Academic journal article Education & Treatment of Children

Toward a Socially Valid Understanding of Problem Behavior

Academic journal article Education & Treatment of Children

Toward a Socially Valid Understanding of Problem Behavior

Article excerpt


As the field of behavior analysis improves its understanding of the environmental and biological causes of problem behavior, the social importance of those findings needs to be established. Currently, research tends toward brief analyses of behavior in highly artificial settings. Although necessary for identifying the reinforcers, stimulus controls and establishing operations associated with problem behavior, much current research is of questionable social validity. A concern is that a technology of behavior analysis is being developed that may be of limited applied utility. The goal of this paper is to present a set of criteria that researchers may want to consider when seeking to establish a socially valid understanding of problem behavior.

Descriptors: problem behavior, applied behavior analysis, social validity


With few exceptions the application of behavioral techniques has been done to humanely improve people's quality of life (cf. Sidman, 1989). Often the focus of these efforts has been on behavioral excesses that lead to self-injury, property damage, or aggression toward others. Behavioral techniques now exist that allow people with behavior problems to live in typical community settings and contribute to those communities in ways that thirty years ago would not have been imaginable (e.g., Homer et al., 1996). The efforts that led to these advances have been referred to as "behavior modification," "applied behavior analysis," "functional assessment," and "positive behavior supports," among other terms.

Such techniques are grounded in the experimental analysis of behavior to the extent that what is known about punishers, rewards, motivation, choice, and other behavioral processes derived from basic research are being successfully applied to behavior problems. Indeed, the experimental analysis of problem behavior has become routine among behavior analysts. That laboratory findings have such real world application is a noteworthy accomplishment for behavior analysis, and is probably the basis for the discipline's growing influence in human services professions. However, those successes have beget a number of questions that remain unanswered.

One result of this research is that we know that many of the reinforcers associated with problem behavior are mediated by others in a person's environment (e.g., Derby et al., 1992; Iwata et al., 1994). For example, it is frequently demonstrated through an experimental analysis of problem behavior that responding often functions to occasion access to preferred activities or events (e.g., food or a computer game). Or, variously, that the response is negatively reinforced by the removal or avoidance of social interactions (e.g., terminating instructional sessions). These are clearly cases in which social interaction is necessary for behavior to be reinforced. And, if the work of Derby et al. (1992) and Iwata et al. (1994) continues to be replicated, it appears that approximately 70% of cases of problem behavior serve a social function. Although behavior analysts are becoming increasingly adept at identifying the causes of problem behavior in clinical settings, it is instructive to remember that those behaviors occur in a larger social context.

However, the current state-of-the-science in behavior analysis focuses on experimental analyses of events occurring within a few seconds of problem behavior in artificial environments over very brief periods of time (e.g., 5 to 15 mm). These experimental analyses have produced important insights into why problem behavior occurs; unfortunately, the analyses bare little resemblance to environments in which people live and behave. In this paper such analyses are referred to as "clinical" research, as opposed to research conducted in "natural" environments (e.g., home, school, community and work settings). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.