Learning potential approaches to the mentally retarded



Journal Title

Journal ISSN

Volume Title



An analysis of current assessment approaches being employed with the Mentally Retarded revealed serious conceptual and pragmatic limitations. A new method of assessing learning abilities (Learning Potential Method), developed by Budoff, relies on sampling actual learning behavior as contrasted with the usual performance measures of ability. The potential usefulness of this approach for assessing the Mentally Retarded was analyzed. The present study extended the Learning Potential Method to include a measure of learning on a Paired-Associate task as well as on the usual perceptual-organization task (Kohs Blocks). In addition, measures of Verbal, Performance, and Full Scale IQ were obtained in order to evaluate the comparative effectiveness of the Learning Potential Method in predicting adjustment and competence in the vocational and residential environment. Finally, the study was the first to employ a representative sample of adult retarded in evaluating the Learning Potential assessment approach. The results indicated that the Learning Potential approach could not be extended easily to include a verbal task such as Paired-Associate learning but that it was possible to identify individual differences among the retarded in acquisition of a general strategy of problem solution on the Kohs task. However, such differences in acquisition of a problem solving strategy proved unrelated to differences in vocational competence as measured by type of employment (competitive-noncompetitive) and by salary received (% of minimum wage). It was also not positively related to adjustment in the residential environment as measured by ratings on the AA1ID Adaptive Behavior Scale. The IQ measures were found once again to be ineffective indicators of differences in competence and adjustment among the Mentally Retarded. The need for competency based and criterion validated measures of adaptive functioning among the retarded was discussed.