|
|
 E-mail this message to a friend
|
|
Title:
|
Prosody, Animacy and Syntax in the Perception of Speech: Immediate and subsequent analyses of main verb/reduced relative constructions
|
|
Author:
|
Christine Kenneally
|
|
Email:
|
click here to access email
|
|
Degree Awarded:
|
University of Cambridge
, Department of Linguistics
|
|
Degree Date:
|
1997
|
|
Linguistic Subfield(s):
|
Psycholinguistics
Syntax
|
|
Director(s):
|
Peter Matthews
Paul Warren
|
|
|
Abstract:
|
|
This thesis reviews the psycholinguistic literature which investigates written and spoken sentence processing, and it explores in four on-line experiments the interaction of variables such as prosody, animacy and lexical-syntactic information at a point of potential ambiguity in the main verb/reduced relative (MV/RR) construction.
It is found in the literature review that the arguments for modularity and syntactic autonomy, traditionally explained in terms of processing architecture, rest primarily on evidence found in the processing of the written signal. This evidence and the theory of modularity are considered inadequate in the light of recent work indicating the use of prosodic, semantic and verb specific information in processing. The probabilistic constraint-based model of MacDonald, Pearlmutter and Seidenberg (1994a, 1994b) is compared to the garden path model of Frazier and colleagues (Frazier, 1990). It is found that the constraint-based model is more able to accommodate semantic effects, verb-specific subcategorisation and collocation effects, as well as effects of prosody in spoken language processing. This model is also considered to be more attractive as a processing model because it describes processing in terms of probabilistic constraint-based interaction instead of rigid structural rules, and also because of its ability to explain individual differences in processing (and the accompanying potential the model has to explain differences between speaker populations in time). Although this model is deemed to be the better alternative, its current emphasis upon the lexical item means that there are many issues which must be addressed before it can claim to be a model for all language processes (i.e. spoken as well as written).
The four on-line experiments used the cross-modal naming paradigm. These experiments examined the use of prosody, animacy of the subject NP, and lexical-syntactic information (as encoded by the probe word and the structural environment of the fragment). It was found that the subjects immediate assignments of constituency to the fragments are explicable in terms of the interactions between prosodic structure and the lexical-syntactic information inferred from single lexical items and from relationships between larger constructions. Animacy is also shown to have a weak effect in on-line interpretation, but its primary influence is found in the off-line ratings assigned to probe words (to indicate how well they fit the preceding fragment).
|
|
|
|
|
|
Page Updated: 29-Nov-2009
Please report any bad links or misclassified data
LINGUIST Homepage | Read
LINGUIST | Contact us
While the LINGUIST List makes every effort to ensure the linguistic relevance of sites listed
on its pages, it cannot vouch for their contents.
|
|
