Animals learn a variety of ways: trial and error, association, social contexts, etc. When we are actively training pets skills, we focus primarily on 1.) Association training and 2.) Consequence training.
Association Training
Also known as “Classical Conditioning,” this learning occurs when the individual develops an emotional response to a specific situation, item, person, etc.
Positive Conditioned Emotional Response (+CER) = They associate the stimulus with something good, happy, fun, etc.
Negative Conditioned Emotional Response (-CER) = They associate the stimulus with something scary, painful, intimidating, etc.
Most of our patients have a -CER when they start working with us. The parking lot, the hospital, certain equipment, the way a person stands or touches them, etc. All of these things can become associated with fear or pain over time. Our goal is to change the emotional response to these stimuli to something more positive.
Consequence Training
Also known as “Operant Conditioning,” this type of learning occurs when the individual actively performs a behavior and then experiences some type of consequence as a result. We focus primarily on positive reinforcement. This means we provide a stimulus the pet likes to encourage them to perform that behavior again. Rewards and punishments are in the eyes of the learner, so we need to develop a Reinforcement Hierarchy for each individual. Some animals don’t find being touched by a stranger a positive experience. In fact, petting a dog to “reward” them could be punishing. Now they don’t want to do that again!
Marker Training
Marker training refers to teaching the pet that a specific word, sight, or sound means whatever they
just did was the rewardable behavior, and a reinforcer is forthcoming. Clicker training is a type of marker training. The clicker is a small device that makes a metallic click noise when the button is pressed. The idea behind the clicker is that it is a unique noise – it is unlikely the pet will hear it in other contexts outside of learning. The closer a marker and reward is 1:1, the stronger the learning. Not all trainers feel comfortable using a clicker for marking. In that case, we may use a specific word “yes,” or “yip” instead. Ultimately, the marker “takes a picture” of the behavior for the pet so it is more clear what earned them reinforcement.
Example: “Bailey, sit!”
Bailey sits.
Mom clicks as her bottom hits the ground.
Bailey recognizes the act of bottom-on-ground is what she wanted.
Mom gives a cookie as the reinforcement.
A cue is any stimulus that the pet can perceive that is meant to encourage a behavior to occur. In traditional training, cues are also known as commands. Cues can come in a variety of forms: a word, a hand gesture, the way a person stands, which foot you step off with, the lights turning on, etc. We can make a cue be pretty much anything as long as the pet realizes it indicates a specific behavior can be rewarded when present.
Practical Training
There are a few different ways to teach cued behaviors. Each have their list of pro’s and con’s and each have their place in training animals.
Luring – This is how most people train cued behaviors. You use a treat and move it in such a way that the pet follows it and almost accidentally does what you want. For example, if I want to teach a dog to stand for a sit using a lure, I will hold the treat in front of their nose and move it just out of reach. Now that dog needs to stand up to reach it. They stand, I mark and give the treat. We typically teach come, sit, down, and stand with his technique.
The Good: This is a quick easy way to teach several skills in a short amount of time. Most dog owners are already familiar with this technique.
The Bad: Busy dogs will frequent mouth the hand with the treat. This also doesn’t work if a scary person is trying to train the dog – they are too close and risk eliciting an aggressive response.
Targeting – Targeting is similar to luring, but the target is the lure, and it is a more conscious learning process. We start by teaching the pet to target an object – hand, wand, etc. We hold the item in front of the pet’s nose, they show interest and sniff it. We mark the sniff, remove the target, and reward with a treat. This was an accident at first, but with practice, the pet learns that when the hand or target is presented, it is an invitation to come and touch it with their nose. This behavior is then used as a foundation for others: standing, redirecting, moving around an obstacle, etc. Hand targets (nose to hand) are the most common targets, but you can also have paw-to-board, body-to-platform, etc.
The Good: This is a powerful tool that can be used in a variety of situations and as a foundation skill for multiple others. This is also commonly used in most dog sports.
The Bad: It can take a little practice for newcomers to teach their pet this skill.
Capturing – This technique requires the teacher set up the learning environment so that it is likely the pet will naturally offer the behavior. The teacher is not actively asking for anything. They wait for the pet to perform the behavior on their own, the teacher marks, and then rewards. An example of this might include teaching a down. Set up a quiet, boring room: just you, the dog, and their treats. At first the dog might sense a training session and nudge the person. The teacher ignores the dog by reading a book, playing on their phone, etc. Eventually, the dog loses interest, gets bored, wanders off, then lies down to rest. The teacher marks the lying down and tosses a reward. With capturing, it works best to toss the treat, so the pet needs to stand up to retrieve it. This resets their ability to offer the behavior again. After the first repetition, the dog will likely come up and ask for treats again. The teacher ignores, the dog loses interest, and lies down again. Mark, reset with a treat. With repetitions, the dog will make the connection that “down” is what earns the reinforcement.
Up to this point, the teacher has not provided a specific word or gesture to indicate they want a “down.” For now, the cue is the learning context. Once the behavior becomes predictable, the teacher can start adding a cue to the action of lying down. As the dogs starts to bend their elbows into the down movement, say “down” or give your down hand gesture, then mark when finished, and reset with a treat. Through repetition and association, the dog learns that “down” means lie down on the ground.
The Good: Capturing helps develop good default behaviors and teaches the learner that their decisions matter. It helps them learn to make good choices on their own without needing constant supervision from their humans.
The Bad: Timing of the teacher is crucial. If a mark is missed on the mark indicates the wrong behavior, it can be difficult to course correct.
Shaping – This technique is similar to capturing, but rather than expecting the full behavior in one sitting, the teacher rewards closer and closer approximations to the goal behavior. If we are teaching a “down” using shaping, the teacher might mark sitting first. Once the pet understands “sitting” is rewardable, they change the criterium. Now the pet must sit and bob their head down to earn the treat. Next, they must move their head halfway to the ground. Next, they touch the ground with their nose. Finally, they might fall into a down. With each change in criterium, the previous behavior is no longer adequate. It doesn’t earn reinforcement. Only “trying harder” or trying different things will get them closer to their mark.
The Good: Shaping allows the dog to slowly learn highly complicated behaviors that might not otherwise be possible. It allows the dog the freedom to try things and learn their decisions matter. It also allows animals to learn from a distance without social pressure or invasion of personal space.
The Bad: Similar to capturing, timing is everything. The teacher needs to be observant and have a strong ability to predict behavior patterns to make their marks and reinforcers clear.
Getting Started
It’s best to practice practical training skills with another human before moving on to your pet. Below are some exercises to help your observation, timing, and muscle memory.
Marker Timing
Person 1 – in charge of cueing and marking
Person 2 – in charge of behaving
Person 1 – Lift your palm up towards Person 2 as if you’re going to “high-five” them.
Person 2 – Using your pointer finger, poke Person 1’s palm.
Person 1 – Mark at the moment the finger touches your hand. The goal is for them to occur simultaneously.
Practice 3-5 times, then Person 2 can start making it more difficult by hesitating before actually touching, moving their finger around and then touching the palm, etc. Switch roles after a few minutes.
Shaping Practice
Gather a supply of 5-6 random objects from around the house. Some examples might be a key, a small box, a die, a card, some small rubber toy, etc. Develop a fairly straight forward goal behavior:
Pick up yellow frog and put it on the ace of spades.
Roll the toy car across the table.
Stack the purple block on top of the green one.
Whatever it is, it needs to be clear and concise. Consider writing it down.
Person 1 – Responsible for marking and shaping the behavior.
Person 2 – Responsible for responding to the marker and changing behavior accordingly.
Person 2 will likely start by moving their hand new the objects. Mark moving forward the items to indicate that the movement is a good start. Mark as they get closer to the appropriate object. Ignore touching incorrect items. Mark when they pick up the item. Mark as they move it in the right direction. Mark as they put it where it is supposed to go.
This is a great exercise not only for Person 2 to practice timing and observation needed for shaping, but also for Person 1 to develop empathy for their pet. It can be so frustrating learning something when you can’t talk about it!
Comments