ABSTRACT
125This chapter completes the aversive control contingencies in the adaptive learning schema. These consist of removing or preventing appetitive events (response cost and time out) as well as presenting, removing, or preventing aversive events (punishment, escape, and avoidance). The effects of deprivation, timing, intensity, and contingency on punishment, escape, and avoidance are described.
Response cost and time out procedures have been found effective in suppressing behavior with a wide range of individuals. Although presentation of an aversive stimulus (positive punishment) can be immediately effective, it is difficult to implement under free-living conditions. It is usually impossible to deliver an aversive stimulus immediately, at sufficient intensity, in a consistent manner. There are also problematic side effects of punishment: the punishing agent may become aversive; deceptive avoidance behaviors may be acquired; generalized suppression beyond the punished response may occur; aggression may be elicited; conditioned masochism can result from following punishment with reinforcement; and, negative modeling may occur whereby the individual imitates the very behavior one would like to suppress. Punishment is most effective when a desirable alternative behavior is followed by an appetitive stimulus (positive reinforcement).
Schedules in which reinforcement is contingent upon a low rate of responding (DRL), the non-occurrence of a behavior (DRO), the occurrence of an alternative (DRA) or incompatible (DRI) response are effective alternatives to punishment to reduce behavioral excesses. Functional analysis enables determination of the reinforcer maintaining a behavior. This permits the implementation of extinction to reduce or eliminate the behavior. Non-contingent delivery of appetitive stimuli has also proved effective in reducing problematic behavior.
At first glance, avoidance appears to violate the Law of Temporal Contiguity in that a successful response results in the absence of an event. Two-factory theory emerged as an attempt to explain avoidance by essentially converting it into escape, either from fear (Mowrer) or the warning stimulus (Schoenfeld). Subsequent research led to the conclusion that avoidance learning does not violate the Law of Temporal Contiguity. Immediate feedback is necessary for avoidance learning to occur. A cognitive adaptive learning model describes avoidance as involving two expectancies: an aversive event will not occur if one makes a specific behavior; otherwise, the aversive event will occur. Such expectancies can explain the difficulty in eliminating avoidance responding.
