Association for Behavior Analysis International

Aversive events are ubiquitous. Yet we know far less about aversive control than we do about positive reinforcement. Indeed, even when environments are deliberately arranged to contain only stimuli considered to be positively reinforcing, evidence suggests that such environments will inevitably contain stimuli that can function aversively. In this symposium, current research and theory regarding the subtle and pervasive effects of aversive control will be presented, along with the translational implications of such effects.

Much of what we know about variables affecting punishment (e.g., frequency, intensity, establishing operations, etc.) comes from research in which unconditioned punishers (e.g., brief electric shocks) were used. Given the frequent use and acceptance of other punishment procedures, such as timeout from positive reinforcement and response cost, across a wide range of situations and populations, it is somewhat surprising that the empirical basic research is lacking. Lerman and Vorndran (2002), lamenting this lack of research, called for more systematic and thorough investigations of punishment contingencies. In this presentation, recent data from studies in which parameters of timeout from positive reinforcement (e.g., duration) and other nonconventional "punishers" (e.g., stimuli associated with fixed-ratio schedules of positive reinforcement) will be presented.

The behavior analysis of self-control has emphasized procedures in which a subject chooses between a small immediate reinforcer and a larger delayed reinforcer. The value of the larger reinforce is discounted as a function of delay, tipping the balance in favor of the smaller reinforcer ("impulsive" choice rather than "self-control"). Experimental results have been orderly, amenable to quantitative analysis, and generalizable across species, and they have provided insight into a range of human ills (e.g., drug addiction, obesity). These benefits are preserved even when real outcomes (e.g., 1 food pellet now vs. 3 pellets after 10 s) are replaced with hypothetical ones (e.g., $100 now vs. $1,000 after a year). But experiments involving reinforcer magnitude and delay do not exhaust the factors implicated in a behavior analysis of self-control. Of particular interest are choices between outcomes that involve immediate reinforcers and, in some cases, delayed aversive events. As a prosaic example, consider a diner's choice between two meals: a bland one (small immediate reinforcer) and a rich one always followed later by heartburn (large immediate reinforcer plus a delayed aversive event). This presentation will explore ways in which laboratory models of self-control may be enhanced by incorporating aversive events.

Sidman's (1994, 2000) theoretical treatments of equivalence expanded original conceptualizations considerably in that all positive elements of a reinforcement contingency are viewed as potential class members, and equivalence is held to result from reinforcement contingencies involving any number of terms. This talk will describe data from our lab demonstrating that reinforcers can serve as the basis for equivalence classes generated by three-term contingencies. Typically developing children received simple discrimination training in which selections of S+ resulted in a class-specific reinforcer (e.g, selecting A1, B1, or C1 produced R1; selecting A2, B2, or C2 produced R2, etc.), while selecting either of two S- stimuli unique to each trial type produced a buzzer. Conditional discrimination probe trials revealed that all stimuli related to the same reinforcer were interchangeable. These procedures then raised the question- what of the S- stimuli, which were also related to a common consequence? This presentation will outline experimental approaches to the exploration of possible equivalence-class formation based on aversive consequences. Results of such studies stand to have important implications for issues ranging from inadvertently establishing stimulus classes with aversive functions (e.g., in classroom applications of discrimination training) to understanding clinical phenomena characterized by complex avoidance patterns.

Chronic, severe, disruptive and destructive behaviors such as self-injury, physical aggression, property destruction, and tantrums are a major problem in persons with Intellectual and other developmental disabilities such as autism. These "challenging" behaviors present barriers to habilitation and independent living, and they are a long-standing treatment challenge. Escape and avoidance behaviors make up the single largest function underlying problem behaviors in the clinical literature, suggesting that this population is particularly sensitive to aversive stimulation. Persons with IDD are also likely to emit such challenging behaviors during transitions from one activity to another. The behavioral mechanisms that make transitions aversive in this population are not known, but treatments generally assume that environmental unpredictability is aversive to people with IDD and transitions are aversive when the upcoming event is unpredictable. Research in this area is sparse and little is known about the behavioral processes that make transitions aversive or how to reduce their aversive properties. This paper will present data from translational research program validating a laboratory model for the study of behavioral and biological variables that may underlie transition-induced challenging behaviors in IDD. A distinguishing feature of this approach is that we have demonstrated equivalent behavioral processes across pigeons, rats, and people with developmental disabilities. This increases the potential that experimental treatments derived from testing in animals and controlled experimental studies will translate to the understanding and treatment of clinically important challenging behaviors.