Short bursts of computer-generated Gaussian noise were rated by observers for the presence or absence of a 500-Hz signal tone burst. A multiple regression analysis found for each observer the linear combination of the energies in narrow bands around the tone frequency that best predicts his total ratings. The estimates of the regression coefficients provide graphs of the “frequency responses” of the observers. Most of the reliable variance in the total ratings was accounted for by the regression analysis in terms of energy in narrow bands. Differences among observers are explained in terms of differential weighting by observers of features labeled “tone presence,” “pitch,” and “loudness.”
It is difficult to explain the extraordinary capacity of the ear in discriminating pitch without assuming the existence of laterally inhibitory neural nets fed by cochlear potentials. On such an assumption, we have sought to show edge effects in direct auditory masking analogous to Mach bands found in vision. Noise bands having theoretically infinite attenuation rates outside the passband were generated by computer from 56 sinusoids spaced randomly by frequency. Monaural pure-tone-masked audiograms were obtained for each of four subjects for each of two such noise bands (480–580 Hz and 960–1160 Hz) at sensation levels of 20, 30, 40, 50, and 60 dB. Edge effects, as measured by contours of threshold shift, were revealed for all subjects, at every loudness, and were similar for both noise bands. Sharpening was greater at the low frequency end of a band and grew nonlinearly with loudness, as did upward spread of masking. Sharpening was greater for an interrupted than for a continuous masked tone and may be maximal for relatively low interruption rates. The data are consistent with the existence of laterally inhibiting neural nets and are discussed in relation to their neural basis and to other phenomena such as the pitch of noise bands and the regions of increased sensitivity often found in abrupt hearing loss.
Earlier attempts to fit a bandpass-filter energy-detector model to signal-detection data indicate that the energy passed by narrow filters best predicted the observers' responses on signal-plus-noise trials, but that wide filters best predicted the performance on noise trials [A. Ahumada, Detection of Tones Masked by Noise, doctoral dissertation; Psychology, UCLA (1967).] We have replicated this finding by presenting computer-generated tones and noises in random sequences to eliminate the possibility that the earlier results were an artifact of the particular stimulus sequence used. A version of the Bell Telephone Laboratory's Music IV Program was used to generate the stimuli: noise bursts synthesized from random amplitude sine and cosine waves. If the stimulus was to contain a signal, a constant was added to the amplitude of the sine member of the 500-Hz component of the noise. The observer made rating responses that were correlated with the energy in rectangular frequency bands of varying width centered at 500 Hz.
A new model and digital simulation program for basilar membrane motion is presented with several improvements over previous models. The model attempts to account for both earlier data (e.g., those of von Békésy) and more recent data [e.g., those of Rhode, J. Acoust. Soc. Amer. (to be published)] relative to basilar membrane dynamics. The model consists of a mixed linear and nonlinear part and represents the basilar membrane as a lumped parameter system. The nonlinear scaling factor and time delays were omitted from Flanagan's original model. Travelling-wave envelopes or instantaneous positions may be displayed for any arbitrary signals input to the membrane. The new model also has sharper tuning for single membrane sections and an 8-φ phase lag range. Details of the digital computer programs will be given and a range of applications of the model suggested. [Work supported by the United States Public Health Service, under Grant MH-7809.]
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.