ABSTRACT

Johnson-Laird, Legrenzi, Girotto, Legrenzi, and Caverni (1999) formulate a mental model theory of what they call extensional reasoning about naive probability. They say that they “take extensional to mean inferring the probability of an event from the different possible ways in which it could occur” (p. 63). The different ways in which an event could occur are the logical possibilities, or models, of the theory. It spite of the reference Johnson-Laird et al. make to the occurrence of events in their statement of what they are doing, probability in their theory is a logical concept. To reason about this concept is to engage in logical thought, about logical possibilities, and not to make inferences, based on actual observations, about physical events that have some relative frequency in the real world. As they themselves say: “This way of reasoning aims to be deductive” (p. 63). “Events” for them are not physical occurrences in the actual world, but rather logically possible states of affairs that satisfy basic propositions or their negations. Mental models are mental representations of these logically possible states of affairs and are used by people to reason deductively, or at least try to do so in a bounded system, about these logical possibilities.