Spoken language understanding requires the organization of a complex auditory pattern into a memory representation that captures the meaning relations intended by the speaker. Current language comprehension models focus on syntactic structure, thematic relations and discourse structure as the determinants of the comprehender's mental representations during sentence processing. The proposed work tests the hypothesis that an alternative linguistic structure, prosodic structure, initially organizes spoken language in memory. Prosodic structure refers to the suprasegmental rhythmic and melodic patterns inherent in a spoken sentence. Although research has established the existence of organized prosodic patterns, and the sensitivity of listeners to metrical and intonational structure, prosodic information has yet to be fully incorporated into psycholinguistic processing models. The proposed experiments examine the influence of prosody on language comprehension, focussing specifically on the relative contributions of prosodic and syntactic information sources to auditory sentence processing. The experimental materials are sentences with temporary syntactic ambiguities, pronounced with different prosodic structures, including Cooperating (prosodic boundaries coincide with syntactic boundaries), Conflicting (prosodic boundaries mark misleading locations in syntactic structure) and Baseline (potential syntactic boundaries are unmarked by prosody). Listeners' response latencies will be measured in three experimental paradigms: 1) End-of-sentence comprehension time, 2) Cross- modal naming, and 3) Location-shift monitoring, a task developed though recent successful piloting in our laboratory. The specific goals of the project are: to determine whether the presence of cooperating prosodic and syntactic structures can facilitate comprehension, to establish the relative time-course of syntactic and prosodic contributions to sentence parsing through the use of a reliable on-line measure of momentary auditory sentence processing difficulty, and to begin to specify the mechanisms by which the sentence comprehension system exploits regularities in the mapping between prosodic and syntactic structures. The broader aim of the project is to develop a more complete model of the sentence comprehension process, one that includes a fundamental and well- specified role for prosodic structure.

Agency
National Institute of Health (NIH)
Institute
National Institute of Mental Health (NIMH)
Type
First Independent Research Support & Transition (FIRST) Awards (R29)
Project #
5R29MH051768-05
Application #
2858024
Study Section
Perception and Cognition Review Committee (PEC)
Program Officer
Kurtzman, Howard S
Project Start
1995-01-01
Project End
2000-12-31
Budget Start
1999-01-01
Budget End
2000-12-31
Support Year
5
Fiscal Year
1999
Total Cost
Indirect Cost
Name
University of Kansas Lawrence
Department
Psychology
Type
Schools of Arts and Sciences
DUNS #
072933393
City
Lawrence
State
KS
Country
United States
Zip Code
66045
Speer, S R; Clifton Jr, C (1998) Plausibility and argument structure in sentence comprehension. Mem Cognit 26:965-78
Speer, S R; Kjelgaard, M M; Dobroth, K M (1996) The influence of prosodic structure on the resolution of temporary syntactic closure ambiguities. J Psycholinguist Res 25:249-71