QUELHAS, Ana Cristina    JOHNSON-LAIRD, P. N.. Conhecimentos, modelos, e raciocínio condicional. []. , 22, 2, pp.309-317. ISSN 0870-8231.

Meaning, reference, and general knowledge can all modulate the interpretation of assertions, and in particular the interpretation of conditionals. The present study examines the pragmatic effects of knowledge on reasoning from pairs of conditional premises. According to the theory of mental models, inferences of the same form but different content should yield different patterns of inference. Consider the following premises, e.g.: If Maria is not in Paris, then she's in France. If Maria is in France, then she's a student. The first premise exploits a spatial inclusion (Paris is in France), and so in one possibility Maria is not in Paris but is in France, and in another possibility Maria is in Paris and so she is also in France. Hence, it is impossible for Maria not to be in France, and so individuals should tend to infer from the second premise that she is a student. In contrast, consider the following premises that have the same form as those in the previous example, but a different content: If João is not in Roma, then he's in France. If João is in France, then he's a student. The first premise exploits a spatial exclusion (Roma is not in France), and so in one possibility João is not in Roma and is in France, and in another possibility he is Roma and so not in France. Hence, individuals have no grounds for the categorical inference that he is a student, and they should tend to infer the conditional conclusion, i.e., that If João is not in Roma, then he's a student. The inclusion problems yield fewer possibilities than the exclusion problems, and so the model theory predicts that the inclusion problems should be easier than the exclusion problems. The paper reports two experiments that corroborated the predictions of the model theory. In Experiment 1, participants drew more categorical conclusion in the inclusion premises, but only a few drew the conditional conclusion from the exclusion premises. In fact, with de exclusion premises, we obtain many "other conclusions"- . To avoid this we have made a second experiment, were participants chose a conclusion from four possibilities: the categorical conclusion, a conditional conclusion, both conclusions, or neither, instead of what happens in Experiment 1, were participants drew the conclusions for themselves from the premises. Experiment 2 replicated the superiority of categorical conclusion in the inclusion premises, and found the superiority of conditional conclusion in the exclusion premises.

: Conditional reasoning; knowledge; mental models.

        · |     · |     · ( pdf )