No two animals are exactly alike, and no two training situations are exactly alike, so my initial response to the question of what to train first is always, “It depends!” However, I do use a specific sequence when I train a new animal, and I will share it with you here.
1. Approach and eat
If an animal does not approach and eat from me willingly, I cannot start training. While this may seem obvious, I do encounter trainers who fail to recognize this fact. Before doing any type of formal training, it is important to teach an animal to approach you and take food from you. The only exception might be a remote training project that does not involve any interaction between the animal and trainer. But, even then, it is impossible to shape behavior until the animal approaches and accepts reinforcers. My priority is always the animal’s well-being, welfare, and comfort, and I work to create an environment that is comfortable and safe.
After the animal approaches and eats from me, I teach “follow.” I continue to feed, and gradually work until the animal follows me when I take a step to the right or left. When the animal stays with me as I back away several steps, that indicates increased comfort and trust and is a sign that the animal is ready for more formal training.
Next, I present a target: a hand target or a physical target, depending on the animal. Some animals are curious and sniff, investigate, or touch new objects presented to them—with those animals targeting is easy. For animals that are cautious, or less likely to touch a target on their own, I proceed more slowly.
I work toward targeting because I find targeting to be an important building block for so many future behaviors (see my letter: The Many Faces of Targeting). Once an animal is targeting readily, I know that I have developed a level of comfort and trust and we are ready to proceed.
Now I am ready to add a clicker to my targeting. Only once the animal is comfortable and is approaching, taking food, and targeting with me reliably do I introduce the clicker. This decision surprises some trainers who are accustomed to using a clicker from the very first interaction with an animal. The idea of using a clicker from the start probably developed from the introduction of clicker training to pet owners who already have a great relationship with their pets. When you start with an animal that already knows, trusts, and eats from his trainer, it makes sense to use a clicker from the beginning. When a clicker is used too soon with animals that do not trust their trainers, however, the click is likely to be associated with scary experiences, and the click does not take on the desired positive properties.
I like to establish some type of default stationing behavior. A default stationing behavior is a behavior an animal does, or a place that an animal goes to, while waiting for the next cue when the animal is anticipating reinforcement or is in doubt about what to do next. Stationing can come in many forms: a sit behavior, eye contact, going to a mat, jumping onto a platform, or moving into a heel position. In some cases, I look to establish a station before targeting because it gives the animal a comfortable place to be, a location or behavior that will always get reinforced. Actually, as soon as the animal starts eating from me I start to create a station, but it is at this point in the training process that I focus on stationing as a formal defined behavior.
6. Follow target
I introduce the idea of following a target to any location and to varying distances early in an animal’s training journey. When the concept of following a target is well established, I use “follow target” to guide an animal into doing hundreds of behaviors: kenneling, scent-detection indications, medical behaviors, research projects, guide-dog work, service-animal skills, agility behaviors, trick training, and so many more.
Tactile interaction is a foundation to training many desirable behaviors. With time, the animals I work with often seek out and enjoy being touched. But at the beginning, I think of tactile as a behavior, and I approximate touch gradually. I pay close attention to the animal’s body language and only touch when the animal truly allows it; I don’t progress if the animal shows any sign of shying away. I frequently use targeting as a great first step that leads to touch. In most cases, I avoid touching an animal’s head and feet at first, as those can be very vulnerable areas. I usually start with the rump, side, and shoulders.
8. Management and husbandry behaviors
Once I have taught the foundational basics described above, I train management and husbandry behaviors. Management and husbandry behaviors include wearing a muzzle, wearing a harness, responding to a recall, riding in a car, standing on a scale, most medical behaviors—in short, anything needed for proper day-to-day care. These kinds of behaviors are needed throughout an animal’s life and are a permanent part of my training plans at all stages of an animal’s life.
Until this point, I have guided or directed the animal in his learning. Next, I teach the animal that he can make the click happen through behavior that he initiates. By this point, the animal has learned a few rules about training, including the need to go to a station or a default behavior. So a capturing session requires a different environmental set-up in order to teach the animal that a “new training game” and a new set of rules are in effect.
I make the environment totally different from a normal session. For example, instead of a side pouch, I hide treats somewhere that I can get to quickly. I try not to make this new stage look like a training session. I make sure there is no place to go to station, and I put some unique prop in the middle of the room. The animal will usually check out the prop, and I click and reinforce his initiative (in capturing sessions, I usually toss the reinforcer on the floor). The first capturing session is the most difficult one because the animal must initiate the action with no guidance from me other than a click. Sometimes I need to move around the room, so that the object is between me and the animal, until the animal discovers that he must initiate the behavior.
I continue to differentiate future capturing sessions from other sessions by setting up the room and the environment in a unique way. For example, with my dogs, I put an object in the middle of the room, leave the room, and sit in the doorway with a baby gate or doggy door separating me from the room where the dog is located. The set-up does not look like any other type of training session. I can see what the dog is doing but he cannot get to me, and that encourages him to investigate the item. After about three of these sessions, I find that most animals recognize the new rules of this set-up and I have a new way to train behavior.
I usually do one “capturing” session out of every four training sessions at the start. Once the animal learns to generate behavior in a capturing session, I then adapt that behavior through small approximations and shape it toward a new variation of the original behavior (this is sometimes referred to as free shaping). I go back to targeting sessions frequently because, initially, I can shape behaviors more quickly with a target. But once the animal has learned to offer behaviors in a capturing session, I want the animal to be comfortable moving back and forth between targeting and capturing.
Suggestions, not a recipe
I share this foundation training plan as food for thought and as an alternative approach for getting animals off to the best possible start. Training is such an important part of developing relationships with the animals in our lives; it is important to start that relationship in the clearest possible way.
Post new comment