ABSTRACT
Spoken sentences are strings of segments that vary not only in spectral quality but in pitch, amplitude, and relative length. To understand what a speaker has communicated, a listener must identify the segments composing the utterance, and the order in which they occur; but a good deal of information is also encoded in the pitch, amplitude, and timing variations, i.e., the prosody of the utterance, and listeners make good use of it. Prosodic processing can be so efficient that listeners will attend to prosodic continuity at the expense of semantic continuity (Darwin, 1975). This paper addresses the question of the extent to which listeners make use of available prosodic cues, and the types of information they extract from them. All of the experimental studies cited in the three main sections below deal with the processing of English only.
