Comodulation masking release in frozen and randomly selected noise.
Item
-
Title
-
Comodulation masking release in frozen and randomly selected noise.
-
Identifier
-
AAI9530900
-
identifier
-
9530900
-
Creator
-
Mackersie, Carol Lee.
-
Contributor
-
Adviser: Harry Levitt
-
Date
-
1995
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Psychology, Experimental | Health Sciences, Audiology
-
Abstract
-
This dissertation examined the effects of combining comodulation masking release (CMR) and frozen-noise masking release (FNMR). It was hypothesized that CMR and FNMR result from essentially the same cues (i.e., envelope reproducibility). This was tested by comparing CMR in frozen and randomly-selected noise. It was expected that if the same cues underlie both phenomena, CMR would not be present in frozen noise.;Subjects were tested using two noise samples (samples A and B) under four masking conditions: single-band frozen- and randomly-selected noise, and seven-band frozen- and randomly-selected noise. During the frozen-noise conditions, the same noise sample was used on each presentation. During the randomly-selected conditions, the tone was masked by either noise A or B and the "noise-alone" sample was randomly chosen from the remaining seven alternatives.;The hypothesis was supported by results of two of the five subjects who did not show CMR in frozen noise. Two of the five subjects showed smaller CMR in frozen than randomly-selected noise. One subject did not show FNMR, suggesting that she was unable to use cues available in frozen noise.;Thresholds were lower for noise sample A for all subjects. CMR and FNMR was absent for noise sample A for the majority of subjects. This was attributed to the low thresholds observed in the reference condition. Differences in the short-term RMS and envelope power spectra of the noise samples provided possible explanations for these results.;In Experiment 2, the single-band stimuli used in Experiment 1 were transposed to 6-kHz. On the basis of evidence that envelope cues may be represented in the auditory system by the pattern of phase locking, it was hypothesized that differences between noise samples A and B would be reduced at 6000 Hz.;Four subjects did not show a significant difference between thresholds for the two noise samples at 6-kHz. No subjects showed FNMR at 6-kHz, whereas four subjects showed FNMR at 1-kHz. These results support the interpretation that envelope cues were responsible for FNMR and the threshold differences between the two noise samples in Experiment 1.
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
