THE (NON) DEFINITION OF AVERSIVE CONTROL IN BEHAVIOR ANALYSIS

The Behavior Analysis literature describes two types of control: aversive and positive. A review of specialized publications revealed an objective definition of control through positive reinforcement, but no definition of aversive control. In this paper, we discuss the scientific meaning of the terms “control” and “aversive” from the viewpoint of behavior analysis. We focus on the relational probabilities between responses and stimuli that occur during the continuous interaction between organisms and the environment. The term “control” is assumed to mean that one event (the controlled one) is changed by the occurrence of another event (the controller). The aversiveness of the control is analyzed as a function of “operations” (addition and subtraction of the stimulus), their “effects” (the increase or decrease of the response probability) and the “nature of the stimulus” involved (aversive or appetitive). We conclude that an analysis of processes, operations, and the nature of the stimulus was unable to identify a factor common to all the behavioral relations defined as aversive. We consider that without clear criteria for classifying a control as aversive, it would be more parsimonious to talk about behavioral control without using the aversive/positive dichotomy. However, if this dichotomy is maintained, the development of an objective analysis of elicited (emotional) responses may offer a way to characterize the aversive/positive distinction. Key words : control; aversive control; positive control; behavior analysis; conceptual questions.