Correspondência fazer-dizer : variáveis de controle e condições de manutenção e generalização

Abstract

The present proposal is based on the assumption that verbal responding is an operant behavior established and maintained by reinforcement contingencies. Study 1 evaluated the effects of three different correspondence training methods on the maintenance of eleven children s accurate self-report. Doing consisted of reading aloud a written word presented on a computer s screen. Saying consisted of reporting on the accuracy of the textual response after the computer s automated feedback. Following report accuracy evaluation, participants were assigned to different correspondence training methods. Five children were exposed to Training A (continuous reinforcement), three to Training B (continuous and intermittent reinforcement plus instructions), and the remaining were exposed to Training C (continuous and intermittent reinforcement only). After training criteria was met, participants were re-exposed to report accuracy evaluation sessions. Follow-up sessions were then conducted 30 and 60 days later. All training methods were effective in maintaining accurate self-reports for eight participants. Low accuracy during initial sessions (report accuracy evaluation) seemed to predict low maintenance over time. Study 2 investigated the role of the audience (computer, experimenter, and peer) on children s self-report accuracy. Participants were six children who exhibited high do-say correspondence during report accuracy evaluation sessions. The Audience Test evaluated the frequency of accurate reports as a function of different audiences. Only the correct and accurate reports were followed by points. The amount of points required to obtain the high preferred item, however, was higher than the amount received during the session. At the end of each session, participants reported how many points they obtained. Self-report accuracy was evaluated in three conditions: while engaging in the task, while reporting to the audiences, and while choosing the item. Five out of six participants self-report accuracy decreased in at least one of the three conditions as a function of the different audience s presence. After correspondence training was conducted, previous do-say correspondence levels were recovered. Audience seemed to be a relevant controlling variable on children s self-report accuracy since the presence of different audiences controlled different patterns of correspondence. Study 3 investigated the effects of errors on six children s self-report accuracy as a function of the nature of the task (academic and nonacademic), and verified generalization of a do-say correspondence training across tasks. Doing consisted of reading aloud written words, playing a computer game, announcing the result of mathematical operations, and naming different auditory or visual stimuli related to music. During baseline, the frequency of corresponding reports was evaluated as a function of the different tasks presented. Correspondence training was conducted, initially, only to the task in which participants exhibited the lowest levels of accuracy. After training criteria was met, generalized dosay correspondence of untrained tasks was then tested. Academic tasks produced the lowest levels of correspondence during baseline sessions. Four out of six children exhibited generalized say-do correspondence after the first correspondence training employed and the remaining, after a second training was conducted. Results indicated do-say correspondence as a generalized operant behavior.