Parametric study of the acoustic change complex elicited by second formant change in synthetic vowel stimuli.

Item

Title
Parametric study of the acoustic change complex elicited by second formant change in synthetic vowel stimuli.
Identifier
AAI9924835
identifier
9924835
Creator
Ostroff, Jodi Maureen.
Contributor
Adviser: Arthur Boothroyd
Date
1999
Language
English
Publisher
City University of New York.
Subject
Health Sciences, Audiology | Biology, Neuroscience | Psychology, Cognitive | Psychology, Behavioral
Abstract
The Acoustic Change Complex (ACC) is an N1-P2 event-related potential elicited by acoustic change that occurs in an ongoing stimulus. As part of an effort to develop an electrophysiologic test of speech discrimination capacity, this dissertation examined the smallest second formant (F2) change within a synthetic vowel stimulus that elicited the ACC and compared those threshold values to behavioral responses. Stimuli were created by generating a synthetic 800 ms 3-formant vowel with 7 separate degrees of F2 change from 1050 Hz. The F2 change, occurring at 393 ms, was either 9, 19, 38, 75, 150, 300, or 600 Hz. A control condition with no spectral change was also included.;Electrophysiologic responses were recorded from 8 adult subjects at 5 scalp locations. The smallest F2 change that elicited an ACC in the group waveform. was: (a) 75 Hz (range 75 to 300 Hz) based on waveform morphology and amplitude criteria, and (b) 28 Hz (range 24 to 131 Hz) based on a curve fit and interpolation technique. Behavioral responses were recorded from 7 of the 8 adult subjects who demonstrated an ACC in the electrophysiologic experiment. Using the same curve fit and interpolation technique used in the electrophysiologic experiment, the binary choice, change/no change detection task revealed a mean just noticeable difference threshold of around 5 Hz (range 4.5 to 24.5 Hz). The mean detection threshold was around 20 Hz when taken from percent correct scores at 75% (range 6.6 to 39.3 Hz). The range of threshold values for the ACC from the group waveform correspond with mean detection probabilities between 86 and 97% correct in the behavioral task. These findings support the conclusion that the ACC can be elicited by F2 changes that are perceived with confidence. An ACC elicited by F2 changes between 30 and 75 Hz is likely sufficient for the detection of F2 changes between English vowels which are on the order of 100 Hz. These data are encouraging in terms of the clinical utility of the ACC as a measure of the potential for the development of speech discrimination capacity.
Type
dissertation
Source
PQT Legacy CUNY.xlsx
degree
Ph.D.
Item sets
CUNY Legacy ETDs