Research question:  A major point of contention about human reasoning is whether or not individuals search for counterexamples to conclusions.  According to theorists who argue that the mind is equipped with tacit rules of inference, the decision that an argument is invalid depends on a failure to find a derivation leading from the premises to the conclusion (see e.g., Rips, 1994).  However, with this procedure, one can never be cer- tain that the space of possible derivations has been searched exhaustively.  Alternatively, reasoners may base their inferences on mental models (Johnson-Laird and Byrne, 1991).  This theoretical account rests on the semantic principle of validity: a conclusion is valid if and only if it allows for no counterexamples, i.e., possibilities in which the premises are true but the conclusion is false.  Hence, by constructing a counterexample, reasoners are able to know that an inference is invalid. Method:  To search for a search for counterexamples, we carried out an experiment in which the participants had to evaluate eight inferences based on non-standard quantifiers.  Quite common in everyday life, such inferences call for a higher-order predicate calculus, which is incomplete. Results:  A striking phenomenon was the participants’ spontaneous use of a variety of strategies.  In 80% of all trials, they constructed a single specific instance of the premises.  Typically, the participants constructed a diagram in which they tried to minimize the overlap between the given sets.  Indeed, every single participant came up with at least one counterexample.  Our study has shown (…) that logically naive individuals do spontaneously search for counterexamples for at least one sort of deduction.

Keywords:  Logic, thinking and reasoning, syllogisms with non-standard quantifiers, mental model theory, counterexamples.

Related:   Suppression effects | Diploma thesis

Resources:  Download PDF | Google Scholar