This investigation evaluated the use of conditional discrimination (CD) instruction and multiple exemplar instruction (MEI) to establish derived relational responding in accordance with hierarchical frames with school aged children. The first experiment used a multiple probe design to evaluate the effectiveness of MEI to teach participants to correctly respond to BELONGS TO and INCLUDES relations between academically relevant stimuli in the target hierarchy. The protocol was presented via an automated computer program. Written and oral intraverbal pre and posttest were administered to determine if CD instruction and MEI were sufficient for academically relevant behaviors to emerge. Transformation of stimulus functions was assessed using a property inheritance task. A retrospective protocol analysis was used to evaluate the covert verbal behavior the participants were engaging in when responding to the CD across the hierarchy task. The second experiment also used a multiple probe across participants design to assess hierarchical responding. An ABABCB withdrawal design was used to assess the functional relation of covert verbal behavior and the CD across the hierarchy task. The target stimuli and procedures for Experiment 2 were identical to the first experiment with the exception of using a concurrent protocol analysis as opposed to the retrospective protocol analysis to assess the role of the participants' covert verbal behavior on task performance. That is, the second experiment used the silent dog method (Hayes, White, & Bissett, 1998) to assess if self-talk is functionally related to the transformation of stimulus function task.

Made available in DSpace on 2016-06-02T19:46:08Z (GMT). No. of bitstreams: 1 2588.pdf: 959710 bytes, checksum: fe057f9dee825e64b6a9897674590534 (MD5) Previous issue date: 2009-03-27 / Financiadora de Estudos e Projetos / This study aimed to develop the procedure for study and analysis of the covert event and to verify verbal behavior emergency under covert stimulus control. Two children and two adolescents with intellectual disabilities, students at a specialized institution, were submitted to matching to sample tasks. Preliminarily, some preferred edibles were selected for each participant and then conditioned reinforcing value of tokens was taught to make it possible to use them as teaching consequences. There was an array of experimental procedures involving two stimulus sets: set A, composed by dictated words and set B, by pictures. There was also set A composed by participants spoken words. Sets A and B were composed by four stimuli. The experimental design consisted of (1) teaching relations AB and BA , (2) teaching arbitrary relations B3B1 and B4B2 and testing symmetry, and (3) testing relations A4A 2 and A3A 1. When systematic incorrect responses were observed during relations B3B1 and B4B2 teaching, blocked trial procedure was applied. Other additional teaching procedures consisted of withdrawing the S-, teaching symmetry (B1B3 and B2B4) and providing verbal cues. However, even after applying these additional procedures, performance were above criterion, characterized by instability and deterioration. The literature emphasizes the difficulties in teaching conditional discriminations with visual stimuli for participants with intellectual disabilities. The results lead to a discussion of the intervenient variables, such as: i. Difficulties found by persons with intellectual disabilities in the matching to sample procedure; ii. What type of instruction should be given by the experimenter to help them perform the task and have access to consequences; iii. The computer as a resource to present the tasks; iv. The modality of stimuli and stimulus relations; v. Different performances presented by the participants during the exposure to the same experimental condition; vi. The consequences of reinforcement; vii. The criterion of performance established for each experimental condition and for additional procedure and; viii. Interruption in data collection. We conclude that although the participants did not reach performance criterion in relations B3B1 and B4B2 teaching, this study provides contributions of procedure design, and raises methodological questions that advance the understanding of covert stimulus tact acquisition by persons with intellectual disabilities. Through the discussion on these possible intervenient variables, we may infer that participants errors may be a matter of procedure instead of being related to their learning disabilities. / Este estudo objetivou desenvolver um procedimento para estudo e análise do evento privado e, por meio deste, verificar a emergência de comportamento verbal sob controle de estímulos encobertos. Duas crianças e dois adolescentes com atraso no desenvolvimento intelectual, alunos de uma instituição especializada, foram submetidas a tarefas de escolha de acordo com o modelo. Preliminarmente foram selecionados itens de preferência comestíveis, dos participantes, e ensinado valor reforçador condicionado de fichas para que servissem como conseqüências ao ensino. Realizou-se um arranjo de procedimentos experimentais que envolveram dois conjuntos de estímulos: o conjunto A, composto por palavras ditadas; o conjunto B por figuras; e o conjunto A de respostas, constituídas por palavras correspondentes aos demais conjuntos, faladas pelos participantes. Cada um dos dois primeiros conjuntos era composto por quatro estímulos. O delineamento experimental constituiu-se de: ensino das relações AB e BA , ensino e, em seguida, teste de simetria das relações arbitrárias B3B1 e B4B2, e, teste das relações A4A 2 e A3A 1. Diante das sistemáticas respostas incorretas apresentadas durante o ensino das relações B3B1 e B4B2, aplicou-se o procedimento adicional de ensino de discriminação condicional com tentativas em bloco e estratégias de ensino de retirada de S-, simetria dos blocos 1 e 2 e dicas verbais. Mesmo com estas estratégias, as respostas incorretas se mantiveram, caracterizadas por instabilidade e deterioração de desempenhos. A literatura enfatiza as dificuldades em ensinar discriminações condicionais com estímulos visuais para pessoas com atraso no desenvolvimento intelectual. Algumas possíveis variáveis intervenientes são discutidas, como: i. Dificuldade encontrada por pessoas com atraso no desenvolvimento intelectual ao serem expostas a tarefas de escolha de acordo com o modelo; ii. A dificuldade de compreensão das instruções para executar a tarefa e para ter acesso às conseqüências reforçadoras; iii. A utilização do computador como recurso para apresentar as tarefas; iv. O tipo de estímulo e a relação de arbitrariedade entre os estímulos modelo e comparação; v. Diferentes desempenhos apresentados pelos participantes frente a exposição de uma mesma condição experimental; vi. As conseqüências de reforçamento; vii. O critério de desempenho estabelecido em cada condição experimental e no procedimento adicional e; viii. Interrupção na coleta de dados. Conclui-se que apesar de não ter sido possível alcançar critério de desempenho no ensino das relações B3B1 e B4B2, o presente estudo fornece contribuições de procedimento, e levanta questionamentos metodológicos, que avançam na compreensão do ensino de comportamentos verbais sob controle de estímulos encobertos, por pessoas com atraso no desenvolvimento intelectual. Depreende-se daí que os erros apresentados pelos participantes podem ser muito mais uma questão de procedimento do que função das dificuldades de aprendizagem.