THE SOCIAL VALIDITY OF AN ARGUMENT SUPPORTING A BAN ON AVERSIVE PROCEDURES

There is currently a controversy regarding the use of non-aversive and aversive procedures in the treatment of severe behaviour problems in individuals with mental retardation. One specific criticism directed a t professionals who support the non-aversive position is that they hav e taken quotations from scientific articles out of context in order to give the impression of empirical support for their position. The pres ent authors addressed this issue by assessing the social validity of r eferences made to selected scientific articles in support of a specifi c argument in a monograph by Guess et al. (1987). Three groups of indi viduals were surveyed in this study: (1) psychologists working with in dividuals with mental retardation at six state schools in Texas (Group I); (2) selected professionals with expertise in the area of self-inj urious behaviour and in the use of behaviour modification techniques w ith individuals who are mentally retarded (Group II); and (3) the auth ors of the six journal articles that were selected for the present stu dy (Group III). The study focused on references made to journal articl es in support of the 'depersonalization hypothesis' that aversive ther apies serve to depersonalize the recipients of these treatments. Resul ts indicate that the authors of the monograph were not completely accu rate in their references to the six articles. Sixty per cent of the re spondents from Group I, 71 % of respondents from Group II, and two out of the three respondents from Group III rated the citations made in t he monograph as inaccurate. In addition, the respondents rated the art icles as not supportive of the depersonalization hypothesis for which they were cited. Over half of Group I respondents, and all respondents from Groups II and III rated the articles as non-supportive of the de personalization hypothesis. These findings suggest that the authors of the monograph have used references selectively and incorrectly in sup port of their views in at least some instances. This emphasizes the im portance of critical reading and hypotheses based on bodies of knowled ge rather than selected sources. Furthermore, this study indicates the need for operationally defined hypotheses which may be examined empir ically.