Schematic spectra of stimuli for shaped, non-shaped, and -discrim stimulus types. Reference (unshifted) stimuli are shown in the bottom row, and shifted stimuli are shown in the top row. The was and was equal to 11.
Air conduction thresholds for the test ears of the hearing-impaired subjects. The age of each subject is also shown.
Excitation patterns ( Moore et al., 1997 ) for shaped stimuli with and , 11, and 18, presented in pink noise with a spectrum level of at . This noise was designed to give roughly the same excitation pattern as the TEN noise used in the experiment (see the text). The patterns are plotted only over the frequency range where the shaped stimuli produced excitation comparable to or above that produced by the noise. Patterns for harmonic and frequency-shifted stimuli are plotted as solid and dotted lines, respectively. The frequency shift was (the maximum shift).
Mean values for normal-hearing subjects for discrimination of harmonic and frequency-shifted complex tones, plotted as a function of . Each panel shows results for one . Open and filled circles show values for shaped and non-shaped stimuli, respectively. Error bars show standard deviation of the mean. The square symbol indicates that the value is not significantly different from zero. This point is plotted at zero.
As Fig. 4 , but showing individual values for the hearing-impaired subjects. The unfilled and filled square symbols indicate values that are not significantly different from zero for shaped and non-shaped stimuli, respectively. These points are plotted at zero.
values for discrimination by normal-hearing and hearing-impaired subjects. Individual data for three hearing-impaired subjects are plotted (HI 1, HI 2, and HI 5, filled symbols), and average data for four normal-hearing subjects are shown in each frame for comparison (open symbols). Error bars indicate standard deviation of the mean.
The filled symbols show values for discrimination by normal-hearing subjects of harmonic and frequency-shifted tones with components added with random starting phases (shaped-RP stimuli). The open symbols show values for discrimination of shaped stimuli by normal-hearing subjects, for comparison (data from the main experiment). The square symbols indicate values that are not significantly different from zero. These points are plotted at zero.
Largest differences in excitation level between harmonic and frequency-shifted shaped tones using the maximum frequency shift of . Excitation patterns were calculated using a model for normal hearing ( Moore et al., 1997 ).
Conditions for which the nonadaptive procedure (na) was used for the normal-hearing subjects. Stimulus types shaped (SH) and non-shaped (N-SH) are shown only, as all subjects completed the adaptive procedure for all conditions in the -discrim task.
As Table II , but for the hearing-impaired subjects. Dashes indicate conditions that were not tested.
Summary of results for normal-hearing subjects for conditions where all subjects completed the adaptive procedure. Geometric means of the thresholds measured using the adaptive procedure are shown in Hertz. The values that would have been measured for discrimination of tones with a frequency shift of 0.5 Hz were estimated by extrapolation from these thresholds (see Sec. III E for details).
Summary of the ANOVA outcomes for hearing-impaired subjects tested with shaped and non-shaped stimuli.
Summary of conditions for which performance was not significantly better than chance, as indicated by x. Only data for shaped (SH) and non-shaped (N-SH) stimuli are shown. A dash indicates that a condition was not tested for that subject.
Article metrics loading...
Full text loading...