ALBUQUERQUE, Luiz Carlos de; PARACAMPO, Carla Cristina Paiva; MATSUO, Gilsany Leão    MESCOUTO, Wandria de Andrade. Variáveis combinadas, comportamento governado por regras e comportamento modelado por contingências. []. , 21, 3, pp.273-283. ISSN 0188-8145.

^lpt^aInvestigando os efeitos de variáveis combinadas sobre o comportamento humano, 30 universitários foram expostos a um procedimento de emparelhamento com o modelo; a tarefa era apontar cada um dos três estímulos de comparação em sequência. A Sessão 1 era de linha de base. As contingências na Sessão 2, iniciada com a regra correspondente, eram alteradas na Sessão 3, e as contingências na Sessão 3, eram mantidas inalteradas na Sessão 4, iniciada com a regra discrepante. As condições diferiam quanto à complexidade da tarefa, à história de reforço para o seguimento da regra correspondente; e, à apresentação, ou não, de pergunta acerca das contingências. Na Sessão 3, a tarefa simples, a história de reforço intermitente e a não apresentação de pergunta, tenderam a favorecer o seguimento de regra. Já a tarefa complexa, a história de reforço contínuo e a apresentação de pergunta, tenderam a favorecer o abandono do seguimento de regra. Na Sessão 4, a manutenção de seguimento de regra dependeu da sensibilidade do comportamento alternativo ao especificado pela regra às contingências da Sessão 3. Propõe-se o conceito de variáveis combinadas. Os resultados também têm implicações práticas e os conceitos de entendimento e adesão ao tratamento são revistos.^len^aTo assess the effects of combined variables on following or not following rules, 30 college students exposed to a matching to sample task had to point to each of three stimuli in a given sequence. Session 1 was baseline. Session 2 begun with a rule corresponding to contingencies. Contingencies effective in Session 2 were changed without signaling in Session 3. Contingencies in Session 3, were effective through Session 4, beginning with discrepant rule.Participants were allocated to 6 conditions differing in respect to 1) task complexity - the simple sequence had 3 responses, the complex task had 6 responses -, 2) history of reinforcement- continuous or intermittent (FR 2) -for following the rule in Session 2, and 3) the presentation or not of question asking which behavior produced reinforcement. The results were compared with previous studies results. Together, these results showed that for the simple task, the history of intermitent reinforcement for rule following and no questions asked favored rule following after contingency change, while for the complex task the history of continuous reinforcement of rule-following and question asked favored the abandon of rule-following after contingency change. Data showed also that discrepant rule following 1) tend to be maintained when before presentation of the rule the alternative behavior to the specified by the rule shows to be insensitive to the contingencies - e. g. is under control of the rule, not of imediate contingencies; and 2) tend to stop occurring when before presentation of discrepant rule this alternative behavior shows sensitivity to the contingencies -e.g. is under control of the immediate consequences, not under control of rule. In summary, the combined variables effective in each of Sessions 2 and 3 determined the behavior in Session 3. Also, the combined variables effective in Sessions 2, 3, and 4 determined the behavior in Session 4. The concept of combined variables is advanced to describe the control by rules and by contingencies. The concepts of understanding is revised. The results have also practical implications and the concept of commitment to treatment is revised.

: .

        · | |     · |     · ( pdf )