Ut wisi enim ad minim veniam

II. Theory and Terminology

Terms associated with training seem to be a source of confusion, frustration, and in some cases, incentive for great debate. Ideally, terms can be used to provide a common language, facilitate good communication, and enhance our ability to train animals. We suggest that your primary focus should be on the training you are doing and the response of the animals to the training. With regard to terminology, we will take a very "minimalist" approach here in hopes that the fewer the terms, the smaller the opportunity for confusion.

As described above, animal training has its roots in experimental and comparative psychology. Within this psychology literature are descriptions of how animals learn: through habituation, classical conditioning, operant/instrumental conditioning, and complex learning. The literature also provides insights into schedules of reinforcement, the roles of positive reinforcement, negative reinforcement, and punishment in learning and training. We think it is important here to remind the reader that in the laboratory, these concepts seem very clear cut and unambiguous. However, in a more "real world" situation, when watching an animal learn how to shift into and out of a barn, it is much more difficult to ascertain which types of learning are involved or whether positive reinforcement, negative reinforcement or punishment played a role in the learning of that behavior. In all likelihood, an animal is learning in a multitude of ways and receiving a combination of reinforcement types. Many new trainers seem to get bogged down by the terminology. We suggest trainers focus on understanding the broad concepts, the most important of which is that training is a process where animals are making associations. Your job as a trainer is to facilitate the animal making those associations. Many new trainers tend to focus on too many things at once (see Pryor, 1984).

To add to the confusion, animal trainers who work in a much more complex environment than the laboratory have developed an additional set of terms that are used to describe nuances of training. This jargon includes "bridge," "cues," and "desensitization." Ramirez (1999) provides a good explanation of these training terms. A list of the basic terminology associated with learning and training is given in Table 2. We suggest here that understanding the concepts of training are more important than the specifics of each and every term and definition.

Terminology
Table 2. Definitions of terms associated with learning and training. These terms were taken from or are based on: Blasko et al., 1996; Dewsbury, 1978; Kazdin, 1994; Mellen & Ellis, 1996; Pryor, 1984, 1995; Ramirez, 1999; Reynolds, 1997; Wilkes, 1994.

Reinforcement	Anything that occurs with a behavior that tends to increase the likelihood that the behavior will occur again. The presentation of something that the animal finds "pleasant" [called a positive reinforcer] increases the likelihood of a particular behavior’s occurrence; the removal of something the animal finds "unpleasant" [called a negative reinforcer] increases the likelihood of a particular behavior’s occurrence.
Primary Reinforcer	Any stimulus whose reinforcing effect is immediate (e.g., food, water, warmth) and not a function of previous experience.
Secondary Reinforcer	Any stimulus whose reinforcing effect is a function of its association with a primary reinforcer.
Punishment	An action whose purpose is to decrease or suppress a behavior. Punishment follows a behavior and decreases its frequency of occurrence.
Extinction	Method of eliminating a behavior by not reinforcing it any longer.
Bridge	Any stimulus that tells the animal that it just did something correct and reinforcement is on the way. Two examples are shown on the left.
Continuous Reinforcement	A schedule of reinforcement where the desired response is reinforced every time it occurs. Trainers typically use a continuous reinforcement schedule when the animal is in the process of learning a new behavior.
Intermittent Reinforcement	Any schedule of reinforcement which does not provide reinforcement following each desired response. Behaviors which are reinforced on an intermittent schedule are more resistant to extinction.
Discriminative Stimulus(cue)	A conditioned stimulus which, when presented, results in the occurrence of a specific response. Some audio examples shown on the left.
Target (noun)	Any object which identifies a location for an animal (e.g., a trainer teaches a rhino to touch its lip to the end of a stick). When using this technique to shape a behavior, it is called "targeting." Examples shown on the left.
Shaping	Using selective reinforcement (approximations) to modify a general response to a specific response. It is building a behavior by dividing it into small increments or steps and then teaching one step at a time until the desired behavior is achieved. Steps become a series of intermediate goals.

back home next