Metaphor comprehension as attributive categorization.

The class-inclusion model claims that metaphors (e.g., That exam is a filter) are comprehended by inclusion of the topic (or subject) as a member of an attributive category named after and exemplified by the vehicle (or predicate). In three experiments, participants rated the extent to which a topic concept (e.g., exam) was a member of a vehicle category (e.g., filters). To test the scope of the class-inclusion model, both conventional and novel metaphors were included in Experiments 1 and 2, and both high and low apt metaphors were included in Experiment 3. Class inclusion was higher following metaphorical primes than following either literal primes (Experiment 1) or no primes (Experiments 2 and 3). Moreover, this metaphor-induced categorization occurred equally for conventional and novel metaphors, but occurred to a greater extent for the high apt than for the low apt metaphors. Thus, aptness rather than conventionality mediated categorization in metaphor comprehension.