the yellow cube is reliably chosen when other colors are present, we start switching all the colors.

Comprehend the essence of the task

 

Then, we can put other colors near the horse in order to increase the level of difficulty, or we can also change the way the cubes are positioned, for example, transform a straight line of cubes into a triangular shape.

 

A small tip from us: The best moment to rearrange the lying cubes for the next round is when, after the click, we reward the horse with a treat, and reposition it by moving the horse back to the starting position and away from the cubes. The horse understands this game if it reliably choses the right cube even if many more cubes in different colors are present.

PHASE 4

 

 

Repetition

 

Analogous to the process established in phases one and two we start including further colors. It is best to add one color first, and later another color because adding more colors at the same time can be confusing.

 

Each of the target cubes receives a corresponding match, and the horse learns to autonomously apply the predominant concept of sameness and difference when it always walks to the blue cube when it sees the color blue. Bit by bit, the horse will be able to transfer the categories to other tasks, and will gain security in applying them.

Pay attention to balance

 

So, the horse can learn to discriminate colors and combine them with specific actions like we have shown in our previous example. What is worth noting, it is crucial to dedicate the same amount of time for practice to all colors and objects. For example, training the third color only 20% of the time we have used to work on the first color can quickly lead to a greater appeal of the first color when compared to the other colors. Consequently, this increases the probability for mistakes in the following learning process. There can also be cases when a horse strongly prefers a certain color or object to all the others, and always privileges this color or object. Then, we can increase the proportional appeal of all the other colors and objects through a series of many repetitions and rewards.

 

Apply the concept

 

Once the horse has understood the concept, we can add more colors or objects in an analogous manner; for example, we can add more matching objects as well as look for a completely new object in an already known color.

 

Brain games for horses

 

Such discrimination tasks serve very well to increase the cognitive abilities of horses. Yet, it is definitely worth checking and observing whether our horse has really internalized the task or whether it only appears to have understood what we want it to learn because horses are really clever. So, it easily happens that our horse does not know how to discriminate colors but has simply learned to follow our gaze to the target object. Consequently, it touches the object we are observing with its nose instead of choosing the right category.

Prevent mistakes right from the start

 

From a scientific point of view, we have to try to prevent such mistakes right from the start by not giving away unintentional clues like staring at the object, turning towards it or pointing at the target. To practice this we can, for example, ask a friend to observe our horse’s reaction when we turn our back towards the horse, and present the sample stimulus. Our friend can then tell us whether the horse still knows to discriminate the colors even if we are not (intentionally) looking or whether the rate of mistakes increases considerably.

What does this mean?

 

Furthermore, we do not even need a real sample object as signal. What does this mean? Our horse can learn to act upon a spoken signal, for instance a specific word, to indicate the right object. Consequently, this also happens even if we do not present the corresponding sample stimulus, the one that signals to indicate the matching sample, in our hands.

Cognitive high performance

 

Tasks like discriminating colors are a lot more challenging and complex than they appear in the first place. It makes sense to establish clear milestones, and to make a lot of breaks in order to find the right pace of learning. It is also worth taking back a step if it becomes clear that one part of the task has not been fully understood.

Proceed in small steps

 

It is easiest for us to maintain the overview for ourselves in this challenging exercise if we plan small and manageable learning sessions. For example, we chose to repeat each round only four times, and then we take a break. In this break we also document the rate of successful versus unsuccessful rounds, and we only increase the level of difficulty and proceed to the next phase if we achieve a rate of 3:1 or 4:0. Ideally, we have a set of four repetitions without mistake or a maximum of one mistake in total. This task is intellectually challenging for our horses, and we should not ask for more than they are able to perform. Too much enthusiasm can easily lead to inadvertent mistakes, or even worse, we might unintentionally make excessive demands our horse cannot fulfill.

Considerable potential for development

 

Now, that we have dealt with all the minute steps of this rather untypical practical lesson, let’s step back mentally, and reflect upon our doing from a bit of a distance. Here, horse and human do not simply learn a spectacular trick to impress others. In fact, this exercise helps to develop a profound understanding of our horses’ marvelous ability to perceive, their cognitive potential that can be developed, and this exercise also teaches about the power of positive communication in our relationship.

Marlitt Wendt & Conny Ranz

AUTHOR: Conny & Marlitt