Soon after learning that a click marks the exact behavior and tells the animal what earned its reward, newcomers to clicker training wonder how to go about getting that clickable behavior. On this point, the answer they hear depends on whom they ask.
"Get the behavior" is one of the adages of clicker training. The saying reinforces the creativity enabled by the method: however you get there is fine; just get there. Or, as Karen Pryor points out, "There are as many ways of training a behavior as there are trainers." In general, however, trainers tend to turn to one of four methods to get the behavior: shaping, capturing, targeting, or luring.
When working with newcomers to clicker training, some teachers will suggest they begin with luring their animal into the desired position, then click and reward. Others ask their students first to teach their animal to touch a target with its nose (or paw or hoof or beak) and then to move that target to put the animal into the clickable position or motion. Using a lure or a target can jump-start a behavior, getting a trainer from point A to point M, T, or X in a few short steps. (Shaping is rich with its own benefits, but relies on clicking and rewarding incremental steps, and can be derailed by jumping over those steps. Capturing, a one-step training method, relies on a behavior being frequent and observable.) Both a food lure and a touched target lead an animal into position. Both "get the behavior." Is there any real difference between the two methods? Why do some trainers rely on luring while others opt for targeting?
We asked trainers and teachers with experience with a range of animals and training goals these questions. Their consensus is that the methods have some similarities and important differences.
Both methods require an eventual fade of the lure or the target so that they don't become the final cue for the behavior. Both carry a risk of the trainer becoming as dependent on a physical prompt as is the animal to complete the behavior. Pet class teachers often hear the lament, "I can't get him to lie down without food in my hand. If I don't have it, he won't do it." These beginning trainers fear losing the progress they have made toward a desired behavior without the method that got them there. (Experienced trainers are not immune to this worry and may also hesitate in fading a lure or a target.) The speed with which luring got the trainer to see the desired behavior is reinforcing to the trainer, and reinforced behaviors endure in trainers as well as learners.
Despite the apparent similarities, however, there is an essential difference between luring and targeting. As Ken Ramirez, ClickerExpo faculty member and Training Director for the Shedd Aquarium in Chicago, explains: "In luring, the animal is focused on the food. The trainer uses the food to guide the animal toward a desired behavior, just as a trainer would use a target to guide the animal. What goes on in the animal's head, however, may be significantly different. Luring keeps the animal thinking about the reinforcement or the treat, while targeting gets the animal thinking about the task."
While both methods may succeed in getting the behavior, a lured animal may be so focused on the treat that it is not aware of what behavior it has just accomplished to earn the reinforcement. Less learning, therefore, has been accomplished. An animal that follows a target, on the other hand, may still be working to receive a treat, but because the treat is not right in front of him, the animal must think about the actual behavior. The result is an animal that is more engaged in the process, has accomplished more learning, and is more able to apply that learning to any number of other behaviors.
Kay Laurence, of Learning about Dogs and Teaching Dogs magazine, prefers targeting to luring because of the flexibility it offers. She asks her Border collies and Gordon setters to touch targets with their noses, and their paws, and their tails, and their hips. The result is an ability to combine a series of cues to direct the dogs into desired positions: left back paw on step, nose on right front paw, and so on in infinite variety. Kay's fine tuning of the use of a target underscores an important principle for newcomers and experienced trainers alike: Many behaviors we humans think of as a single behavior are, in fact, combinations of multiple behaviors. A down is a down, but there are many ways in which a dog can lie down (front feet first, back legs first, on its side, or legs tucked under its belly). Creative use of a target can help a trainer to guide the components that make up a completed behavior, rather than a broad approximation of the behavior.
At its essence, targeting engages the animal's mind, rather than its appetite. It allows the use of food to be maintained as a reinforcer of behavior, rather than a stimulant of behavior. And it enables trainers to be as creative as they wish, and to extend the applications of clicker training as far as they can imagine.