Clicker Training

by Rebecca on November 7, 2010

in Animal Training

From Wikipedia:

Clicker training is an operant conditioning method for training an animal using a clicker, or small mechanical noisemaker, as a marker for behavior. The method uses positive reinforcement – it is reward based. The clicker is used during the acquisition phase of training a new behavior, to allow the animal to rapidly identify that a behavior is sought and also the precise behavior of interest.

File:ClickerDogFxwb.jpg 

Clicker training was originated through Marian Bailey (neé Kruse) and Keller Breland, who as graduate students of psychologist and eminent behaviorist B.F. Skinner taught wild-caught pigeons to bowl while participating in military research. According to their work, animal training was being needlessly hindered because traditional methods of praise and reward did not inform the animal of success with enough promptness and precision to create the required cognitive connections for speedy learning. Similar methods were later used in training at least 140 species including whales, bears, lions and domestic dogs and cats, and even humans.

Advantages

One of the challenges in training an animal is communicating exactly when the animal has done the behavior that the handler is attempting to reinforce. As a simple example, consider teaching a dog to turn in a circle (spin). At the instant that the dog completes the turn, the handler must let the dog know that it has done the correct thing. However, the traditional “good dog!” takes so long to say that the dog might already have moved on to some other behavior. At the least it is not immediately obvious that the “good boy” is earned at the precise moment of completing a circle. By the time the dog realizes it is being praised, it might be sitting and scratching or looking for something else to do.

In the laboratory behavioral researchers including Norm Guttman, Marian Kruse and Keller Breland, realized that rats always stop what they are doing when they hear the hopper make a sound indicating it was beginning to deliver food, and they tend to do more of what they were doing when the sound occurred. Under the instruction of B.F. Skinner, they decided to try using a sound to mark behavior outside the operant chamber. Toy crickets, the earlier equivalent of today’s clicker, were common in those days, and served the purpose very well. The clicker is likened to the surgeon’s scalpel; it allows for precise timing and clear communication about what specific behavior is being reinforced, and enables the trainer to teach complex and difficult skills to the animal without the use of force or punishment.

At least one study has shown that the clicker can reduce training time by 1/3.

As this type of training was practiced and improved upon, it became apparent that the variability of the human voice, and its presence during all activities make it a less than salient tool for marking behavior. Besides the imprecision in timing, using the trainer’s voice for feedback means that the actual sounds for feedback will vary. A handler’s voice, pronunciation, tone, loudness, and emphasis may change even during the same training session. Clicker trainers believe that it is better to use a “click” sound to avoid variations in sound. Many trainers opt to use clickers for training that requires precision and continue to use their voices in the form of praise for behaviors that do not need to be precise.

There is also some circumstantial evidence which suggests that the sound of the clicker is the kind of stimulus — like a bright flash of light or a loud, sudden sound — that reach the amygdala (the center of emotion in the brain) first, before reaching the cortex (the thinking part of the brain). Clicker trainers often see rapid learning, long retention and a “joy” response to the sound of the click in the learning animal. This idea is not universally accepted, and no known research has confirmed it. Any reinforcer can produce joyful behaviors in learners if delivered correctly.

Tasks learned with the clicker are retained even years after the fact and with no additional practice after the initial learning has taken place. This is probably due to the fact that the animal participates fully in the learning process and applies itself to it, learning by trial and error rather than acting out of habit or a momentary response to a situation. Clicker–trained animals become great problem–solvers, develop confidence, and perform their work enthusiastically. This retention of learning is present in positive reinforcement training (including but not exclusive to clicker training), but does not happen with any regularity with correction-based training.

The marker can be any signal that the animal can perceive, so long as the signal is brief (to prevent the problem of imprecise timing) and consistent (to prevent the problem of variations that may confuse the animal). For large sea animals the marker is usually a whistle rather than a clicker. However, not all conditioned reinforcers are sounds. Goldfish and birds such as falcons and hawks can be trained using a quick flash of a flashlight as their “clicker”.] Deaf dogs can be trained with a vibrating collar.

As pointed out by Lindsay the advantages of the clicker may be particularly strong in some situations: “…the clicker’s simplicity and clarity provide a significant advantage for some training activities…”


Previous post:

Next post: