Independence training vs modeling procedures for teaching phone conversation skills to the mentally retarded

Forty-five mildly mentally retarded adults were assessed for deficits in telephone conversational skills. They were ranked on the basis of their pretest scores and matched into triplets on skill level. One member from each of these groups was then randomly assigned to either a no-treatment control group, a modeling condition or independence training—a treatment that incorporated social reinforcement, instructions, performance feedback, modeling, shaping by successive approximations and self-monitoring and evaluation. For the two latter conditions, treatment was provided three times weekly for 2 months in group sessions (N = 5) of 1 hr duration. The dependent variable, conversational skills of subjects was assessed at pretest, posttest and at a 3-month follow-up based on verbal responses that subjects made to nine questions asked by an experimenter and related social responses during a phone call. Independence training proved to be significantly more effective than the other two experimental conditions at posttest and follow-up. In addition, modeling proved to be significantly more effective than the no-treatment controls at both of these assessment points.