1. An Ethical Concern in Behavior Therapy

Behavior therapists are becoming increasingly concerned with the ethico-legal issues involved in applying learning-theory principles to the modification of deviant behavior.  One of the authors who has been instrumental in sensitizing others to these concerns is Perry London, who discussed the morals of psychotherapy (1964) and the issues involved in behavior control (1969).  Examples of more recent efforts include: the development and legal regulation of coercive behavior modification techniques with offenders (Schwitzgebel, 1971); consent, selection of strategy, contractual problems, accountability, and supervision and control (Martin, 1975); certification of procedures (Risley, 1975); behavioral control (Begelman, 1973); civil liberties (Davison & Stuart, 1975); malpractice liability (McConnell, 1974); treatment goals (Davison, 1976); and ethical issues on research and intervention with behavior therapy (Stuart, 1975a, 1975b; Golfried & Davison, 1976; Stolz, 1975).

An ethico-legal issue that is yet to be widely explored is that which is inherent in the typical pretest-posttest design, O 1 x O 2 (Campbell & Stanley, 1963 / 1966), or in any of its variants when used in clinical intervention with behavior therapy.  The issue referred to is, namely, the withholding of the more effective, known techniques and procedures in an effort to obtain the baseline incidence of target behaviors by the systematic observation and recording of maladaptive behaviors.

In measuring behavior for a baseline, Rimm and Masters (1974) indicate that sources of information such as the client’s self-report, or the report by significant others to the client, are at best incomplete estimates of the frequency or intensity of the behavior in question.  This situation is due to a number of possible reasons, among which are the following: the finding that people are not objective observers of their own behaviors (Robbins, 1963; Mischel, 1968); the inadequate preparation of the client as an observer, a task which requires extensive training (Patterson & Cobb, 1971); or the likely possibility that, in observing one’s own behavior, the behavior under observation may be altered in some manner (Lindsley, 1966, Note 1; Peine, 1970).  Rimm and Masters (1974) thus conclude that using trained observers is the most effective technique of behavior measurement, but this creates economic limitations which deter the use of the technique in certain settings (e.g., private practice).

The observation and recording of behaviors, as used when obtaining the baseline frequency of behaviors, modify the observed behavior.  This effect on behavior was studied by O. R. Lindsley (1966), who found, in training parents of children who exhibited deviant behaviors, that the counting of behaviors served to significantly decrease 20% of the target behaviors.  This phenomenon was similarly demonstrated in a laboratory setting by Peine (1970), who found that parental observation and recording of the child’s behavior produced short-term changes in the child’s deviant behavior.

Concerning the effectiveness of measuring behavior as a therapeutic technique, it should be noted that it yields only very limited results, as evidenced by the reports of Lindsley (1966) and Peine (1970) previously cited.  In any event, the baseline measure of the behaviors is typically carried out in order to assess the treatment procedure, and not as a treatment procedure by itself (with the noted exception of some investigators, among them Lindsley and his collaborators at the University of Kansas).

In view of the evidence presented above, the systematic observation and recording of behaviors, as an assessment technique of clinical intervention, is the technique of choice in that it yields precise baseline measurements, but is also instrumental in delaying the onset of effective clinical intervention programs.  This situation is largely due to both the conditions of stability and sensitivity, as well as to the need to control extraneous processes (Sidman, 1960) which are required for the adequacy of the baseline measure.

To better understand the extent of the delay involved in establishing a formal baseline of behavior incidence, it is only necessary to examine some of the current literature on behavioral family therapy.  For example, Patterson (1971) reports that, in his University of Oregon laboratories, observation data are collected for a period of two weeks in order to establish a baseline describing family interaction.  In another study on behavioral family therapy in which the data provided was based on 10 hours of baseline observation, Patterson, Ray, and Shaw (1968) studied seven families with deviant children.  Mealiea (1976), in his study on conjoint-behavior therapy, used baselines of behavior based on data collected in a period of 1 to 2 weeks prior to the onset of treatment.

What follows is a formal statement of the problem under study, as well as a review of the relevant literature and a description of interaction therapy.  A hypothetical case in interaction therapy is also included to facilitate the description of the therapy process under study.  In conclusion this chapter presents the research design used in this investigation, and a formal statement of the hypotheses to be tested.

Statement of the Problem

The purpose of this investigation is to study the feasibility of implementing a behavior therapy procedure without first obtaining a formal baseline of the incidence of the target behavior(s).  The behavior procedure presently studied, which is interaction therapy, must also meet specified criteria for modifying the target behaviors within a brief period of time (efficiency), and for yielding meaningful empirical data that can be communicated to other investigators.


© 1976 Angel Enrique Pacheco, Ph.D., C.Psych.  All Rights Reserved.




 We are here to help you…

 Learn to Live Better ®