LINGUIST List 19.1929
Wed Jun 18 2008
Diss: Phonology/Psycholing: Shelton: 'An Experimental Approach to ...'
Editor for this issue: Evelyn Richter
<evelynlinguistlist.org>
1. Michael
Shelton,
An Experimental Approach to Syllable Weight and Stress in Spanish
Message 1: An Experimental Approach to Syllable Weight and Stress in Spanish
Date: 17-Jun-2008
From: Michael Shelton <msheltonoxy.edu>
Subject: An Experimental Approach to Syllable Weight and Stress in Spanish
E-mail this message to a friend
Institution: Pennsylvania State University
Program: Department of Spanish, Italian and Portuguese
Dissertation Status: Completed
Degree Date: 2007
Author: Michael Shelton
Dissertation Title: An Experimental Approach to Syllable Weight and Stress in Spanish
Linguistic Field(s):
Phonology
Psycholinguistics
Subject Language(s): Spanish (spa)
Dissertation Director:
Chip Gerfen
Judith F Kroll
Nicolás Gutiérrez Palma
Richard Page
John M Lipski
Dissertation Abstract:
This work examines the cognitive representation of phonotactic constraintson Spanish stress via the collection of behavioral data. Critical stimuliin four experiments consist of nonwords that violate the Spanish stresswindow. Experiment 1 finds statistical differences among latency andaccuracy data for stimuli that represent theoretically proscribedsequences, theoretically licit sequences that are unattested for diachronicreasons, and fully licit gaps. Experiments 2 and 3 find differentialpatterning of rising and falling diphthongs. Experiment 4 tests the timecourse of phonological encoding and evidences differential treatment ofboth diphthong types across delays in a delayed naming task. The data areinterpreted as evidence in favor of sensitivity to a continuum of weightacross the lexicon. The findings challenge traditional approaches tosyllable weight which call for a categorical, binary light/heavydistinction. Rather, the results accord with a stochastic or probabilisticconception of the lexicon in which speakers of a language are able to trackstatistical patterns of extant combinations of sounds in order to constructa phonological grammar that is gradient in nature. Implications for twoprominent psycholinguistic models of speech production (Levelt et al. 1999;Dell 1986, 1988) are also discussed. Specifically, the findings challengethe proponents of each model to account better for the interaction ofstress and syllable-level encoding, both as represented in the lexicon andacross time during processing.
|