Comprehending spoken metaphoric reference: a real-time analysis

Speakers and writers often use metaphor to describe someone or something in a referential fashion (e.g., The creampuff didn't show up for the fight to refer to a cowardly boxer). Research has demonstrated that readers do not comprehend metaphoric reference as easily as they do literal reference (Gibbs, 1990; Onishi & Murphy, 1993). In two experiments, we used a naming version of the cross-modal lexical priming (CMLP) paradigm to monitor the time-course of comprehending spoken metaphoric reference. In Experiment 1, listeners responded to visual probe words of either a figurative or literal nature that were presented at offset or 1000 ms after a critical prime word. Significant facilitatory priming was observed at prime offset to probes consistent with the metaphorical interpretation of the figuratively referring description, yet no priming was found for either probe type at the downstream location. In Experiment 2, we partially replicated Experiment 1 results at prime offset and found no priming at a probe point placed 1000 ms upstream from prime onset. Taken together, the data from these two experiments indicate that listeners are able to comprehend metaphoric reference faster than literal reference. Moreover, the effect appears to be strongest at prime offset, suggesting that activation of the nonliteral interpretation is closely tied to the relationship between the figuratively referring description and the intended referent. Implications for theories of metaphor comprehension, as well as for research in spoken metaphor, are discussed.