Clicker training

From Wikipedia, the free encyclopedia - View original article

 
Jump to: navigation, search
Clicker-training a dog.

Clicker training (also known as magazine training) is a method for training animals that uses positive reinforcement in conjunction with a clicker, or small mechanical noisemaker, to mark the behavior being reinforced. The clicker is used during the acquisition phase of training a new behavior, to allow the animal to rapidly identify the precise behavior of interest.

Clicker training originated with Marian Bailey (née Kruse) and Keller Breland, who as graduate students of psychologist and eminent behaviorist B.F. Skinner taught wild-caught pigeons to "bowl" (push a ball with their beaks) during military research.[1] 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, chickens and domestic dogs and cats, and even humans (TAGteach).[2][3][4][5]

A clicker is just one example of a conditioned reinforcer (secondary reinforcer) or "bridge".[6] Technically a stimulus from any sensory mode may become a conditioned reinforcer (ex. light, smells).

Co-founders[edit]

B. F. Skinner first identified and described the principles of operant conditioning.[7][8] Marian Kruse and Keller Breland, two of Skinner’s first students (and later married couple) first saw the potential of the new technique for animal training business.[9]

After participating as research students with Skinner in pigeon behavior and training projects during World War II,[10] the Brelands left graduate school and formed the first company to intentionally use operant conditioning, Animal Behavior Enterprises (ABE).[2] They created the first free-flying bird shows and a host of commercial animal exhibits.[11]

Bob Bailey was the US Navy's first Director of Training[2] and later came to work at ABE in 1965. Keller Breland died in 1965 and Marian married Bob Bailey in 1976. Together they continued the pioneering work at ABE. Radio-carrying cats were steered through cities and into buildings under a contract with the CIA. Dolphins located targets many miles from their trainers, at sea. Ravens and other birds, carrying cameras and directed by lasers, could fly to a specific window of a skyscraper and photograph the people inside. Gulls, expert sea searchers by nature, could locate and report life rafts and swimmers far offshore.[12]

Controversy[edit]

There are several common objections posed to clicker training. Proponents assert that while most of these can be a problem for the unskilled clicker trainer, these are all avoidable.

  1. "The dog will never perform the behavior without the clicker." The clicker should be used to identify correct behavior during training, not to maintain behavior once the behavior has been learned. Once a behavior is performed each time the animal hears a specific cue (known as a command in traditional training), the clicker is discontinued.[13][14]
  2. "Dogs will become distracted by the clicks of other trainers in a class or public setting." This is very short-lived problem. Participants in clicker classes find that dogs are easily able to discriminate that only the clicks from their handler pay off. Clicks that don't pay off are soon ignored by animals in learning situations.
  3. "Dogs become fat with clicker training because they get too many treats." Part 1 of the solution to this problem is either to use a portion of the dog's regular diet as the training treats or to use reinforcers other than food.[15] Part 2 is to remember that a training treat for a dog the size of a Labrador Retriever should be about the size of a pea. Smaller dogs get even smaller treats. Larger dogs get only slightly larger treats. Food is not the only reinforcer that can be used in training. A "reinforcer" is anything the animal is willing to work for in the current situation.[16] Common non-food reinforcers include toys, attention, and the opportunity to do something the dog wants.[17][18][19] For example, for a dog who wants to go for a walk, putting on the leash can reinforce sitting, going through the door can reinforce the dog who wants to go outside, and being greeted can reinforce a dog seeking attention.
  4. "You can't clicker train in noisy environments." The influence of environmental reinforcers is a challenge sometimes. Training for distractions is done by first training without distractions and then gradually adding complexity to the training environment.[20][21]
  5. "A dog may grow into adulthood and only listen and obey if the owner is carrying treats. If the owner does not have treats, often is the case that the dog is distracted and paying attention to whoever may have treats and food rewards available." This is actually a potential problem with the "Lure Reward" method of training where food is visible. In clicker training the food should not be visible to the animals until the behavior is completed. This could also happen when the trainer uses only one type of reinforcer. If the trainer uses only food, then the dog clearly learns that if food isn't present, then there can be no reinforcement. This is a trainer error. The solution is to use a variety of types of reinforcers and to hold training sessions where food isn't present. Also, you can include running to get the reinforcer into the reinforcement sequence.[22]
  6. "There are some situations where a clicker may not be loud enough, such as in hunting or retrieving when the dog is 'working away' from the handler." The clicker is not magic; it is just one type of marker. If the dog can't hear the click, use a different marker such as a whistle or a tone on a collar.[23] Deaf dogs are frequently trained with a flash of light or a hand signal.[24]
  7. "Some dogs are sensitive to noise and frightened by a clicker, so clicker training won't work for them." If your dog is afraid of the clicker, then simply choose a different marker—perhaps even just a word, the clicking of a retractible pen, or a juice cap.[25][26][27]

Methodology[edit]

The first step in clicker training is to teach the animal that the clicker sound means that they will get a primary reinforcer, usually food. To do this, some trainers "charge" or "load" the clicker. To do this the trainer clicks the clicker and immediately thereafter gives the animal a reward, usually a tasty treat, one small enough to be consumed almost instantly. Some animals tend to learn the association much more quickly than others. Progress may be tested by waiting until the dog's attention is elsewhere and then clicking. If the dog immediately looks toward the trainer as though expecting a reward, it is likely that the dog has made the association.

Other trainers, including Bob Bailey and the ABE Trainers, simply start training a behavior and following desired approximations with a click. ABE conducted experiments that demonstrated that for their purposes, where they may be training many animals at the same time, this method was more efficient.[28] Today many clicker trainers use this method of introducing the clicker.

After that, the trainer uses the clicker to mark desired behaviors as they occur. At the exact instant the animal performs the desired behavior, the trainer clicks and promptly delivers a food reward or other reinforcer. One key to clicker training is the trainer's timing; clicking slightly too early or too late rewards and therefore may reinforce whatever behavior is occurring at that instant. The saying goes, "you get what you click for."

Clicker trainers often use the process of "shaping," which means gradually transforming a specific behavior into the desired behavior by rewarding successive approximations to it. A successive approximation is "a behavioral term that refers to gradually molding or training an organism to perform a specific [completed] response by [first] reinforcing responses that are similar to the desired response."[29] Clicker trainers learn to split behavior instead of lumping it, i.e. to look for and reward small steps in the right direction rather than waiting for the whole, "perfect" behavior to appear on its own. It is important to create opportunities for the animal to earn rewards very frequently. A reinforcement rate of one click/treat (C/T) every two to three seconds is common among professional dog trainers. Criteria for receiving the click is tightened gradually, at the rate the animal is comfortable with and so that it will remain successful.

See also[edit]

References[edit]

  1. ^ Peterson, G. (2000). The Discovery of Shaping or B.F. Skinner’s Big Surprise. The Clicker Journal: The Magazine for Animal Trainers, 43, 6-13.
  2. ^ a b c Bailey's website historical notes
  3. ^ Burghardt, G.M. (1975). Behavioral research on common animals in small zoos. In Institute of Laboratory Animal Resources (Ed.), Research in zoos and aquariums (pp.103-133). Washington, DC: National Academy of Sciences.
  4. ^ Clicker training aids shelter adoption rates. (1996, August). The Don't Shoot the Dog News, 1(2), 2.
  5. ^ Mellen, J. D., & Ellis, 5. (1996). Animal learning and husbandry training. In D. G. Kleiman, M. E. Allen, K. V. Thompson, & S. Lumpkin (Eds.), Wild mammals in captivity: Principles and techniques (pp.88-99). Chicago: The University of Chicago Press.
  6. ^ Donaldson, Jean. "Culture Clash 2nd Edition". James and Kenneth Publishers Berkeley,CA. 2005 p.130
  7. ^ Skinner, B.F. (1951). How to teach animals. Scientific American, 185, 26-29.
  8. ^ Skinner, B.F. (1938). The behavior of organisms: An experimental analysis. New York: Appleton-Century-Crofts.
  9. ^ Bailey and Gillaspy, Operant Conditioning Goes to the Fair,The Behavior Analyst 2005, pp 143-159
  10. ^ Peterson, G. (2000). The Discovery of Shaping or B.F. Skinner’s Big Surprise. The Clicker Journal: The Magazine for Animal Trainers, 43, 6-13.,
  11. ^ PATIENT LIKE THE CHIPMUNKS-Version 2: The Story of Animal Behavior Enterprises.
  12. ^ The Don’t Shoot the Dog! News, (1996, November), p. 1.
  13. ^ Alexander, Melissa. Click for Joy. Waltham, MA: Sunshine Books, 2003. p. 97
  14. ^ Spector, Morgan. "Clicker Training for Obedience." Waltham, MA: Sunshine Books, 2005. p. 11, 38
  15. ^ Miller, Pat. "The Power of Positive Dog Training." New York: Hungry Minds, Inc., 2001. p. 24
  16. ^ Alexander, Melissa. Click for Joy. Waltham, MA: Sunshine Books, 2003. p. 198
  17. ^ Alexander, Melissa. Click for Joy. Waltham, MA: Sunshine Books, 2003. p. 12-13
  18. ^ Spector, Morgan. "Clicker Training for Obedience." Waltham, MA: Sunshine Books, 2005. p. 84
  19. ^ Miller, Pat. "The Power of Positive Dog Training." New York: Hungry Minds, Inc., 2001. p. 214
  20. ^ Alexander, Melissa. Click for Joy. Waltham, MA: Sunshine Books, 2003. p. 100-102
  21. ^ Barry, Jim (et al). Positive Gun Dogs. Waltham, MA: Sunshine Books, 2007. p. 43
  22. ^ Alexander, Melissa. Click for Joy. Waltham, MA: Sunshine Books, 2003. p. 37
  23. ^ Alexander, Melissa. Click for Joy. Waltham, MA: Sunshine Books, 2003. p. 12
  24. ^ Spector, Morgan. "Clicker Training for Obedience." Waltham, MA: Sunshine Books, 2005. p. 11
  25. ^ Alexander, Melissa. Click for Joy. Waltham, MA: Sunshine Books, 2003. p. 24
  26. ^ Spector, Morgan. "Clicker Training for Obedience." Waltham, MA: Sunshine Books, 2005. p. 12
  27. ^ Miller, Pat. "The Power of Positive Dog Training." New York: Hungry Minds, Inc., 2001. p. 56
  28. ^ private conversation with Kellie Snider, 2006
  29. ^ http://www.alleydog.com/glossary/definition.cfm?term=Successive%20Approximations