Processing Intonationally Implicated Contrast versus Negation in American English

Certain English intonational contours facilitate a conversational implicature that a relevant alternative to the stated proposition does not hold true. We evaluated how frequently and how quickly naïve participants achieved such pragmatically enriched meanings when their attention had not already been drawn to a set of alternatives. Sentences with L+H* L-H% intonational contours, along with broad focus affirmative and negative counterparts, were tested in a pair of experiments. Experiment 1 revealed that most interpretations of the L+H* L-H% sentences evidenced the expected implicature, but a substantial number did not. Experiment 2 mapped the activation levels across time for the asserted state and a contradictory/implicated alternative for the same three sentence types, using a picture-naming paradigm. The results revealed that lexical negation produced a contrast in activation levels between the two alternatives at an earlier time point than the L+H* L-H% contour, and that the relative activation of the two states shifted over time for L+H* L-H% sentences, such that an intonationally implicated alternative was highly activated at a time point when the activation for the asserted meaning had declined. These results further our understanding of the pragmatic processes involved in the interpretation of negation and intonation.