Understanding the nature and the processes involved in how people make inferences is a critical question in cognitive psychology. Unfortunately, there is currently no consensus as to what is involved in the inferential process. Complicating the situation is the fact that there is no uniform definition of what an inferential task is; current competing models both suggest and use very different task paradigms. Interpreting the consequent results is thus difficult, since despite the identical labeling of tasks, there is no guarantee that participants necessarily deploy the same processes when important parameters of inferential tasks are varied.In the present article, we concentrate on conditional reasoning, which involves making inferences with a given major premise of the form "P implies Q" and one of four possible minor premises. Modus ponens (MP) is the logical principle that involves reasoning with the premises "P implies Q and P is true" and leads to the logically correct conclusion "Q is true." Modus tollens (MT) involves reasoning with the premises "P implies Q and Q is false" and leads to the logically correct conclusion "P is false." These two principles are valid logical forms, since they both lead to a single logically correct conclusion. Affirmation of the consequent (AC) involves reasoning with the premises "P implies Q and Q is true." Denial of the antecedent (DA) involves reasoning with the premises "P implies Q and P is false." In both cases, the implied conclusions-"P is true" for AC and "Q is false" for DA-are not logically correct. Neither of these forms leads to a single logically correct conclusion, and the correct response would be to deny the implied (biconditional) conclusion in both cases.Currently, there are two major theories that attempt to explain how people make conditional inferences and what kinds of processes they use to do so. Mental model theory (Johnson-Laird & Byrne, 1991 supposes that reasoners will generate a representation of the premises using symbolic tokens. Tokens represent classes of possibilities, and a conclusion will be accepted if there is no counterexample available in the representation. This theory is constructed specifically to explain reasoning on standard deductive tasks, which generally require a dichotomous response (i.e., a conclusion must be judged as certain or as uncertain). Probabilistic theories suppose that a key factor in making a conditional inference is the subjective conditional probability of the conclusion given the premises (Oaksford, Chater, & Larkin, 2000). Importantly, people are assumed to hold variable degrees of belief in conditional statements, which clearly has an impact on the strength of the inferences that they are prepared to make. These theories are constructed specifically to explain reasoning on a probabilistic inference task, in which a natural response is one that is on a scale from unbelievable to believable.Of course, the simplest way of reconciling these two theories would be to postulate the existence of two sep-
485© 2010 ...