Printer-friendly version


Browse the glossary using this index

A | B | C | D | E | F | G | H | I | J | K | L | M | N | O
P | Q | R | S | T | U | V | W | X | Y | Z |

Page:  1  2  (Next)

H

:
A reduction, over repeated presentations, in the respondent behavior elicited by a stimulus. Cf. ADAPTATION, POTENTIATION.
:
See LEARNED HELPLESSNESS.
:
A heterogeneous chain requires different responses for each link of the schedule. Dog trainers make use of heterogeneous chains when they teach complex behavioral sequences to their animals. In going for a walk, a seeing-eye dog stops at intersections, moves forward when the traffic is clear, pauses at a curb, avoids potholes, and finds the way home. Each of these different responses is occasioned by specific stimuli and results in conditioned reinforcement. See also chain schedule.
:
a ranking of response classes on the basis of their relative probabilities. A more probable class is said to be higher in the response hierarchy.
:
A class that includes within it other classes that can themselves function as operant classes, as when generalized imitation includes all the component imitations that could be separately reinforced. Higher-order classes may be a source of novel behavior (e.g., as in the generalized indication of behavior that the imitator had not seen before). They also have the property that contingencies may operate differently on the higher-order class than on the classes that are its components. For example, if all instances are reinforced except imitations within one component class (e.g., jumping whenever the model jumps), that class may change with the higher order class rather than with the contingencies arranged for it (i.e., imitations of jumping may not extinguish, even though no longer reinforced)- Control by the contingencies arranged for the higher-order class rather than for component classes defines these classes; the component classes are sometimes said to be insensitive to the contingencies arranged for them. A higher-order class is sometimes called a generalized class, in that contingencies arranged for some component classes within it generalize to all the others. Generalized matching and rule governed behavior are examples of higher order classes.
:
Respondent conditioning in which the stimulus that functions as the US in producing one conditioned reflex is itself the CS of another.
:
Establishing a conditioned stimulus by pairing a neutral stimulus with an already established conditioned stimulus.
:
A graphic display in which each value is represented by a bar or column (usually vertical) whose height indicates the value. Also called a bar graph.
:
VI schedules used intervals selected in irregular order from a set of intervals, often described by a mathematical progression (e.g., arithmetic, geometric). Current practice favors schedules with a constant reinforcement probability over time within the interval (with probability measured by reinforcers per opportunity, i.e., probability that a response will be reinforced at a given time in an interval, given that the organism has reached that time). Such conditions are met by a type of VI schedule called random interval (RI), which arranges a setup (makes the next response eligible to produce a reinforcer) with a fixed probability every t seconds. In RI schedules, the average interval equals t divided by the probability (e.g., arranging a setup once per second with a probability of 0.02 produces RI 50-s). In one version, the schedule stops operating after a setup until the scheduled reinforcer is produced, so that low response rates make the obtained reinforcement rate lower than what had been scheduled; in another, the schedule continues to operate and successive setups accumulate, so that obtained and scheduled reinforcement rates remain about equal even with low response rates.
:
One kind of threat to internal validity is called history. History refers to conditions that are changed at the same time as the manipulation of the independent variable. See also internal validity.

Page:  1  2  (Next)